What Does Customary International Law Say About Halkbank’s Immunity?


Halkbank” by erdalde

is licensed under CC BY-NC-SA 2.0.

Tomorrow, the Second Circuit will hear argument in United States v. Turkiye Halk Bankasi A.S. to consider whether Halkbank, a Turkish state-owned bank (but not its central bank), is immune from criminal prosecution for violating U.S. sanctions on Iran. Halkbank claimed immunity under both the Foreign Sovereign Immunities Act (FSIA) and federal common law. The U.S. Supreme Court rejected Halkbank’s FSIA claim last year, holding that the statute does not apply to criminal proceedings. The Court remanded Halkbank’s common law claim to the Second Circuit for further consideration.

In discussing the federal common law rules, the U.S. brief on remand does not rely on customary international law. The brief argues, first, that courts must obey executive branch determinations of common law immunity and that, by bringing this prosecution, the executive has determined that Halkbank is not immune. (Disclosure: Ingrid (Wuerth) Brunk and I filed an amicus brief arguing against blind deference to the executive.) As a fallback, the brief argues that federal common law does not immunize foreign state-owned corporations for commercial activities. But the brief does not look to customary international law to inform federal common law.

That is a shame. In a prior post, I suggested that customary international law supports the government’s position that Halkbank is not immune from prosecution. By not even discussing customary international law, the government abandons that field to the other side. Halkbank’s own brief invokes customary international law opportunistically, citing it alongside U.S. cases whenever it seems helpful. An amicus brief by Roger O’Keefe takes a more systematic approach to customary international law, arguing that foreign states are absolutely immune from criminal jurisdiction, that the restrictive theory applies only to civil proceedings, and that the state’s absolute immunity extends to state instrumentalities (including state-owned companies) when they exercise state authority.

In this post, I examine what customary international law has to say about Halkbank’s immunity. I draw on an excellent amicus brief that Mark Feldman and Chimène Keitner submitted to the Supreme Court about international law, which the Second Circuit would be well advised to read in conjunction with O’Keefe’s brief. The bottom line, on which everyone seems to agree, is that Halkbank is not immune under customary international law unless its prosecution is for exercising state authority.

Identifying Customary International Law

“Customary international law,” says § 102(2) of the Restatement (Third) of U.S. Foreign Relations Law, “results from a general and consistent practice of states followed by them from a sense of legal obligation.” Conclusion 2 of the International Law Commission’s (ILC) 2018 Draft Conclusions on Identification of Customary International Law agrees: “To determine the existence and content of a rule of customary international law, it is necessary to ascertain whether there is a general practice that is accepted as law (opinio juris).” In other words, a customary international law rule exists if and only if states generally act in a certain way and understand that they are following a legal rule when they do so.

The second requirement is frequently called opinio juris (“opinion of law”). To correct a misunderstanding I have heard some judges express, opinio juris refers to the opinions of states not scholars. Opinio juris is not a way of claiming that customary international law is whatever scholars think it should be. Rather, it is a requirement that states understand that their practice is required or permitted by a rule of international law. Opinio juris is a way of distinguishing general and consistent state practices that are not legally required (such as rolling out a red carpet for visiting heads of state) from general and consistent state practices that are legally required (such as not throwing visiting heads of state in jail).

The question, therefore, is not whether Professor O’Keefe thinks that state instrumentalities such as Halkbank should be immune or whether I think that they should not be. The question is what state practice and state views tell us. I will discuss below a few specific questions on which I disagree with O’Keefe, but I begin with a few propositions on which there appears to be consensus.

Foreign States Themselves Are Immune from Suit

The U.S. government conceded at oral argument before the Supreme Court that “there is a strong customary international law principle against prosecuting a state qua state” (Transcript p. 51). Certainly, there is a general and consistent practice of not criminally prosecuting foreign states themselves. I am aware of no country that has. The harder question is whether this practice exists because most countries’ criminal laws simply do not apply to foreign states or rather because countries feel legally prohibited from prosecuting foreign states.

U.S. practice in the Halkbank case may strengthen the argument that customary international law prohibits criminal prosecutions of foreign states themselves. Part II of the Supreme Court’s opinion in Halkbank specifically held that Congress’s grant of subject matter jurisdiction over criminal offenses against the United States does not exclude foreign states. Additionally, the U.S. government’s concession at oral argument seems to establish opinio juris by acknowledging the existence of a customary international law prohibition.

But the prohibition acknowledged by the U.S. government is only against prosecuting foreign states themselves. As Feldman and Keitner note in their amicus brief (p. 5), “Halkbank is a company, not a country.” O’Keefe (p. 12) finds only one decision, from the French Court of Cassation, holding that a foreign state’s immunity from criminal prosecution extends to state instrumentalities exercising state authority. Feldman and Keitner (pp. 12-13), on the other hand, point out that the same decision held that an entity to which the government had delegated non-sovereign functions was not immune. The critical question, then, is when a foreign state’s immunity attaches to a state instrumentality like Halkbank under customary international law.

Ownership and Control

Halkbank asserts (p. 33) that it is entitled to the same immunity as the state “because it is owned and controlled by the Republic of Türkiye.” As a matter of customary international law, this is clearly not true. As Feldman and Keitner explain at length (pp. 5-14), state practice generally treats companies and countries as separate entities. Although the FSIA considers state-owned corporations to be instrumentalities of foreign states based on ownership alone, the United States appears to be the only country to do so. There is no general and consistent practice of states immunizing corporations from civil suit, let alone prosecution, based simply on state ownership and, thus, there is no customary international law requiring such immunity.

O’Keefe does not argue otherwise. Rather, he says (p. 3) that “under customary international law a foreign state instrumentality enjoys absolute immunity from the criminal jurisdiction of a forum state when it acts in the exercise of state authority, as opposed to in a private capacity” (emphasis added). In other words, O’Keefe agrees with Feldman and Keitner that a state-owned corporation is not entitled to immunity from criminal prosecution unless it is exercising state authority.

Exercising State Authority

The indictment alleges that proceeds from Iran’s sales of oil and gas to Turkey’s national oil company were deposited with Halkbank, which conspired with high-ranking Turkish officials to transfer some of these funds back to Iran in violation of U.S. sanctions. O’Keefe characterizes (p. 16) Halkbank’s duties to hold the proceeds and comply with U.S. sanctions as “public functions.” But it is hard to see how these functions differ from those of ordinary commercial banks that hold funds for customers and comply with applicable laws. Halkbank’s conduct seems indistinguishable from that of other, non-state-owned banks, like HSBC and BNP Paribas, that the United States has charged with violating sanctions. The fact that Turkish officials directed some of Halkbank’s conduct (some of them allegedly in return for bribes) does not convert these actions into exercises of state authority.

Halkbank argues (pp. 35-36) that it performs other “sovereign functions” such as collecting customs duties, but that is irrelevant. Halkbank is not being prosecuted for collecting customs duties but rather for transferring money to Iran in violation of U.S. sanctions. Even under Professor O’Keefe’s view of customary international law, Halkbank is entitled to immunity only to the extent that it was exercising state authority.

A Few Disagreements with Professor O’Keefe

The customary international law rules that one derives from state practice may depend on how that practice is interpreted. Professor O’Keefe’s brief fits state practice together in a way that leads to broad immunity of foreign state instrumentalities from criminal prosecution. For those readers who are still with me, I’d like to discuss three points on which I disagree.

The Baseline Question

O’Keefe’s brief starts by assuming immunity and then requires a general and consistent practice of states to establish exceptions. I have previously referred to choosing one’s starting point as “the baseline question”: Should one begin by assuming a prohibitive rule (no immunity) and require state practice to create an exception? Or should one begin by assuming a permissive rule (plenary jurisdiction) and require state practice to establish immunity?

O’Keefe invokes the U.S. Supreme Court’s decision in The Schooner Exchange v. McFaddon (1812) and the International Court of Justice’s (ICJ) decision in Jurisdictional Immunities (2012) to support his prohibitive baseline. But they do not. The Schooner Exchange actually begins with the propositions that “[t]he jurisdiction of the nation within its own territory is necessarily exclusive and absolute” and that all exceptions to jurisdiction “must be traced up to the consent of the nation itself.” Chief Justice Marshall went on to hold that U.S. consent to foreign state immunity should be implied, but he quickly added that “the sovereign of the place is capable of destroying this implication” and that “[h]e may claim and exercise jurisdiction.” (Contrary to the U.S. brief, this does not mean that the executive alone makes the decision. As I explained, and notwithstanding the word “he,” Marshall clearly understood Congress and the President acting together to be the U.S. “sovereign.”)

The ICJ in Jurisdictional Immunities recognizes two fundamental principles of the international order that are in some tension: “the principle of sovereign equality of States” and “the principle that each State possesses sovereignty over its own territory.” “Exceptions to the immunity of the State represent a departure from the principle of sovereign equality,” the Court notes, whereas “[i]mmunity may represent a departure from the principle of territorial sovereignty and the jurisdiction which flows from it.” In Jurisdictional Immunities, as I have explained, the ICJ did not begin with a baseline of immunity and look for state practice establishing an exception. Instead, it looked for (and found) state practice and opinio juris establishing state immunity with respect to the acts of armed forces during armed conflict.

If the proper baseline in this area of customary international law is a permissive rule of plenary jurisdiction, then the proper question is whether there is a general and consistent practice of states and opinio jurisestablishing that state instrumentalities like Halkbank are immune from prosecution based on the activities alleged in the indictment. Neither Halkbank nor O’Keefe points to such practice, and I would thus conclude that customary international law provides no such immunity.

The Relevance of Exceptions to Immunity in Civil Cases

O’Keefe also argues (p. 9) that the restrictive theory of state immunity, which holds that foreign states are not immune from suit based on their non-governmental acts, applies only to civil cases and not to criminal cases. He notes that the U.N. Convention on the Jurisdictional Immunities of States and Their Property, which follows the restrictive theory, was adopted by the General Assembly on the understanding that it does not cover criminal proceedings. He further notes (p. 9 & n.5) that domestic statutes codifying the restrictive theory extend only to civil proceedings.

Conceding this point for the sake of argument, it is not clear what difference it makes for state instrumentalities. Under the restrictive theory, a state instrumentality—like the state itself—would be immune only with respect to its governmental acts. That is essentially the same position that O’Keefe takes in his brief: that state instrumentalities are immune when they exercise state authority. (Under Article 2(1)(b) of the U.N. Convention, which O’Keefe relies on here (p. 13), instrumentalities are part of a foreign state only “to the extent that they are entitled to perform and are actually performing acts in the exercise of sovereign authority of the State.”).

In other words, whether one applies the restrictive theory to criminal proceedings against state instrumentalities and concludes that they are not immune from prosecution for their non-governmental activities, or whether one maintains (as O’Keefe does) that state instrumentalities enjoy absolute immunity when they act in the exercise of state authority, one ends in the same place, at least in this case. Instrumentalities like Halkbank, acting in a commercial capacity and not in the exercise of state authority, are not immune from criminal prosecution.

Acting in the Exercise of State Authority

Finally, I disagree with the way O’Keefe would determine whether an instrumentality exercises state authority. Relying on the ILC’s Draft Articles on State Responsibility, O’Keefe asserts (p. 15) that an instrumentality exercises state authority even if it “manifestly exceeded its competence” so long as it acted with “apparent authority.”

This is mixing apples and oranges. As I have explained previously in the context of foreign official immunity, the fact that an official’s act is attributable to the state does not establish that the act was done in an official capacity for purposes of immunity. What is true for foreign official immunity is similarly true for state instrumentalities. State responsibility and state immunity are different things. The fact that an instrumentality’s act is attributable to a state does not establish that an instrumentality was exercising state authority for purposes of immunity.

The question, then is whether Halkbank was actually exercising Turkey’s state authority when it transferred funds to Iran in violation of U.S. sanctions, not whether it was apparently exercising such authority. As noted above, Halkbank is being prosecuted for actions that other, non-state-owned banks have previously been prosecuted for.

Translating Customary International Law into Federal Common Law

If, as I have argued, Halkbank is not immune under customary international law, how should that translate into federal common law? In his Halkbank concurrence, Justice Gorsuch expressed concern that, in order to apply the customary international law of immunity, lower courts might have to confront a “long-running debate” about the status of customary international law in the U.S. legal system. But as I wrote previously, he need not worry.

To apply the customary international law of state immunity in this case, the Second Circuit need not decide that all customary international law is federal common law for all purposes. It need decide only that complying with customary international law rules on state immunity provides sufficient justification for creating federal common law rules on immunity. The Supreme Court recognized in Sabbatino that foreign relations can justify the creation of federal common law. The need to avoid violations of international law certainly justifies creation of such rules.

In translating customary international law rules on immunity into federal common law, however, courts should act with appropriate caution. Specifically, as Keitner and I have argued, courts should not go beyond the immunity that customary international law clearly requires. Immunity prevents courts from enforcing law. In this case, if Halkbank is immune, the United States will be unable to enforce criminal statutes that Congress has passed to punish violations of sanctions that the executive has promulgated. It is appropriate for courts in the United States to recognize federal common law rules of immunity to the extent necessary to avoid U.S. violations of international law (for example head-of-state immunity), even if they sometimes prevent enforcement of federal law. But it is not appropriate for courts to extend federal common law rules of immunity beyond what international law requires. That is a question for Congress.


The Second Circuit should decide this case by applying a federal common law rule based on customary international law. By doing so, it would avoid becoming “potted plants” (in Justice Gorsuch’s phrase), bound to obey whatever the executive decides with respect to immunity.

Under customary international law, Halkbank is not immune from criminal prosecution. Halkbank is not a foreign state. The fact that it is state-owned does not create immunity under customary international law. And Halkbank is not being prosecuted for exercising state authority. Halkbank faces the same sort of charges that the United States has previously brought against non-state-owned banks, and it is no more entitled to immunity than they are.