13 May Guest Post: Official Act Immunity-Keeping the Questions Straight
[William S. Dodge is Professor of Law and Associate Dean for Research at the University of California, Hastings College of the Law. From August 2011 to July 2012, he served as Counselor on International Law to the Legal Adviser at the U.S. Department of State, where he worked on the amicus brief of the United States to the Fourth Circuit in Yousuf v. Samantar. The views expressed here are his own and do not necessarily reflect the views of the State Department or of the United States.]
In Yousuf v. Samantar, the U.S. Court of Appeals for the Fourth Circuit held that a former Somali official was not entitled to official act immunity for alleged violations of jus cogens norms. In a recent post, Professor Ingrid Wuerth takes issue with that conclusion, arguing that state practice and ICJ jurisprudence establish that allegations of jus cogens violations “do not generally deprive conduct of its ‘official’ nature for immunity purposes.”
As I have previously explained, all immunity doctrines involve three basic questions: (1) who is covered, (2) what is covered, and (3) whether there is an exception. For example, head of state immunity (a status-based immunity) generally applies only to sitting heads of state, heads of government and foreign ministers, covers all acts (even purely private ones), and is not subject to a jus cogens exception. See Case Concerning the Arrest Warrant of 11 April 2000 (Dem. Rep. Congo v. Belg.), 2002 I.C.J. 3 (Feb. 14). On the other hand, official act immunity (a conduct-based immunity) applies to lower level officials and to former officials, covers only acts taken in an official capacity, and may or may not be subject to a jus cogens exception.
In a recent article, Pinochet’s Legacy Reassessed, 106 Am. J. Int’l L. 731 (2012), Ingrid looks exhaustively at the third question in the context of official act immunity, concluding that state practice does not support a jus cogens exception to such immunity. That conclusion may or may not be correct—recent state practice including Samantar and the decision of the Swiss Federal Criminal Court in Nezzar may cast some doubt—but I will assume for present purposes that it is correct. Ingrid’s AJIL article did not, however, focus the same attention on the question of what constitutes an official act to which immunity attaches in the first place. Instead, she simply assumed that the second and third questions were the same. See id. at 732 n.9. In her recent post and in an interesting article on the Nezzar case, Ingrid answers the “what acts” question by adopting what one might call the “atributability theory” of official act immunity—that any act attributable to the state for purposes of state responsibility must be deemed an “official act” for purposes of conduct-based immunity. The problem is that there is no general and consistent practice of states supporting that position.
That is not to say that there is no authority at all in support of the attributability theory. The former rapporteur of the ILC’s project on the immunity of state officials from foreign jurisdiction adopted this theory in paragraph 24 of his Second Report. But his assertion has proved very controversial at the ILC and is flatly contradicted by another final ILC report, the 2001 Draft Articles on State Responsibility, article 58 of which expressly states that its rules on attributability to the state are “without prejudice to any question of the individual responsibility under international law of any person acting on behalf of a State.” The Bush Administration at one point advanced a similar attributability theory, but that theory has not been repeated by the Obama Administration and is in fact contradicted by the Obama Administration’s determination of non-immunity in Samantar with respect to human rights violations that were clearly attributable to Somalia under international law.
If one looks to state practice to determine the customary international law rule, one finds much evidence that states have not treated gross human rights violations as official acts to which conduct-based immunity attaches. See, e.g., Attorney General of Israel v. Eichmann, 36 I.L.R. 277, 309-10 (Israel S. Ct. 1962) (holding that crimes against humanity “in point of international law . . . are completely outside the ‘sovereign’ jurisdiction of the State that ordered or ratified their commission, and therefore those who participated in such acts must personally account for them and cannot shelter behind the official character of their task or mission”); Regina v. Bow Street Metropolitan Stipendiary Magistrate ex parte Pinochet Ugarte (No.3),  1 A.C. 147, 205 (H.L. 1999) (“How can it be for international law purposes an official function to do something which international law itself prohibits and criminalises?”); see also 1 Trial of the Major War Criminals Before the International Military Tribunal 223 (1947) (“The principle of international law, which under certain circumstances, protects the representatives of a state, cannot be applied to acts which are condemned as criminal by international law.”). And then, of course, there is the abundant U.S. practice in cases brought under the Alien Tort Statute and Torture Victim Protection Act (TVPA), which have consistently held that jus cogens violations are not official acts entitled to conduct-based immunity. See, e.g., Enahoro v. Abubakar, 408 F.3d 877, 893 (7th Cir. 2005); Hilao v. Estate of Marcos, 25 F.3d 1467, 1472 (9th Cir. 1994); Doe v. Qi, 349 F. Supp. 2d 1258, 1287 (N.D. Cal. 2004); Cabiri v. Assasie-Gyimah, 921 F. Supp. 1189, 1198 (S.D.N.Y. 1996); Xuncax v. Gramajo, 886 F. Supp. 162, 175-76 (D. Mass. 1995). The fact that these cases were decided before the Supreme Court’s Samantar decision, on the assumption that the Foreign Sovereign Immunities Act applied to foreign officials, does not negate the value of these decisions as evidence of state practice for purposes of customary international law.
