The Administrative Law of Nations: A New Perspective on Sosa, the Alien Tort Statute, and Customary International Law

by: Josh Goodman

Download PDF of Article

I. Introduction

Through no fault of its own, the Supreme Court recently missed a chance to clarify important questions about the role of international law in U.S. courts. In American Isuzu Motors v. Ntsebeza,_^[1] an appeal of a split decision from the Second Circuit,_^[2] the Court could have ended the confusion surrounding the scope of “aiding and abetting” liability under the Alien Tort Statute (“ATS”), a law that allows aliens to sue in U.S. federal courts for violations of international law._^[3] The Supreme Court’s ruling in the case could have also provided general guidance on how courts should apply customary international law in the United States. Unfortunately, the Court lacked the quorum necessary to hear the case, leaving the issue unresolved._^[4]

This question is particularly significant because modern customary international law now touches many areas of policy that were once the sole domain of domestic politics._^[5] As we enter a period in which international law plays an increasing role in our domestic legal system, we must ensure that we apply international law in the United States in a way that respects the federal constitutional structure. Incorporation of modern customary international law into the American legal system – through vehicles like the Alien Tort Statute – should occur in a way that meaningfully reflects the functional design of the U.S. Constitution.

This brief article suggests that because customary international law formation has come to resemble, in certain respects, global administrative rulemaking, the application of customary international law in the U.S. raises “non-delegation” concerns about Congress delegating its lawmaking authority away to international bodies. Administrative law doctrine, therefore, may offer a helpful way of thinking about this challenge. 

First, this article surveys the recent debate surrounding aiding and abetting liability under the Alien Tort Statute, with a focus on the 2008 case the Supreme Court was unable to hear, American Isuzu Motors v. Ntsebeza (the “Apartheid Case”)._^[6] This article then examines how the customary international law claims raised in the case reflect a modern theory of customary international law formation that bears certain similarities to domestic administrative lawmaking. Finally, it argues that constitutional concerns about separation of powers suggest that clear legislative authorization should be required before courts can apply a particular area of international law domestically – and that, absent such authorization, liability should not be extended by an aiding and abetting theory.

II. The Alien Tort Statute, Sosa, and the Apartheid Case

The Alien Tort Statute was originally enacted as part of the first Judiciary Act in 1789. It provides federal courts with the power to hear suits brought by an “alien for a tort only, committed in violation of the law of nations.”_^[7] Despite its antique origins, the ATS was rarely invoked until a 1980 case in which the family of a Paraguayan torture victim successfully relied on it to sue the torturer in New York for violations of an international law prohibition against torture._^[8] Since then, the ATS has generated numerous lawsuits seeking damages for alleged violations of international human rights norms. While these cases initially targeted the actual alleged perpetrators of the offenses, a wave of more recent cases has invoked theories of indirect liability in order to hold large corporations responsible for “aiding and abetting” human rights abuses._^[9]

The Supreme Court weighed in on the ATS in the 2004 case of Sosa v. Alvarez-Machain._^[10] The Sosa court observed that there are “limited enclaves in which federal courts may derive some substantive law in a common law way”_^[11] and held that Congress, by enacting the ATS, created one such enclave for claims based on a “narrow set of common law actions derived from the law of nations”—i.e., customary international law._^[12] The ruling left the precise scope of the “narrow set” of actions in which courts may recognize claims derived from international law ambiguous. According to the Court, only three relevant types of law of nations violations existed when the ATS was enacted in the 18^th Century—offenses against ambassadors, violations of safe passage, and piracy. The Court held, however, that modern federal courts may recognize additional violations under modern customary international law, provided that these 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 . . . .”_^[13] Ultimately, the Court rejected the claim in Sosa, holding that the international law norm invoked by the petitioner (a prohibition against arbitrary arrest) was insufficiently well-defined in customary international law to create a cause of action under the ATS.

The Sosa ruling left it unclear as to whether a corporation could be sued under the ATS for aiding and abetting a customary international law violation. Does Sosa mean that there needs to be an international law norm against “aiding and abetting” that is “defined with a specificity comparable to the features of the 18th-century paradigms,” or is it enough for a plaintiff to show that there is a sufficiently well-defined international law norm against whatever conduct has allegedly been aided and abetted (say, torture)? And if the latter is the case, what aiding and abetting law should be applied? International law? Federal law?  

The Apartheid Case addresses potential answers to these questions. In the case, plaintiffs from South Africa sued various major corporations, including Ford Motor Company, Daimler-Chrysler, General Motors, IBM, Coca-Cola, Nestle, Shell Oil, and others, on the theory that they aided and abetted the racist and repressive policies of apartheid by doing business with the former South African regime: “South African police shot demonstrators ‘from cars driven by Daimler-Benz engines,’ . . . the regime tracked the whereabouts of African individuals on IBM computers, . . . [and] the military kept its machines in working order with oil supplied by Shell.”_^[14] The U.S. government petitioned the Supreme Court to dismiss the suits._^[15] When the Second Circuit heard the case, the U.S. government argued in an amicus brief that there could be no liability for aiding and abetting under the ATS unless Congress created it._.^[16] Citing Central Bank of Denver v. First Interstate Bank,_^[17] a securities law case, the government claimed that “whether or not to permit a civil aiding and abetting claim is a legislative choice.”_^[18]

Although the district court accepted the government’s argument and dismissed the case,_^[19] the Second Circuit rejected the argument on appeal. The Second Circuit distinguished Central Bank because it involved a court creating a novel aiding and abetting claim where the legislature had already enacted a detailed statutory scheme. The Alien Tort Statute, on the other hand, is brief and seemingly open-ended, and had already been interpreted in Sosa as giving courts some common law power to recognize new causes of action.

Instead of relying on Central Bank, each judge on the Second Circuit panel took a distinct approach to deciding the Apartheid Case. Judge Katzmann asked whether aiding and abetting international law violations is itself a violation of the law of nations that is sufficiently well-defined to pass the Sosa test._^[21] After analyzing diverse international law sources—from the Nuremberg tribunals to the International Criminal Tribunal for the former Yugoslavia—he concluded that it was, and ruled that the apartheid suits could go forward. Judge Korman, in dissent, also looked to international law, asking whether there was a well-defined international legal norm for holding a corporation vicariously liable for violating international human rights laws._^[22] He concluded that there was not, and he therefore ruled that the suits should be dismissed. Finally, Judge Hall opined that “[l]acking the benefit of clear guidance . . . a federal court should resort to its traditional source, the federal common law” _^[23] to derive the appropriate standard of liability._^[24] Judge Hall then identified a standard for aiding and abetting liability under federal common law, and held that the allegations of the plaintiffs, if proven true, could meet the standard. Therefore, he ruled that the suits could go forward.

The divergent approaches of the three judges on the Second Circuit panel suggests that a strictly doctrinal and judicial approach to resolving the question of corporate aiding and abetting liability under the ATS is unlikely to bear fruit. A better approach would be to step back and consider the role and the scope of the ATS in light of the Sosa ruling, the broader framework of modern customary international law, and its interplay with U.S. federal law and the U.S. Constitution. Such an approach suggests that further legislative authorization should be required for corporate aiding and abetting claims to proceed under the ATS.

III. Modern Customary International Law and U.S. Courts

Traditionally, U.S. case law limiting the application of international law has often focused on preventing courts from interfering with foreign affairs. For example, in the leading case of Banco Nacional de Cuba v. Sabbatino,_^[25] the Supreme Courtrefused to rule that Cuba had violated customary international law by expropriating American property. The Sabbatino ruling relied on the Act of State doctrine, which holds that courts will not judge the domestic policies of another country. The Court observed that this limitation on courts getting involved in foreign affairs questions had “constitutional underpinnings” in the separation of powers because the Constitution makes the conduct of foreign affairs the prerogative of other branches._^[26] The Sabbatino court suggested that courts should apply (or decline to apply) customary international law pragmatically in order to preserve the constitutional separation of powers on foreign affairs questions and to maintain appropriate respect for the domestic policies of other sovereign countries._^[27] Increasingly, though, modern customary international law is being invoked in cases that do not seriously implicate the government’s foreign affairs powers or the sovereignty of foreign states. For example, in the 1999 case of Domingues v. Nevada,_^[28] the petitioner argued that his death sentence for a double murder in Nevada was barred by a customary international law prohibition against the juvenile death penalty._^[29] The Supreme Court denied certiorari in Domingues, but later ruled that the juvenile death penalty was unconstitutional in Roper v. Simmons, the opinion noted for its controversial citation of international and foreign law to inform its Eighth Amendment holding on the meaning of “cruel and unusual punishment.”_^[30]

This trend is driven by the novel nature of modern international law—particularly customary international law—as compared to international law as it existed prior to the Second World War._^[31] Traditional customary international law governed relationships between states, but the modern version claims to regulate private actors as well._^[32] The content of customary international law has also shifted to embrace many areas that were once the exclusive domain of domestic regulation; this is particularly apparent in the realm of human rights,_^[33] which, defined broadly, can touch many traditionally domestic political questions.

The speed with and manner in which modern customary international law norms are supposedly produced is unprecedented. “Custom” is defined as a “common tradition or usage so long established that it has the force or validity of law;”_^[34] customary international law is defined as the practice of states adhered to out of a sense of legal obligation (opinio juris). This meaning of customary international law is evident, for example, in the famous case of The Paquete Habana,_^[35] which surveys 600 years of legal history and state practice in order to support the rule of customary international law it invokes to exempt a fishing vessel from capture as a prize of war. In contrast, Professor Ralph Steinhardt illuminates the novel, modern meaning of “customary international law” when he writes: “The process of customary law formation has been accelerated for example by the concentration of state practice in multilateral inter governmental organizations and the proliferation of non governmental organizations [sic].”_^[36] Thus, modern customary international law, in the view of some scholars, emerges rapidly from various international institutions and sources, some of which are even private organizations (although that view is hardly mainstream). This understanding is quite different from the original meaning. If one accepts this view, something like an international administrative state—composed of international courts and tribunals, regional and international human rights bodies, NGOs and UN bodies—is now responsible for the production of modern customary international law norms.

The claims in the Apartheid Case exemplify many of these aspects of modern customary international law. The defendants stand accused of aiding and abetting various international human rights violations—some well-established and some more controversial—including genocide, torture, forced labor, sexual assault, unlawful detention, extrajudicial killings, war crimes, and racial discrimination. The suits attempt to place international legal liability on private actors (corporations), rather than states. The ultimate ambition of these suits is regulatory in nature, in that the suits attempt to use the legal system to create monetary disincentives for certain types of corporate (and, indirectly, state) behavior—in effect creating a transnational method of corporate regulation by using established domestic legal mechanisms to enforce norms of international law directly on private actors. Like modern domestic tort law, the goal is to deter harm-causing behavior and transfer wealth to those who have suffered harms from those who bear responsibility for causing them.

When it comes to sources of law, the Apartheid Case plaintiffs point in many directions to evidence the customary international law norms allegedly violated by the defendants. They invoke the jurisprudence of the ad hoc International Criminal Tribunals for Rwanda and the former Yugoslavia, widely non-ratified treaties (e.g., the “Apartheid Convention”), and non-binding U.N. General Assembly resolutions._^[37] Formally, these sources do not have legal effect in the United States: the ad hoc tribunals have limited criminal mandates; treaties do not bind states that have not ratified them; and General Assembly resolutions are not legally binding at all. The idea that these sources may nonetheless shape customary international law is a modern development that brings them in the back door of U.S. domestic law._^[38]

The application of customary international law will increasingly impact matters of U.S. domestic policy. Much of the academic debate about the treatment of this law has dealt with whether, and to what extent, customary international law is federal common law._^[39] This article suggests a method of analyzing questions of international law rooted in the Sabbatino approach, with its functional emphasis on upholding the constitutional separation of powers, but which extends that approach beyond foreign affairs to uphold the separation of powers underlying general domestic legislation as well.

Internationally created norms, which are judicially incorporated into federal law to regulate the behavior of domestic parties and preempt domestic policies, appear to undermine principles of democratic governance and infringe on the constitutional mechanisms for federal lawmaking via bicameralism and presentment._^[41] According to the Constitution, the legislative power is vested in Congress._^[42] Therefore, the U.S. should cautiously incorporate the new customary international law into its legal system in a manner that functionally preserves the separation of powers in the Constitution.  

Domestic administrative rulemaking provides a useful point of comparison. Regulation by domestic administrative agencies has been an increasingly important source of U.S. domestic law since the New Deal—so much so that scholars have argued that a sort of unwritten constitutional amendment has occurred._^[43] The challenges of governance in a modern, industrial state like the post-New Deal United States necessitated this type of administrative agency rulemaking. Similarly, the challenges of governance in a world of rapidly increasing global interaction, migration, and economic integration may point to a need for an increased reliance on mechanisms of global governance and coordination, including guarantees of rights that transcend state borders. The United States should meet this global governance challenge in a way that functionally preserves its constitutional structure and the principles of democratic government that underlie it.

Even though regulation by the modern administrative state was not contemplated by the Framers of the U.S. Constitution,_^[44]the administrative state can still be considered consistent with the U.S. constitutional system of government on a functional level._^[45] It can be considered consistent with the constitutional structure of federalism because agency rules that displace state law regulation theoretically derive from authority granted in a congressional statute._^[46] Similarly, it is consistent with a functional understanding of separation of powers because Congress and the President must pass agency organic statutes by bicameralism and presentment. Further, administrative agencies are subject to pressure and oversight by Congress and the Executive through appointment and removal powers, supervisory powers, executive orders, congressional notification and funding powers, congressional hearings, and other formal and informal avenues of political branch control, in addition to judicial review. On the other hand, the Supreme Court has declared that novel lawmaking structures that appear to remove inter-branch constraints and empower a single branch to act alone—like the legislative veto_^[47] or the line-item veto_^[48]—are unconstitutional.

Although federal courts applying international law are never technically “making” that law the way agencies make rules—international processes are doing that—the courts are incorporating it and defining it within the U.S. system. Courts, therefore, have a high level of control over its application and interpretation, as the opposite conclusions reached by Second Circuit judges in the Apartheid Case illustrate. Thus, federal courts applying customary international law, like domestic administrative agencies, should be subject to functional constraints that reflect the separation of powers inherent in federal lawmaking. Where customary international law has the potential to alter the domestic legal rights and obligations of private parties or to displace democratically adopted policies, Congress should not be allowed to write the courts a blank check to import international law; there should be a requirement of relatively specific statutory authorization to enable federal courts to engage in federal common lawmaking based on customary international law. This requirement would preserve the integrity of the lawmaking functions of the political branches.

In order for this requirement to be robust, something like an “intelligible principle” for limiting the judicial lawmaking to a particular subject-matter enclave should be required in the statutory delegation. Under the “non-delegation doctrine,” the Supreme Court has held that a Congressional grant of power to a government agency must contain an “intelligible principle” to define the scope and limits of the agency’s regulatory authority._^[49]This doctrine has been enforced rather loosely; an administrative law delegation is typically upheld unless it fails “to articulate any policy or standard that would serve to confine the discretion of the authorities to whom Congress had delegated power.”_^[50] However, the “intelligible principle” analogue here should be applied more rigorously because courts are more insulated from the pressure and control of the political branches than administrative agencies. Courts will be policing themselves in this context, and the processes that generate customary international law may not be subject to the same accountability and due process standards as domestic agencies._^[51] Therefore, an “intelligible principle” in this context would require Congress to identify particular subject-matter areas of international law in which courts can exercise common law powers._^[52]

The “intelligible principle” approach provides a useful perspective for evaluating the Sosa opinion. In Sosa, the Court found statutory authorization for the incorporation of customary international law into federal common law in the ATS—a good first step. Unfortunately, there was no “intelligible principle” in the ATS beyond a vague reference to the “law of nations.” This left the Court to create the applicable “intelligible principle” on its own. In the administrative law case of Whitman v. American Trucking, the Supreme Court ruled that an agency receiving rulemaking authority from Congress should not also be empowered to determine the intelligible principle that limits the scope of that authority._^[53] Applying that logic to courts themselves, it would have been preferable if the Sosa opinion had directed Congress to identify an intelligible principle of subject-matter limitation for suits brought under the law of nations. Presumably, Congress would have no trouble doing so. For example, in 1992 Congress passed the Torture Victim Protection Act, which essentially codified international law norms against torture and extrajudicial killing into federal law. Instead, Congress could have amended the ATS to identify available causes of action—like torture, extrajudicial killing, and other serious offenses. Requiring the legislative branch to make that choice is more consistent with the constitutional separation of powers than having courts try to decide which international law norms are actionable on the basis of the vague, court-created Sosa test. That conclusion also suggests that Courts should defer to Congress all the more so before endorsing an ATS aiding and abetting theory, which will extend the scope of such ill-defined international legal liability to a wide range of private actors.

V. Conclusion

Ultimately, if globalization and the expansion of international law continue on their current track, political realities will likely lead to the introduction of new international law doctrines into U.S. law one way or another. This outcome will not only result from international and domestic political pressures, but also from micropolitical pressures, such as judges’ personal knowledge of international legal trends and concern about their reputations. It is therefore essential to bring international law into the domestic legal system through relatively transparent mechanisms that reflect the checks and balances of traditional federal lawmaking.

Absent further instruction from Congress, the Supreme Court should hold not to extend liability to corporations when the issue of aiding and abetting liability under the ATS finally reaches the Court again. This conclusion does not mean that lawsuits against corporations for complicity in human rights violations abroad should, in principle, be eliminated. On the contrary, within appropriate limits, these types of suits can serve important purposes. They promote the protection of human rights, benefit the U.S. government and its standing in the world, and, indeed, can benefit the U.S. business community. Even so, in order to preserve the functional integrity of U.S. lawmaking in an age of rapidly expanding international law claims, the Court should require at least minimally clear instructions from Congress before applying modern customary international law norms against private parties in the United States.