1. Introduction
As a matter of principle, states are internationally responsible only for the acts of their organs, while they cannot be called upon to answer for private misconducts.Footnote 1 Under certain circumstances, however, private wrongs may be attributed to states, thereby triggering the latter's international responsibility. It is well known that ‘secondary’ norms on attribution of conducts have been codified by the International Law Commission (ILC) in Part One, Chapter II of the 2001 Articles on the Responsibility of States for Internationally Wrongful Acts (‘the Articles’).Footnote 2 The criteria envisaged by the ILC do not need to be enumerated: with an acceptable degree of approximation, it can be said that private conducts are regarded as acts of state when they are carried out on the state's behalf or under its tight control (the so-called ‘agency paradigm’).Footnote 3
At a closer look, this legal framework appears unable to deal with state involvement in human-rights (or environmental) abuses perpetrated by powerful non-state actors, such as terrorist groups or transnational corporations (TNCs). Such wrongs, indeed, are often put in place with the fundamental contribution of – though not on behalf of (or under the control of) – a state, with the consequence that they cannot be attributed to the latter.
Against this backdrop, the present paper argues that there has been developing a new secondary norm, according to which private wrongs are to be imputed to a state if the latter knowingly facilitated (or otherwise co-operated in) their commission. In other words, it is here contended that, besides the traditional criteria enshrined in the Articles, another criterion of attribution of conduct has been forming: complicity.Footnote 4 Since, as we will see, such an emerging rule has been applied only in some fields of international law (human-rights and environment protection, fight against terrorism), it should prevail over the Articles as lex specialis in accordance with Draft Article 55.Footnote 5
Of course, state complicity in private wrongs can be (and has actually been) conceptualized in other ways. First, it can be reframed as an infringement of the so-called ‘due-diligence’ rule. Second, it may be contended that state complicity is forbidden by an autonomous primary norm. Third, complicity in private facts can be read as a criterion of attribution of responsibility, in the same way as Draft Article 16 does with reference to inter-state complicity.Footnote 6 It is my stance, however, that such approaches are legally flawed and/or normatively undesirable. Furthermore, I contend that they do not match the recent practice on the issue.
In my analysis, although I will also linger on international practice, particular attention will be paid to the US case law concerning corporate liability for international human-rights violations. For reasons that I will explain more clearly below, US courts have, in fact, been called upon to rule on the attribution to states of international wrongs perpetrated by TNCs, and have fulfilled this task by resorting to criteria that clearly recall the notion of complicity.
The choice to focus on this case law is due to several reasons. First, until now, authors dealing with issues of complicity have mainly considered international practice, while overlooking the fundamental role that can be played in this regard by the domestic case law. Second, the same scholarship has mainly taken into account practice concerning state involvement in paramilitary and terrorist activities and not that concerning TNCs’ abuses. Third, and last, although US jurisprudence on corporate human-rights violations has been studied in great depth, no author has yet assessed its impact on secondary norms on attribution.
On these premises, this paper will be divided into two parts. In the first part (section 2), I will develop some general remarks on the issue. In particular, in subsection 2.2, I will explain why the other possible ways to conceptualize state complicity in private wrongs are unsuitable. In subsection 2.3, I will provide a quick framework of the international practice from which the emergence of complicity as a criterion of attribution of conduct can be inferred. Section 3, on the other hand, will be devoted to US case law. After some introductory observations (subsection 3.1), in subsections 3.2 and 3.3, I will clarify why US courts are called upon to rule on attribution to states of TNCs’ misconducts. Subsequently, having described the criteria of attribution applied by American judges (subsection 3.4), I will focus on the so-called ‘joint-action’ test, which closely resembles the notion of complicity (subsection 3.5). In the conclusion (section 4), I will tackle some possible objections that might be raised against my reasoning.
2. General remarks
2.1. Three (unsuitable) ways to conceptualize state complicity in private facts
As mentioned in the introduction, although I will deem state complicity in private facts as a criterion of attribution of conduct, there are other ways to conceptualize it. In particular, it can be read (i) as a violation of the due-diligence rule, (ii) as a violation of an autonomous primary norm, and (iii) as a criterion of attribution of responsibility. In this paragraph, I will explain why these ways to frame state complicity in private wrongs are unsuitable.
