Melbourne Journal of International Law
[Since the late 1990s, plaintiffs have increasingly used the Alien Tort Claims Act to attempt to hold multinational corporations to account for their alleged misconduct outside the United States. In Sosa v Alvarez-Machain (‘Sosa’), decided in 2004, the US Supreme Court ruled on the proper scope of the Act in a decision many thought would clarify the viability of such litigation. The Court, however, issued a decision characterised by ambiguity. This commentary examines and critiques the federal court jurisprudence on corporate accountability under the Act that has developed subsequent to Sosa, and reflects on the impact that the currently uncertain state of the law will have on multinational corporate conduct abroad.]
In 2004, in Sosa v Alvarez-Machain the United States Supreme Court ruled for the first time on the proper reach of the Alien Tort Claims Act. Although the decision dealt with a state-sponsored kidnapping, many thought it would answer the question of the Act’s applicability to corporations alleged to have committed human rights abuses abroad. From the mid-1990s onward, the Act had increasingly been used by plaintiffs as a tool to hold multinational corporations to account, albeit with little success.
While the human rights community supported the use of the Act, those in the business community were stridently opposed to ATCA litigation. Two commentators dubbed the Act an ‘awakening monster’. The Supreme Court’s decision was therefore eagerly awaited. Yet the decision was marked by ambiguity, which has been reflected in the divergent reactions that followed it. Some saw the decision as a victory for US-based corporations; others were not convinced, reading the decision as favourable to potential plaintiffs. The purpose of this commentary is to examine the Supreme Court’s decision and to assess the impact it will have, and has already had, on the effectiveness of the ATCA as a tool to redress or prevent human rights violations perpetrated by multinational corporations.
To assess the impact of the Supreme Court’s decision in Sosa, in Part II I sketch the development of ATCA jurisprudence leading up to Sosa, particularly as it relates to corporations, and note three disputed issues: whether corporations can be held civilly liable under international law; the true meaning of the ‘law of nations’ as that phrase is used in the ATCA; and whether aiders and abettors to human rights violations can be held liable under the Act. In Part III, I examine the Sosa decision and assess the extent to which it addresses these issues and raises additional complications. Part IV examines and critiques post-Sosa case law on these issues, and I conclude by reflecting on the decision’s potential impact on corporate conduct abroad.
The ATCA, enacted in 1789, provides that ‘[t]he district courts shall have original jurisdiction of any civil action by an alien for a tort only, committed in violation of the law of nations or a treaty of the United States’. Three cases illuminate how such a simple statute has given rise to potential corporate accountability.
The first is Filartiga v Pena-Irala, in which the family of a Paraguayan man who had been tortured and killed in Paraguay brought an ATCA suit against the alleged torturer, a Paraguayan police official who was then living in New York. Although the District Court dismissed the suit, holding that a state’s treatment of its citizens did not violate the law of nations, the Second Circuit Court of Appeal reversed this decision, holding that ‘an act of torture committed by a state official against one held in detention violates established norms of the international law of human rights, and hence the law of nations’. Importantly, the Court interpreted the ‘law of nations’ as meaning customary international law, an interpretation the US Government as amicus curiae supported and noted that courts must ‘interpret international law not as it was in 1789, but as it has evolved and exists among the nations of the world today’ Further, in finding that international law was part of federal common la and that the ATCA ‘open[ed] the federal courts for the adjudication of the rights already recognized by international law’ the Second Circuit Court of Appeal implied that the Act created a private cause of action.
Filartiga was important insofar as it resurrected an all but forgotten statute but limited to its facts, its ratio was narrow: it found that a state actor could breach the international norm prohibiting torture. The question of whether a non-state actor, such as an individual or a corporation, could be held responsible for violating international law was not addressed. The next important case, therefore, was Kadic v Karadžić, which squarely tackled this question.
The Court in Kadic considered whether victims of genocide, war crimes, torture and summary executions committed by Bosnian-Serb military forces, or their representatives, could bring an ATCA suit against the commander of these forces, Radovan Karadžić. Karadžić had been President of Srpska at the time, a self-declared but non-recognised republic within Bosnia-Herzegovina. The District Court had dismissed the suit for want of jurisdiction, holding that ‘acts committed by non-state actors do not violate the law of nations’, but the Second Circuit Court reversed this decision. Finding support in the US Government’s Statement of Interest, the Court held that ‘certain forms of conduct violate the law of nations whether undertaken by those acting under the auspices of a state or only as private individuals’. In particular, genocide and war crimes were among the offences for which the Court found private actors could be held responsible. Thus, the Second Circuit Court, albeit obliquely, raised the possibility of corporate accountability by signalling its intent to allow certain ATCA claims made by those whose human rights had been violated by private actors and, in the wake of Kadic, lawsuits against corporations burgeoned.
One of the first significant cases against a US-based corporation was brought in 1996. In Doe I v Unocal Corp, Burmese villagers sued, among others, the US-based company Unocal for its joint involvement with the Burmese Government in an oil pipeline project in Burma, which had allegedly resulted in human rights violations committed by the Burmese military, including forced labour, torture and crimes against humanity. Unocal had been warned before committing to the project that the Burmese Government and military were notorious human rights abusers and, in 1995, after the project had commenced, a consultant to Unocal concluded that:
egregious human rights violations have occurred, and are occurring now, in southern Burma. The most common are forced relocation without compensation of families from land near/along the pipeline route; forced labour to work on infrastructure projects supporting the pipeline … and imprisonment and/or execution by the army of those opposing such actions. Unocal, by seeming to have accepted [the Burmese Government’s] version of events, appears at best naïve and at worst a willing partner in the situation.
Although the District Court awarded summary judgment for Unocal, a three-judge panel of the Ninth Circuit allowed the case to proceed. Assuming the facts as pleaded were true, the Court found that Unocal could be liable for aiding and abetting the Burmese military in its human rights violations. On the standard for third party liability, however, the Court was split. Two judges adopted the standards of aiding and abetting liability from the jurisprudence of the ad hoc international criminal tribunals, namely ‘knowing practical assistance, encouragement, or moral support which has a substantial effect on the perpetration of the crime’. Judge Reinhardt, however, would have preferred to apply federal tort law principles, such as agency, joint venture or reckless disregard. Soon after this decision, the Ninth Circuit agreed to rehear the case en banc. A rehearing never took place, however, as subsequent to the Supreme Court’s decision in Sosa, the parties settled out of court.
Unocal is important for holding that corporate liability is possible under the ATCA when corporations are complicit with states in human rights violations. This is significant given that cases alleging corporate liability without some kind of state involvement are uncommon. But because the parties settled, the appropriate standard for determining third party liability was never conclusively determined.
Prior to the Supreme Court’s decision in Sosa, then, it appeared that corporations could potentially be held liable under the ATCA even if such suits had yet to be successfully concluded. This assumption, however, was not completely free from challenge; some commentators continued to argue that only states and individuals, not corporations, could be held civilly liable under international law. This was one issue that many were hoping the Supreme Court would clarify in its decision, but two other complications with important ramifications for corporations also cried out for clarification.
The first was that suits against corporations were concerned not only with alleged violations of clear norms of international law, such as the prohibitions on forced labour and torture, but also with unusual torts, such as environmental pollution and ‘cultural genocide’. Although courts were able to dismiss relatively easily the more unusual claims, the wide range of allegations being brought meant that the correct standard to be applied in determining which violations of international law were actionable under the Act became an increasingly important issue. Although the Court in Filartiga advocated a customary international law standard, other courts adopted a higher threshold. In Kadic and Unocal, for example, the courts favoured international norms that were ‘universally recognized’, or ‘specific, universal, and obligatory’. Yet, as Paust and others have argued, customary law need not have a truly ‘universal’ consensus, instead, ‘normative content need only be based in generally shared patterns of legal expectation or acceptance’.
Finally, while the Ninth Circuit in Unocal — albeit confusingly — tried to distinguish its ‘specific, universal, and obligatory’ test from the test for jus cogens, some courts had gone so far as to require breaches of peremptory norms before allowing ATCA actions to proceed. The meaning of ‘law of nations’ for ATCA purposes was therefore an issue that the Supreme Court was expected to address.
The second complication concerned indirect corporate complicity in human rights violations and the standard for third party liability. As noted above, the Ninth Circuit lost an opportunity to determine this issue, as the parties settled before the scheduled rehearing. While the three-judge panel determined that corporations could be held liable under the ATCA as aiders and abettors, there was no consensus on the proper standards to apply. This was the third issue it was thought the Supreme Court would clarify.
Far from offering clarification, the Supreme Court in Sosa issued a decision marked by ambiguity. The relevant facts of the case were as follows: the Mexican plaintiff, Dr Alvarez-Machain, had been abducted in Mexico, held overnight in a motel, and brought to the US in 1990 by a group of Mexicans (including Sosa) who had been hired for this purpose by the US Drug Enforcement Administration (‘DEA’). The DEA suspected Alvarez-Machain of helping to torture and kill a DEA agent in 1985, but had been unable to secure his extradition from Mexico. Once in the US, Alvarez-Machain was charged and tried, and then acquitted of any wrongdoing. On his return to Mexico, Alvarez-Machain sued Sosa under the ATCA for, inter alia, an alleged arbitrary arrest and detention contrary to international law.
Unlike in many previous cases, where it had supported actions brought under the ATCA, here the US Government filed a brief in support of Sosa in which it argued that the ATCA was strictly jurisdictional and that without further statutory authorisation, no cause of action could be brought for a violation of international law. Whilst this type of argument was not novel, no federal court had ever endorsed it. Justice Souter, writing for the majority, began his judgment by rejecting it. While accepting that the Act was purely jurisdictional, Justice Souter found that it could not have been ‘stillborn’ given that when the ATCA was enacted, ‘torts in violation of the law of nations would have been recognized within the common law’. The majority therefore concluded that in 1789, Congress would have intended the ATCA to cover the three offences known to violate the law of nations at the time: violations of safe conduct, infringements of the rights of ambassadors, and piracy. Importantly, the majority recognised that no development in the law since 1789 had ‘precluded federal courts from recognizing a claim under the law of nations as an element of common law’. The majority, although counselling caution, left open the possibility that claims today based on violations of modern customary international law norms could be brought: ‘the judicial power should be exercised on the understanding that the door is still ajar subject to vigilant doorkeeping, and thus open to a narrow class of international norms today’.
What is meant by ‘narrow class of international norms’? The majority explained itself thus:
federal courts should not recognize private claims under federal common law for violations of any international law norm with less definite content and acceptance among civilized nations than the historical paradigms familiar when [the ATCA] was enacted.
Moreover, the majority warned that a court’s assessment of ‘definite content’ ‘should (and, indeed, inevitably must) involve an element of judgment about the practical consequences of making that cause available to litigants in the federal courts’.
The Court found that Alvarez-Machain’s claim foundered on both these limbs. Not only was the prohibition on arbitrary detention, defined by the Court as requiring more than a ‘relatively brief detention in excess of positive authority’, too broad a rule to fulfil the Court’s specificity requirements, but admitting the claim would have had ‘breathtaking’ implications.
While the Court purported to clarify the standard of actionable international law violations under the ACTA with its historical test, it had less to say about whether corporations could be held liable pursuant to the Act. Despite amicus curiae briefs from the US Government and business groups exhorting the Court to preclude corporations from the Act’s reach, the Court alluded to the possibility of corporate liability in a footnote. Justice Souter stated that in addition to inquiring into the specificity of the relevant international norm:
A related consideration is whether international law extends the scope of liability for a violation of a given norm to the perpetrator being sued, if the defendant is a private actor such as a corporation or individual. Compare [Tel-Oren,  USCADC 53; 726 F 2d 774, 791–795 (DC Cir, 1984)] (Edwards J concurring) (insufficient consensus in 1984 that torture by private actors violates international law), with [Kadic USCA2 16; , 70 F 3d 232, 239–241 (2nd Cir, 1995)] (sufficient consensus in 1995 that genocide by private actors violates international law).
Thus the Court appeared to recognise that some international norms are applicable to private actors, including corporations.
In support of this interpretation, one can point to the Court’s reference to ATCA pending class actions against corporations alleged to have participated in, or aided and abetted, the human rights violations perpetrated by the South African Government under apartheid. As Steinhardt has noted, rather than deciding that these cases were simply beyond the ATCA’s purview on the basis that corporations cannot be held liable under international law, the Court indicated that foreign policy considerations should determine whether or not these claims should proceed.
In this regard, the Court noted that the US Government had previously agreed with its South African counterpart that such cases would interfere with the policy embodied by the South African Truth and Reconciliation Commission, which ‘deliberately avoided a “victors’ justice” approach to the crimes of apartheid and chose instead one based on confession and absolution, informed by the principles of reconciliation, reconstruction, reparation and goodwill’.
As for whether the ATCA extends to third party liability, and if so, whether federal tort law or international law standards should govern, the Court provided no guidance.
Finally, the Court casually raised two further potential limitations on claims under the ATCA. Firstly, the Court stated that ‘in an appropriate case’ there might be a requirement that domestic remedies be exhausted before a claim can be made under the Act. Secondly, the Court noted that in cases such as those involved with the South African apartheid litigation, ‘there is a strong argument that federal courts should give serious weight to the Executive Branch’s view of the case’s impact on foreign policy’.
The Supreme Court’s decision has been criticised for its opacity and lack of guidance. Not only did the Court obliquely refer to the question of corporate liability and remain silent on the issue of third party liability — points which, although not directly in issue in the case, were in need of clarification, and on which it was thought the Court might comment — it set out a historical test to determine which torts constitute actionable violations of international norms for ATCA purposes that was impractical. As one commentator objected, it is
hard to imagine how to compare the degree of specificity of one rule to that of another rule in a completely different substantive area, as it existed over two hundred years ago — especially given the disagreement over the bounds of the eighteenth-century norms in the first place.
Even the federal courts have expressed their displeasure at the Supreme Court’s test. One District Court opined that ‘it would have been unquestionably preferable for the lower federal courts if the Supreme Court had created a bright-line rule’. One can query whether ascertaining new customary norms can ever be easy given the disputes around customary law and the difficulty involved in collating the necessary information, but the essence of the District Court’s complaint is clear. The Supreme Court did litigants and lower courts no favours. The only point most commentators concede is that the Court intended to halt the expansion of ATCA claims.
The Supreme Court stressed the ‘great caution’ and ‘vigilant doorkeeping’ that courts should exercise when determining whether to allow claims based on modern customary norms. But although the Court’s specificity requirements seem to contemplate a subset of customary norms, it would be erroneous to equate these norms only with jus cogens, given that specificity, rather than egregiousness, lies at the heart of the Court’s test.
While the Supreme Court may well have raised the threshold for actionable claims, it has left courts with ‘a number of difficult and unsettled questions in a controversial area of the law’.
How, then, have federal courts interpreted the Supreme Court’s decision in Sosa? This part examines the post-Sosa case law and highlights the three principal points of contention that have particular relevance to corporations.
As a result of the Supreme Court’s failure to address clearly the issue of corporate accountability under the ATCA, soon after Sosa, many corporate defendants tried with renewed vigour to have corporations excluded from the Act’s ambit. The arguments against holding corporations liable are essentially twofold. The first argument, based on policy, is that recognising corporate liability under the Act would have harmful economic consequences. The contours of this argument were set out in an amici curiae brief submitted to the Supreme Court in Sosa by various business groups:
The threat of [ATCA] lawsuits can result in higher insurance costs [for corporations], difficulties accessing capital markets, and negative effects on shareholder confidence and stock prices. The foreseeable result is less investment and trade with developing countries, and less international trade overall.
What is more, the economic downside of [ATCA] litigation uniquely and adversely affects the US economy. … Foreign companies without a US presence … need not fear [ATCA] suits, and their associated costs. This means that US companies (or companies with a US presence) are at a significant competitive disadvantage against their foreign competitors … They either have to absorb these costs, or cede profitable ventures to other nations’ companies.
Although courts have been reluctant to consider explicitly such policy-based arguments, many commentators have argued that the likelihood of damaging economic ramifications as a result of possible ATCA suits is exaggerated.
The exacting standards that courts have demanded of plaintiffs suing under the ATCA mean that corporations have little to fear if they are conducting themselves appropriately.
It is the second argument against corporate liability under the Act, purportedly based on law, that has been used most frequently by corporate defendants post-Sosa and thus examined by courts. Corporations, these defendants have argued, simply cannot violate norms of international law, and therefore cannot be held accountable under the ATCA. This argument is grounded on a traditional understanding of international law, namely that it deals primarily with nation states and, in certain cases, individuals, but does not impose duties on other actors such as corporations.
