Jump to Content Jump to Main Navigation

s.V Avoiding or Exiting Treaty Commitments, 25 Terminating Treaties

Laurence R Helfer

From: The Oxford Guide to Treaties (1st Edition)

Edited By: Duncan B. Hollis

A newer edition of The Oxford Guide to Treaties is available. Latest edition (2 ed.)
Next Edition: 2nd Edition Latest edition (2 ed.)

From: Oxford Public International Law (http://opil.ouplaw.com). (c) Oxford University Press, 2023. All Rights Reserved.date: 26 May 2024

Treaties, invalidity, termination, suspension, withdrawal — Vienna Convention on the Law of Treaties

(p. 634) 25  Terminating Treaties


An old adage says that no one likes to talk about divorce before a wedding. Yet that is, in effect, precisely what States do when they negotiate new treaties. Buried in the back of most international agreements are provisions that describe procedures for the treaty parties to end their relationship. In addition, no fewer than thirteen articles of the 1969 Vienna Convention on the Law of Treaties (VCLT) contain termination, denunciation, or withdrawal rules that apply when States do not negotiate treaty-specific rules on these topics.1 These ‘exit’ provisions share a distinctive attribute: they authorize one treaty member acting unilaterally or all treaty parties acting collectively to end their obligations under an international agreement.2 The act of exiting pursuant to these provisions is thus distinguishable from a termination or withdrawal in response to breach by another treaty party.3

(p. 635) The structure and operation of treaty exit provisions were long overlooked by most legal scholars and political scientists.4 Over the last decade, that silence has ended as commentators in both fields have devoted fresh attention to the design and use of international agreements in general and treaty flexibility mechanisms in particular. This chapter reviews the findings of this research as it applies to treaty exit rules and discusses its practical, theoretical, and normative implications.

Part I provides an overview of the international law rules governing exit from multilateral and bilateral treaties, including key provisions of the VCLT. Part II highlights the wide variations in the design and invocation of treaty termination, denunciation, and withdrawal clauses using illustrations from a range of subject areas. Part III sets forth a theory of treaty exit. It argues that termination, denunciation, and withdrawal clauses are tools for managing risk—a pervasive feature of international affairs.5 A concluding section briefly identifies avenues for future research on treaty exit that may aid scholars and practitioners alike.

I.  The International Law of Treaty Termination, Withdrawal, and Denunciation

It is helpful to begin with a definition of key terms. Denunciation and withdrawal are used interchangeably to refer to a unilateral act by which a nation that is currently a party to a treaty ends its membership in that treaty.6 In the case of multilateral agreements, denunciation or withdrawal generally does not affect the treaty’s continuation in force for the remaining parties.7 For bilateral agreements, in contrast, denunciation or withdrawal by either party results in the termination of (p. 636) the treaty for both parties. The termination of a multilateral agreement occurs when the treaty ceases to exist for all States parties.8

It is also useful to situate denunciation, withdrawal, and termination within a broader group of mechanisms and doctrines concerning treaty dissolution. For example, Article 59 of the VCLT describes situations in which a treaty ‘shall be considered as terminated if all the parties to it conclude a later treaty relating to the same subject matter’.9 Another temporal incompatibility provision appears in Article 64 of the VCLT, which provides that a treaty terminates if it conflicts with a newly emerged peremptory norm.10 In addition, Article 61 of the VCLT authorizes a party to ‘invoke the impossibility of performing a treaty as a ground for terminating or withdrawing from it if the impossibility results from the permanent disappearance or destruction of an object indispensable for the execution of the treaty’.11 The VCLT’s exit provisions, together with those set forth in the agreement itself, are intended to be exhaustive.12 In practice, however, a treaty may end in other ways, such as upon the performance of all of its obligations, by implication, or by falling into desuetude.13

