Turkiye Halk Bankasi A.S. v. United States, Part 1: The FSIA and Criminal Prosecutions

Editor's Note: This is the first in a three-part series on oral arguments in Turkiye Halk Bankasi A.S. v. United States. Click here for part two and here for part three. 

On Jan. 17, the Supreme Court will hear oral argument in Turkiye Halk Bankasi A.S. v. United States, a case that raises the question whether the U.S. government can criminally prosecute corporations owned by foreign states. The petitioner, Türkiye Halk Bankasi A.Ş. (Halkbank), a corporation owned and controlled by Turkey, maintains that the federal criminal jurisdiction conferred by 18 U.S.C. § 3231 does not permit prosecutions of foreign sovereigns or their instrumentalities, and that the Foreign Sovereign Immunities Act (FSIA), 28 U.S.C. § 1330 et seq., precludes criminal jurisdiction over foreign sovereigns and their instrumentalities. The respondent, the United States, argues primarily that 18 U.S.C. § 3231 confers criminal jurisdiction over foreign sovereign instrumentalities, that the FSIA applies only to civil actions and does not confer any immunity in criminal cases, and that the government has the authority to determine sovereign immunity in criminal cases. 

We will preview Halkbank in two parts. In this article, we lay out the background of the case and assess the proper interpretation of the FSIA as it applies to criminal prosecutions. In a second article, we discuss how federal courts should address the question of foreign sovereign immunity in criminal cases if, as the government argues, foreign sovereigns and their instrumentalities lack immunity under federal statutory law.

Background

We assume that Lawfare readers have a basic familiarity with the FSIA, so we will be brief in describing it. Four provisions of the FSIA are most relevant in Halkbank:

• 28 U.S.C. § 1603 establishes that, for purposes of the FSIA, a “foreign state” includes corporations, like Halkbank, that are majority owned by foreign states. 
• 28 U.S.C. § 1604 establishes the presumptive immunity for foreign states in both federal and state courts. It provides that “a foreign state shall be immune from the jurisdiction of the courts of the United States and of the States except as provided in sections 1605 to 1607” of Title 28. 
• 28 U.S.C. § 1330 establishes the contours of federal jurisdiction over foreign states and ties such jurisdiction to the immunities of foreign states. It provides that the “district courts shall have original jurisdiction without regard to amount in controversy of any nonjury civil action against a foreign state … to any claim for relief in personam with respect to which the foreign state is not entitled to immunity … under sections 1605–1607” of Title 28. 
• 28 U.S.C. § 1605 establishes the relevant exceptions to the jurisdictional immunity of a foreign state in both federal and state court. It provides (among other things) that a “foreign state shall not be immune from the jurisdiction of courts of the United States or of the States in any case … in which the action is based upon a commercial activity carried on in the United States by the foreign state” (emphasis added).

Also relevant in Halkbank is 18 U.S.C. § 3231, which establishes federal jurisdiction in federal criminal cases. It provides that the “district courts of the United States shall have original jurisdiction, exclusive of the courts of the States, of all offenses against the laws of the United States.”

The United States indicted Halkbank in 2019 in connection with its alleged laundering of billions of dollars’ worth of Iranian oil and natural gas proceeds in violation of U.S. sanctions against Iran and related entities. Halkbank argued that it was immune from criminal prosecution. The district court rejected this argument and the U.S. Court of Appeals for the Second Circuit affirmed. The court of appeals ruled that 18 U.S.C. § 3231 conferred criminal jurisdiction over Halkbank. It then rejected Halkbank’s immunity claims. After assuming (but not deciding) that § 1604 of the FSIA confers sovereign immunity in criminal cases, the court ruled that the § 1605 exceptions to immunity also applied in criminal cases. It further ruled that Halkbank lacked immunity in the case because its alleged conduct satisfied the commercial activities exception of § 1605(a)(2). Finally, the court rejected Halkbank’s argument that it was immune from suit under the “common law.” It noted that the common law, which would not apply if the FSIA applied, had a commercial activities exception. And it stated that “at common law, sovereign immunity determinations were the prerogative of the Executive Branch; thus, the decision to bring criminal charges would have necessarily manifested the Executive Branch’s view that no sovereign immunity existed.”

Before the Supreme Court, the parties disagree about whether 18 U.S.C. § 3231 establishes criminal jurisdiction over foreign sovereigns. On its face, the statute’s conferral of exclusive federal jurisdiction over “all offenses against the laws of the United States” (emphasis added) applies to all defendants who commit such offenses, including foreign sovereign instrumentalities. Halkbank relies on Schooner Exchange v. McFaddon (1812) and subsequent cases, including Berizzi Brothers Co. v. SS Pesaro (1926), for the proposition that generally worded jurisdictional statutes do not extend to foreign sovereigns or their instrumentalities. The government counters with a different line of cases that it contends recognizes “that a commercial enterprise owned or controlled by a sovereign generally lacks immunity from suit.” 

We will not assess this argument here, in part for lack of space, and in part because we think the Court will likely address the FSIA issue in any event. The Court need not address the FSIA issue if it agrees with Halkbank that § 3231 should not extend to the prosecution of foreign sovereigns. But even if the Court rules that way, which we doubt, it might go on to address the proper scope of § 1604 of the FSIA, just as Halkbank does. And if the Court rejects Halkbank’s reading of § 3231, it will have to address whether § 1604 independently confers criminal immunity on foreign states. For reasons stated below, we are persuaded that § 1604 does confer such criminal immunity—a conclusion that, if right, forces one to address the complex question of how the FSIA in its entirety applies to criminal cases.

FSIA Section 1604 and Criminal Cases

As noted above, § 1604 provides that “a foreign state shall be immune from the jurisdiction of the courts of the United States and of the States except as provided in sections 1605 to 1607” of Title 28. On its face this provision appears to confer immunity from all federal jurisdiction, civil and criminal, subject to the exceptions in (as relevant here) § 1605. This statute is also more specific (and was enacted later in time) than 18 U.S.C. § 3231 and thus would seem to take precedence in the event of a conflict.

There are other indications in the FSIA that § 1604 should be interpreted, as its plain meaning suggests, to include immunity from criminal prosecution. Although not emphasized in the briefs in the case, § 1604 expressly applies to state courts as well as federal courts. This makes sense: A state court action against a foreign state or its instrumentality would raise the same foreign policy and international law issues as one in federal court.

The possibility of a state court proceeding against a foreign state draws attention to other elements of the FSIA. First, the statute contemplates at least some form of concurrent jurisdiction in state courts in FSIA cases, subject to removal, and some FSIA cases are in fact litigated in state court. Second, and more importantly, the FSIA’s removal provision, 28 U.S.C. § 1441(d), provides for removal only of “[any] civil action” (emphasis added) brought in state court against a foreign state or instrumentality. Neither § 1441 nor any other provision of federal law provides for removal of a state criminal case against a foreign state or instrumentality. 

As a result, if a state prosecutor brings an indictment against a foreign state or instrumentality, that case cannot be removed and can proceed in state court. Such a prosecution would not be preempted by 18 U.S.C. § 3231, which establishes exclusive federal jurisdiction over federal crimes, but does not speak in any way to state prosecutions for state crimes. Nor would it obviously be subject to nonstatutory federal preemption, and, in any event, the assertion of a preemption defense would not be a basis for removal. In addition, without a basis for federal preemption or some other federal question in the case, the state prosecution might not even be subject to review in the Supreme Court.

If the FSIA does not prohibit state criminal prosecutions of foreign states, the lack of any statutory basis for removing such prosecutions would be a highly anomalous result for a statute that seeks to establish federal judicial control over suits against foreign sovereigns. The House report on the bill that became the FSIA captures what the structure of the law clearly suggests: “In view of the potential sensitivity of actions against foreign states and the importance of developing a uniform body of law in this area, it is important to give foreign states clear authority to remove to a Federal forum actions brought against them in the State courts.” 

The limitation on removal to civil actions in § 1441(d) is especially anomalous when compared to 28 U.S.C. § 1442, which provides for removal of both a “civil action” and a “criminal prosecution” commenced in state courts against the United States, its agencies, or officers. It would be strange enough for Congress to allow state criminal prosecutions to proceed against foreign sovereigns in state courts, possibly without Supreme Court review. But it would be stranger still in light of § 1442 (the relevant version of which predates the FSIA), which shows that Congress knows how to use removal in situations in which state prosecutions will implicate federal interests and, more specifically, the federal interest in having disputes against sovereigns, possibly involving immunity, resolved in federal court. 

However, this anomaly dissipates entirely if § 1604, which expressly applies in state courts, is read in accord with its plain meaning to immunize foreign states in all cases, including criminal ones. On this natural reading of § 1604, state courts would have to dismiss any state prosecution against a foreign sovereign or instrumentality, and their failure to do so would present a clear federal question for Supreme Court review. Reading § 1604 to confer criminal immunity thus helps explain why Congress in § 1604 provided immunity for foreign sovereigns in state court at the same time that it permitted removal in civil actions. 

We believe that this analysis provides powerful textual and structural support for the conclusion that § 1604 accords immunity to foreign states in both federal and state courts in criminal cases. The government argues against this conclusion by maintaining at a high level of generality that the entire FSIA, including § 1604, applies only to civil cases. We will consider the government’s main arguments when we assess below whether the exceptions to immunity in § 1605 apply in criminal cases. For now, we simply note that the government’s position that the FSIA does not apply in any criminal cases implies that states can prosecute foreign states under state law without limitation, very possibly without Supreme Court review. 

The government does not address this implication, but its likely and perhaps only response would be that the federal executive branch could, without statutory authorization, compel a state court to dismiss a case against a foreign sovereign or instrumentality. We address this claim about executive branch control—which in the preemption context would go far beyond the government’s claim that it can make immunity determinations as a component of federal prosecutorial discretion—in our second article. 

The point for now is that if the Supreme Court does not apply § 1604 to confer foreign sovereign immunity in criminal cases, the Court would likely be required either to allow states to prosecute foreign states or to accept a strong form of federal executive branch control over state criminal prosecutions. (We explain in the second article why a third possibility, the automatic incorporation of the international law of sovereign immunity as federal common law, is ruled out by a variety of considerations, including the case law governing post-Erie federal common law.)

To be clear, our claim here is about the best reading of the statutory text. We make no claim about what Congress (or individual members of Congress) “really intended,” in part because we do not know, and also because that is not how the Supreme Court typically approaches matters of statutory interpretation. It is possible that, because there were virtually no criminal actions against foreign states before 1976, Congress simply did not think about the issue or the implications of its scheme for state court prosecutions. The bottom line, though, is that Congress voted to approve specific statutory text, and the text of § 1604 applies to both criminal and civil cases.

Do the FSIA Section 1605 Exceptions to Immunity Apply in Criminal Cases?

The conclusion that § 1604 confers immunity on foreign sovereigns in criminal prosecutions does not resolve the case. FSIA § 1604 confers immunity subject to the exceptions (as relevant here) in § 1605. One thus must ask whether the § 1605 exceptions apply in criminal cases. Halkbank argues that they do not. The Second Circuit below ruled that they do. The United States takes the same position, albeit as a brief alternative argument to its broader claim that the entire FSIA is inapplicable in criminal cases. 

The text and structure of § 1605 do not, in our view, provide a clear answer to the question whether its exceptions to immunity apply in criminal cases. The relevant exception in Halkbank applies “in any case … in which the action is based upon a commercial activity carried on in the United States by the foreign state” (emphasis added). The Second Circuit reasoned that “any case” can include criminal cases. Considering the phrase “any case” alone, that contention is very plausible. Halkbank points out, however, that some of the exceptions in § 1605—concerning “money damages,” “rights in property,” and arbitration agreements—do not apply in criminal cases. It concludes that it would be “odd to read section 1605 to apply only sometimes in criminal cases, on an exception-by-exception basis, without any indication in the provision’s text that it applies to criminal cases at all.” 

The phrase “the action” in the above-quoted exception in § 1605 might also suggest that Congress was referring to a civil case. In the federal officer removal statute we mentioned above, for example, Congress referred separately to “criminal prosecutions” and “civil actions.” It is also worth keeping in mind that, as the Supreme Court has noted, the FSIA largely codified the “restrictive approach” to sovereign immunity that had developed under international practice. That restrictive approach, unlike the historic absolute approach to immunity, allowed for some civil suits against foreign states but, at least when the FSIA was enacted, did not appear to contemplate the allowance of criminal prosecutions. 

There are other anomalies that would arise if the § 1605 exceptions applied in criminal cases, and that we believe support a reading of § 1605 as applicable to civil cases only. Some of these anomalies are fleshed out in the government’s brief in support of its argument that the entire FSIA (including § 1604) should be construed to be limited to civil cases. Among other things, the government notes that the FSIA sets forth a “reticulated procedural scheme that relates to only civil cases”—the removal provision discussed above, the venue provision (28 U.S.C. § 1391(f)), and other procedural and choice of law rules—“without any similar procedural provisions for criminal cases.”

We have already explained why, contrary to the government, we believe that § 1604 is best construed to confer immunity in both civil and criminal cases. Yet the FSIA’s dominant textual focus on civil matters outside of § 1604 counts against reading the § 1605 exceptions to apply in criminal cases. Importantly, though, that focus is still consistent with reading § 1604 as covering criminal as well as civil cases: There was no need for Congress to make provisions for venue, choice of law, service of process, and the like in criminal cases if, as its plain language suggests, § 1604 does not allow for the exercise of jurisdiction in those cases.

Another argument in support of the conclusion that the § 1605 exceptions are limited to civil suits is that the state court anomaly that we noted above would persist, albeit in less severe form, if § 1605 applied to criminal cases. If § 1605 applied to such cases, then a state prosecutor could indict a foreign state or its instrumentality for a crime that fell within a statutory exception to immunity, and there would be no way for the foreign sovereign to remove the case to federal court. The odd result of an unremovable state prosecution of a foreign sovereign would be tempered somewhat, compared to scenarios sketched above, by the fact that the Supreme Court would clearly have appellate jurisdiction to review the state court interpretation of the exceptions under § 1605. But this outcome makes no sense under the overall FSIA scheme, especially in light of the statute’s broad allowance of removal of civil cases brought against foreign sovereigns.

The government makes much of the fact that the FSIA in 28 U.S.C. § 1330(a) grants subject matter jurisdiction to the federal courts solely over civil cases. But there is nothing incoherent or anomalous about Congress expressly conferring civil jurisdiction over suits against foreign sovereigns in federal court (subject to exceptions) and focusing the FSIA structure on civil suits, while at the same time conferring absolute immunity from criminal jurisdiction in § 1604. Otherwise one must conclude that Congress was approving unremovable criminal prosecutions against foreign sovereigns in state court, a consequence that evaporates if § 1604 provides immunity from criminal prosecution. 

We do not claim that this is the only possible reading of the scope of § 1605, and we acknowledge that the “any case” language in that provision, by itself, suggests that the exceptions apply in civil and criminal cases. But extending the § 1605 exceptions to criminal cases is not compelled, and indeed seems odd, once the “any case” language is read in conjunction with the overall civil case focus of § 1605 and other FSIA provisions. Such an extension also leads to incongruities that seem inconsistent with the FSIA’s structure.

To be sure, this reading of §§ 1604 and 1605 would mean that the federal government would not currently be able to bring criminal prosecutions against foreign state-owned instrumentalities. Such prosecutions have been very rare, as the briefs make clear, and it is unclear how necessary they are to government law enforcement, especially given the government’s ability to criminally prosecute individual foreign officials (who are not covered by the FSIA). In any event, Congress has amended the FSIA a number of times, often at the government’s request, and could do so again if needed. In the process, Congress could specify precisely when prosecutions against foreign sovereign entities can and cannot be brought.

Finally, reading the text of the FSIA to confer immunity on foreign sovereigns and instrumentalities in criminal cases would allow the Court to avoid difficult nonstatutory questions. If the government prevails on its argument that the FSIA (including § 1604’s conferral of immunity) does not apply to criminal cases, then the Court would need to determine the proper scope of nonstatutory immunity (if any) that a foreign sovereign should receive in criminal cases. We will examine possible answers to this question in the next article, but for now it suffices to say that none of the answers to this avoidable question are free from difficulty.