Although there is much debate in the literature over the extent of legal personality corporations possess in international law, federal courts since Sosa have tended to reject this legal argument without serious question. As one judge stated, the ‘argument that corporate liability under international law is not supported by sufficient evidence and is not sufficiently accepted in international law to support an [ATCA] claim is misguided’. This view amongst the judiciary reflects a growing body of scholarship — driven by the increasingly evident power of multinational corporations and the vulnerability of weak states, as well as the inadequacies of international law as traditionally conceived — that has documented how international law has held and continues to hold corporations directly liable under international law.
Moreover, it is a view that has been justified logically: ‘[l]imiting civil liability to individuals while exonerating the corporation directing the individual’s action through its complex operations and changing personnel makes little sense in today’s world’.
Only one judge in the case law subsequent to Sosa has been sympathetic to the idea that corporations might be outside the sphere of international law. In Khulumani, a case concerning corporate conduct in South Africa during apartheid, Judge Korman, dissenting, opined that at the time of the alleged violations, international law did not recognise corporate responsibility for breaches of norms proscribing crimes against humanity. Moreover, Judge Korman was reluctant to acknowledge that custom has moved on since the fall of the apartheid regime, conceding only that there has been ‘some movement’ toward the recognition of corporate liability. This view was not shared by the majority. Indeed, neither the corporate defendants nor the US Government in Khulumani objected in principle to the liability of corporations under the ATCA, suggesting that the argument against corporate liability is fast losing any traction it might once have had.
To date, no courts have engaged with the Supreme Court’s ‘transhistorical test’ to determine which norms fall within the ATCA’s ambit. This abstention likely reflects the impracticality of the Supreme Court’s test, but the consequence is that courts’ decisions appear to be based on instinct and impression, and can foster uncertainty.
At the outset it is important to recognise that there are two ways in which indefiniteness can play a role when assessing ATCA claims. First, a ‘norm’ that has allegedly been violated may be indefinite insofar as it lacks recognition among states as an international norm. Federal courts have always been wary of allowing claims based on such ‘norms’. Thus, courts have rejected claims based on alleged violations of the right to free speech, conversion, fraud, sexual violence and intra-national environmental damage.
The second way in which indefiniteness may arise is that, although the norm in question may be sufficiently recognised as an international norm, its precise content may be unclear. An example is the prohibition on torture: the norm itself is not in doubt, but conduct giving rise to its breach may be. This problem best describes what occurred in Sosa. The question for the Court was not so much whether a norm relating to the prohibition on arbitrary arrest and detention existed, but more whether the actual detention of Alvarez-Machain was sufficient to fall within the scope of the norm. It is in respect of this second category of norms — those that are widely recognised but whose exact content is unclear — that problems arise and confusion can reign. In particular, there is a worrying trend towards dismissing claims based on these norms on the ground that their content is inadequately defined, and that they are therefore too indefinite to support a claim under the ATCA, irrespective of the conduct in question. Two recent decisions evince this trend.
The first is Vietnam Association for Victims of Agent Orange v Dow Chemical Co, a case concerning claims arising from the use of toxic herbicides by the US during the Vietnam War. ‘Agent Orange’ supplied by the defendants was used to defoliate jungle with the aim of exposing enemy forces. The plaintiffs in the Second Circuit argued that the corporate defendants had aided and abetted violations of customary norms prohibiting the use of ‘poisoned weapons’ and the infliction of unnecessary suffering. In respect of the ‘poisoned weapons’ claim, the Court held that inasmuch as the toxic herbicide was intended for the destruction of foliage only and not calculated to harm human populations, its use did not breach the international norm invoked.
The Court’s finding in respect of the infliction of unnecessary suffering claim, however, was more troubling. The plaintiffs had based their claim on art 23(e) of the 1907 Hague Regulations, prohibiting the use of ‘arms, projectiles, or material calculated to cause unnecessary suffering’; art 6 of the Charter of the International Military Tribunal (‘Nuremberg Charter’), proscribing ‘wanton destruction of cities, towns or villages, or devastation not justified by military necessity’; and art 147 of the Geneva Convention (IV) relative to the Protection of Civilian Persons in Time of War, defining ‘grave breaches’ as ‘wilfully causing great suffering or serious injury to body or health’ as well as ‘extensive destruction … of property, not justified by military necessity’. The Court rejected this claim on two separate bases, the first of which is problematic. The Court stated:
These norms are all simply too indefinite to satisfy Sosa’s specificity requirement. As [one] expert opined, ‘norms that depend on modifiers such as “disproportionate” or “unnecessary” … invite a case-by-case balancing of competing interests … [and] black letter rules become vague and easily manipulated. They lose the definite and specific content that Sosa seems to demand for recovery under the [ATCA]’.
Thus, the Court suggests that any norm defined without specified limits, no matter how well-recognised in the international community, will not support an actionable ATCA claim, irrespective of the actual conduct in question. The Court considered it irrelevant that the principles of unnecessary suffering and proportionality are widely accepted as cardinal in international humanitarian law. The Court’s approach appears to have two stages. The first — the threshold stage — would involve examining the international norm allegedly breached, to ascertain whether it has definite content. If not, then the claim will fail, regardless of the conduct giving rise to the alleged violation. If so, then only at that point does the court examine whether the conduct in question falls within the specified content of the norm.
The Eleventh Circuit Court in Aldana v Del Monte Fresh Produce also appears to have adopted this approach. Aldana concerned the detention, in some cases following abduction, of trade union leaders in Guatemala by security forces who threatened detainees with death at gunpoint. The plaintiffs brought a suit alleging, inter alia, violations of the international norm prohibiting cruel, inhuman and degrading treatment. Upholding the District Court’s dismissal of the plaintiffs’ claim, the Court of Appeal disregarded the actual facts of the case, and simply stated that ‘based largely on our reading of Sosa … [w]e see no basis in law to recognize [the] Plaintiffs’ claim for cruel, inhuman, degrading treatment or punishment’.
It is of course true that the norm prohibiting cruel, inhuman and degrading treatment is somewhat vague; its boundaries are unclear, and its violation requires ‘suffering’ on the part of the victim, a term that is admittedly slippery. But its vagueness does not necessarily preclude its recognition as a universal norm. Further, to say that no action lies under the ATCA for cruel, inhuman and degrading treatment on the basis that its precise content is unclear is tantamount to saying that no action lies for torture either, because what separates the one norm from the other lies in many cases on the elusive degree of suffering involved. As one Court put it, while
it may present difficulties to pinpoint precisely where on the spectrum of atrocities the shades of cruel, inhuman, or degrading treatment bleed into torture [this] should not detract from what really goes to the essence of any uncertainty: that, distinctly classified or not, the infliction of cruel, inhuman or degrading treatment by agents of the state, as closely akin to or adjunct of torture, is universally condemned and renounced as offending internationally recognized norms of civilized conduct.
The Eleventh Circuit Court’s problematic approach did not go unnoticed. The Aldana plaintiffs applied for a rehearing en banc. While the Court denied the application, one judge dissented. Finding that the Court’s conclusion that no cause of action lies for a claim of cruel, inhuman or degrading treatment under the ATCA was a ‘precedent-setting error of exceptional importance … in direct conflict with precedent of the Supreme Court’, Judge Barkett would have ordered a rehearing. Notably, Judge Barkett set out what he regarded as the correct approach:
The specific content requirement of Sosa is not one of categorical specificity — it does not require defining every possible instance of cruel, inhuman, or degrading treatment or punishment, but rather compels a determination of whether the facts alleged in a particular situation sit within the universal prohibition against cruel, inhuman, or degrading treatment or punishment: Sosa … (avoiding the larger question of whether a plaintiff could ever bring a claim of arbitrary detention under the ATCA, and settling instead on the narrower issue of whether the plaintiff … had alleged events that could support a claim under the ATCA).