The foundational principle of State consent governs the design and operation of all treaty exit clauses. At the negotiation stage, State representatives have free reign to choose the substantive and procedural rules that will govern the future cessation of their relationship. Once those rules have been adopted as part of the final text, however, a State that ratifies or accedes to the treaty also accepts any conditions or restrictions on termination, withdrawal, or denunciation that the treaty contains.14 Unilateral exit attempts that do not comply with these conditions or restrictions are ineffective. A State that ceases performance after such an attempt remains a party to the treaty, albeit one that may be in breach of its obligations.15 However, the treaty parties may waive these conditions or restrictions and permit unilateral withdrawal, or terminate the treaty, ‘at any time by consent of all the parties after consultation with the other contracting States’.16

In sum, States are the undisputed masters of treaty exit rules. As illustrated in Part II, they have utilized that power to negotiate a diverse array of termination, (p. 637) denunciation, and withdrawal clauses and to invoke those clauses in a wide variety of circumstances. But what if a treaty omits such clauses entirely? In such a situation, the VCLT provides default rules to govern the end of the parties’ relationship.

A.  Treaties with no provision for termination, denunciation, or withdrawal

The most important—and the most controversial—of these exit default rules is Article 56(1) of the VCLT, which provides that a treaty that contains no provisions for termination, denunciation, or withdrawal ‘is not subject to denunciation or withdrawal unless: (a) it is established that the parties intended to admit the possibility of denunciation or withdrawal; or (b) a right of denunciation or withdrawal may be implied by the nature of the treaty’.17 Article 56(2), in turn, requires twelve months’ notice before a withdrawal or denunciation effectuated pursuant to either of these clauses takes effect.18

Article 56 reflected an uneasy compromise among the members of the International Law Commission (ILC) as to whether States may exit from treaties that do not contain an express denunciation or withdrawal clause. In his 1957 report to the ILC, Sir Gerald Fitzmaurice wrote that such treaties should be assumed to be of ‘indefinite duration, and only terminable … by mutual agreement on the part of all the parties’.19 Fitzmaurice also acknowledged, however, the possibility of several exceptions:

This assumption, however, may be negatived in any case (a) by necessary inference to be derived from the terms of the treaty generally, indicating its expiry in certain events, or an intention to permit unilateral termination or withdrawal; (b) should the treaty belong to a class in respect of which, ex naturae, a faculty of unilateral termination or withdrawal must be deemed to exist for the parties if the contrary is not indicated—such as treaties of alliance, or treaties of a commercial character.20

Sir Humphrey Waldock revisited the issue in a subsequent report to the ILC. The report included a detailed draft article on ‘treaties containing no provisions regarding their duration and termination’.21 Waldock disagreed with Fitzmaurice that there was a presumption against exit from treaties that lack (p. 638) a withdrawal or denunciation clause, and he reviewed State practice to identify the types of agreements for which exit was or was not permitted. The former category included:

(i) a commercial or trading treaty, other than one establishing an international regime for a particular area, river or waterway; (ii) a treaty of alliance or of military co-operation … ; (iii) a treaty for technical co-operation in economic, social, cultural, scientific, communications or any other such matters … ; (iv) a treaty of arbitration, conciliation or judicial settlement [and] ‘a treaty which is the constituent instrument of an international organization’.22

In contrast, Waldock asserted that a treaty ‘shall continue in force indefinitely’ if it:

(a) is one establishing a boundary between two States, or effecting a cession of territory or a grant of rights in or over territory; (b) is one establishing a special international regime for a particular area, territory, river, waterway, or airspace; (c) is a treaty of peace, a treaty of disarmament, or for the maintenance of peace; (d) is one effecting a final settlement of an international dispute; (e) is a general multilateral treaty providing for the codification or progressive development of general international law.23

Treaties not referenced in either list would be subject to a presumption against withdrawal ‘unless it clearly appears from the nature of the treaty or the circumstances of its conclusion that it was intended to have only a temporary application’.24 Waldock’s proposed typology divided the ILC and the VCLT’s drafters.25 The result was the compromise reflected in Article 56(1), quoted above, which refers to the treaty’s (frequently undefined) nature and the parties’ (often ambiguous) intent.

In the years following the ILC reports, scholars have continued to debate the types of treaties whose nature implies a right to withdraw as well as the evidence needed to demonstrate that the parties recognized the possibility of unilateral exit even if they failed to memorialize such an option in the treaty.26 State practice has also been divided on these two issues. Several States have purported to quit multilateral conventions, including those establishing international organizations, notwithstanding the absence of an express exit clause.27 Others have withdrawn without providing the one-year notice that Article 56(2) requires.28 Some of these actions (p. 639) triggered objections from other treaty parties.29 In the case of international organizations, the withdrawing States soon rejoined the organizations, acquiesced in the characterization of their conduct as a temporary cessation of participation, and paid a portion of the dues assessed against them during their absence.30

A recent and high profile dispute involving Article 56 of the VCLT concerns North Korea’s attempt to denounce the International Covenant on Civil and Political Rights (ICCPR) in 1997. In response to the State’s action, the UN Human Rights Committee (HRC) issued a General Comment concluding that the ICCPR was not capable of denunciation or withdrawal.31 Tracking Article 56’s two-part inquiry, the Committee first explained that the absence of an exit clause was not an oversight, inasmuch as the ICCPR’s First Optional Protocol and other contemporaneously negotiated human rights conventions expressly provided for withdrawal.32 It then reasoned that the rights protected by the ICCPR ‘belong to the people living in the territory of the State party’ and cannot be divested by changes in government or State succession.33 As a result, the treaty ‘does not have a temporary character typical of treaties where a right of denunciation is deemed to be admitted, notwithstanding the absence of a specific provision to that effect’.34

(p. 640) The UN Secretary-General also rejected North Korea’s purported denunciation, although he relied on a different legal theory. In the Secretary-General’s view, unilateral exit from the ICCPR was precluded by Article 54 of the VCLT, which he interpreted as permitting North Korea to withdraw only with the consent of all of the other treaty parties.35 The UN Treaty Section referred to this interpretation in a notification sent to these States in response to North Korea’s action, and ‘[a]t least one State, Denmark, sent a Notification to the Secretary-General agreeing with his understanding of Article 54 and stating that it did not consent to [North Korea’s] withdrawal’.36 North Korea ‘appears to have accepted’ that unilateral withdrawal from the ICCPR is not legally permissible.37 In 2000, the country ‘submitted its long overdue second periodic report’ to the HRC and ‘participated in the examination of that report’ in the following year.38

B.  The legal effects of exit

In addition to providing default exit rules for treaties that lack express exit provisions, the VCLT sets forth important principles concerning the legal consequences of exit. Article 70 provides that ‘the termination of a treaty under its provisions or in accordance with the present Convention … releases the parties from any obligation further to perform the treaty’.39 Termination does not, however, ‘affect any right, obligation or legal situation of the parties created through the execution of the treaty prior to’ the date that the termination takes effect.40 Nor does it ‘impair the duty of any State to fulfil any obligation embodied in the treaty to which it would be subject under international law independently of the treaty’41—an implicit reference to customary international law. These limitations are equally applicable to a State that unilaterally withdraws from or denounces a multilateral treaty.42

(p. 641) These rules function as a deterrent to exit. As explained below, the overwhelming majority of denunciation and withdrawal clauses require prior notice to other treaty parties. Notice is also required when a State asserts a basis for terminating or withdrawing from a treaty pursuant to the VCLT.43 During the notice period, the legal obligations of all States parties—including the nation that seeks to withdraw from or terminate the agreement—continue unabated. States also remain responsible for any breaches that occur prior to or during the notice period, a responsibility that survives the State’s withdrawal or the treaty’s end.44 Taken together, these provisions restrict States from using exit to avoid accountability for past violations of international law. They also discourage precipitous and opportunistic withdrawals in which a State seeks to exit and then immediately violate a rule that it previously accepted as binding.45

II.  The Design and Invocation of Termination, Withdrawal, and Denunciation Clauses

In contrast to issues relating to when an international agreement implicitly precludes exit, scholars have devoted less attention to express denunciation, withdrawal, and termination clauses. This Part reviews the findings of several recent studies that reveal a wide variation in the design of these clauses and in the situations in which States invoke the clauses to end their treaty-based relationships. This variation suggests that treaty exit provisions are not mere boilerplate provisions but rather a tool for States to manage the risks of international cooperation.

Treaty provisions that authorize unilateral denunciation and withdrawal are pervasive. They are found in a wide array of multilateral and bilateral agreements governing key transborder regulatory issues, including human rights, arms control, trade, investment, and environmental protection. A 2010 study based on a random sample of 142 international agreements published in the United Nations Treaty Series (UNTS) found that 60 per cent of treaties surveyed contain an exit clause. However, the incidence of these clauses ‘varies by issue area, with human rights (p. 642) agreements almost always incorporating them but more than half of the security agreements in the sample failing to do so’.46

More intriguingly, denunciation clauses impose different types and degrees of restrictions on a State’s ability to withdraw from a treaty and from the obligations it imposes. Handbooks and model treaty rules published by the UN and other international organizations (IOs) on the ‘final clauses’ of treaties demonstrate wide variation in express exit provisions.47 A review of these drafting guides reveals that denunciation and withdrawal clauses cluster around five ideal types:

  1. (1)  treaties that may be denounced at any time;

  2. (2)  treaties that preclude denunciation for a fixed number of years, calculated either from the date the agreement enters into force or from the date of ratification by the State;

  3. (3)  treaties that permit denunciation only at fixed time intervals;

  4. (4)  treaties that may be denounced only on a particular occasion, identified either by time period or upon the occurrence of a particular event; and

  5. (5)  treaties whose denunciation occurs automatically upon the State’s ratification of a subsequent agreement.48

Examples of each type of clause can be found in Section VI(21) of this volume.

Divergences also exist as to the procedures for providing notice of a denunciation, including the period of time that must elapse before a denunciation takes effect, to whom notice must be given, and whether the denouncing State’s obligations continue after the withdrawal takes effect. For some categories of treaties, such as humanitarian law conventions, the effective date of withdrawal is contingent upon external events, such as the cessation of an existing armed conflict.49 Others, most notably bilateral investment agreements (BITs), ‘contain a continuing effects (p. 643) clause that provides that investments made, acquired, or approved prior to the date of the termination of the treaty will be protected by the treaty’s provisions for a further period of ten, fifteen, or twenty years’.50

The most common unilateral exit clauses require advance notice (most often of twelve or six months)51 of a decision to withdraw, sometimes with the additional condition that the treaty have been in force for a specified number of years.52 The large majority of exit provisions do not, however, require a State to justify its decision to withdraw. To the contrary, notices of denunciation and withdrawal are generally short, stylized letters of two or three paragraphs that inform the treaty depository that a State is quitting a particular agreement on a specified future date.53 A few treaties—most notably arms control agreements—require States to explain a decision to withdraw,54 although they generally allow the denouncing nation to decide whether the factual predicate for withdrawal has been satisfied.55 In addition, States often provide explanations when denouncing international labour conventions, although the treaties do not require them to do so.56

Treaty termination clauses are also highly diverse. Negotiators can implicitly address the issue of termination by specifying a treaty’s duration. Common examples include agreements that have a fixed term of years, often with a presumption of renewal or an expectation of renegotiation.57 At the other end of the (p. 644) spectrum are multilateral conventions that are intended to continue in force indefinitely.58 For treaties that include express termination clauses, common provisions include: termination upon the occurrence of a particular event; the entry into force of a later treaty; prior written notice (with cessation of the agreement to take effect after a specified period of time); and the decision of a body established pursuant to the treaty.59 A treaty that does not contain an express termination clause is considered to continue indefinitely, although it may be terminated at any time by consent of all the parties.60 Examples of clauses governing treaty termination and duration are included in Section VI(23) of this volume.

As with unilateral withdrawal provisions, the incidence and type of termination clauses vary by issue area and by type of agreement. Multilateral human rights and environmental protection treaties, for example, often do not include express termination provisions.61 In contrast, many bilateral agreements contain two modes of termination: (i) an initial term after which the treaty ends unless the parties have expressly or tacitly extended it, and (ii) termination upon notice. These ‘flexible provisions enable the parties to keep their options open’.62 A 2005 study based on a random sample of 146 treaties in the UNTS found that two-thirds have a finite duration, but that the percentage of finite treaties varied across subject areas, ranging from a high of nearly 80 per cent of economic agreements to a low of 44 per cent of human rights agreements.63

In contrast to the design of denunciation, withdrawal, and termination clauses, far less attention has been devoted to how often or in which circumstances States actually invoke these provisions.64 The conventional wisdom holds that unilateral exit is an extremely rare event, a supposition based on anecdotal evidence of a few high-profile denunciations and withdrawals. A 2005 study provided a more comprehensive empirical analysis using data collected from the treaty offices of several IOs. The study identified 1,546 instances of denunciation and withdrawal from 5,416 multilateral agreements registered with the UN between 1945 to 2004.65 It also found that, although older treaties are denounced more frequently than (p. 645) recently adopted ones, the rate of exit ‘has held relatively constant or declined only slightly over the last fifty years, even after controlling for the large increase in ratifications and the emergence of new nations in the 1960s and 1970s’.66 Based on these findings, the study concluded that ‘denunciations and withdrawals are a regularized component of modern treaty practice—acts that are infrequent but hardly the isolated or aberrant events that the conventional wisdom suggests’.67

Data from the 2005 study, supplemented with more recent examples, reveal that denunciations and withdrawals can be grouped into four broad categories. These categories are not mutually exclusive. There may be more than one explanation for a State’s decision to exit in a particular instance, and multiple States that exit the same treaty may have different reasons for doing so. Nevertheless, the four categories provide a basic framework for reviewing the empirical landscape of treaty denunciations and withdrawals.

The most high profile and often the most controversial of these involve States that quit a treaty to challenge disfavoured international legal rules or rebuke international institutions. In the late 1990s, for example, three Caribbean States denounced human rights treaties and withdrew from the jurisdiction of international human rights bodies in response to treaty interpretations that resulted in the de facto abolition of the death penalty in those countries.68 More recently, several Latin American States denounced investment agreements and their associated dispute settlement mechanisms, charging that the international investment regime ‘is not transparent, … does not account for the disparity in economic situation of regime members’, is staffed by arbitrators who ‘have an investor bias [and whose] decisions infringe on the legitimate exercise of sovereignty by host countries’.69 These and other examples70 illustrate how States use unilateral exit to disengage from or radically reconfigure existing forms of international cooperation.71

(p. 646) Second, withdrawing from an agreement (or threatening to withdraw) can increase a denouncing nation’s negotiating leverage with other States parties and its influence in IOs. The United States’ denunciation in the 1970s and 1980s of the agreements establishing the International Labour Organization (ILO) and the United Nations Educational, Social and Cultural Organization (UNESCO) follow this pattern. In each instance, the United States used exit and threats of exit—and the loss of organizational support and funding these entailed—to pressure the organizations’ members to change their behaviour, after which it rejoined the treaties. The Soviet Union and its allies pursued a similar approach in the 1950s, temporarily withdrawing from but later rejoining the World Health Organization (WHO), UNESCO, and the ILO. In the mid-1990s, the United States and the European Communities used an exit strategy to close the Uruguay Round of trade talks that created the World Trade Organization (WTO). They withdrew from the old General Agreement on Tariffs and Trade—a treaty that gave special benefits to developing States—and then ratified the new WTO Agreement as a ‘single undertaking’, forcing developing States to accept a broad package of obligations favourable to US and European interests.72 These examples reveal how States use exit and threats of exit to increase their voice within treaty-based negotiating forums and to reshape treaty commitments to more accurately reflect their interests.73

