Cross-posted from Just Security.

The Justice Against Sponsors of Terrorism Act (JASTA) is now the law of the United States, Congress having overridden President Obama's veto of the bill. Among other things, JASTA amends the Foreign Sovereign Immunities Act (FSIA) by adding a new terrorism exception that is not limited to designated state sponsors of terrorism. The European Union has claimed that JASTA "conflict[s] with fundamental principles of international law and in particular the principle of State sovereign immunity." Because the version of JASTA that is now law differs significantly from the version I considered back in April, it is worth taking a fresh look at whether JASTA violates international law.

The United States has had a terrorism exception in the FSIA since 1996, the current version of which is found at Section 1605A of Title 28 of US code. Section 1605A provides that a foreign state shall not be immune from suits seeking money damages for personal injury or death caused by certain acts like torture and extrajudicial killing-or material support for such acts-by foreign government officials. But this provision is limited to countries designated by the United States as state sponsors of terrorism (currently Iran, Sudan, and Syria).

The new terrorism exception added by JASTA is not limited to state sponsors of terrorism, but it is limited in other ways. The new Section 1605B provides that a foreign state shall not be immune from suits seeking money damages for personal injury or death, or for injury to property, occurring in the United States that is caused by (1) "an act of international terrorism in the United States;" and (2) a tortious act of a foreign state or its officials "regardless where the tortious act or acts of the foreign state occurred." The tortious act of a foreign state may not, however, be an omission or "constitute mere negligence." 

It is clear that customary international law requires states to recognize foreign sovereign immunity in at least some cases. In the Jurisdictional Immunities Case (Germany v. Italy), for example, the International Court of Justice (ICJ) held that international law requires immunity with respect to torts committed by armed forces during an armed conflict (para. 78). Customary international law rules of immunity-like customary international law rules more generally-must be based on a general and consistent practice of states followed out of a sense of legal obligation or opinio juris (para. 55). States may, of course, go further than international law requires and grant foreign states more immunity from suit as a matter of comity. But "the grant of immunity in such a case is not accompanied by the requisite opinio juris" and therefore does not establish rules of customary international law. See id. Looking at the practice of states with respect to foreign sovereign immunity, it is not always easy to tell where international law stops and international comity begins.

Like many other nations, the United States follows a restrictive theory of foreign sovereign immunity, under which the immunity of foreign states does not extend to their private and commercial acts (acta jure gestionis) but generally does extend to their governmental acts (acta jure imperii). But "generally" does not mean "invariably." The FSIA contains a number of exceptions to immunity that may apply to the governmental acts of foreign states, including the expropriation exception (Section 1605(a)(3)) and the territorial tort exception (Section 1605(a)(5)). The ICJ has also been careful not to hold that the line between immunity and no immunity neatly tracks the line between governmental and non-governmental acts. In Jurisdictional Immunities, it noted at paragraph 64 that "none of the national legislation which provides for a 'territorial tort exception' to immunity expressly distinguishes between acta jure gestionis and acta jure imperii."  And at paragraph 65, the ICJ limited its holding in that case to armed forces during armed conflict, leaving open the question whether other governmental acts might not be covered by immunity. So even if acts of terrorism or providing material support for acts of terrorism were properly considered governmental, such a classification would not by itself entitle those acts to immunity under international law.

Focusing on the new terrorism exception more specifically, there appears to be no general and consistent practice of states followed out of a sense of legal obligation establishing that foreign states are entitled to immunity for acts of terrorism or material support of such acts. To be sure, most states that have statutes governing foreign sovereign immunity do not have exceptions for terrorism. But it is not clear that the states extending foreign sovereign immunity to cover terrorist acts do so out of a sense of legal obligation. Again, as the ICJ noted at paragraph 55 in Jurisdictional Immunities, unless state practice is "accompanied by the requisite opinio juris," it does not establish a rule of customary international law. Significantly, there are two states-the United States and Canada-that do have terrorism exceptions in their foreign sovereign immunity laws. A terrorism exception has been part of U.S. law since 1996 and part of Canadian law since 2012, and neither exception, to my knowledge, has provoked the sort of widespread protests from other nations that one might expect in the case of a clear violation of customary international law. Perhaps that will change with JASTA's new terrorism exception, and such protests would provide new evidence relevant to the international law question. But the lack of protests prior to JASTA is more evidence that a terrorism exception does not violate customary international law.

Critics might point out that Canada's terrorism exception and the old U.S. exception were both limited to designated state sponsors of terrorism, while JASTA's new exception is not. Certainly this difference may be relevant to whether the new exception is good policy, as President Obama pointed out in his veto message. And this difference might also provoke new protests from other states, which would provide more evidence of customary international law regarding terrorism exceptions. But it is hard to see how this difference determines whether JASTA violates customary international law or not. Foreign sovereign immunity typically turns on the nature of the act, and international law does not typically dictate the particular processes a state must use to grant or deny such immunity. If customary international law allows the United States and Canada to deny foreign sovereign immunity when they have designated a particular country as a state sponsor of terrorism, it is not because the United States and Canada have satisfied some customary international law requirement with respect to designation. It is rather because customary international law does not require foreign sovereign immunity for terrorist acts in the first place.

Although new Section 1605B is written as a terrorism exception, it also finds support in the exception-well established under customary international law-for territorial torts. Recall that while Section 1605B does permit a suit for damages to be based on the tortious acts of a foreign state or its officials outside the United States, it also requires both conduct and injury inside the United States-specifically, "an act of terrorism in the United States" and injury or death "occurring in the United States." In Jurisdictional Immunities, the ICJ recounted the extensive state practice establishing an exception to foreign sovereign immunity for torts occurring in the forum state, although the Court also found that this exception did not extend to the activities of armed forces during armed conflicts. (See paragraphs 64-79.) Specifically, the ICJ noted that while the territorial tort exception had "originated in cases concerning road traffic accidents and other 'insurable risks,'" national legislation codifying the exception was written in more general terms (para. 64). Prior to JASTA, U.S. courts had adopted an "entire tort" interpretation of the FSIA's territorial tort exception (§ 1605(a)(5)), requiring that not just the injury but also all of the tortious conduct have occurred in the United States. But it is not clear that such a limitation is required by customary international law. Article 12 of the United Nations Convention on Jurisdictional Immunities of States and Their Properties, for example, would apply the territorial tort exception if the act or omission occurred "in whole or in part" in the territory of the state exercising jurisdiction. And it may be that the U.N. Convention is in fact more generous with immunity than customary international law requires.

In a previous post, I noted that there are various approaches to organizing state practice with respect to foreign sovereign immunity. The results may depend heavily on the baselines from which one begins and the levels of generality at which one reads state practice. Fully addressing those questions would require far more space than this post allows, but perhaps one observation may be made. In determining the customary international law of foreign sovereign immunity with respect to armed forces during armed conflict in Jurisdictional Immunities, the ICJ considered that "the most pertinent State practice is to be found in those national judicial decision which concerned the question whether a state was entitled to immunity in proceedings concerning acts allegedly committed by its armed forces in the course of an armed conflict" (para. 73). The Court found an almost unbroken practice of judicial decisions extending such immunity, even when the acts were committed on a state's own territory (paras. 73-77). There is no similarly unbroken practice of forum states extending immunity to foreign states that provide support for terrorist acts causing injury and death within the forum state.

Powerful arguments have been made that JASTA is bad policy, that it will not in fact help the victims of the 9/11 attacks, and that it will hurt our relations with important allies. I am not disputing those points. My only claim here is that JASTA does not clearly violate customary international law.

Cross-posted from Just Security.

In April, Curt Bradley and Jack Goldsmith wrote in The New York Times that the Justice Against State Sponsors of Terrorism Act (JASTA) then under consideration in the Senate - a bill that would make it easier for victims of the 9/11 terrorist attacks to sue Saudi Arabia - would violate international law and hinder United States' ability to claim sovereign immunity in other nations' courts. I argued in response that whether JASTA would violate international law was far from clear. Since then, the Senate passed a much changed version of JASTA. The revised bill would, among other things, create a new terrorism exception to the Foreign Sovereign Immunities Act (FSIA) for suits claiming damages for injury or death in the United States caused by the combination of an act of international terrorism in the United States and the act of a foreign state or official anywhere in the world.

Curt and Jack write in a new post that this exception would harm US interests because (1) it might lead to unanticipated suits against countries other than Saudi Arabia, and (2) it will "create a broad precedent that can be used against the United States and its allies as an excuse for 'reciprocal' or 'analogous' reductions in immunity even if no suit is brought against those countries in the United States." They suggest that Congress could reduce the damage to US interests by limiting the exception to Saudi Arabia alone. Doing so, they write, "would confine most of the impact of the statute to US-Saudi relations and thereby minimize collateral consequences." I doubt that the more targeted statute Curt and Jack propose would help with either of the problems they identify.

With respect to the foreign relations difficulties that suits under the new terrorism exception might cause, it is true that limiting the exception to Saudi Arabia would avoid the possibility of suits against other countries, but that possibility seems remote. It is hard to think of other countries whose acts contribute to international terrorism in the United States. And it hard is argue that any countries whose acts contribute to international terrorism in the United States should be immune from suit. On the other hand, singling out Saudi Arabia is likely to increase the affront to that country. 

With respect to the precedent such an exception might create for reciprocal legislation in other countries, it is hard to see how a statute targeted at Saudi Arabia alone would provide less of an excuse for reciprocal reductions in the sovereign immunity of the United States and its allies. If another country wanted to allow suits against the United States in its courts for "international terrorism," a US exception aimed at a single country would be all the precedent it would need. Indeed, even without JASTA, a precedent for such foreign legislation may be found in the United States' existing exception for state sponsors of terrorism which allows suits against Iran, Sudan, and Syria and which is not limited to terrorism in the United States. While I do not think that terrorism exceptions violate international law (Canada has one too), I do think they are problematic. As Curt and Jack wrote back in April, "terrorism is often in the eye of the beholder." It would not be surprising for the United States to find itself targeted someday by another country's terrorism exception to sovereign immunity.

If Congress wants to reduce the adverse impacts on the United States that JASTA might cause, I would suggest another option. Rather than create a new terrorism exception to the FSIA, Congress might amend the territorial tort exception not to require that the "entire tort" have occurred in the United States. This was the option proposed in the version of JASTA that I wrote about in April, and it would be just as effective in removing sovereign immunity as a barrier to the 9/11 suits. The territorial tort option would have several advantages.

First, as I explained in my previous post, the territorial-tort exception is well supported by state practice. Although US courts currently interpret the existing exception in the FSIA to require that the "entire tort" have occurred in the United States, it does not appear that customary international law requires this limitation. Building on the well-established territorial tort exception is likely to be less controversial internationally than expanding the more politically charged terrorism exception.

Second, the territorial tort exception is subject to an important exception of its own for military activities during armed conflict. In 2012, the International Court of Justice (ICJ) concluded in the Jurisdictional Immunities Case (Germany v. Italy) that customary international law requires "that a State be accorded immunity in proceedings for torts allegedly committed on the territory of another State by its armed forces and other organs of State in the course of conducting an armed conflict" (para. 78). A number of the specific examples that Curt and Jack worried about in their April piece - like arming Syrian rebels or airstrikes against al-Qaeda and the ISIL - would fall in this category. Using the territorial tort option in JASTA would allow the United States to continue to claim immunity under customary international law for military activities even if JASTA eliminated immunity for non-military activities.

Third, it is important to remember that the territorial tort exception is a territorial tort exception. The only countries that would be able to exercise jurisdiction over the United States reciprocally by enacting a similar exception are those countries in which tortious injury occurs. In April, Curt and Jack raised the possibility that the United States might become subject to suits based on financial support of Israel that results in displacing or killing Palestinians in the West Bank. But under the territorial tort exception, it is only the courts of the Palestinian Authority that would be able to exercise such jurisdiction. The territorial tort option would create no risk of expanded jurisdiction for torts in the courts of third countries.

Of course, the same is technically true of JASTA's new terrorism exception, which requires an act of international terrorism "in the United States." But it is not true of the existing FSIA exception for state sponsors of terrorism, which contains no such limitation. If Congress is really worried about reciprocal legislation by other countries that might strip the United States of its sovereign immunity, the territorial tort option is a safer one than the terrorism option passed by the Senate.

Although I am less concerned than Curt and Jack about the adverse impacts of passing JASTA, they have certainly identified some genuine concerns. The territorial tort option would be a better way of addressing those concerns than limiting the bill to Saudi Arabia. Whatever speculative damage to US relations with other countries might be avoided by limiting the bill as Curt and Jack suggest is likely to be more than offset by the offense that singling out Saudi Arabia would cause. As for the precedent that JASTA would set for reciprocal legislation in other countries, a territorial tort exception would build on a firmer foundation, exempt military activities, and be territorially limited in ways that a terrorism exception - even one limited to a single country - would not.

Cross-posted from Just Security.

Writing in The New York Times last Friday, Curt Bradley and Jack Goldsmith argued that the Justice Against State Sponsors of Terrorism Act (JASTA) would "violate a core principle of international law," the principle of foreign sovereign immunity. At Lawfare, former State Department Legal Adviser John Bellinger seconded their assertion. Earlier in the week, White House Press Secretary Josh Earnest expressed similar concerns. (For a good primer on JASTA and its background, see Steve Vladeck's post here.) The amount of legal analysis one can fit into an op-ed is necessarily limited, and it would be welcome to see Curt, Jack, John, or others flesh out the argument. But, in my view, there are serious problems with the assertion that JASTA would violate customary international law governing sovereign immunity, problems that raise more general questions of how one identifies rules of customary international law.

Curt and Jack summarize their international law argument this way: 

A nation's immunity from lawsuits in the courts of another nation is a fundamental tenet of international law. This tenet is based on the idea that equal sovereigns should not use their courts to sit in judgment of one another. Many nations have tacitly agreed to limit immunity in specified contexts, such as when they engage in certain commercial activities. But apart from those exceptions (or where a binding treaty or Security Council resolution otherwise dictates), international law continues to guarantee immunity, even for alleged egregious crimes.

The first question to ask is where this fundamental tenet of international law comes from. I believe it is common ground that - as the Restatement (Third) of Foreign Relations Law puts it - "[c]ustomary international law results from a general and consistent practice of states followed by them from a sense of legal obligation." If one looks at state practice with respect to foreign sovereign immunity, one finds some situations in which states are consistently held to be immune from the jurisdiction of the courts of other states, other situations in which states are consistently held not to be immune, other situations in which the practice is mixed, and still other situations in which there is no practice at all. How should one make sense of this practice?

Curt and Jack's approach with respect to foreign sovereign immunity seems to be to infer a general rule of immunity based on state practice granting immunity and to treat the state practice denying immunity as establishing exceptions to the general rule. Where the practice is mixed or non-existent, the general rule of immunity would govern because there is not a "general and consistent practice of states" sufficient to create an exception. Of course, this is not the only possible way to read the existing state practice. One could instead infer specific rules of immunity only in those situations where there is a general and consistent practice of granting immunity. Under this approach, where the practice is mixed or non-existent, a general rule of non-immunity would govern. As Curt has recently written on page 34 in his excellent contribution to the book Custom's Future, one must "necessarily make choices about how to describe [state practice], which baselines to apply in evaluating it, and whether or when to extend or analogize it to new situations." I agree. My point is simply that the choices Curt and Jack have made in their analysis of sovereign immunity are choices - and they need to be defended.

One way to defend their approach would be to invoke the International Court of Justice's 2012 decision in the Jurisdictional Immunities Case (Germany v. Italy), which took a similar approach to questions of sovereign immunity. But the ICJ took this approach because the state parties to the dispute both agreed on it. See paragraph 61 ("Both Parties agree that States are generally entitled to immunity in respect of acta jure imperii."). In light of the parties' agreement, the ICJ was certainly justified in starting with a general rule of immunity and then looking for state practice sufficient to support exceptions. But as Curt again has reminded us in his contribution to Custom's Future, "ICJ decisions are technically binding only on the parties and thus should not automatically be treated as the last word by the international community on the content of [customary international law]" (p. 59).

Even if one adopts Curt and Jack's basic approach to sovereign immunity, there remains the question of how broadly or narrowly to read the state practice creating exceptions to the general rule. There is lots of state practice supporting a territorial tort exception to sovereign immunity - that is, an exception for torts that occur in the nation that would exercise jurisdiction over the foreign state. (See Jurisdictional Immunities paragraphs 62-79.) This is what allows Americans injured in traffic accidents by a foreign government employee to sue the foreign state for damages. One might argue that this state practice should be read narrowly to apply only in these sorts of situations. But states that have codified the exception have done so in general terms applicable to any tort.

The US Foreign Sovereign Immunities Act (FSIA) also codifies the territorial tort exception in general terms. (See 28 U.S.C. § 1605(a)(5).) But US courts have interpreted it to require that the "entire tort" occurs within the United States. (See the Second Circuit Court of Appeals 2013 decision from In re Terrorist Attacks on September 11, 2001.) It is this limitation that JASTA would remove. JASTA would still require that there be "physical injury or death, or damage to or loss of property, occurring in the United States," but it would make clear that the territorial tort exception applies "regardless of where the underlying tortious act or omission occurs."

Customary international law does not seem to require the "entire tort" limitation. Article 12 of the United Nations Convention on Jurisdictional Immunities of States and Their Properties would apply the territorial tort exception if the act or omission occurred "in whole or in part" in the territory of the state exercising jurisdiction. Most nations that have codified the exception appear to require some act or omission in their territories, but it is not clear that these nations have done so from a sense of international legal obligation rather than from comity. Even if customary international law were properly read to preclude a nation from applying the territorial tort exception solely on the basis of death and damage within its territory, the application of JASTA to the 9/11 cases would still not violate international law, since the 9/11 attacks clearly involved tortious acts in the United States.

Another, more controversial, path would be to expand the FSIA's terrorism exception, so that it covers state-sponsored terrorism even when the foreign state has not been designated by the State Department as a state-sponsor of terrorism, as is currently required under the FSIA. JASTA would not do this, but Curt and Jack discuss it at some length, so it is worth considering. They assert that the current exception "is almost certainly contrary to international law." If this is true of the existing exception, then it would also be true of an expanded exception.

But I am not so sure that the terrorism exception violates customary international law. First, the United States is not alone in having adopted such an exception - Canada has done so too in Section 6.1 of its State Immunity Act. Second, to my knowledge, these exceptions have not provoked the sorts of widespread protests one might expect from other nations in the event of a clear violation of customary international law. Curt and Jack anticipate this point, explaining that "[t]he controversy has been muted, however, because the exception applies to only a few nations designated as bad actors by the executive branch." This is true, but explaining away the absence of protests is not the same thing as having such protests as evidence of state practice. Third, Curt's and Jack's argument here necessarily depends on the choices they made at the outset about how to organize existing state practice. This is a situation in which the state practice is mixed - two states have such exceptions but most do not. Whether customary international law requires immunity for state-sponsored terrorism depends on whether one begins from a baseline of immunity (as Curt and Jack do) or from a baseline of non-immunity. To repeat what I said above, this is a choice and must be defended.

Of course, Curt and Jack are trying to make a policy argument based on reciprocity too. They write: "If the United States reduces the immunity it accords to other nations, it exposes itself to an equivalent reduction in its own immunity abroad." But their reciprocity argument against JASTA depends on several propositions. First, it depends on the proposition that other states would view the immunity that the United States currently extends (and that JASTA would take away) as required by international law. If not, then they are already free to reduce the immunity they extend the United States, whether JASTA passes or not. Second, it depends on the proposition that other countries would read JASTA broadly to authorize exceptions to sovereign immunity in non-identical situations. Curt and Jack write: "It might appear that the United States has little to fear in lawsuits abroad for acts of terrorism akin to 9/11. But terrorism is often in the eye of the beholder, and reciprocity need not be precise." A broad reading of JASTA is possible, but certainly not inevitable, and the United States would have strong arguments that its practice should be read more narrowly. Finally, Curt and Jack's reciprocity argument depends on the proposition that international law influences the behavior of other states. I believe that to be true, but Jack has co-written an entire book disputing the proposition. See The Limits of International Law.

In the end, I am not certain whether Congress should pass JASTA. Much depends on how one weighs the benefits of providing legal redress for victims of terrorism against the impact on relations with countries like Saudi Arabia that currently cooperate with US counterterrorism efforts. But as a legal matter, the argument that JASTA would violate international law is far from clear.