The European Journal of International Law Vol. 22 no. 3 © EJIL 2011; all rights reserved
Immunities of State Officials, International Crimes, and Foreign Domestic Courts: A Reply to Dapo Akande and Sangeeta Shah
Alexander Orakhelashvili*
1 Introduction
The present contribution will not provide yet another analysis of the law of immunities in relation to international crimes; this has been done elsewhere.1 It is instead a response to certain views put forward by Dapo Akande and Sangeeta Shah.2 Akande and Shah disagree with my own conclusion that jus cogens can, and does, prevail over state immunity. They however advance an alternative approach favouring the denial of immunity, and their conclusion as to the lifting of immunity in civil proceedings manifests that the disagreement is not as wide as it could seem.3 The aim of this contribution is to clarify whether, in attacking my views, Akande and Shah have moved the debate forward, or made an original case against the primacy of jus cogens. The following analysis will demonstrate that these objections to the primacy of jus cogens over immunities rely only on factors and evidence that support the conclusions reached in that contribution, disregard the evidence that would stand in their way, and ascribe to some authorities the impact they have never been intended to produce.
LLM cum laude (Leiden); PhD (Cantab.); Lecturer, School of Law, University of Birmingham. Email: orakhelashvili@hotmail.com.
A. Orakhelashvili, Peremptory Norms (2006); Orakhelashvili, 'State Immunity and Hierarchy of Norms: Why the House of Lords Got it Wrong', 17 EJIL (2007) 915; Orakhelashvili, 'State Immunity and International Public Order Revisited', 49 German Yrbk Int'l L (2006) 32 7.
Akande and Shah, 'Immunities of State Officials, International Crimes and Foreign Domestic Courts', 21 EJIL (2011) 815. Ibid., at 839, 852.
EJIL (2011), Vol. 22 No. 3, 849-855
doi: 10.1093/ejil/chr065
850    EJIL 22 (2011), 849-855
2 The Scope of Immunities and Jus Cogens
This writer agrees that, in terms of restrictive immunity theory, whether or not an act is jure imperii or sovereign does not depend on its legality, but on whether it is 'intrinsically governmental'.4 But it is then argued that if the relevant acts are associated with the policies of the state, and carried out using state apparatus, they should be considered official acts. The approach that 'international crimes may never be regarded as official acts' is consequently opposed.5
This conclusion, one is bound to emphasize, is straightforwardly false. Instead, an act is 'intrinsically governmental' - 'intrinsic' mattering here just as 'governmental' does - when it is performed as part of the governmental authority that only a government can exercise. The purpose of restrictive immunity is not to protect everything that the state does by using its resources or furthering its policies. Akande and Shah argue, pursuant to the House of Lords' approach in Jones, that immunity applies to 'acts of the state for the purpose of imputing state responsibility'.6 But this reveals a structural confusion. State responsibility is about what the state has done as a factual matter; state immunity is about how what the state has done should be characterized. If one accords immunity to acts because they are imputable to states, the restrictive immunity doctrine will collapse, because anything that is imputable to states, including commercial and other non-sovereign acts, would be immunized.
The House of Lords in I Congreso held that state conduct is not a sovereign act and attracts no immunity, even if performed to further state policies, if it could be performed by any private actor, and the state invokes no governmental authority, even if it relates to a highly contingent political context.7 Thus, the category of acta hire imperii encompasses a narrow category of acts inherent to the sovereign authority of a state.8 This fundamental test has been followed in later jurisprudence.9 In Arrest Warrant, the Joint Separate Opinion also made it clear that the scope of official acts does not shield international crimes, because they are outside state functions; State-related motives, or the use of the power invested in the state, are not the proper test for determining what constitutes public state acts.10 The House of Lords in Pinochet held that acts of torture, hostage-taking, and crimes against humanity cannot be official functions of a public official,11 not least due to the jus cogens status of these crimes.12
4 Ibid., at 830-831.
5 Ibid., at 8 3 0-8 3 2; in ibid., at 8 31, Akande and Shah refer to Saudi Arabia v Nelson. But Nelson relates to the commercial activities exception under the 1976 FSIA, 100 ILR 551-554, not the qualification of relevant acts under international law.
6 Akande and Shah, supra note 2, at 832.
7 I Congreso [1983] 1 AC 268 (HL).
8 R. Higgins, Problems and Process (1996), at 84.
9 Lord Millett in Holland v. Lampen-Wolfe [2000] 3 AU ER 845.
10 Joint Separate Opinion of Judges Higgins, Koojmans, andBuergenthal, Arrest Warrant [2002] ICJ Rep 63, at paras 74 and 85.
11 LordNicholls in Pinochet [1998] 4 AUER 897, at 939-940; Lord Steyn, in ibid., at 945-946; LordHutton in Pinochet [1999] 2 AUER 97, at 165-166; Lord Millett, in ibid, at 179.
12 Lord Browne-Wilkinson, in ibid., at 113-114.
Immunities of State Officials, International Crimes, and Foreign Domestic Courts 851
A similar result was reached by the US Court of Appeals, refusing to immunize torture, killing, and disappearance performed by, under the direction of, or in connivance with a head of state, with the systematic use of state machinery, because no public official can claim these as his functions.13 The result is the same in civil and criminal cases.
Akande and Shah argue that the lifting of immunities cannot be defended on the basis of the 'implied waiver' doctrine, addressed by US courts in the context of the Foreign State Immunities Act (FSIA) as a domestic statute.14 Waiver is an act of will; even if implied, it refers to the will and position of the state.15 When murdering, torturing, or imprisoning arbitrarily, the state does not express any will and legal qualification as to the legal nature of these acts, let alone waive any privilege; it just commits them. Whether or not 'implied waiver' works in the US law is without prejudice to the scope of immunities under international law.
3 Immunities, Jus Cogens and Extraterritorial Jurisdiction
While objecting to the primacy of jus cogens over immunities, Akande and Shah advance three arguments. Their first argument suggests that not all humanitarian rules are peremptory.16 This argument is not crucial - if the relevant rule is not peremptory, no case for its primacy over immunities arises.
The second argument claims that, for immunity to come into conflict with jus cogens norms prohibiting international crimes, there has to be an obligation on third states to prosecute the crime in their domestic courts (or in civil cases to provide a civil remedy), and this obligation itself must be part of jus cogens. It is then argued that '[s]econdary norms which emerge as a consequence of violations of norms of jus cogens are not themselves necessarily of overriding effect'.17
This view constitutes a (frequently repeated) fallacy. It portrays peremptory norms as mere aspirations regarding substantive conduct. But then, all implications of jus cogens, whether the voidness of treaties under Article 53 of the 1969 Vienna Convention, the duty not to recognize situations created through a breach of jus cogens under the ILC's Articles 40 and 41 on state responsibility, the impact on waiver or acquiescence by states to wrongful acts, or in other areas, relate to what happens after a wrongful act.18 By and large, thus, jus cogens is about effects and consequences as much as it is about substantive prohibitions. There is no reason why it should be otherwise in relation to jurisdiction and immunities. This outcome becomes even more compelling as jus cogens norms relating to international crimes do not just prohibit the relevant
13 Hilao v. Marcos, US Court of Appeals (Ninth Circuit), 104 ILR 119, at 122-125; these violations were 'as adjudicable and redressable as would be a dictator's act of rape'.
14 Akande and Shah, supra note 2, at 828-830 (referring to cases like Princz and Siderman).
15 On waivers see Orakhelashvili, Peremptory Norms, supra note 1, at Ch 11; A. Orakhelashvili, Interpretation (2008), at Ch. 13.
16 Akande and Shah, supra note 2, at 833-834.
17 Ibid., at 836-837.
18 See the general analysis in Ch 3 of Orakhelashvili, Peremptory Norms, supra note 15.
852    EJIL 22 (2011), 849-855
conduct but also criminalize it with peremptory effect. Once the criminality of conduct is part of jus cogens, so are the rules regarding prosecution.19 As for civil cases, allowing immunity claims for breaches of jus cogens makes the relevant substantive jus cogens norm inoperative as a norm. When an ordinary rule of international law is breached, the injured entity can demand remedies; when a peremptory rule is breached, the injured entity can demand remedies even if that contradicts another legal rule. In either case, the entitlement to demand remedies before a court that has jurisdiction is indispensable if the breached rule is to operate as a legal rule. Preventing, through immunity, the injured entity from claiming remedies for the breach of jus cogens is therefore substantially more than erecting a procedural bar - it is essentially a denial of the normative status of the substantive rule that has been violated. An abstractly valid prescription that cannot produce legal effect in relation to violation is simply not a legal rule. The distinction between substantive and secondary norms is not conceptually and normatively feasible. It remains a much-repeated misconception.