Because states have generally not treated gross human rights violations as official acts, they have tended to reach the question of whether to recognize a jus cogens exception only in cases where a statute extends immunity for foreign officials beyond what customary international law requires. Jones v. Saudi Arabia  1 A.C. 270 (H.L. 2006), Zhang v. Zemin,  NSWSC 1296 (New South Wales S. Ct.), and Bouzari v. Iran, 71 OR (3d) 675 (Ontario Ct. of App. 2004), are examples. But precisely because these cases arose in the context of statutes that affirmatively granted immunity, they do not answer the question of what sort of conduct constitutes an official act for the purposes of immunity under customary international law.
To be sure, there are decisions that mix up the questions of whether immunity attaches to particular acts in the first place and whether to recognize a jus cogens exception to an existing immunity. I have criticized the Fourth Circuit’s Samantar decision for making this mistake, and the same might be said of the recent Swiss decision in Nezzar. But the fact that courts sometimes get confused does not mean that scholars should follow them into error.
Finally, I should say a few words about Ingrid’s proposals to soften the harsh results to which her analysis may lead. I am sympathetic to her position that international law does not require a state to grant official act immunity in the absence of an invocation by the foreign state. I would only add that this position does not depend on adopting an attributability theory for official act immunity. I am less sanguine about her suggestion that national legislation like the TVPA (or presumably the Swiss war crimes legislation in Nezzar) provides the grounds to overcome immunity in particular contexts. If it were true that any act attributable to a foreign state is covered by conduct-based immunity under customary international law, then suits under the TVPA for torture and extrajudicial killing under color of foreign law would violate international law. While Congress certainly has authority to do this within the U.S. legal system, it is not a happy result. I would instead view the TVPA as additional evidence of state practice showing that jus cogens violations are not official acts entitled to conduct-based immunity in the first place. As the Senate Report accompanying the TVPA explained “because no state officially condones torture or extrajudicial killings, few such acts, if any, would fall under the rubric of ‘official actions’ taken in the course of an official’s duties.” S. Rep. No. 102-249, at 8.
I find your analysis very intriguing but, in my view, it lacks the consideration of a relevant aspect as regards the relationship (and possible conflict) between jus cogens norms and the law of immunity.
In Jurisdictional Immunities, the ICJ held that (albeit addressing State immunity as opposed to immunity of State officials) the two sets of rules are different in nature: while the question of immunity is procedural in nature, an alleged violation of a jus cogens norm necessary raises substantive issues (see para. 93 of the Judgement). Hence, immunity functions as an “on” or “off” switch with respect to the forum court’s exercise of jurisdiction, and must be separated from the merits phase, when the lawfulness of the conduct is determined (i.e., whether there has been a violation of a jus cogens norm).
In light of your analysis, the question will be whether this is true with respect to official act immunity (or is it confined merely to State immunity)? If it is, functional immunity for a State conduct will remain active even when that conduct is, allegedly, a breach of jus cogens rules, since immunity (being a procedural matter) will effectively bar any further consideration of that allegation.
Janos, thanks for your comment. I accept the distinction between procedural and substantive rules articulated by the ICJ. In the context of official act immunity, however, immunity does “switch on” unless the act is official. State practice on this point shows that states do not generally and consistently treat jus cogens violations as official acts. This, to repeat, is a separate question from whether to recognize an exception to an immunity that has already attached (to flip the switch off again, if you will).
There is some overlap between the determination of whether an act is official and the ultimate merits. But this is not unusual in immunity law (consider the quesiton whether an act is commercial for purposes of state immunity, for example). Limits on discovery and other procedural devices can protect the potentially immune defendant from the burdens of unwarranted litigation.
I agree that one cannot so easily analogize foreign sovereign (and official status-based) immunities to official conduct-based or act immunity. After all, international law itself tells us they are different. For a somewhat different argument that conduct-based immunity can be deemed a substantive as opposed to a jurisdictional defense, and therefore would conflict with jus cogens, I recently posted this forthcoming piece on SSRN: