Three Ways to Leave the INF Treaty

Editor’s note: This piece has been expanded to address the possibility that the United States views the Intermediate-range Nuclear Forces (INF) Treaty as a multilateral, not bilateral, treaty.

President Donald Trump’s Oct. 20 announcement that he intends to remove the United States from the Intermediate-range Nuclear Forces (INF) Treaty with Russia—the latest in a rash of recent treaty withdrawals—has triggered new concerns that the United States’s long-standing commitment to nonproliferation may be flagging. The Trump administration, however, has not provided a lot of details about its full array of intended actions, stating only that it intends to respond to alleged Russian violations by exiting the INF Treaty and resuming some development of the ground-based intermediate-range nuclear and conventional missile systems that are prohibited by the treaty. To date, the United States has not taken steps to formally end U.S. participation in the INF Treaty, though National Security Adviser John Bolton said at a recent bilateral meeting in Moscow that this would come “in due course.”

Writing on Lawfare, Hilary Hurd and Elena Chachko provided an excellent primer on Trump’s decision, the policy context in which it was made and the relevant legal frameworks governing U.S. actions. This post supplements their analysis by highlighting a specific ambiguity in how, precisely, the Trump administration intends to exit the INF Treaty. International law provides for at least three possible exit mechanisms: withdrawal, termination for breach, and the suspension of some or all of the INF Treaty’s obligations. Congress has explicitly endorsed the latter and appears to oppose disposing of the INF Treaty as a whole, as would result from withdrawal or termination for breach. Which mechanism the Trump administration ultimately chooses will reveal a great deal about its intentions and will have implications for future U.S. nonproliferation policies. And if the president acts contrary to Congress’s apparent wishes, he may also lay the foundation for a legal fight over his authority to do so—one that he is not certain to win.

Three Ways Out

The United States originally entered into the INF Treaty as a bilateral agreement with the Soviet Union in 1987. When the Soviet Union dissolved in 1991, several of its successor states—the Russian Federation among them—agreed to assume the Soviet Union’s obligations under various U.S.-Soviet bilateral treaties, including the INF Treaty.

How such circumstances affect treaty obligations is an unsettled area of international law. The Vienna Convention on Succession of States in Respect of Treaties (VCSSRT) asserts that bilateral treaties with a predecessor state become bilateral treaties with each accepting successor state, rather than a multilateral treaty involving all the successor states. That said, neither the Soviet Union nor the United States was a party to the VCSSRT around the time of the Soviet Union’s dissolution. And the VCSSRT is not widely considered to be a reflection of customary international law; as a result, individual state views may vary. By seeking to clarify the status of U.S.-Soviet bilateral treaties through bilateral discussions with each successor state, the United States appears to have embraced a similar view as the VCSSRT. Nevertheless, the current U.S. position on whether the INF Treaty at issue is bilateral or multilateral is not entirely clear.

Given these points, this analysis assumes that the INF Treaty that Trump has threatened to exit from is a bilateral agreement between the United States and Russia—and that obligations between the United States and other Soviet successor states arising from the INF Treaty would thus remain unaffected. The final section below, however, addresses how this analysis might change if the INF Treaty were seen as a multilateral treaty.

If the INF Treaty at issue is a bilateral treaty between the United States and Russia, then international law provides the United States with at least three means by which it may escape the obligations imposed by the document. The most straightforward is withdrawal, which the INF Treaty itself authorizes. Specifically, Article XV of the treaty provides each party with “the right to withdraw ... if it decides that extraordinary events related to the subject matter of this Treaty have jeopardized its supreme interests.” The only condition is that the withdrawing state must provide the other party with six months’ advance notice as well as a description of the “extraordinary events” justifying withdrawal. The treaty provides no limits on what may constitute “extraordinary events” for this purpose, nor any mechanism for contesting a party’s assertion that they exist. This means that both appear to be at the discretion of the invoking party. For bilateral treaties such as the INF Treaty, the withdrawal of either party effectively terminates the treaty, permanently removing any mutual legal obligations it may impose.

Customary international law, meanwhile, provides two additional options. Both are described in Article 60 of the Vienna Convention on the Law of Treaties (VCLT), which is generally seen as an accurate statement of customary international law (including by the United States, which is not a party to the VCLT). As described by Article 60, when one party commits a “material breach” of a bilateral treaty—defined as a “violation of a provision essential to the accomplishment of the object or purpose of the treaty”—the other party may choose to respond in two ways: by “terminating the treaty,” or by “suspending its operation in whole or in part.” Like withdrawal, the former—often called “termination for breach” to distinguish from other possible grounds for termination—renders a bilateral treaty null and void. The latter, meanwhile, leaves the treaty in place but allows the invoking party to suspend some or all of the obligations arising from it until the material breach ends. Subsequent provisions of the VCLT normally require certain procedures before terminating or suspending a treaty, including a three-month waiting period. Unlike Article 60, however, these provisions are generally not viewed as reflections of customary international law, meaning they do not govern nonparties to the VCLT, like the United States.

The Trump administration has been unclear on which of these three options—suspension, termination for breach or withdrawal—it intends to pursue. Both Trump and Bolton have used the terms “terminate” and “withdrawal” in discussing their intent regarding the INF Treaty. Yet they have framed U.S. actions as a response to Russian breaches of the treaty, a precondition that is most directly relevant to suspensions and terminations for breach. And both have implied that they would prefer for Russia to come back into compliance—a possibility that makes sense only in relation to suspension, as both termination for breach and withdrawal would nullify the INF Treaty altogether.

The United States does not appear to be limited in which option it may choose. Since 2014, the United States has openly accused Russia of developing and deploying ground-based missiles and launchers whose range places them in violation of the INF Treaty. These assertions, which most policy experts find credible, could well qualify as material breaches for the purposes of Article 60 of the VCLT. They strike at the treaty's very objectives. Additionally, as Hurd and Chachko describe, there are a number of recent developments that the United States could plausibly identify as “extraordinary events” justifying withdrawal under Article XV of the INF Treaty, including not just Russian violations but also a growing Chinese arsenal of intermediate-range missiles that are not covered by the INF Treaty’s terms. As a result, all three options seem like plausible exit strategies.

Policy Implications

Which exit strategy the Trump administration chooses to pursue could have substantial implications for the future of U.S. nonproliferation policy. Both termination for breach and withdrawal would have the same basic effect: the nullification of the treaty and any international legal obligations arising from it. This would eliminate the INF Treaty’s restrictions, allowing the United States to develop and deploy prohibited weapons systems—as Trump has stated he intends to do. If Russia and the United States were to later decide to renew their commitment to limitations on these weapons systems, however, they would have to negotiate and ratify an entirely new treaty. For the United States, this would almost certainly require the advice and consent of two-thirds of the Senate—a potentially heavy political lift. (While certain types of international agreements may be concluded by the executive branch alone or with the approval of Congress, nonproliferation agreements generally require Senate advice and consent.)

Suspension, meanwhile, could be used to create more tailored responses. Assuming that the Trump administration chooses to suspend provisions restricting U.S. weapons development—the same provisions that Russia has allegedly violated—then suspension would similarly allow the United States to develop ground-based intermediate-range missiles and other covered weapons systems. But limiting suspension to these provisions would leave in place other elements of the INF Treaty, including its various notification requirements (Article IX), inspection and monitoring provisions (Articles XI and XII), and the bilateral Special Verification Commission it establishes (Article XIII). Russia may well respond to suspension by accusing the United States of its own material breach and further suspending or terminating the INF Treaty, but it might hold back if it sees these provisions as beneficial or does not wish to bear the political consequences of ending the treaty altogether. And if both parties were to wish to resume compliance with the INF Treaty in the future, then they would not need to negotiate a new treaty. Instead, as soon as both parties ceased any material breaches—or simply stopped invoking them as a basis for suspending their treaty obligations—the INF Treaty would once again be fully binding.

If the Trump administration chooses to pursue termination or withdrawal, the route endorsed by Bolton before he joined the administration, that is probably a sign that the administration is uninterested in resuming INF Treaty obligations in the future or even in leaving open the possibility for future administrations. This may indicate that Trump and his senior advisers believe that the strategic logic underlying the INF Treaty has permanently and irrevocably shifted or that the treaty was never in U.S. strategic interests. Or it may reflect other motivations altogether. Suspension, meanwhile, could be a sign that the Trump administration’s stated desire to see Russia comply with the terms of the treaty is more than just rhetorical—and that it is willing to leave the INF Treaty’s obligations in place to be picked back up in the future, either by Trump or his successors.

What about the possibility of expanding the INF Treaty to incorporate China’s intermediate-range missiles to which Trump has alluded? While this remains largely hypothetical, doing so would require the same amount of effort regardless of which exit strategy the United States pursues. Negotiating a new, trilateral INF Treaty would involve the same domestic and international procedures as amending the current INF Treaty. The United States may feel that it needs to credibly threaten production of its own intermediate-range missiles and other prohibited equipment to encourage China to enter into such a treaty, but it could do so through suspension so long as Russia is in material breach. Or if Russia and the United States were to once again both come into compliance with the INF Treaty, they could mutually agree to a voluntary suspension in order to increase their mutual weapons production up to certain levels, giving China even more incentive to pursue a negotiated arrangement. Either way, a desire to expand or replace the INF Treaty to address Chinese threats would not favor withdrawal or termination for breach over suspension.

Disagreeing With Congress

Of course, the Trump administration is not the only branch of government with a view on whether and how the United States should exit the INF Treaty. Congress has expressed its position repeatedly, most recently in the National Defense Authorization Act (NDAA) for fiscal 2019, which was enacted in August. Section 1244 of the NDAA actually parrots the language of Article 60 of the VCLT in order to express the sense of Congress that Russian actions “constitute a material breach of the INF Treaty” and that, as a result, the United States “is legally entitled to suspend the operation of the INF Treaty in whole or in part for so long as” Russia’s material breach continues. And Section 1243 directs the president to report to Congress on whether Russia is “in material breach” of its INF Treaty obligations and if the United States remains bound by “the prohibitions set forth in Article VI of the INF Treaty”—i.e., those restricting weapons production—by Jan. 15 of next year. This is a transparent effort to push the Trump administration toward the suspension option that Congress supports.

This support does not, however, appear to extend to termination for breach or withdrawal. Section 1244 also calls for “actions to encourage the Russian Federation to return to compliance with the INF Treaty” and recommends several related measures to accomplish this objective—an effort that would be rendered moot if the Trump administration were to nullify the INF Treaty through termination for breach or withdrawal. Several other recent NDAAs have included similar calls for restoring Russian compliance. The NDAA for fiscal 2018 even contained an “Intermediate-Range Nuclear Forces (INF) Treaty Preservation Act of 2017” within its provisions, which underscores Congress’s ongoing commitment. Together, these provisions send a clear, if not express, message that Congress supports the continued existence of the INF Treaty, not its nullification through termination for breach or withdrawal.

While the president’s authority to exit from the INF Treaty is not contingent on congressional authorization, Congress’s views could still matter. As Hurd and Chachko describe, the Constitution establishes clear procedures for entering into treaties but is silent on who may suspend, terminate or withdraw from them. Recent presidents have asserted the authority to do so unilaterally without Congress’s input—a practice that is now widely accepted, despite a mixed historical record. Members of Congress have occasionally challenged such executive actions in the courts, but their lawsuits have thus far failed to reach the merits due to concerns over standing, ripeness and political-question doctrine. As a result, the executive branch’s assertion that it has the authority to act unilaterally in this area has remained unperturbed.

Yet these litigation barriers are not insurmountable. The contemporary Supreme Court has adopted a narrow view of the political-question doctrine, particularly when it comes to interbranch conflicts. And here Congress has repeatedly enacted legislation that implicitly opposes termination and withdrawal, which seems likely to be sufficient to overcome ripeness concerns. If a private plaintiff could not somehow establish standing, Congress as a whole—or, alternatively, the House or Senate separately—could authorize litigation on its behalf, a strategy that has proven effective at establishing standing in other litigation contexts. One bipartisan group of senators has already proposed legislation aimed at overcoming these barriers in order to launch a legal challenge if Trump were to attempt to withdraw from NATO. If enough members of Congress feel strongly about the INF Treaty, they could take similar action.

If a challenge to Trump’s termination or withdrawal from the INF Treaty were to reach the merits, then the courts would almost certainly view it through the lens of the Youngstown framework. In that context, consistent implied congressional opposition to the termination or withdrawal of the INF Treaty, though substantially less weighty than explicit opposition or an outright statutory prohibition, could well push Trump’s authority to pursue either option into the “lowest ebb” of executive power. There Trump’s action could be sustained only if the court concludes that the Constitution does in fact give him exclusive control over treaty exit. And while the executive branch could argue that the Constitution does so, this case is far from airtight.

None of this means that Trump is barred from withdrawing from the INF Treaty or terminating it for breach. But doing so contrary to the wishes of Congress could bring with it a real, if remote, risk that a legal challenge to his actions could succeed. That litigation might not only result in reversal but also create precedent that sets limits on his authority over U.S. treaty relationships moving forward. For this reason—as well as the strategic considerations discussed above—a prudent administration might merely seek to suspend the INF Treaty for the time being, at least until it can secure a clear sign of congressional support for termination for breach or withdrawal. Whether the Trump administration will pursue this path of least resistance, however, remains to be seen.

A Multilateral INF Treaty?

Of course, it’s not certain that the assumption that the INF Treaty continued as a bilateral U.S.-Russian agreement after the dissolution of the Soviet Union is correct. The United States and Russia may very well depart from the VCSSRT on their views regarding what happens to bilateral treaties during state succession. Even if Moscow and Washington do not, the United States has indicated that it sought to “multilateralize” the INF Treaty through subsequent agreements following the Soviet Union’s collapse, including by involving several successor states other than Russia in the treaty’s implementation. These follow-on arrangements are not readily available, however, leaving it unclear whether and how they changed the INF Treaty’s underlying international legal obligations.

In a similar context, the United States attempted to use a follow-on agreement to convert the Anti-Ballistic Missile (ABM) Treaty with the Soviet Union into a multilateral agreement with various Soviet successor states. That follow-on agreement was never ratified, however, leaving the status of the ABM Treaty ambiguous—but most likely bilateral—until the United States’s 2002 withdrawal. While unlikely, it is possible that there was a similar effort in relation to the INF Treaty that has simply not been widely discussed or readily available. This could in turn change the exit options available to the United States.

If the INF Treaty were in fact multilateral, the United States’s ability to withdraw pursuant to Article XV of the treaty would remain the same, but the consequences would be different. U.S. withdrawal would not terminate a multilateral INF Treaty as it would a bilateral one. Instead, once withdrawal was complete, it would only remove any reciprocal treaty obligations between the United States and other parties. Any obligations between the other parties to the treaty would remain unperturbed.

U.S. options in the event of material breach, however, would change substantially. Article 60(2) of the VCLT provides parties to a multilateral treaty with two different sets of possible responses to material breach by another party, depending on whether the remaining parties act together or individually. If all the non-breaching parties unanimously agree, they may either suspend the treaty in whole or in part; terminate it in its entirety; or terminate it only in relation to the breaching party, effectively expelling that party from the treaty relationship. A party may only respond unilaterally, however, if it is “specially affected” by the breach or if the breach “radically changes the position of every party” with respect to the treaty. In the former circumstances, the party may suspend the operation of the treaty in whole or only as it relates to the relationship between the party itself and the defaulting state, leaving in place reciprocal obligations with other parties under the same treaty. In the latter, a party may suspend the treaty in whole or in part as it relates to its interactions with all the other parties as well.

Assuming that Russia’s violations constitute a material breach, most of these options would appear to be available to the United States. The requirement that the U.S. coordinate with other treaty parties, however, is likely to make termination or full suspension of the treaty or the expulsion of Russia more difficult. Instead, the United States would most likely have to choose between withdrawing and suspending its treaty obligations, assuming that it can credibly claim that it is “specially affected” by Russia’s material breach or that the breach “radically changes” the treaty parties’ relationships—both of which seem plausible. The United States would also have to choose between suspending its treaty obligations in whole or in part, and as they relate to Russia or as they relate to all the other treaty parties.

Under any of these scenarios, the INF Treaty itself would survive the United States’s exit. Yet if the United States were to withdraw, it would still be difficult for it to resume participation at some point in the future, as rejoining the INF Treaty would almost certainly require Senate advice and consent. For this reason, withdrawal would still be a signal that the Trump administration is disinterested in pursuing restrictions of the sort in the INF Treaty in the future (and does not believe future U.S. administrations should do so either). Suspension, meanwhile, would leave open the possibility of rejoining the INF Treaty in full if Russia were to come back into compliance in the future.

As for Congress, its express preference for suspension stands regardless of whether the INF Treaty is bilateral or multilateral. If the Trump administration were to instead withdraw or—in coordination with the other non-breaching parties—terminate the INF Treaty, then it would be inviting the same legal risks that those options would entail for bilateral treaties. For these reasons, suspension would remain the most prudent choice, albeit one that the Trump administration seems no more likely to pursue.