According to the due-diligence rule, a state incurs international responsibility when it fails to take all reasonable measures to prevent or to punish unlawful conducts by non-state actors.Footnote 7 While it has developed in the context of the treatment of aliens and foreign diplomats,Footnote 8 over the years this principle has conspicuously widened its scope and nowadays finds application in all main branches of international law.Footnote 9
It is, therefore, unsurprising that scholarly works on state responsibility usually deal with this rule after listing the criteria of attribution of conduct envisaged by the Articles.Footnote 10 Yet, it would be a mistake to consider it as a rule of closure, covering every hypothesis of state involvement in private facts left out by the ILC's criteria. In fact, the due-diligence principle turns to be inadequate when the state's behaviour shifts from culpable inaction to more intense forms of co-operation.
On the one hand, indeed, since the due-diligence principle imposes upon states an obligation of conduct,Footnote 11 its infringement may only consist of an omissive demeanour. Accordingly, complicit conducts, which are often commissive (e.g., material or financial support), largely fall outside its scope of application.Footnote 12 On the other hand, this principle cannot provide an adequate legal answer to cases in which state inaction amounts to connivance with private wrongdoers, namely when the state repeatedly and knowingly fails to prevent and to punish the unlawful conducts carried out by private entities under its jurisdiction.Footnote 13
In other words, notwithstanding its wide scope of application, the due-diligence rule is unable to govern state complicity in private conducts. To fill this normative gap, two routes can be followed. It may be argued (i) that state complicity in private abuses is forbidden by an autonomous primary norm, or (ii) that state complicity in private wrongs determines the imputation of the latter to the state, in accordance with a secondary norm of attribution.
Both legal reasonings are equally acceptable on a theoretical plane,Footnote 14 but they lead to fairly different results in terms of degree of state responsibility. The infringement of a primary norm prohibiting state complicity in private wrongs, in fact, entails a lesser degree of responsibility than that arising when such wrongs are directly carried out by the state.Footnote 15 On the contrary, conceptualizing complicity as a criterion of attribution implies that the accomplice state is held responsible in the same way as if it had committed the unlawful act through its own organs.
It goes from the foregoing that the choice between the two options is to be grounded on reasons of legal policy rather than on theoretical arguments. In this connection, it can be argued that the need to guarantee a stronger protection to the fundamental values of the international community (peace, environment, human rights) should lead to hold that, at least when it comes to the infringement of such values, states must be deemed directly responsible for private activities that they have knowingly helped to bring about. Indeed, as one author noted with regard to state responsibility for terrorist activities, ‘Limiting the State's responsibility to a failure to . . . abstain [from complicit conducts] in cases where its breaches have been essential to the terrorists’ success seems to unfairly absolve the State of its full measure of responsibility’.Footnote 16
From a slightly different perspective, it may be added that the ‘primary-norm’ approach fails to grasp the current reality of the relationships between state and non-state actors. Notably, it overlooks that, nowadays, powerful non-state actors are able to commit the most heinous atrocities, thanks to the support provided by states but without operating under their control. Against this backdrop, it appears overly unjust to consider state complicity in private wrongs as a source of marginal responsibility.Footnote 17 This explains why – as we will see in the next paragraph – in recent practice regarding state involvement in terrorism and human-rights abuses, complicity has been regarded more as a criterion of attribution than as an autonomous unlawful act.
Once established that, in some areas of international law, complicity should be conceived as a criterion of attribution, a last clarification is needed. It has to be specified, in fact, whether complicity in private wrongs operates as a criterion of attribution of responsibility or as a criterion of attribution of conduct.
This distinction can be made clear, as follows. The first notion refers to the cases in which a legal subject is held responsible for acts committed by another legal subject and for which also the latter is responsible (so-called responsabilité derivée).Footnote 18 The second notion, on the other hand, concerns the conditions under which an act of an entity is deemed as the act of a legal subject, regardless of the responsibility (and thus the subjectivity) of such an entity. Although not even mentioned in the travaux préparatoires, this distinction is clearly reflected by the structure of the Articles that deal with the two hypotheses in different chapters of Part One. In particular, criteria of attribution of conduct are listed in Chapter II (‘Attribution of Conduct to a State’), while criteria of attribution of responsibility are set forth in Chapter IV (‘Responsibility of a State in Connection with the Act of Another State’).
Some scholars seem to suggest that state complicity in private conducts should operate as a criterion of attribution of responsibility. I refer to those authors that contend that Draft Article 16 (‘Aid or Assistance in the Commission of an Internationally Wrongful Act’) could be applied by analogy to the relations between state and non-state actors.Footnote 19 For the sake of clarity, it is well to recall that, according to this provision:
A State which aids or assists another State in the commission of an internationally wrongful act by the latter is internationally responsible for doing so if a) that State does so with knowledge of the circumstances of the internationally wrongful act; and b) the act would be internationally wrongful if committed by that State.
On closer inspection, an analogy to Draft Article 16 so as to allow for a state's direct responsibility in cases of assistance to private entities appears to be legally unsound. Indeed, the possibility of widening the scope of such a provision is more asserted than demonstrated. Notably, scholars propounding this view fail to consider, on the one side, that the ILC adopted a strict attitude towards the issue of attribution of private conducts and, on the other side, that Article 16, by its own terms, is expressly limited to co-operation between states.Footnote 20
Furthermore, this approach is likely to lead to unsatisfactory outcomes. As Draft Article 16 makes clear, in fact, in order to hold the accomplice state responsible for the act carried out by the main perpetrator, it is required that the latter (i) is considered an international legal subject, and (ii) is addressed by the same international norm allegedly breached by the accomplice state. Indeed, it cannot be taken for granted that both requirements are always met. It suffices to recall, in this regard, that the international subjectivity of private corporations has been recently questioned in both domesticFootnote 21 and internationalFootnote 22 case law. Accordingly, if we frame complicity in private conducts as a criterion of attribution of responsibility, there is a serious risk of making this criterion largely unworkable.
In light of the foregoing, it can be assumed that state complicity in private wrongs threatening the fundamental values of the international community should be conceptualized as a criterion of attribution of conduct. Now, all that remains is to verify whether such a view finds support in international practice.
2.2. Moving towards complicity as a criterion of attribution of private conducts: an overview
By way of premise, it should be stressed that, as a matter of principle, the ILC was not contrary to the inclusion of complicity as a criterion of attribution of private conducts. In his Fourth Report, Special Rapporteur Ago admitted – at least theoretically – that states and non-state actors may be accomplices in the commission of internationally wrongful acts, distinguishing this situation from both the violation of due-diligence obligation and the other criteria of attribution.Footnote 23 Although Ago ultimately set aside this criterion, assuming it had never been applied in practice,Footnote 24 the concept of complicity has occasionally emerged in the ILC's subsequent debates on the notion of de facto organ.Footnote 25 Moreover, even if the ILC eventually adopted the restrictive approach propounded by Special Rapporteur Crawford, it has been suggested that the criterion of attribution referred to in Draft Article 11 (‘Conduct Acknowledged and Adopted by a State as Its Own’) can be read as envisaging a sort of ex post facto complicity.Footnote 26
Before embarking on the analysis of recent practice, it is worth noting that, despite the ILC's contention, the resort to complicity as a criterion of attribution of conduct can be found in two old arbitral decisions, both taken into account by Ago in his Fourth Report: the Poggioli award delivered by the Italian–Venezuelan Commissions in 1903Footnote 27 and the Janes award rendered by the Mexico–United States General Claims Commission in 1926.Footnote 28
In Poggioli, one of the claims concerned the failure by the Venezuelan authorities to apprehend and punish the four men who had attempted to kill an Italian citizen, Silvio Poggioli. According to the complainants, this was due to the connivance of the local authorities with the culprits (evidenced inter alia by the fact that the former notified the latter when they were in danger of being arrested). Dealing with Venezuela's responsibility for these events, Umpire Ralston expressly resorted to complicity as a criterion of attribution of conduct:
when the authorities of the State of Los Andes have acted in apparent conjunction with criminals, and have with them and under the circumstances heretofore detailed joined in the commission of offences against private individuals, and no one has been punished therefore and no attempt made to insure punishment, the act has become in a legal sense the act of the government itself.Footnote 29
The Janes case is less clear in this regard, but it is nonetheless noteworthy. On that occasion, the Mexican state had been accused of not having taken proper steps to apprehend the man who had murdered an American citizen. In particular, the United States contended that Mexico's inaction amounted to complicity in such a crime. The Commission did not endorse this view, holding Mexico responsible only for the infringement of the due-diligence rule. However, it deemed it necessary to underline that ‘[a] reasoning based on presumed complicity may have some sound foundation in cases of non-prevention where a Government knows of an intended injurious crime, might have averted it, but for some reason constituting its liability did not’.Footnote 30 In this way, it is evident, that the Commission, while considering the notion of complicity to the case as not applicable, somehow left the door open for other cases where this notion could be relevant.
In recent times, the emergence of complicity as a criterion of attribution of conduct has found support in international case law on human rights as well as in the practice of states and international organizations concerning the fight against terrorism. Since this issue has been already dealt with by other authors,Footnote 31 I will limit myself to sketching some brief remarks.