In addition, immunities cannot stand in the face of the duty of states not to recognize situations created through violations of jus cogens.20 The grant of immunity before foreign courts validates precisely the situation created through such violation, by attaching to it a sovereign privilege available to states in relation to jure imperii acts proper.
Akande and Shah then argue that violations of jus cogens 'do not automatically confer jurisdiction on international courts'.21 But the use of the DRC v. Rwanda decision, which dealt with international tribunals when they have no jurisdiction in the matter in the first place, for justifying immunities for violations of jus cogens before national courts facing immunity arguments in the context of their pre-existing jurisdiction is problematic. These are two different issues having little to do with each other.
One principal point Akande and Shah advance is that, for immunity to be denied in criminal (and presumably also civil) proceedings, there has to be a rule of international law establishing extra-territorial jurisdiction over the relevant wrongful act.22 This argument attempts to introduce an extra element into the normative hierarchy: rule X cannot prevail over rule Y only unless rule Z additionally provides for jurisdiction over the breach of rule X. This conclusion is somewhat at odds with the conclusion Akande and Shah reach regarding the impact of jus cogens over immunities: as we saw above, they deny the primacy of jus cogens unless jurisdiction established over the breach is also part of jus cogens; while here they admit that jurisdiction can, by itself, prevail over immunities, even if its peremptory status is not demonstrated.
Furthermore, as was clarified in Arrest Warrant, jurisdiction is separate from, and antecedent to, immunities.23 Jurisdiction is about whether a national court can judge
19 Evidence is extensively discussed in ibid., at Ch 9.
20 [2001] ILC Rep, A/56/10 (Arts 40-41).
21 Akande and Shah, supra note 2, at 834-835 (referring to East Timor [1995] ICJRep 102; Armed Activities (DRC v. Rwan(fa)[2006]ICJ Rep 4. For analysis of this decision see Orakhelashvili, 'Case Review on DRC v. Rwanda', 55 Int'l CompLQ (2006) 753).
22 Ibid., at 817, 839ff.
23 Supra note 10, Judgment, at para. 59.
Immunities of State Officials, International Crimes, and Foreign Domestic Courts 853
the case in the first place. Immunities are about two qualitatively distinct questions as to the identity of the respondent and the relationship of the act impugned to the respondent's sovereign powers. If extraterritorial jurisdiction of a national court were established over the relevant act, yet the defendant succeeded in demonstrating the grounds on which it has immunity, the court would be compelled to decline the exercise of jurisdiction in that particular case. In relation to other defendants, jurisdiction would continue operating. Immunity would not deprive jurisdiction of all practical meaning.
Akande and Shah suggest that, in Arrest Warrant, the Court focused only on the immunity ratione personae available to serving senior state officials. As for immunity ratione materiae, they suggest that where a subsequent jurisdictional rule is practically co-extensive with a prior rule according immunity, that is where both rules apply 'in large measure' to the same set of circumstances, there will be a conflict between the two; and that 'where extra-territorial jurisdiction exists in respect of an international crime and the rule providing for jurisdiction expressly contemplates prosecution of crimes committed in an official capacity, immunity ratione materiae cannot logically co-exist with such a conferment of jurisdiction'.24
However, Arrest Warrant is constrained to its context: by applying its reasoning to immunities ratione personae, the International Court should not be seen as rejecting the same reasoning in relation to immunities ratione materiae. That was simply outside its judicial task in that case. The Court specified in general terms that treaty-based jurisdiction does not displace immunities,25 and it is difficult to see how it approved that it does displace immunities ratione materiae. Furthermore, much as the Lords in Pinochet mentioned jurisdictional arrangements under the 1984 Torture Convention, this case was about more than that; it was also about torture falling outside the sovereign powers of the state; and about the effect of jus cogens to which a number of the Lords alluded. It was less about jurisdiction displacing immunities; it was more about exercising treaty-based jurisdiction over the crime that by its nature was not covered by immunities. Last but not least, both the ILC Analytical Group and Special Rapporteur on immunities were quite sceptical regarding the immunities being displaced by extraterritorial jurisdiction.26 The 2005 Institut de Droit International (IDI) Resolution did not exclude immunities for crimes falling under universal jurisdiction either.27 Therefore, the argument of Akande and Shah does not get round the range of relevant opinions and objections.
The 2009 Naples Resolution of the Institute of International Law on the Immunity from Jurisdiction of the State and of Persons Who Act on Behalf of the State in case of International Crimes28 does not expressly refer to jus cogens when it upholds the lifting
24 Akande and Shah, supra note 2, at 841-843.
25 Supra note 10, Judgment, at para. 59.
26 Immunity of State Officials, Memorandum by the Secretariat, A/CN/4/596, 31 Mar. 2008, at paras 204-207; Second Report, RA Kolodkin, A/CN.4/631, 10 June 2010, at 50-51.
27 IDI, Krakow Session, 2005.
28 IDI, Naples Session, 2009.
854    EJIL 22 (2011), 849-855
of immunity in civil proceedings for conduct that constitutes international crime (Article III). But it is closer to the jus cogens reasoning than it is to the jurisdiction-based reasoning. It relies on 'the underlying conflict between immunity from jurisdiction of States and their agents and claims arising from international crimes', and intends to contribute to 'a resolution of that conflict' (preamble). The resolution does not require that for immunity to be lifted extraterritorial jurisdiction over the relevant conduct has to be established. It merely refers to the existence of normative conflict between the two normative requirements. On what basis other than jus cogens would, one wonders, the criminality of the relevant act prevail in this normative conflict.
The third argument Akande and Shah advance is that the primacy of jus cogens was rejected by the European Court of Human Rights in Al-Adsani and the International Court of Justice in Arrest Warrant.29 However, both Al-Adsani and the House of Lords' decision in Jones have claimed that, while no immunity is available for the breaches of jus cogens in relation to criminal proceedings, the same position was not yet in place in relation to civil proceedings.30 What, therefore, these two decisions have essentially done is to undermine the legitimacy of their own reasoning, as it is indeed difficult to see what makes civil cases so special as to prevent the primacy of jus cogens that obtains in criminal cases.31
The Arrest Warrant case, which upheld the immunity of an incumbent foreign minister, cannot be seen as denying the impact of jus cogens on immunities either; for that would read too much into it, as the Court did not engage with jus cogens. Even if, in the unlikely case, the opposite was true, it would still not justify the blanket position taken in Al-Adsani and Jones. Arrest Warrant is a more nuanced decision, based on the balance of interests which underlies the hierarchy of norms issue. This underlying policy is reflected in the Joint Separate Opinion in Arrest Warrant that courts have to balance conflicting interests of individual victims and states.32 As the Court specified, the functionally and temporarily limited immunity of the Congolese foreign minister was not to tolerate impunity. Options to prosecute nationally, extraterritorially, and internationally remained intact.33
In civil cases, the outcomes are less adequate. In Al-Adsani the UK Government did not afford the claimant diplomatic protection against Kuwait, and then argued before the Strasbourg Court that the proper way of obtaining remedies was diplomatic representation.34 Al-Adsani was left without any remedy, and so were victims in Jones, much as the Law Lords asserted that all they decided was to divert settlement to other means,35 which is said to be the purpose of immunities.36 This approach rang hollow,
29 Akande and Shah, supra note 2, at 837-838.
30 For the fallacy of that position see supra note 1.
31 Joint Dissenting Opinion, Al-Adsani, 34EHRR 11(2002) 273, at 297-298.
32 Joint Separate Opinion, supra note 10, at para. 75.
33 Judgment, supra note 10, at paras 60-61.
34 Al-Adsani, supranote 10, at 280, 288 (paras 19, 50-51).
35 Jones v. Saudi Arabia [2006] UKHL 16, at paras 44-45.
36 Akande and Shah, supra note 2, at 826.
Immunities of State Officials, International Crimes, and Foreign Domestic Courts 855
for it ensured that the victims never got any remedy and Saudi Arabia was never held responsible. The outcome was a state of impunity, which cannot be defensible under the balance of interests approach. It would be an odd way of balancing interests if the defendant state always got away with its conduct, and the victim never obtained any remedy.
4 Conclusion
The argument against the primacy of jus cogens over immunities has no consistent basis and is not supported by evidence. Unfortunately, Akande and Shah have not advanced any new original argument, and thus not taken matters further than the position one of them rather impressively presented seven years ago.37 Willingly or not, adopting a position near the medium in the doctrinal spectrum inevitably exposes that position to objections from both sides of that spectrum. A moderate position does not always secure a right outcome.
37   Akande, 'International Law Immunities and the ICC, 98 A/IL (2004) 407.