Home EJIL Analysis Head of State Immunity is not the same as State Immunity: A Response to the African Union’s Position on Article 98 of the ICC Statute

Head of State Immunity is not the same as State Immunity: A Response to the African Union’s Position on Article 98 of the ICC Statute

Published on February 13, 2012        Author: 

Jens Iverson is a researcher for the Jus Post Bellum Project at the Grotius Centre for International Legal Studies, University of Leiden, the Netherlands.

On 9 January 2012, the African Union (AU) Commission issued a press release responding to the decisions (see here and here) issued by Pre-Trial Chamber I of the International Criminal Court (ICC) last December regarding the “alleged” failure by Chad and Malawi to comply with the cooperation requests with respect to the arrest and surrender of President Al Bashir of Sudan (For previous EJIL:Talk! Commentary on the ICC decisions and the AU response see here and here). In the press release, the AU makes the  assertion that the decision has the effect, inter alia, of:

“Rendering Article 98 of the Rome Statute redundant, non-operational and meaningless.”

There has been much discussion (including by Dapo Akande on this blog and elsewhere) regarding the meaning and effect of Article 98 of the Rome Statute. Paragraph 1 of that provision states that:

 “The Court may not proceed with a request for surrender or assistance which would require the requested State to act inconsistently with its obligations under international law with respect to the State or diplomatic immunity of a person or property of a third State, unless the Court can first obtain the cooperation of that third State for the waiver of the immunity.”

 Unfortunately, the following points are often not emphasised in discussions or rulings on Article 98(1) and are entirely overlooked by the AU Commission:

i.          Article 98(1) only covers “State or diplomatic immunity of a person or property of a third State”;

ii.         Head of state immunity is not the same thing as either a) state immunity or b) diplomatic immunity; and

iii.       Head of state immunity is the relevant immunity in this case. 

Nowhere does the Rome Statute explicitly recognize head of state immunity as a reason not to comply with obligations under the Statute.  As is well known, Article 27(2) clearly and unambiguously states that immunities which may attach to the official capacity of a person do not bar the Court from exercising jurisdiction over such a person.  Particularly in light of Article 27(2) and the obvious importance of the question of head of state immunity, if the framers of the Rome Statute intended Article 98(1) to apply to heads of state, one might have expected that explicit language to that effect would have been negotiated at the Rome Conference.  It was not.  Rather, as this analysis will briefly elaborate, it appears that Article 98 was crafted not to interfere with States qua States and with the efficient performance of the functions of diplomatic missions, while retaining the capacity to hold heads of state to account. 

As pointed out by Prof. Paola Gaeta in her 2009 JICJ article Does President Al Bashir Enjoy Immunity from Arrest?, it is important to distinguish between the question of what is legal for the ICC under the Rome Statute (i.e., Is the ICC authorized to issue to States Parties a request for surrender of the President of Sudan?) and the question of whether it is legal for states other than Sudan to enforce the warrant against Al Bashir under customary international law (i.e., Would a state commit a wrongful act vis-à-vis Sudan should it decide to arrest and surrender President Al Bashir?).  Prof. Gaeta argues forcefully that the ICC is not authorized to issue such a request for surrender, and that a state would commit a wrongful act should it decide to honour the request.  The authors of the AU press release clearly concur, although perhaps for different reasons.  I respectfully disagree with both Prof. Gaeta and the authors of the AU press release, particularly with respect to the first question (regarding the effect of Article 98).  The second requires a more extensive review of the evolution of customary international law than is possible in this brief post, and has been attempted by Pre-Trial Chamber I in the decisions at issue.

Regardless of how one interprets laws or secondary materials on particular national viewpoints, the burden of persuasion clearly lies with those who would read an immunity into Article 98(1) which is not explicitly present in the language of the Article.  Such an argument, to be persuasive, must demonstrate that under contemporary customary international law either a) diplomatic immunity necessarily includes head of state immunity or b) state immunity necessarily includes head of state immunity, not merely the selective incorporation of head of state immunity into domestic legislation meant to cover diplomatic or state immunity.  I have yet to see such a demonstration.

In an LJIL article written in 1999, Jürgen Bröhmer helpfully notes that state immunity, diplomatic immunity, and head of state immunity are separate concepts. (Diplomatic Immunity, Head of State Immunity, State Immunity: Misconceptions of a Notorious Human Rights Violatorsee also here).  Perhaps Article 98(1) could have been written more explicitly to clearly indicate that it did not include head of state immunity.  Of course, listing one excluded type of immunity might imply that it was the only excluded immunity, which I do not believe is the case.  For example, Article 98(1) also makes no mention of and does not cover privileges and immunities relating to International Organizations or Institutions. This is a pertinent example in the light of the (in my mind) ultra vires attempt to pre-emptively extend immunity to United Nations peacekeepers (see UNSC 1422). 

While the laws on diplomatic immunity and state immunity are complex, the short answer to the application of Article 98(1) to President Al Bashir can be summarized as follows: he is neither a state, nor a diplomat.  The trend in authority, particularly since the demise of the Soviet Union, is towards a restrictive view on State Immunity.  The Restatement (Third) on Foreign Relations Law of the United States has dropped any reference to State immunity covering anything other than states or state instrumentalities (see Foreign Relations Law of the United States, Restatement of the Law Third § 451 (1987), including Comment A).  Diplomatic immunity can cover more than recognized diplomats—for example, it can cover the family of diplomats.  Domestic legislation may combine several types of immunity under one statute or understanding, but as a matter of principle, diplomatic immunity and head of state immunity should be kept separate from one another.  One type of immunity does not attach to the object of another type of immunity.  Diplomats are not heads of state, nor (unless specifically recognized as such) heads of state diplomats.  As well put by Jürgen Bröhmer’s 2000 LJIL article, Immunity of a Former Head of State General Pinochet and the House of Lords: Part Three: “Diplomatic immunity is enjoyed by (former or present) diplomats only, head of state immunity is tied to being or having served as head of state and state immunity is tied to being a state.”

To determine whether President Al Bashir is covered by diplomatic immunity, one can look at Article 1 of the Vienna Convention on Diplomatic Relations, widely considered to codify customary international law on diplomatic immunity.  Particularly helpful are Article 1(e) (“A ‘diplomatic agent’ is the head of the mission or a member of the diplomatic staff of the mission”) and Article 31(1) (“A diplomatic agent shall enjoy immunity from the criminal jurisdiction of the receiving State.”).

What is the purpose of diplomatic immunity?  There are a variety of potential answers, but perhaps the most widely accepted answer is present in the Chapeau of the Vienna Convention on Diplomatic Relations, namely: “the purpose of such privileges and immunities is not to benefit individuals but to ensure the efficient performance of the functions of diplomatic missions” (emphasis added). 

There is no evidence that President Al Bashir was formally recognized as a diplomat by Chad or Malawi.  This is unsurprising, as his presence would not be necessary to ensure the efficient performance of the functions of diplomatic missions.  The accredited Sudanese diplomatic corps can do that for Sudan without President Al Bashir’s personal presence.  Neither the letter nor the spirit behind diplomatic immunity supports lending President Al Bashir such immunity. 

One can clearly imagine what the actual text of Article 98(1) might mean in practice without an overbroad interpretation.  Article 98(1) does not refer to all “obligations” or all “obligations under international law” but rather specifically “obligations under international law with respect to the State or diplomatic immunity of a person or property of a third State[.]”  The phrase “with respect to the State or diplomatic immunity of a person or property of a third State” must have had meaning, or it would not have been included.  That meaning is readily explainable.  For example, one might suggest that the ICC may not order that the bank accounts and other property of states and diplomats would be subject to seizure, if that violated the specific obligations under international law in question.  Any attempt by the Court to request the arrest or surrender of accredited diplomatic agents of a third state (widely interpreted as a non-State Party) would be subject to a legitimate objection under Article 98(1).