With regard to international case law, some interesting points emerge from the case law of the International Criminal Tribunal for the Former Yugoslavia (ICTY), the Inter-American Court on Human Rights (IACHR), the European Court on Human Rights (ECtHR), and the African Commission on Human Rights (AfCHR), as well as – although with some reservations – the International Court of Justice (ICJ).
In the first place, it has been carefully noted that, in the post-Tadić case law, the ‘overall-control’ testFootnote 32 has been applied by the ICTY so as to encompass forms of co-ordination between states and paramilitary groups that lack hierarchical nature.Footnote 33 This appears very clearly in the judgement rendered by the ICTY Trial Chamber in the Lašva Valley case.Footnote 34 On that occasion, in fact, the Court contended that a paramilitary group acts under the ‘overall control’ of a state when the latter (i) provides it with ‘financial and training assistance, military equipment and operational support’; and (ii) participates ‘in the organisation, co-ordination or planning of [its] military operations’.Footnote 35 It is evident how, in this test, there is no reference to a hierarchical relationship: in this way, the notion of ‘control’ is expanded well beyond its logical limits, transmuting into a form of complicity.
The IACHR followed the same path in some judgments regarding paramilitary activities in Colombia. In these rulings, in fact, the Court, clearly relying on the notion of complicity, held the Colombian government responsible ‘for the violations committed by paramilitary groups who have acted with the support, acquiescence, involvement, and cooperation of State security forces’.Footnote 36
The concept of complicity has also been employed by the ECtHR in Ilascu v. Moldova and Russia.Footnote 37 In that case, the applicants were complaining of violations that had been mainly perpetrated by the authorities of the breakaway government of Transdniestria – an entity that controls part of the Moldovan territory with the military support of the Russian Federation and whose international subjectivity is generally denied. The Court attributed such abuses to Russia in light of the following reasoning:
the events which gave rise to the responsibility of the Russian Federation must be considered to include not only the acts in which the agents of that State participated, like the applicants’ arrest and detention, but also their transfer into the hands of the Transdniestrian police and regime, and the subsequent ill-treatment inflicted on them by those police, since in acting in that way the agents of the Russian Federation were fully aware that they were handing them over to an illegal and unconstitutional regime. In addition, regard being had to the acts the applicants were accused of, the agents of the Russian Government knew, or at least should have known, the fate which awaited them.Footnote 38
Thus, even in this case, the human-rights abuses perpetrated by a non-state actor (the Transdniestrian regime) were imputed to a state (the Russian Federation) because the latter knowingly facilitated their commission.Footnote 39
Finally, it is worth mentioning the communication delivered by the AfCHR in Social and Economic Rights Action Centre (SERAC) and Another v. Nigeria (2001).Footnote 40 Discussing Nigerian responsibility for environmental devastations carried out by oil companies, in fact, the African Commission seemingly resorted to complicity as a criterion of attribution of conduct when it noted that:
the government of Nigeria facilitated the destruction of the Ogoniland. Contrary to its Charter obligations and despite such internationally established principles, the Nigerian government has given the green light to private actors, and the oil companies in particular, to devastatingly affect the well-being of the Ogonis.Footnote 41
Against this backdrop, the ICJ appears to be somehow the black sheep. In both the Nicaragua Footnote 42and the Bosnian Genocide Footnote 43 cases, indeed, it showed a restrictive attitude towards the issue of attribution.Footnote 44 Actually, both these judgments are not bereft of ambiguities. In Nicaragua, in fact, while strictly applying the ‘effective-control’ test in order to rule out direct US responsibility for Contras's activities, the Court seems to have employed a rather different test when dealing with the acts of the Unilaterally Controlled Latino Assets (UCLA). The ICJ, indeed, by resorting to an (embryonic) notion of complicity, attributed them to the United States in the light of the fact that ‘agents of the United States participated in the planning, direction, support and execution of [UCLA's] operations’.Footnote 45
Even more significantly, in the Bosnian Genocide case, the ICJ wondered whether the Serbian state could have been condemned for complicity with the Army of the Serbian Republic in the commission of the Srebrenica genocide.Footnote 46 Admittedly, in that case, the Court was authorized to carry out such inquiry by Article III(e) of the Genocide Convention, which expressly forbids ‘complicity in genocide’. Nevertheless, it has been convincingly argued that the ICJ has regarded ‘complicity’ more as a criterion of attribution than as a wrong envisaged by an autonomous primary norm.Footnote 47
As far as terrorist acts are concerned, it has been noted that, since the end of the Cold War, a principle has been emerging according to which states supporting or harbouring terrorist groups are internationally responsible for terrorist acts, independently of the degree of their involvement in the specific act perpetrated.Footnote 48 This principle has been endorsed by UN organs and finds support in state practice as well.Footnote 49
The most important (and dramatic) example is represented by the US reaction after 9/11. As is well known, in fact, the United States regarded the Taliban regime as directly responsible for the terrorist attack on the Twin Towers, notwithstanding such an attack had been actually carried out by al Qaeda – that is, by a private entity clearly distinguishable from the Afghanistan government. Such an imputation, which was not disputed by the greater part of the international community, has been justified by the US government in light of the fact that the Taliban knowingly harboured the agents of al Qaeda, namely because the former were the latter's accomplices.Footnote 50
Therefore, it seems arguable that customary international norms on attribution are developing so as to consider complicity as a criterion to impute private facts to states, at least when it comes to conducts that threaten the fundamental values of the international community (peace, human rights, environment).Footnote 51 In the next part, I will try to show that such a view also finds support in US case law on corporate liability for human-rights abuses.
3. The attribution to states of corporate misconducts in US case law
3.1. The Alien Tort Statute and its applicability to corporate human-rights violations
The US judicial system is characterized by the most conspicuous case law concerning the liability of private companies for violations of internationally recognized human rights.Footnote 52 This feature is due to several factors,Footnote 53 the most important of which is the presence in the US legal order of a juridical tool that is, for many reasons, unique: the Alien Tort Statute (ATS).Footnote 54
As is known, the ATS is an ancient US statute (it dates back to 1789) granting jurisdiction of federal courts over civil actions brought by aliens for torts ‘committed in violation of the law of nations or a treaty of the United States’.Footnote 55 Despite being one of the first statutes adopted by Congress, the ATS was successfully invoked for the first time only in 1980 (Filàrtiga case) when the Second Circuit ruled that customary international law on human rights is to be considered ‘law of nations’ under the ATS and thus its infringement is able to trigger the jurisdiction of federal courts.Footnote 56 Since it was generally understood that international law addressed only states, in the years following Filàrtiga, the ATS was invoked against former foreign officials with regard to acts perpetrated in the exercise of public functions.Footnote 57
However, in 1995 (Kadic case), the Second Circuit went further and affirmed its jurisdiction also over non-state actors (in that case, the leader of a non-recognized state).Footnote 58 This evolution opened up the door to lawsuits against private companies. Only two years later, in fact, a Californian District Court, confronted with a civil action concerning the abuses committed by a mining corporation in Myanmar, stated that ATS jurisdiction covers not only natural, but legal persons as well (Unocal case).Footnote 59 Since Unocal, several multinational enterprises have been brought before federal courts to answer for abuses allegedly perpetrated in countries where they carry out their activities.Footnote 60
The possibility of invoking the ATS against corporations has been recently questioned by the Second Circuit in 2010 (Kiobel case).Footnote 61 On that occasion, it was held that legal persons are not subjects of international law and cannot infringe customary international law, as evidenced by the fact that no international criminal tribunal has jurisdiction over them. It is not possible here to embark upon a critical analysis of this controversial ruling. For our purposes, it suffices to note that the line of cases that I will take into account is not affected by the Kiobel precedent, since they are not grounded on the international legal personality of transnational corporations.
3.2. Holding corporations liable for the violation of state-addressed norms: the ‘state-action’ requirement
The applicability of the ATS to wrongs committed by private actors has been justified by federal courts on a double basis. In the first place, it has been recognized that international law imposes duties directly on individuals.Footnote 62 Yet, since this can be affirmed only with regard to norms forbidding the most egregious violations of human rights (e.g., genocide, forced labour, or war crimes), courts have worked out ways to link private facts to states so as to hold the former responsible for the infringement of norms that typically address the latter.
Generally speaking, two routes have been followed by US judges. Some courts have accepted that private parties may be sued for the violation of state-addressed norms if they aided and abetted a state that directly infringed those norms.Footnote 63 Even though the concept of ‘aiding and abetting’ is clearly contiguous to (if not synonymic with) that of ‘complicity’,Footnote 64 this case law does not serve our purposes and will not be dealt with in this paper. Courts adopting this approach, in fact, do not rule upon the attribution to states of private conducts. On the contrary, they consider as given that a state perpetrated a human-rights abuse and limit themselves to considering whether the corporate respondent aided and abetted the state in committing it.
Other courts, on the other hand, have maintained that private actors may be deemed liable for the violation of state-addressed norms when their conduct can be regarded as state conduct (the so-called ‘state-action’ requirement).Footnote 65 In order to assess compliance with this requirement, US courts have looked at the jurisprudence relating to a domestic provision: 42 USC 1983. This section was introduced by the Civil Rights Act of 1871 and provides redress for constitutional (or federal) violations perpetrated under the colour of law – that is, committed by actors clothed with state or federal authority.Footnote 66 Although section 1983 was originally deemed to be applicable only against state officials, US courts have subsequently recognized that there are circumstances in which private entities also may be regarded as acting under the colour of law.Footnote 67
The same principle has been applied, by analogy, in ATS cases in order to hold private entities responsible for the infringement of international norms that only address states.Footnote 68 This approach is of striking interest for our purposes. Indeed, as I will try to demonstrate, when dealing with the ‘state-action’ requirement, US courts are confronted with the attribution to states of private conducts.
3.3. ‘State-action’ jurisprudence and international norms on attribution
International norms on attribution are based on the premise that states, like every legal person, necessarily act (and may do wrongs) through individuals. Accordingly, they identify the cases in which the wrongful conduct of an individual can be referred to a state, so determining the latter's international responsibility.Footnote 69
The US ‘state-action’ jurisprudence performs a function that is, from certain aspects, very similar. As I have recalled, section 1983 provides whoever has suffered the violation of a civil right with a cause of action against the wrongdoer. Under US constitutional law, however, civil rights may not be infringed by private parties, but only by federal and state authorities. Section 1983, therefore, requires that the wrong committed by the defendant is regarded as an act of a state, namely that his (or her) conduct is ‘fairly attributable to the State’.Footnote 70 Against this backdrop, the ‘state-action’ canons – like international norms on attribution – indicate when and under which conditions such an imputation is possible.
Thus, it seems at least arguable that the ‘state-action’ jurisprudence constitutes the US domestic counterpart of the international norms on attribution.Footnote 71 Accordingly, when US courts resort to such case law in order to link TNCs’ international wrongs to states, they are actually attributing the former to the latter under international law.Footnote 72 While the familiarity between these two sets of norms has never been admitted by US courts, a clear indication in this sense is provided by the judgment rendered by the Eastern District Court of the Louisiana in Beanal v. Freeport-McMoran.Footnote 73 In its ruling, in fact, the court made reference both to domestic jurisprudence and to international-law principles in order to assess whether the ‘state-action’ requirement had been met.Footnote 74
One might wonder, at this juncture, why US courts rely on section 1983 case law instead of referring exclusively to international principles on state responsibility. Arguably, this choice stems from the fact that US courts consider such principles inadequate to deal with ATS corporate cases. As we have seen above, in fact, traditional international norms on attribution require that states wield a tight control on private actions, while, in most of the ATS cases, plaintiffs contend that TNCs and government officials co-operated at an equal level in the perpetration of the alleged human-rights violations. As we will see, ‘state-action’ canons have proved much more useful, since they provide courts with criteria of attribution that are definitely more comprehensive.Footnote 75
3.4. The ‘state-action’ tests
US courts employ three basic tests to determine ‘state action’: the ‘public-function’ test, the ‘state-compulsion’ test, and the ‘joint-action’ test.Footnote 76
According to the ‘public-function’ test, ‘a private party who performs a function that has traditionally been the exclusive prerogative of the State may be deemed a state actor’.Footnote 77 The ‘state-compulsion’ test, on the other hand, requires that ‘the State creates, coerces, or significantly encourages the challenged activity to the extent that the decision is deemed to be the State's’.Footnote 78 Under ‘joint action’, finally, state action is present ‘where there is a substantial degree of cooperative action between the state and private actors in effecting the deprivation of rights’.Footnote 79
The first two tests find clear matches in the Articles. The ‘public-function’ test, in fact, presents more than one analogy with the criterion of attribution envisaged by Draft Article 5 (‘Conduct of Persons or Entities Exercising Elements of Governmental Authority’), while the ‘state compulsion’ test, despite some differences in the formula employed, closely recalls the ‘effective-control’ test under Draft Article 8.Footnote 80 The application of these tests in order to impute private facts to states, therefore, does not innovate traditional standards on attribution.
The same cannot be said with reference to the ‘joint-action’ test. The formula ‘substantial degree of cooperative action’, indeed, expresses the idea, almost unknown to the Articles, according to which it is possible to attribute to states acts committed by private entities not placed under state control. For this reason, I will linger only on this test.
3.5. The ‘joint-action’ test and the notion of ‘complicity’
Like the whole jurisprudence relating to Section 1983, the ‘joint-action’ case law is not ‘a model of consistency’Footnote 81 and its formula has been used to encompass very different forms of co-operation. Nevertheless, at least as far as ATS corporate cases are concerned, some general remarks could be made.Footnote 82
A first observation is that US courts did not consider the ‘joint-action’ requirement to be met when the host state failed to prevent (or to punish) the unlawful acts committed by private actors.Footnote 83 In Aldana v. Del Monte,Footnote 84 for instance, the Eleventh Circuit held that police inaction in response to crimes committed by private security forces hired by Del Monte was not enough to find state action. In the court's view, it was crucial that such an omissive behaviour, though culpable (as the crimes occurred very near to the police station), was not intentionally aimed to assist the private wrongdoers.Footnote 85 This approach is perfectly in line with the remarks we put forward in subsection 2.1, since it confirms that the mere lack of due diligence does not determine the attribution to states of private wrongs but, at the same time, clarifies that things stand quite differently when state officials omit to intervene with the purpose of facilitating abuses by non-state actors.
In the second place, it is worth highlighting that the state's failure to counter harmful corporate activities abroad has never been taken into account by US courts in order to establish ‘state action’. In this hypothesis, indeed, state involvement in corporate abuses is too loose to justify attribution, even on the ground of complicity. This is not to say, however, that the home state cannot be held responsible on other bases. As some authors argue, under certain circumstances, international law imposes on states an extraterritorial duty to monitor and govern their corporate nationals.Footnote 86 This duty particularly arises whenever the home state is unable or unwilling to exercise effective control on corporate activities (e.g., because it is a failed state or is governed by a repressive regime).Footnote 87
Turning to the cases in which state action was held to be present, a distinction should be drawn. In a line of cases, state action has been found under the ‘joint-action’ test because state agents directly committed the alleged violations. This happens, for instance, when corporations have paid public authorities to obtain protection and the latter have committed abuses. In Mujica v. Occidental Petroleum,Footnote 88 the court deemed the state action requirement to be met because the wrongs that Occidental Petroleum was accused of had in fact been carried out by state agents (the Colombian Air Force), who were on the defendant's payroll.Footnote 89 Since, in these cases, the unlawful acts are directly carried out by state agents, their attribution to states is unproblematic and the same outcome would have been reached by resorting to the Articles.
In other cases, courts held ‘state action’ to be present even though state agents did not take part directly in the alleged violations. So far, this has happened in two circumstances.
On the one hand, private human-rights violations have been regarded as state actions when a state knowingly facilitated their commission. In Abdullahi v. Pfizer,Footnote 90 for instance, a well-known pharmaceutical company was accused of having tested an experimental drug on Nigerian children without seeking their parents’ informed consent, so causing their death or serious diseases. The Second Circuit attributed these abuses to the Nigerian state, arguing that the Nigerian government (i) provided a letter of request to the US Food and Drug Administration (FDA) to authorize the export of the drugs incriminated, (ii) granted the exclusive use of two hospital wards to Pfizer as well as the use of the hospital's staff and facilities to conduct the tests, (iii) silenced Nigerian physicians critical of the test, and (iv) back-dated an ‘approval letter’ that the FDA and international protocol required to be provided prior to conducting the medical experiment.Footnote 91
Similarly, in Arias v. Dyncorp,Footnote 92the Columbia District Court accepted that the spraying of fumigants by Dyncorp onto cocaine and heroin poppy plantations in Colombia that caused unlawful damages also to Ecuadorian plantations could be attributed to both the United States and Colombia, since (i) the corporation was contracted with the US State Department to engage in aerial spraying of cocaine and heroin fields in Colombia, (ii) the compensation for this task was to come from funds approved by Congress under Plan Colombia, and (iii) Dyncorp acted in co-ordination with the Colombian and US governments.Footnote 93
It is clear that none of these conducts can be deemed as a direct participation by the states in the abuses committed, nor do they constitute evidence that the respondent corporations acted as state organs. The basis of such an attribution, instead, has been found in the practical (in the case of Nigeria and Colombia) or financial (in the case of the United States) assistance provided by the states to the perpetrators – in other words, in their ‘complicity’.Footnote 94
On the other hand, private abuses have been imputed to states when they were part and parcel of a more general plan set up by the private entity together with state officials. In Wiwa v. Royal Dutch Petroleum,Footnote 95for instance, the claimants alleged that Royal Dutch met with Nigerian officials in England and the Netherlands to organize a joint action to suppress the Ogoni rebels. In pursuing this plan, both Royal Dutch and Nigerian agents perpetrated several human-rights abuses. During the trial, the corporate defendant argued that state action could not have been found with regard to the violations, which were not committed under state control. The District Court for the Southern District of New York rejected this argument, maintaining that, for the purposes of attribution, it was enough that state and non-state actors conspired to commit those violations, while it was not required that they ‘acted in concert to commit each specific act that violates plaintiffs’ rights’.Footnote 96
Therefore, private facts have been regarded as state actions, since they have been carried out in furtherance of a plan arranged along with state officials, regardless of whether each act had been put in place under state agents’ direction or control, namely because of the complicit conspiracy between state and non-state actors.
4. Conclusions
Traditional norms on attribution are currently questioned by the convergence of two factors. First, the hierarchical paradigm enshrined by classical criteria of attribution of conduct is made obsolete by the fact that well-organized private actors are able to co-operate with states in the commission of international wrongs without being absorbed in their structure. Second, the growing attention paid by the international community to the protection of fundamental values like peace, environment, and human rights calls for the expansion of the number of cases in which states may be held liable for private activities threatening such values.Footnote 97
In this connection, it has been suggested that a customary norm is developing according to which states are directly responsible for complicity in private wrongs when the latter affect the fundamental values of the international community. As we have seen, however, authors advancing such a thesis have mainly looked at international practice concerning state involvement in paramilitary and terrorist activities.
In the present paper, I have tried to demonstrate that the emergence of a customary norm envisaging complicity as a criterion of attribution of conduct finds support also in US case law on corporate human-rights abuses. In particular, in the case law considered above, this criterion assumes two declinations. According to the first one, a private abuse is attributed to a state when the latter knowingly facilitated its commission (complicity stricto sensu). Under the second one, private facts are to be regarded as state actions when they have been carried out in furtherance of a plan arranged along with state officials, regardless of whether each act had been put in place under state agents’ direction or control (complicit conspiracy).
Against the line of reasoning I have developed so far, (at least) two objections might be raised. First, it may be said that US case law is unlikely to be followed by other courts (either domestic or international), which, of course, cannot make reference, in their judgments, to a US domestic provision such as Section 1983. Second, one could claim that the ATS corporate cases have nothing to do with the attribution to states of wrongful acts, since they concern only the liability of private corporations and not that of states (which, in any case, would benefit from state immunity).
As to the first objection, it is worth premising that the judicial application of the Section 1983 jurisprudence to the ATS cases does not derive from confusion between domestic and international norms. Rather, US courts apply domestic-law principles by analogy in order to fill a perceived gap in the international norms on attribution. Granted, the use of a domestic provision does not encourage the adoption of the same approach by other courts.Footnote 98 However, it is worth noticing, on the one hand, that some courts dealt with the ‘state-action’ requirement without resorting to ‘Section 1983’ jurisprudence. This happened, for instance, in Iwanowa v. Ford, in which a New Jersey district court analysed the relationship between the corporate defendant and the Nazi regime by adopting autonomous criteria of attribution.Footnote 99 On the other hand, also courts applying ‘Section 1983’ case law are becoming aware of the need to contextualize such an approach in a more appropriate international-law framework. In Sinaltrainal v. Coca-Cola, for instance, the Florida Southern District Court invoked the intervention of the Eleventh Circuit in order to clarify ‘the extent to which section 1983 jurisprudence constitutes an established norm of international law’.Footnote 100
As to the second point, it suffices to note that it is not the first time that the development of international norms on attribution has been propelled by judgments not strictly related to state responsibility. In Tadić, as is known, the ICTY questioned the ‘effective-control’ test worked out by the ICJ.Footnote 101 Yet, the ICTY did not reach this legal finding in order to assess the international responsibility of a state. On the contrary, the legal question was whether the Bosnian war was, for the purposes of the application of the Geneva Conventions, an international conflict. To this end, the court assessed whether the army of the Serbian Republic in Bosnia, a non-state actor, could have been considered as a de facto organ of the Federal Republic of Yugoslavia.Footnote 102 Similarly, in the cases analysed above, US courts had to find whether the alleged human-rights violations were attributable to a state for the purposes of the application of the ATS. The fact that such judgments do not end up with a finding of state responsibility does not exclude that US courts are actually dealing with issues of attribution.