Barkett J’s approach is undoubtedly correct. The counter approach would overlook the fact that some norms by their very nature admit the possibility of violation in various ways; therefore, a detailed definition of the norm is impractical and undesirable. Having established that a norm has sufficient consensus, even if it has indefinite boundaries, one must then turn to the facts in question to determine whether those specific facts fit within the well-established content, as opposed to the unclear periphery, of the norm. This approach was adopted by the Court in Roe v Bridgestone Co, a case concerning allegations of, among other things, forced labour and cruel, inhuman or degrading treatment occurring at a Liberian rubber plantation. In respect of the claim based on the norm prohibiting forced labour, the Court recognised that the norm was indefinite insofar as it contained a continuum ranging from ‘those clear violations of international law (slavery or forced labour at the point of soldiers’ bayonets) to more ambiguous situations involving poor working conditions and meager or exploitative wages’. But such indefiniteness did not rule out the norm as one that could found an ATCA claim. For Judge Hamilton, the key was to assess the particular facts in order to see whether they fell within the well-established content of the norm. Thus, having found that the facts as alleged — essentially low wages and difficult working conditions — lay ‘at a point on a continuum far from the forced labor of Nazi Germany, Japanese labor camps, or the workers rounded up more recently by the Burmese military’ only then was the Court able to dismiss the claim. Similarly, in respect of the cruel, inhuman and degrading treatment claim, the Court rejected the defendants’ argument that ‘any international norm against cruel, inhuman and degrading treatment fails the Sosa test of being specific, universal, and obligatory’ Rather, the Court stressed the need to focus ‘on the particular conduct in question [in order] to decide whether the customary international norm against cruel, inhuman, and degrading treatment is sufficiently specific, universal and obligatory as applied to that conduct’
As is evident from the emerging jurisprudence post-Sosa, there is confusion over the appropriate method to adopt when determining whether a norm is sufficiently determinate to fall within the scope of the ATCA. The erroneous approach adopted by the Second and Eleventh Circuits as detailed above, is unlikely to win the day for, if stringently applied, its absurdity would become manifest: it would spell the end of claims based on certain jus cogens norms, such as torture and forced labour, whose boundaries are notoriously vague. Denying such claims, however, would undermine precedent the Supreme Court in Sosa cited with approval. Moreover, even the Second Circuit Court has recognised on occasion that jus cogens violations ‘form the least controversial core of modern day ATCA jurisdiction’. Be that as it may, the answer to the broader question of which norms fall within the proper scope of the ATCA remains unclear, and many future claims will have to come before the courts before this uncertainty is diminished.
A few months after Sosa, the District Court for the Southern District of New York in In Re South African Apartheid Litigation rejected the argument that defendants could be held liable under the ATCA for aiding and abetting. Three reasons underpinned its conclusion. Firstly, the Court did not regard aiding and abetting liability under international law as being sufficiently determinate or recognised so as to fall within the specificity requirements laid down by the Supreme Court in Sosa. Secondly, the Court was wary of the ‘foreign relations repercussions’ that would result from recognising such liability. Thirdly, the Court relied on Supreme Court precedent, albeit in the context of securities legislation, cautioning against accessorial liability in civil causes of action where Congress had not explicitly provided for it.
This decision, which was overturned in late 2007 by the Second Circuit Court of Appeal in Khulumani, found little support among district courts. Although one court accepted its reasoning in full, Judge Weinstein in In Re ‘Agent Orange’ Product Liability Litigation declared that there ‘is simply no question that the [ATCA] provides for aiding and abetting liability’ under international law. Similarly, another court, rejecting the argument that secondary liability standards are too imprecise to support an ATCA claim, held that aiding and abetting was among the ‘core principles that form the foundation of customary international legal norms — principles about which there is no disagreement’. These courts and others have relied on the jurisprudence of the ad hoc international criminal tribunals to support their view that international law does indeed recognise to the requisite degree of specificity aiding and abetting liability.
The most thorough examination of secondary liability, however, was made by the Second Circuit Court of Appeal in Khulumani. As noted earlier, this appeal from the District Court’s decision in In Re South African Apartheid Litigation concerned the actions of multinational banks and corporations which had allegedly collaborated with the South African Government during apartheid. Although the Second Circuit Court had an ideal opportunity to clarify accessorial liability standards, there was little agreement among the three judges who heard the case.
By majority, the Court held corporations could be liable under the ATCA for aiding and abetting human rights violations. But the majority judges disagreed on the appropriate standard. Judge Katzmann, citing consistency with prior case law as a reason, turned to international law. By tracing the law on accessorial liability from its origins in the Nuremberg Charter through the ad hoc international criminal tribunals to the Rome Statute, he adopted a test based on the latter:
[that a defendant] may be held liable under international law for aiding and abetting the violation of that law by another when the defendant (1) provides practical assistance to the principal which has a substantial effect on the perpetration of the crime, and (2) does so with the purpose of facilitating the commission of that crime.
Judge Katzmann concluded that ‘aiding and abetting liability, so defined, is sufficiently “well-established and universally recognized” to be considered customary international law for the purposes of the ATCA’.
By contrast, Judge Hall, finding that both customary international law and federal common law included standards of accessorial liability, preferred to use federal common law standards. Invoking § 876(b) of the Restatement (Second) of Torts, Judge Hall proposed that a party should be liable for aiding and abetting under the ATCA if that party ‘knows that the [principal’s] conduct constitutes a breach of duty and gives substantial assistance or encouragement’ to the principal. Such a standard would cover circumstances ‘where the alleged aider and abettor is accused of having purchased security services with the knowledge that the security forces would, or were likely to, commit international law violations in fulfilling their mandate’. Further, liability would arise ‘when a defendant provides “the tools, instrumentalities, or services to commit [human rights] violations with actual … knowledge that those tools, instrumentalities or services will be (or only could be) used in connection with that purpose”’.
The dissenting judge, Judge Korman, was loath to accept corporate responsibility under international law; it followed, a fortiori, that a corporation could not be held liable for aiding and abetting international law violations. However, one could not rule out accessorial liability under the ATCA altogether. For Judge Korman, the ATCA was drafted broadly enough to cover aiding and abetting liability; however, the key inquiry in every case would be whether the international norm in question provided for such liability. In other words, one would have to engage in a ‘norm-by-norm’ analysis. If the norm did so provide, then the appropriate international law standard would be the standard set out in the Rome Statute, as Judge Katzmann identified.
Given the terse wording of the ATCA and the paucity of historical evidence casting light on the intentions of the Act’s drafters, there is, as one commentator noted, no easy answer as to which accessorial liability standard should be used in ATCA litigation.
Further Supreme Court guidance is needed, and is likely to be provided eventually, given that aiding and abetting liability is the preferred approach of those seeking to hold multinational corporations to account under the Act. For now, some brief points can be made about the international law standard for aiding and abetting that the two judges in Khulumani advocated.
Judges Katzmann and Korman chose the standard for aiding and abetting liability set out in the Rome Statute. The mental element of that definition requires that assistance be given to the principal ‘for the purpose of facilitating the commission’ of the crime.
As Eser has explained, ‘[t]his means more than mere knowledge that the accomplice aids the commission of the offence, … rather he must know as well as wish that his assistance shall facilitate the commission of the crime’. This mens rea requirement is more stringent than that which has been articulated in the jurisprudence of the ad hoc international criminal tribunals, a fact which was not lost on the two judges in Khulumani. The ad hoc tribunals have held that mere knowledge or awareness that one’s acts will assist in the commission of the offence is sufficient, but for Judges Katzmann and Korman, the higher standard articulated in the Rome Statute was preferable. Both judges stressed that much of the law on aiding and abetting stated by the ad hoc tribunals took the form of obiter dicta, and was articulated for the purpose of providing ‘clarification [that] might have some value for the future development of international criminal law’. Thus, for both judges, the jurisprudence on the matter from the ad hoc tribunals lacked sufficient consensus and recognition in the international community to form the basis of an ATCA claim.
This reasoning, however, is unconvincing. As Judge Katzmann himself conceded, the ad hoc tribunals consider themselves authorised to apply only rules of customary international law. The former UN Secretary-General Annan has explained why:
the application of the principle nullum crimen sine lege requires that the international tribunal [for the former Yugoslavia] should apply rules of international humanitarian law which are beyond any doubt part of customary law so that the problem of adherence of some but not all States to specific conventions does not arise.
In view of this fact alone, the international tribunals’ jurisprudence must be considered a persuasive source when ascertaining clear customary law. This is especially so with respect to their statements on aiding and abetting liability, given the tribunals’ thorough analysis of such liability in international law since the Nuremberg trials. Furthermore, art 25(3)(c) of the Rome Statute, which regulates aiding and abetting liability and which was relied on in Khulumani, was not intended to reflect custom. Although over 100 states have signed and ratified the Rome Statute — a point both judges stressed in reaching their conclusions — this fact is immaterial given that customary and conventional law can exist simultaneously on the same matter, even if the content of each differs. Finally, the two judges ignored the view of the majority of the Ninth Circuit Court of Appeal in Unocal, which found that the mens rea for aiding and abetting liability in customary law was that set out in the international tribunals’ jurisprudence, namely that knowledge is required. Although the two judges were not formally bound by the Ninth Circuit Court’s opinion, their failure to refer to or engage with the Court’s opinion perhaps suggests a pre-determined viewpoint.
In practical terms, the standard set in Khulumani that requires a plaintiff to prove purpose on the part of the corporation providing assistance will be all but impossible to meet. The likelihood that US corporations will actively seek to further human rights violations in those places where they are doing business is low. This fact might well have been the real reason behind the decisions of Judges Katzmann and Korman to adopt the Rome Statute test. It effectively shields corporations from ATCA claims, an outcome favoured by many who are sympathetic to big business. But, as a matter of law, their conclusion is weak and of dubious precedential value. It further complicates the issues with which, in the absence of Supreme Court authority, many federal courts have to grapple. The irony, as Heller has noted, is that if federal courts adopt the true customary law standard as articulated by the ad hoc tribunals, then the question of whether to apply domestic tort law or international law standards on aiding and abetting liability loses its importance, as the mens rea requirement of both is the same.
An examination of the post-Sosa jurisprudence illustrates the difficulties that courts have had in interpreting and applying the Supreme Court’s decision and brings into relief the divergent streams of opinion on corporate accountability under the ATCA. Some of the difficulties and divergent standards can perhaps partly be explained by the inexperience of lawyers and judges when it comes to arguing and adjudicating international law. Doubtless they are more at home with purely domestic law disputes. As one federal Court of Appeal judge has freely admitted, international human rights law is ‘not a major, or even a minor, component of the business of federal courts: it is a miniscule part of what we do’. However, it is likely that the political persuasion of judges also plays a part. As I have argued elsewhere, isolationists, those sympathetic to the business community and those who fear that the ATCA may obstruct the US and its allies’ efforts in the war on terrorism, are more likely to interpret the ATCA in a restrictive manner than judges more receptive to international law.
But surely, exacerbating the problem for lawyers and courts is the Sosa decision itself, which provides scant practical guidance.
Whatever the reasons for the confused state of the law, in light of the post-Sosa jurisprudence, four general points can be made.
Firstly, there is a growing consensus that corporations can breach norms of international law and can be held responsible under the ATCA. Arguments to the contrary are beginning to sound anachronistic and are likely soon to die out.
Secondly, the same can be said about aiding and abetting liability: the Second Circuit in Khulumani has confirmed and effectively endorsed the now widespread view among District Courts that corporations can be held liable as accessories under the Act.
Thirdly, uncertainty surrounds the question of which norms fall within the ATCA’s ambit, and no courts have grappled with the Supreme Court’s impractical historical test. The Second and Eleventh Circuit Courts of Appeal, in particular, appear to have misunderstood the test, rejecting claims based on well-recognised norms whose outer limits are unclear. The fallacy of this approach is evident when it is realised that many norms, including the prohibition on torture, have limits that are difficult to ascertain; yet no court would suggest excluding claims based on torture. The correct approach, which several courts have adopted, is undoubtedly to ascertain whether a norm has core content that is widely accepted by states to be beyond any doubt, and then to determine whether the facts in question fall within it. But even this test is complex in its application. For the moment, perhaps all that can be said is that jus cogens violations ‘form the least controversial core of modern day ATCA jurisdiction’.
Fourthly, there is uncertainty concerning the correct standard to be applied in determining aiding and abetting liability. The majority in Khulumani held that the proper standard is that set out in the Rome Statute, which requires proof of the accomplice’s purpose or intent. However, this commentary proposes that this standard does not reflect customary international law. The more appropriate standard is that set out by the ad hoc international criminal tribunals. If this standard, which requires mere knowledge on the part of the accomplice, is adopted, then the question of whether to apply international law or federal tort law standards loses its significance.
In conclusion, it is worth reflecting on the ramifications of Sosa and the uncertainty that has emerged in its wake. By raising the threshold for actionable claims, the Supreme Court theoretically made it harder for plaintiffs to bring successful suits. It is therefore understandable that some commentators viewed Sosa as a ‘win’ for corporations. Yet it has never been easy for plaintiffs to bring successful ATCA suits. Courts have consistently applied a stringent standard to filter out claims based on inchoate norms of international law. As Stephens has pointed out, prior to Sosa, some 35 corporate lawsuits had been brought, yet fewer than a dozen had survived preliminary legal motions. Therefore, it is questionable whether the Supreme Court’s test will have any noticeable impact; one can expect that the types of ATCA claims that were strong enough to receive preliminary judicial sanction prior to Sosa will successfully meet the Sosa criteria also.
Furthermore, it is suggested that despite the higher threshold set by the Supreme Court, its failure to provide corporations and courts with a ‘bright-line test’ means that corporations will have to be as cautious as they have always been when engaging in activities abroad. As Waldron has argued, to the extent that legal standards are unclear or indeterminate, those who are governed by those standards may not know what is expected of them, and consequently may not know how to discharge their duty of fidelity to the law. The effect of vague legal standards on a person will therefore be ‘to chill that person’s exercise of liberty as he tries to avoid being taken by surprise by enforcement decisions’. This being so, the Supreme Court’s vague decision, which has in turn led to the articulation of divergent standards by the lower courts, is likely to instil uncertainty in those directing corporate activities. Not knowing the exact limits of what is permissible, corporations will likely err on the side of caution. Of course, ambiguity can influence behaviour only to a degree. It is not being suggested that the Supreme Court’s decision and the resultant ATCA jurisprudence is so vague that it will scare corporations away completely from disfavoured states. As Koh has argued, plaintiffs bringing claims under the ATCA need to show ‘much, much more than simply that the multinational enterprise has chosen to invest in a “troublesome country”’.
But while corporations should not be inhibited from doing business in countries with poor human rights records after Sosa, the lack of clear standards will certainly make corporations think twice about participating in projects with governments where human rights abuses connected to those projects are a possibility. Bright-line standards would have enabled corporations to push the limits of the law in terms of their conduct and its relation to human rights abuses; they would know just how far they could go. Post-Sosa, in the absence of definitive standards, corporations have to be just as circumspect as they were prior to the decision. Ironically, despite setting a more restrictive standard for ATCA claims, the Supreme Court’s decision, marked by its lack of clarity, has, in practical terms, changed little.
[*] LLB (Hons) (Victoria University of Wellington); LLM (Toronto); JSD candidate (New York University School of Law). The author wishes to thank Nehal Bhuta and the anonymous referees for their comments on earlier versions of this commentary.
  USSC 2852; 542 US 692 (2004) (‘Sosa’).
 Alien Tort Claims Act, 28 USC § 1350 (2007) (‘ATCA’).
 See below Part II.
 Sosa USSC 2852; , 542 US 692 (2004) (Amici Curiae Brief: National Foreign Trade Council et al).
 Gary Clyde Hufbauer and Nicholas K Mitrokostas, Awakening Monster: The Alien Tort Statute of 1789 (2003).
 See, eg, Warren Richey, ‘Ruling Makes It Harder for Foreigners To Sue in US Courts’, Christian Science Monitor (Boston, US) 30 June 2004, 3, noting that the decision ‘marks a victory for US-based corporations’ and ‘a setback for international human rights activists’.
 See Lisa Girion, ‘Unocal To Settle Rights Claims’, Los Angeles Times (Los Angeles, US) 14 December 2004, A1, citing Elliot Schrage, who opined that nobody ‘can treat these [corporate] cases as a joke anymore’.
 ATCA, 28 USC § 1350 (2007).
  USCA2 576; 630 F 2d 876 (2nd Cir, 1980) (‘Filartiga’).
 Ibid 880.
 Referring to Supreme Court authority, Judge Kaufman equated the law of nations with the ‘general usage and practice of nations’, and required any international law rule to command the ‘general assent of civilized nations’: ibid 880–1. This ‘general usage’ or ‘general assent’ test is consistent with the approach for finding custom taken at the international level: see North Sea Continental Shelf Cases (Federal Republic of Germany v Denmark; Federal Republic of Germany v The Netherlands) (Judgment)  ICJ Rep 3, 43; Malcolm N Shaw, International Law (5th ed, 2003) 68–84; Jordan J Paust, ‘The History, Nature, and Reach of the Alien Tort Claims Act’ (2004) 16 Florida Journal of International Law 249, 258–60.