The burden must remain on those claiming that Article 98 applies to Chad and Malawi’s failure to comply with cooperation requests to explain specifically how it applies.  They must persuasively explain how, under Article 98(1), President Al Bashir is either a diplomat or a state, or how, under Article 98(2), he is covered under obligations under specific international agreements.  One can understand how confusion may arise given the arguably unsettled state of the law, and the muddled roots of head of state immunity as customary international law.  Head of state immunity may be ultimately rooted in—while still distinguishable from—state immunity or diplomatic immunity, depending on the national tradition.

Article 98(2) does not discuss customary law but instead “obligations under international agreements pursuant to which the consent of a sending State is required to surrender a person of that State to the Court[.]”  The language of “sending State” implies that agreements which use similar language such as the Vienna Convention on Diplomatic Relations or perhaps Status of Forces Agreements were in the minds of those who drove the adoption of Article 98(2) at Rome.  It has no applicability to the customary law norms regarding head of state immunity.  Neither Chad nor Malawi has a specific agreement requiring the consent of Sudan before honouring their obligations to the Court.

The split between the AU and the ICC, and the great confusion surrounding the proper interpretation of Article 98 makes a clear analysis of immunity under Article 98 an important and pressing matter.  This issue will certainly return.  It would be desirable for a consensus on Article 98 to emerge as soon as possible‑particularly a consensus that reflected the text of the Article, the intent of the framers as reflected in other Articles, and the stated goals of the Rome Statute.

Print Friendly

6 Responses

  1. André de Hoogh André de Hoogh

    Just a quick reply: Leaving aside the question who should bear the burden of proof regarding the interpretation of article 98(1) of the ICC Statute, the interpretation offered by Jens Iverson is not persuasive.

    If the proof would have to be that president Al Bashir is a State, then quite obviously this will not be possible. And this would be equally impossible with respect to any other person –natural, legal, corporate, or otherwise–, unless that person is a State. In other words, only the State as a legal person would be covered by the reference to State immunity in art. 98(1).

    Yet the formulation that the ICC Statute uses in art. 98(1) –”the State or diplomatic immunity of a person or property of a third State”– would then imply that State immunity only applies to property and not to persons. But if that were the case, the more logical formulation should have been ‘the diplomatic immunity of a person or State immunity of property of a third State’.

    Though this argument is perhaps not ironclad, one should note that the ICC only possesses criminal jurisdiction over natural persons (art. 25(1)) and that immunities pertaining to natural persons more generally, that is as a category, clearly exist (as the provision of art. 27(2) evidences by its denial of such immunities to natural persons).

    What is less clear then is why art. 98(1) would only relate to the diplomatic immunity of natural persons and not to immunities of heads of State, heads of Government, ministers of foreign affairs, or those of State officials more generally. It seems to me that the burden of proof with respect to the answer to that question lies with Jens Iverson, but one that he –at least in my view– fails to meet.

  2. a few points, after noting that I agree with Jens, but why focus on 27(2) instead of 27(1), which clearly denies immunity before the ICC for a head of state (sitting or former)? Also, 27(2) does not answer the question whether some immunity may attach under international law, but merely states “which may attach” — thus, may or may not attach; and whether or not they otherwise “amy,” they, expressly, “shall not bar the Court”.
    Furthermore, with respect to alleged diplomat immunity for international crimes, recall the famous statement of the IMT at Nuremberg regarding the law of immunity for any person in any capacity if the state moves outside its competence under international law — an ultra vires rationale that should help one to recognize that no person can have authority under international law to violate international criminal law and that any such conduct would be ultra vires and not protectable. I recall that some diplomats were prosecuted for war crimes, e.g., The French case of Abez.
    And why should the Vienna Convention be read as providing anything more that diplomatic immunity for certain diplomats with respect to ordinary offenses under domestic law of the host state (and, thus, not international crimes)?
    Unfortunately, we are experiencing efforts by some state elites to protect themselves from international crimnal responsibility, e.g., by claims to immunity, claims to limit universal jurisdiction under international law, claims to limit aut judicare aut dedera, and the ICJ seems to be supporting the state elites.

  3. sorry for the typos above, here is the para from the IMT:
    “The principle of international law, which under certain circumstances, protects representatives of a state, cannot be applied to acts which are condemned as crimnal by international law…. He who violates the laws of war cannot obtain immunity while acting in pursuance of the authority of the State if the State in authorising action moves outside its competence under international law.”

  4. I thank André de Hoogh for his helpful and well-phrased disagreement. Prompting discussion on the issue was one reason why I wrote the post. I hope there is more.

    Starting with where I think we agree – Article 98 could perhaps have been more clearly written. To take an example not in the piece, “third State” is generally regarded to mean non-state parties, or non-state parties that have not referred the under Article 12(3) – but the language isn’t clear. There was a great deal of disagreements between many NGOs and the US State Department during the first term of the George W. Bush administration over the applicability of Article 98 to Status of Forces Agreements. Of course, sometimes ambiguity in legal language is the result of design rather than error.

    As far as I can tell, we also agree that head of state immunity is not diplomatic immunity. What remains is the question of conflating state immunity and head of state immunity.

    De Hoogh’s main point, as I understand it, is that the presence of the word “person,” in combination with Article 25(1)’s restriction that with respect to individual criminal responsibility, the Court only has jurisdiction over natural persons, means that Article 98(1) must be read to include “head of state immunity” when it says “State […] immunity.” After all, who would surrender a State?

    To me, this logic cuts both ways. If it was obvious that Article 25(1) prevented jurisdiction over the legal person of a State, and that a request surrender of a state is a bit odd, then it seems wholly unnecessary to bother with a baroque construction such as:

    1. The Court may not proceed with a request for surrender or assistance which would require the requested State to act inconsistently with its obligations under international law with respect to the diplomatic immunity of a person or property of a third State, or a request for assistance which would require the requested State to act inconsistently with its obligations under international law with respect to the State immunity of property of a third State, unless the Court can first obtain the cooperation of that third State for the waiver of the immunity.

    De Hoogh’s point would also be more persuasive if Article 98(1) applied only to a request for “surrender.” Article 98(1), however, also applies to requests for “assistance,” a rather open-ended term which may have inspired caution in the drafters. State immunity covers more than criminal or civil liability for the legal person of the state. A baroque construction such as the above might end up causing more problems than it solved, particularly given that the “problem” of being clear that nobody should surrender a state isn’t really much of a problem.

    A more persuasive argument implied by De Hoogh’s response is that the drafters did not wish to list head of government, head of state, the minister of foreign affairs, and perhaps other government officials, preferring to describe their collective immunities as “State immunity.” This seems at least plausible. Perhaps this is an instance of intended ambiguity. Nonetheless, one might expect that, given that the question of the head of state/government was almost inevitably going to arise, the drafters would have been explicit on this point had it been their intention. At a minimum, a clearer phrase including the specific listing of “Head of State or Government,” the language used in Article 27, could have been used. It was not. The phrase used was “State immunity.” That may have been meant in the manner De Hoogh suggests, but I suggest that usage is confusing at best and strained at worst.

    Fundamentally, the problem of differentiating or conflating state immunity and head of state immunity remains unchanged. As recently put by the ICJ, no slouch when upholding the immunities of state officials:

    The Court does not consider that the United Kingdom judgment in Pinochet (No. 3) ([2000] 1 AC 147; ILR, Vol. 119, p. 136) is relevant, notwithstanding the reliance placed on that judgment by the Italian Court of Cassation in Ferrini. Pinochet concerned the immunity of a former Head of State from the criminal jurisdiction of another State, not the immunity of the State itself in proceedings designed to establish its liability to damages. The distinction between the immunity of the official in the former type of case and that of the State in the latter case was emphasized by several of the judges in Pinochet (Lord Hutton at pp. 254 and 264, Lord Millett at p. 278 and Lord Phillips at pp. 280-281).

    While it is true that the contrast drawn is between former heads of state and the state, there is no indication that Pinochet’s retirement is the critical distinction made; rather it seems clear that the ICJ is fully capable of discerning the difference between state immunity and head of state immunity. There is no persuasive reason to believe the drafters of the Rome Statute lacked that capacity.

  5. In response to jpaust:

    Thanks for your comments. I referenced Article 27(2) rather than 27(1) because I read 27(1) as covering jurisdiction and sentencing, while 27(2) covers immunities and procedural bars to admissibility. Article 27(2) seems more on point to me with respect to Article 98.

    With respect to your other points- they track to some degree what the Pre-trial Chamber has said, and others have said. I think the language of the IMT is quite powerful. While some disagree with the quality, substance, and result of the Pre-Trial Chamber’s analysis, a crucial point raised by the African Union (and others) in response to that analysis with respect to Article 98 is… what is Article 98 there for if that analysis is the law? My piece is, in part, my humble attempt to help answer that question. Many others have written more extensively on the same or similar questions, but as I hadn’t come across an argument for or against the analysis I put forth, and it seemed legitimate to me, I wanted to put it forward in hopes that it’s helpful.

  6. Hossein Sadat

    Thanks Jens for your analysis. I had some comments:

    On the burden of proof, the situation is not similar to the one in Eastern Greenland case (available here: , para. 111) and the burden of proof in not on only those who argue that para 2, Article 98 is covering the heads of states. All interpretations are on equal footing, since the meaning of the words employed in this provision are no clear enough.

    Article 98 (2) of the ICC’ Statute was proposed by the delegation of Singapore during the Rome Conference on 26 June 1998 (available here : ) and later on it was adopted in the Working Group on International Cooperation and Judicial Assistance on 16 July 1998 with some minor changes (available here:
    I fully agree with Jens that head of states are not covered in the category of “diplomats” and the reference to diplomatic immunity in article 98 (1) is not relevant to the immunity of heads of state. Even, the foreign ministers were not considered as diplomats in the famous case of Yourodia and for this reason the ICJ did not apply the 1961 Vienna Convention and instead it applied the customary international law. That is the same here.

    The travaux préparatoires of the Rome Statue does not clearly indicate the intent of 98 (1) drafters that what is the exact scope of the “the State or diplomatic immunity of a person or property of a third State” ratio persona. However, if one utilizes the principles and rules governing treaty interpretation in particular those referred in the VCLT or in the ICJ jurisprudence, it would be acceptable to argue that “State Immunity of a person” shall be read as “immunity of state officials” rather than only the “immunity of the State itself”. These are two dimensions of the same coin and that was looked in the same way recently by the ICJ. I had a different interepretation from para. 87 of the Immunity Case (Germany v. Italy). The Court at the beginning of the paragraph says that Pinochet case concerned the immunity of head of State and not immunity of the State “itself”. It means that the immunity of State does have two fold; one to the State “itself” and the other to those who represent the State. This interpretation would gives a role to the word “person” in para 1, Article 98 of the Rome Statute too. It would lead to a coherent and reasonable understanding from this provision. It could be possible to introduce different arguments sustaing such interpretation. If the notion of “Immunity of State” does not include the natural persons, then what practical effects is expected for this paragraph. How a request for “surrender” or even “assistance” would be relevant with State Immunity, in particular jurisdictional immunity. If one would limit this paragraph only to the immunity from execution then what role for the word “person”. And finally, when the immunity for a diplomat has been accepted in the Statute, a priori it shall be the same for higher officials, for sure, not all but at least for heads of states and foreign ministers.

    As it was indicated by Prof Crawford and Prof Sands in the context of para 2, Article 98: “The general object and purpose and of the ICC Statute is, however, subject to limitations which States Parties have accepted. Article 98 identifies two sets of obligations that may lawfully prevent a State Party from acceding to a request to surrender a person to the Court – State and diplomatic immunity (under Article 98(1)), and a certain class of international agreements (under Article 98(2)). On its own terms, therefore, the ICC Statute limits the possibility of the complete realization of the policy of avoiding impunity by ensuring investigation or prosecution of persons within the territory of a State Party.” Available here: