In 2009 and 2010, the International Criminal Court (ICC) issued two arrest warrants for Omar al-Bashir, then president of Sudan, relating to acts of genocide, crimes against humanity, and war crimes committed in Darfur. The ICC requested its member States to arrest al-Bashir, should he enter their territory, and surrender him to the ICC. The warrants and requests raised a number of difficult and interrelated legal questions that preoccupied international lawyers for the next decade.

Does customary international law permit one State to arrest the sitting president of another State pursuant to a request by an international criminal court? The International Law Commission had around a decade to address the question while developing its draft articles on the immunity of State officials from foreign criminal jurisdiction. It didn’t. The question would have then been squarely presented to the Sixth Committee of the United Nations General Assembly for discussion by a wide and representative group of States. It wasn’t.

Did the United Nations Security Council resolution that referred the situation in Darfur to the ICC displace any immunities that Sudanese officials might otherwise enjoy? The Security Council had at least a decade to resolve the question. It didn’t.

Does the ICC Statute, which provides that immunities shall not bar the Court from exercising its jurisdiction, have the “horizontal” effect of displacing any immunities that might otherwise bar its member States from exercising their jurisdiction over persons in their territory? The Assembly of States Parties to the ICC Statute had at least a decade to resolve the question. It didn’t.

Instead, in May 2019, the Appeals Chamber of the ICC answered all three questions in the affirmative. By then, al-Bashir had been removed from office. The question of his immunity from arrest by other States was moot. Last week, Sudan indicated its willingness to surrender al-Bashir to the ICC. We’ll see.

The first question remains the most controversial. What is the scope of Head of State immunity under customary international law? Over at Opinio Juris, Dire Tladi writes that “the time has now come for a united effort by States, from different perspectives, to agree to send this matter to the International Court of Justice for an advisory opinion.” Sarah Nouwen reaches the same conclusion in a recent article.

In my view, this proposal compounds the mistakes of the past. Worse still, a premature advisory opinion may entrench a legal error for the foreseeable future. If the U.N. General Assembly, or another competent U.N. organ, simply requests an advisory opinion from the International Court of Justice (ICJ), then its member States would once again abdicate their own responsibility to resolve—or at least reduce—legal uncertainty of their own making.

Here is the general problem. The existence and content of a rule of customary international law is determined, in the first instance, by the general practice and legal convictions of States. But, of course, there may be no general practice with respect to new or rare circumstances. This may leave the precise content of the rule uncertain. Suppose that widespread, representative, and consistent State practice establishes a customary rule that prohibits acts with features A, B, and C. Now suppose that new circumstances arise, making acts with features A, B, C, and D possible for the first time. Or suppose that acts with features A and B but not C rarely occur. Do such new or rare acts fall within the scope of the prohibition? How can we tell?

We might just wait and see how practice develops. But suppose we can’t. How to proceed? We might reason inductively, generalizing from the limited practice that exists. This is risky, since the available sample may prove unrepresentative. Or we might reason deductively, from the rule’s purpose or function, or from the deeper principles that the rule reflects. This is risky, too, since purposes and principles may be controversial, or may conflict, or may be too abstract to ground reliable inferences.


Return to the specific problem: Does customary international law prohibit States from arresting visiting Heads of State pursuant to arrest warrants issued by international criminal courts? A customary rule exists, but its precise content is uncertain, because international criminal courts are quite new and arrest warrants for sitting Heads of State are extremely rare.

Most of the State practice bearing directly on the question arose in the al-Bashir case itself. Fourteen ICC member States allowed Bashir to visit and declined to arrest him. Several asserted that doing so would violate customary international law. This practice, though substantial, was not exactly widespread. This practice was also not entirely consistent (see here and here). More importantly, this practice was not representative. Indeed, the sample was skewed by al-Bashir and his government. Al-Bashir did not choose his travel destinations at random. His government contacted certain member States and obtained assurances from them in advance. The vast majority of member States—which his government did not contact, or which did not provide such assurances—were not given an opportunity to arrest him or refrain from doing so. To generalize from such selective practice would enable leaders accused of atrocity crimes to gerrymander customary international law in their favor.

In principle, skewed samples of State practice could be corrected by the reactions of a representative cross-section of States. In practice, State silence is deafening. While failure to react can indicate acceptance when some reaction is called for, it is seldom clear which events call for some reaction. States did not generally praise or condemn those States that allowed al-Bashir to visit and refrained from arresting him. Arguably, some reaction was called for, at least from ICC member States, to condemn non-compliance with the ICC’s requests. But perhaps member States felt this was a matter best left to the ICC itself to resolve through its ordinary procedures.

Significantly, States did not generally welcome or deplore the Appeals Chamber’s ruling. At the subsequent meeting of the Assembly of States Parties, the African Union briefly criticized the ruling. As far as I know, no State and no other regional organization said a word about it. Since the ICC is an international organization, not a State, the consistency of its actions with customary international law can only be determined by the reactions of States. In my view, some reaction was called for. After all, the effect of the ruling was that all ICC member States have a treaty obligation to arrest visiting Heads of State at the ICC’s request. If States generally accept a conflicting customary obligation not to arrest visiting Heads of State, then one would expect them to say so.

In any event, silence and skewed practice are not particularly reliable bases for inductively determining the precise content of immunity under customary international law.


The ICJ could try to deduce the precise content of immunity in light of its purpose, function, or underlying principles. But this is risky as well. The purposes, functions, and underlying principles of immunity are controversial, and sufficiently abstract as to permit reasonable minds to draw quite different concrete conclusions.

For its part, the Appeals Chamber found that Head of State immunity derives from “the principle of par in parem non habet imperium, which is based on the sovereign equality of States.” However, international courts “do not act on behalf of a particular State or States. Rather, international courts act on behalf of the international community as a whole.” By implication, States that execute arrest warrants issued by international courts do not act on their own behalf, but instead on behalf of the international court and, ultimately, the international community as a whole. Therefore, one cannot deduce, from the underlying principle, that the customary rule prohibits executing such an arrest warrant against a visiting Head of State. While I’m inclined to agree, I think it is fair to say that this reasoning is neither clearly right nor clearly wrong. The principle is so abstract that its concrete implications are simply hard to draw with great confidence.

In their Joint Concurring Opinion, four judges made short work of the “functional” rationale according to which immunity “is held out as necessary to allow Heads of State to discharge their duties of State without disruption occasioned by prosecution.” Essentially, these judges found that “non-disruption of political mandate does not have an overriding value” that clearly outweighs the competing values of international criminal justice. Therefore, one cannot deduce, from this functional rationale, that the customary rule prohibits arresting a visiting Head of State and surrendering them to an international criminal court. Again, while I’m inclined to agree with the majority, I think it is fair to say that its reasoning is neither clearly right nor clearly wrong. Few would be shocked if States ultimately resolved this conflict of values one way or the other.


Finally, the ICJ could presume that customary international law permits—or prohibits—the conduct in question absent proof to the contrary. Before the Appeals Chamber, Jordan, the African Union, and one amicus curiae essentially argued that the general rule of Head of State immunity applies to any arrest undertaken for any reason unless State practice clearly establishes an exception to that general rule. Other amici accepted this premise, though one argued that such an exception had in fact been established.

In an unexpected turn, the Appeals Chamber found that,

given the fundamentally different nature of an international court as opposed to a domestic court exercising jurisdiction over a Head of State, it would be wrong to assume that an exception to the customary international law rule on Head of State immunity applicable in the relationship between States has to be established; rather, the onus is on those who claim that there is such immunity in relation to international courts to establish sufficient State practice and opinio juris. As further explained in the Joint Concurring Opinion … there is no such practice or opinio juris.

The Appeals Chamber may have reasoned as follows: Jordan is defending its non-compliance with its apparent treaty obligation to arrest al-Bashir by asserting a conflicting customary obligation not to arrest a visiting Head of State. Therefore, Jordan must establish the existence and applicable content of the customary obligation it alleges. Those who assert must prove. Presumably, this line of reasoning would not be available to the ICJ in a non-adversarial, advisory proceeding.

In her article, Nouwen criticizes the Appeals Chamber in harsh and, I think, unfortunate terms:

As a result of this trick, it is easy to establish that there is no immunity before international courts. But its legal soundness can be contested. It can be compared to saying: it may be the case that people are prohibited from killing other people, but show me the evidence that this rule also applies when 122 people together kill one person.

In Nouwen’s comparison, the 122 ICC member States (as of May 2019; Kiribati joined later) are killers, prosecuting genocide is murder, and al-Bashir is the victim. The comparison is morally objectionable but, for that very reason, also revealing.

The purpose of the customary rule prohibiting murder (say, as a crime against humanity) is to give legal effect to the moral rule prohibiting murder. Or, if you prefer, the purpose is to protect the intrinsic value of human life. In contrast, the sovereign equality of States is a legal construct, not a moral rule, and the uninterrupted performance of official functions is an instrumental value, not an intrinsic value. So, while we can reliably deduce that the customary rule against murder applies irrespective of the number of perpetrators, we cannot reliably deduce that the customary rule of immunity applies irrespective of an arrest warrant issued by an international court relating to international crimes.


I have not attempted to show that the Appeals Chamber’s judgment was correct, though I think it was. I have only attempted to show that the risk of error is very high. If States request an advisory opinion from the ICJ, they will put the ICJ in a position to fail. The relevant State practice remains unrepresentative, the underlying purposes and principles controversial and abstract, and the applicable presumptions debatable. The ICJ might have no choice but to decide the question in a contentious case, but to force the question into an advisory proceeding seems irresponsible.

The costs of error are also very high and, I think, asymmetrical. If the ICJ finds that States may arrest visiting Heads of State on behalf of the ICC, then perhaps a few ICC member States currently prepared to defy the ICC will instead comply. However, if the ICJ finds that States may not arrest visiting Heads of State on behalf of the ICC, then no ICC member State will be prepared to do so. State practice will align itself with the ICJ, rather than vice versa. In effect, the ICJ will make the very law that it attempts to interpret.

Instead of requesting an advisory opinion from the ICJ, the U.N. General Assembly should instead provide a forum in which all States, or at least a representative sample of States, express their legal convictions regarding the scope of Head of State immunity under customary international law. Put aside the question of when, if ever, “verbal practice” can determine the content of a rule governing non-verbal conduct. States refrain from arresting visiting Heads of State all the time. We just need States to tell us the legal rule they take themselves to be following when they engage in that practice.

Tladi is right that such an important question of customary international law should never have been left to “the five members of the Appeal Chamber, having heard arguments from a group of professors, two regional organisations and one State.” But Tladi is wrong that “[a]dvisory proceedings before the International Court of Justice are warranted.”

States created this mess and left it to fester. After so many years, they should clean it up.

Image: Sudan’s then-President Omar al-Bashir attends a meeting with his new 20-member cabinet as they take oath at the presidential palace in the capital on March 14, 2019. Photo by ASHRAF SHAZLY/AFP via Getty Images