 Filartiga USCA2 576; , 630 F 2d 876 (2nd Cir, 1980) (Amicus Curiae Memorandum: US) 22.
 Filartiga USCA2 576; , 630 F 2d 876, 881 (2nd Cir, 1980).
 Ibid 886.
 Ibid 887.
 Beth Stephens and Michael Ratner, International Human Rights Litigation in US Courts (1996) 7, noting that since its enactment in 1789, the ATCA had been ‘virtually ignored for almost 200 years’.
 In Tel-Oren v Libyan Arab Republic,  USCADC 53; 726 F 2d 774, 795 (DC Cir, 1984), Judge Edwards held that the ATCA did not cover torture by non-state actors, but added that the ‘trend in international law is toward a more expansive allocation of rights and obligations to entities other than states’. In In Re Estate of Ferdinand E Marcos Human Rights Litigation,  USCA9 3140; 978 F 2d 493, 501–2 (9th Cir, 1992), the Ninth Circuit Court of Appeal stated that ‘[o]nly individuals who have acted under official authority or under color of such authority may violate international law’.
  USCA2 16; 70 F 3d 232 (2nd Cir, 1995) (‘Kadic’).
 Ibid 237, 244–5.
 Doe v Karadžić, 866 F Supp 734, 739 (SD NY, 1994).
 Kadic USCA2 16; , 70 F 3d 232, 240 (2nd Cir, 1995).
 Ibid 239–40.
 Ibid 239.
 Ibid 242–3.
 Gwynne Skinner, ‘Nuremberg’s Legacy Continues: The Nuremberg Trials’ Influence on Human Rights Litigation in US Courts under the Alien Tort Statute’ (2008) 71 Albany Law Review 321, 343.
 The first major cases against corporations brought pursuant to the ATCA were initiated by Ecuadorian and Peruvian plaintiffs in 1993 and 1994 against Texaco Inc, for the latter’s oil exploration and extraction activities in Ecuador and Peru, which had allegedly caused environmental and personal injuries. See Aguinda v Texaco Inc,  USCA2 273; 303 F 3d 470 (2nd Cir, 2002).
 963 F Supp 880 (CD Cal, 1997).
 Doe I v Unocal Co, 110 F Supp 2d 1294, 1297 (CD Cal, 2000).
 Ibid 1299–300.
 Doe I v Unocal Co,  USCA9 708; 395 F 3d 932, 947–56 (9th Cir, 2002) (‘Unocal’).
 Ibid 951.
 Ibid 963.
 Unocal,  USCA9 100; 395 F 3d 978 (9th Cir, 2003).
 See John R Crook (ed), ‘Contemporary Practice of the United States relating to International Law’ (2005) 99 American Journal of International Law 479, 497–8.
 David D Christensen, ‘Corporate Liability for Overseas Human Rights Abuses: The Alien Tort Statute after Sosa v Alvarez-Machain’ (2005) 62 Washington and Lee Law Review 1219, 1242–3.
 Unocal was the first ‘successful’ claim brought against a company in the sense that a settlement was reached out of court, but this occurred after the Supreme Court’s decision in Sosa.
 See generally Harold Hongju Koh, ‘Separating Myth from Reality about Corporate Responsibility Litigation’ (2004) 7 Journal of International Economic Law 263.
 William S Dodge, ‘Which Torts in Violation of the Law of Nations?’ (2001) 24 Hastings International and Comparative Law Review 351, 351–2.
 See, eg, Beanal v Freeport-McMoRan Inc,  USCA5 1266; 197 F 3d 161, 166–8 (5th Cir, 1999).
 Filartiga USCA2 576; , 630 F 2d 876, 880–1 (2nd Cir, 1980).
 Kadic USCA2 16; , 70 F 3d 232, 239 (2nd Cir, 1995).
 Unocal USCA9 708; , 395 F 3d 932, 944 (9th Cir, 2002). See also Flores v Southern Peru Copper Co,  USCA2 373; 343 F 3d 140, 154 (2nd Cir, 2003).
 Paust, ‘The History, Nature, and Reach of the Alien Tort Claims Act’, above n 12, 259–60. See also American Law Institute, Restatement (Third) of Foreign Relations Law of the United States § 102(2) (1987), noting that ‘[c]ustomary international law results from a general and consistent practice of states followed by them from a sense of legal obligation’; Dodge, above n 40, 354–6; Hugh King, ‘Sosa v Alvarez-Machain and the Alien Tort Claims Act’ (2006) 37 Victoria University of Wellington Law Review 1, 5–9.
 In Unocal USCA9 708; , 395 F 3d 932, 945 (9th Cir, 2002), the Court stressed that ‘although a jus cogens violation is, by definition, “a violation of ‘specific, universal, and obligatory’ international norms” that is actionable under the ATCA, any “violation of ‘specific, universal, and obligatory’ international norms” — jus cogens or not — is actionable under the ATCA’ (emphasis in original).
 See, eg, Doe I v Unocal Co, 110 F Supp 2d 1294, 1304 (CD Cal, 2000); Xuncax v Gramajo, 886 F Supp 162, 183 (D Mass, 1995).
 Hoffman and Zaheer note that, by 2003, no court had settled on particular rules for aiding and abetting liability: Paul L Hoffman and Daniel A Zaheer, ‘The Rules of the Road: Federal Common Law and Aiding and Abetting under the Alien Tort Claims Act’ (2003) 26 Loyola of Los Angeles International and Comparative Law Review 47, 48–9.
 This analysis addresses only the part of the decision pertaining to the ATCA and ignores the claims against the US Government under the Federal Tort Claims Act, 28 USC §§ 1346(b), 2671–80 (2007).
 See Sosa,  USSC 2852; 542 US 692, 697–9 (2004) for the Supreme Court’s account of the facts.
 See Filartiga USCA2 576; , 630 F 2d 876 (2nd Cir, 1980); Kadic USCA2 16; , 70 F 3d 232 (2nd Cir, 1995). See also Sarah H Cleveland, ‘The Alien Tort Statute, Civil Society, and Corporate Responsibility’ (2004) 56 Rutgers Law Review 971, 983, noting that prior to the advent of the Bush administration, ‘executive branches consistently have either actively supported [ATCA] litigation or remained neutral in such cases’.
 See Sosa USSC 2852; , 542 US 692 (2004) (Amicus Curiae Brief: US).
 See Tel-Oren v Libyan Arab Republic,  USCADC 53; 726 F 2d 774, 801–23 (DC Cir, 1984) (Judge Bork); Al Odah v United States,  USCADC 52; 321 F 3d 1134, 1146–9 (DC Cir, 2003) (Judge Randolph).
 See Sosa USSC 2852; , 542 US 692 (2004) (Amici Curiae Brief: National and Foreign Legal Scholars).
 Sosa USSC 2852; , 542 US 692, 697 (2004).
 Ibid 714.
 Ibid 715.
 Ibid 725.
 Ibid 729.
 Ibid 732.
 Ibid 732–3.
 Ibid 737.
 Ibid 736. The Court noted that Alvarez-Machain’s purported rule of international law, a general prohibition of ‘arbitrary’ detention defined as officially sanctioned action exceeding positive authorization to detain under the domestic law of some government, regardless of the circumstances[,] … would support a cause of action in federal court for any arrest, anywhere in the world, unauthorized by the law of the jurisdiction in which it took place: at 736.
 See Sosa USSC 2852; , 542 US 692 (2004) (Amici Curiae Brief: National Foreign Trade Council) 4, arguing that the ATCA ‘has become a serious impediment to US companies investing abroad’; Sosa USSC 2852; , 542 US 692 (2004) (Amicus Curiae Brief: US).
 Sosa USSC 2852; , 542 US 692, 733 (2004).
 Beth Stephens, ‘Sosa v Alvarez-Machain: “The Door is Still Ajar” for Human Rights Litigation in US Courts’ (2004) 70 Brooklyn Law Review 533, 556; Ralph G Steinhardt, ‘Laying One Bankrupt Critique to Rest: Sosa v Alvarez-Machain and the Future of International Human Rights Litigation in US Courts’ (2004) 57 Vanderbilt Law Review 2241, 2284. Cf Skinner, above n 26, 341, noting that the ‘US Supreme Court has not yet decided whether private, non-state actors can be liable’ under the ATCA.
 Sosa USSC 2852; , 542 US 692, 733 (2004).
 Steinhardt, above n 66, 2285.
 Sosa USSC 2852; , 542 US 692, 733 (2004).
 See Brad R Roth, ‘Sosa v Alvarez-Machain; United States v Alvarez-Machain’ (2004) 98 American Journal of International Law 798, 804, noting that the decision ‘announces the existence of strict limits to the power of courts to establish international law-based causes of action under the [ATCA], but does relatively little, in practical terms, to specify those limits’; Benjamin Berkowitz, ‘Sosa v Alvarez-Machain: United States Courts as Forums for Human Rights Cases and the New Incorporation Debate’ (2005) 40 Harvard Civil Rights–Civil Liberties Law Review 289, 295, noting the Court’s ‘failure to articulate clear standards to guide litigants and the lower courts as to which customary international legal norms are adequately accepted and defined’; Saad Gul, ‘The Supreme Court Giveth and the Supreme Court Taketh Away: An Assessment of Corporate Liability Under § 1350’ (2007) 109 West Virginia Law Review 379, 405, noting that ‘[a]cademics and courts alike have struggled to comprehend the Sosa holding’.
 ‘The Supreme Court, 2003 Term — Leading Cases’ (2004) 118 Harvard Law Review 248, 454–5.
 In Re South African Apartheid Litigation, 346 F Supp 2d 538, 547 (SD NY, 2004).
 See Shaw, above n 12, 68–72.
 See, eg, Roth, above n 72, 803. See also Laura A Cisneros, ‘Sosa v Alvarez-Machain — Restricting Access to US Courts under the Federal Tort Claims Act and the Alien Tort Statute: Reversing the Trend’ (2004) 6 Loyola Journal of Public Interest Law 81, 97–8, noting that the decision ‘benefits private defendants, including corporations’ and that for potential plaintiffs, it ‘places the bar almost out of reach’.
 Sosa USSC 2852; , 542 US 692, 728 (2004).
 Ibid 729.
 Eugene Kontorovich, ‘Implementing Sosa v Alvarez-Machain: What Piracy Reveals about the Limits of the Alien Tort Statute’ (2004) 80 Notre Dame Law Review 111, 113.
 See King, above n 45, 17–18. See also Sosa USSC 2852; , 542 US 692, 737 (2004), noting that
although it is easy to say that some policies of prolonged arbitrary detentions are so bad that those who enforce them become enemies of the human race, it may be harder to say which policies cross that line with the certainty afforded by Blackstone’s three common law offenses.
 Khulumani v Barclay National Bank Ltd, 504 F 3d 254, 291 (2nd Cir, 2007) (Judge Hall) (‘Khulumani’).
 Other points of contention, a discussion of which is outside the scope of this commentary, include whether there is a requirement that plaintiffs exhaust local remedies before bringing claims and the extent to which courts should defer to the view of political branches of government on the foreign policy ramifications of admitting claims.
 See, eg, Gary Clyde Hufbauer and Nicholas K Mitrokostas, ‘International Implications of the Alien Tort Statute’ (2004) 7 Journal of International Economic Law 245, arguing that recognition of corporate liability under the Act could ‘devastate global trade and investment’: at 245.
 Sosa USSC 2852; , 542 US 692 (2004) (Amici Curiae Brief: National Foreign Trade Council et al) 11–12.
 See, eg, Koh, above n 39, 269; Cleveland, above n 51, 982.
 See below Part IV(B). See also Cleveland, above n 51, 981.
 Gul, above n 72, 407.
 See Vietnam Association for Victims of Agent Orange v Dow Chemical Co, 517 F 3d 104, 115 (2nd Cir, 2008); Bowoto v Chevron Co, No C 99-02506 SI, 9 (ND Cal, 22 August 2006); Presbyterian Church of Sudan v Talisman Energy Inc, 374 F Supp 2d 331, 335 (SD NY, 2005).
 See Louis Henkin et al, International Law: Cases and Materials (3rd ed, 1993) xvii.
 For a good discussion of this issue, see Emeka Duruigbo, ‘Corporate Accountability and Liability for International Human Rights Abuses: Recent Changes and Recurring Challenges’ (2008) 6 Northwestern University Journal of International Human Rights 222.
 Lucien J Dhooge, ‘A Modest Proposal to Amend the Alien Tort Statute to Provide Guidance to Transnational Corporations’ (2007) 13 University of California Davis Journal of International Law and Policy 119, 132.
 Presbyterian Church of Sudan v Talisman Energy Inc, 374 F Supp 2d 331, 335 (SD NY, 2005). See also In Re ‘Agent Orange’ Product Liability Litigation, 373 F Supp 2d 7, 58 (ED NY, 2005); Bowoto v Chevron Co, No C 99-02506 SI, 9 (ND Cal, 22 August 2006).
 Steven R Ratner, ‘Corporations and Human Rights: A Theory of Legal Responsibility’  YaleLawJl 32; (2001) 111 Yale Law Journal 443, 461–5.
 See ibid 475–88; Beth Stephens, ‘The Amorality of Profit: Transnational Corporations and Human Rights’ (2002) 20 Berkeley Journal of International Law 45, 75–8; Koh, above n 39, 264–8; Jordan J Paust, ‘Human Rights Responsibilities of Private Corporations’ (2002) 35 Vanderbilt Journal of Transnational Law 801.
 In Re ‘Agent Orange’ Product Liability Litigation, 373 F Supp 2d 7, 58 (ED NY, 2005). See also Koh, above n 39, arguing that
if states and individuals can be held liable under international law, then so too should corporations, for the simple reason that both states and individuals act through corporations. Given that reality, what legal sense would it make to let states and individuals immunize themselves from liability for gross violations through the mere artifice of corporate formation?: at 265 (emphasis added).
 Khulumani, 504 F 3d 254, 291, 326 (2nd Cir, 2007).
 Ibid 282–3 (Judge Katzmann).
 Ibid 282, noting that it is
perhaps not surprising that neither the defendants nor the United States raised this issue as a bar to liability: We have repeatedly treated the issue of whether corporations may be held liable under the ATCA as indistinguishable from the question of whether private individuals may be.
 Christensen, above n 37, 1231.
 See, eg, Guinto v Marcos, 654 F Supp 276, 280 (SD Cal, 1986).
 See, eg, IIT v Vencap Ltd,  USCA2 317; 519 F 2d 1001, 1015 (2nd Cir, 1975).
 See, eg, Hamid v Price Waterhouse,  USCA9 1064; 51 F 3d 1411, 1418 (9th Cir, 1995).
 See, eg, Doe v Exxon Mobil Co, 393 F Supp 2d 20, 24 (D DC, 2005).
 See, eg, Beanal v Freeport-McMoRan Inc, 969 F Supp 362, 384 (ED La, 1997), noting that the ‘three [international environmental law] principles relied on by [the] Plaintiff, standing alone, do not constitute international torts for which there is universal consensus in the international community’; Beanal v Freeport-McMoRan Inc,  USCA5 1266; 197 F 3d 161 (5th Cir, 1999).
 See, eg, Human Rights Watch, US Vice President Endorses Torture: Cheney Expresses Approval of the CIA’s Use of Waterboarding (2006) <http://www.hrw.org/english/docs/2006/10/26/usdom14465.htm> at 23 September 2008.
 517 F 3d 104 (2nd Cir, 2008).
 Ibid 119–20.
 Ibid 114.
 Ibid 120.
 Hague Convention (IV) respecting the Laws and Customs of War on Land, Annex to the Convention, Regulations respecting the Law and Customs of War on Land, opened for signature 18 October 1907, (1910) UKTS 9 (entered into force 26 January 1910) art 23(e) (‘Hague Regulations’).
 Annexed to the Agreement by the Government of the United States of America, the Provisional Government of the French Republic, the Government of the United Kingdom of Great Britain and Northern Ireland and the Government of the Union of Soviet Socialist Republics for the Prosecution and Punishment of the Major War Criminals of the European Axis, opened for signature 8 August 1945, 82 UNTS 280 (entered into force 8 August 1945).
 Opened for signature 12 August 1949, 75 UNTS 287 (entered into force 21 October 1950).
 Ibid art 147. See Vietnam Association for Victims of Agent Orange v Dow Chemical Co, 517 F 3d 104, 119, 122 (2nd Cir, 2008).
 Vietnam Association for Victims of Agent Orange v Dow Chemical Co, 517 F 3d 104, 119, 122 (2nd Cir, 2008). The second basis was that the plaintiffs had been unable to prove the necessary intent: at 123.
 See Legality of the Threat or Use of Nuclear Weapons (Advisory Opinion)  ICJ Rep 226, 257.
  USCA11 188; 416 F 3d 1242 (11th Cir, 2005) (‘Aldana’).
 Ibid 1245.
 Ibid 1247.
 See, eg, Mehinovic v Vuckovic, 198 F Supp 2d 1322, 1348 (ND Ga, 2002), where cruel, inhuman or degrading treatment is defined as including ‘acts which inflict mental or physical suffering, anguish, humiliation, fear and debasement, which do not rise to the level of “torture” or do not have the same purposes as “torture”’.
 See, eg, de Sanchez v Banco Central de Nicaragua USCA5 1141; , 770 F 2d 1385, 1397 (5th Cir, 1985); Xuncax v Gramajo, 886 F Supp 162, 187 (D Mass, 1995); Wiwa v Royal Dutch Petroleum Co, 96-8386, 9 (SD NY, 28 February 2002); Doe v Qi, 349 F Supp 2d 1258, 1320–5 (ND Cal, 2004). But see Forti v Suarez-Mason, 672 F Supp 1531, 1543 (ND Cal, 1987), holding that the norm’s lack of precision cannot support an ATCA claim.
 Tachiona v Mugabe, 234 F Supp 2d 401, 437 (SD NY, 2002).
 Aldana v Del Monte Fresh Produce,  USCA11 180; 452 F 3d 1284 (11th Cir, 2006) (Judge Barkett).
 Ibid 1284.
 Ibid 1288 (emphasis in original).
 492 F Supp 2d 988 (SD Ind, 2007). This approach was also adopted in Doe v Qi, 349 F Supp 2d 1258, 1322 (ND Cal, 2004), where the Court noted that ‘the fact that there may be doubt at the margins — a fact that inheres in any definition — does not negate the essence and application of that definition in clear cases’.
 Roe v Bridgestone Co, 492 F Supp 2d 988, 1010 (SD Ind, 2007).
 Ibid 1012.
 Ibid 1022.
 Ibid 1023.
 See Sosa USSC 2852; , 542 US 692, 732–3 (2004), where the Supreme Court affirms the approach taken in Filartiga USCA2 576; , 630 F 2d 876 (2nd Cir, 1980).
 Sarei v Rio Tinto, 456 F 3d 1069, 1078 (9th Cir, 2006). See also Sarei v Rio Tinto,  USCA9 204; 487 F 3d 1193, 1202 (9th Cir, 2007).
 346 F Supp 2d 538 (SD NY, 2004).
 Ibid 549–50.
 Ibid 551.
 Ibid 550, citing Central Bank of Denver v First Interstate Bank of Denver,  USSC 5; 511 US 164 (1994).
 See Doe v Exxon Mobil Co, 393 F Supp 2d 20, 24 (D DC, 2005).
 373 F Supp 2d 7 (ED NY, 2005).
 Presbyterian Church of Sudan v Talisman Energy Inc, 374 F Supp 2d 331, 341 (SD NY, 2005).
 See, eg, Bowoto v Chevron Co, No C 99-02506 SI, 3–4 (ND Cal, 22 August 2006).
 In Re South African Apartheid Litigation, 346 F Supp 2d 538 (SD NY, 2004).
 Khulumani, 504 F 3d 254, 260 (2nd Cir, 2007).
 Ibid 269.
 Rome Statute of the International Criminal Court, opened for signature 17 July 1998, 2187 UNTS 3 (entered into force 1 July 2002) (‘Rome Statute’).
 Khulumani, 504 F 3d 254, 277 (2nd Cir, 2007).
 Ibid 287–8, reasoning that ‘when international law and domestic law speak on the same doctrine, domestic courts should choose the latter’.
 American Law Institute, Restatement (Second) of Torts (1979).
 Khulumani, 504 F 3d 254, 287–9 (2nd Cir, 2007).
 Ibid 289.
 Ibid 290.
 Ibid 321–6.
 Ibid 327.
 Ibid 333.
 Anthony J Sebok, ‘More on the Second Circuit’s Recent, Significant Decision regarding Two Suits Involving the Alien Tort Claims Act: Part Two in a Two-Part Series’, FindLaw (US) 6 November 2007 <http://writ.news.findlaw.com/sebok/20071106.html> at 23 September 2008. See generally Hoffman and Zaheer, above n 48; Daniel Diskin, ‘The Historical and Modern Foundations for Aiding and Abetting Liability under the Alien Tort Statute’ (2005) 47 Arizona Law Review 805, 808–9.
 Tarek F Maassarani, ‘Four Counts of Corporate Complicity: Alternative Forms of Accomplice Liability under the Alien Tort Claims Act’ (2005) 38 New York University Journal of International Law and Politics 39, 39.
 Rome Statute, above n 146, art 25(3)(c).
 Albin Eser, ‘Individual Criminal Responsibility’ in Antonio Cassese, Paola Gaeta and John R W D Jones (eds), The Rome Statute of the International Criminal Court: A Commentary (2002) vol 1, 767, 801.
 See, eg, Prosecutor v Furundžija (Trial Chamber) Case No IT-95-17/1-T (10 December 1998) – (Judgment); Prosecutor v Aleksovski (Appeals Chamber) Case No IT-95-14/1-A (24 March 2000)  (Judgment); Prosecutor v Vasiljević (Appeals Chamber) Case No IT-98-32-A (25 February 2004)  (Judgment); Prosecutor v Bagilishema (Trial Chamber I) Case No ICTR-95-1A-T (7 June 2001)  (Judgment).
 Khulumani, 504 F 3d 254, 278 (2nd Cir, 2007) (Judge Katzmann), 337 (Judge Korman). Both judges cite Antonio Cassese, ‘The ICTY: A Living and Vital Reality’ (2004) 2 Journal of International Criminal Justice 585, 589.
 Khulumani, 504 F 3d 254, 279 (2nd Cir, 2007) (Judge Katzmann), 337 (Judge Korman).
 Ibid 274–5.
 Machteld Boot, Genocide, Crimes Against Humanity, War Crimes: Nullum Crimen Sine Lege and the Subject Matter Jurisdiction of the International Criminal Court (2002) 305.
 UN Secretary-General, Report of the Secretary-General pursuant to Paragraph 2 of Security Council Resolution 808 (1993), UN Doc S/25704 (3 May 1993)  (emphasis added).
 See, eg, Prosecutor v Furundžija (Trial Chamber) Case No IT-95-17/1-T (10 December 1998) – (Judgment).
 See Robert Cryer, ‘The Boundaries of Liability in International Criminal Law, or “Selectivity by Stealth”’ (2001) 6 Journal of Conflict and Security Law 3, 23; Kai Ambos, ‘Article 25: Individual Criminal Responsibility’ in Otto Triffterer (ed), Commentary on the Rome Statute of the International Criminal Court: Observers’ Notes, Article by Article (1999) 475, 483.
 Khulumani, 504 F 3d 254, 279 (2nd Cir, 2007) (Judge Katzmann), 337 (Judge Korman).
 Military and Paramilitary Activities in and against Nicaragua (Nicaragua v US) (Merits)  ICJ Rep 14, 93–4.
 Unocal USCA9 708; , 395 F 3d 932, 950–1, 953 (9th Cir, 2002).
 See Kevin Jon Heller, The Second Circuit’s Incorrect Interpretation of Customary International Law (2007), available from <http://www.opiniojuris.org> at 23 September 2008.
 John M Walker Jr, ‘Domestic Adjudication of International Human Rights Violations under the Alien Tort Statute’ (1997) 41 Saint Louis University Law Journal 539, 539.
 See King, above n 45, 9–10.
 Sarei v Rio Tinto, 456 F 3d 1069, 1078 (9th Cir, 2006).
 Kadic USCA2 16; , 70 F 3d 232 (2nd Cir, 1995).
 Beth Stephens, ‘Corporate Liability Before and After Sosa v Alvarez-Machain’ (2004) 56 Rutgers Law Review 995, 1001–2.
 Jeremy Waldron, ‘Vagueness in Law and Language: Some Philosophical Issues’ (1994) 82 California Law Review 509, 534.
 Jeremy Waldron, ‘Torture and Positive Law: Jurisprudence for the White House’ (2005) 105 Columbia Law Review 1681, 1699.
 Koh, above n 39, 269.