A third circumstance concerns what might be termed ‘forced exit’, which occurs when one State or group of States requires another nation to withdraw from a treaty as a condition of joining or retaining membership in an IO. The most striking example of forced exit occurred in the mid-2000s, when the European Union (EU) demanded that States seeking EU membership denounce BITs with the United States that had been in force since the early 1990s. The EU ‘announced that the treaties, which broadly prohibited, among other things, discrimination against foreign investment, violated European (protectionist) laws that had governed the region’s economic policies for nearly fifty years’.74 Commentators have noted the possibilities of similar forced exits from bilateral trade and investment agreements between the United States and the members of Mercosur, South America’s largest regional trading block.75 These examples starkly illustrate that exit sits at the intersection of law and power in international relations.76

(p. 647) A fourth and very different type of exit occurs when the denunciation of one treaty is linked to joining a later-negotiated agreement that relates to the same subject matter. In the ILO and the International Maritime Organization, for example, the ratification of certain revising conventions or protocols triggers the automatic or compulsory denunciation of earlier agreements. Similarly, a few Council of Europe treaties that supersede earlier agreements on the same topic require ratifying States to denounce the earlier agreements as a condition of membership. Such paired treaty actions update a State’s international obligations without diminishing its overall level of commitment. Unlike the three circumstances discussed above, denunciations and withdrawals of this type are also fundamentally cooperative in nature. They often occur in groups or waves, a pattern which suggests an attempt to shift to a new equilibrium point that benefits all or most States parties.77

III.  Exit Clauses as Risk Management Tools

The wide variation in the design and use of termination, denunciation, and withdrawal clauses suggests that States pay close attention to the conditions and contours of exit, both when they negotiate international agreements and when they evaluate the costs and benefits of continuing to comply with those agreements over time. To many commentators anxious to demonstrate that States obey international law, the pervasiveness of these exit options is not something to be advertised, let alone celebrated.78 For risk-averse governments, however, exit clauses are a rational response to a world plagued by uncertainty, one in which States negotiate commitments with imperfect information about the future and the preferences of other treaty parties.

To see why this is so, consider the perspective of government officials negotiating a treaty. In an ideal world, the negotiators would hammer out an agreement that maximizes joint gains and induces all affected States to join the treaty and invest the material resources and political capital needed to comply with its terms. In practice, however, numerous types of uncertainty limit the ability of negotiators to achieve such a salutary result. These include uncertainty about the preferences of other States, uncertainty about their behaviour, and uncertainty about future events such as ‘unanticipated circumstances or shocks’, or ‘new demands from domestic coalitions or clusters of States wanting to change important rules or procedures’.79(p. 648) Negotiators must also contend with the fact that treaties are voluntary instruments; even for States that actively participate in the drafting process, ratification is never guaranteed. The consensual nature of international agreements means that States will join a treaty only if the anticipated benefits of doing so outweigh the expected costs.

In terms of their benefits, denunciation, withdrawal, and termination clauses reduce the uncertainties that are pervasive in international affairs. They do so by providing what is, in effect, an insurance policy—a low cost option for States to end treaty-based cooperation if an agreement turns out badly. All other things being equal, exit clauses encourage ratification by a larger number of States than would join the treaty in the absence of such a clause.80 Such clauses also enable the negotiation of deeper or broader commitments than would otherwise be attainable.81 And they encourage treaty parties to address openly the consequences of changed circumstances rather than remaining as parties but committing surreptitious violations.82 Taken together, these factors counsel negotiators to include broad and permissive exit provisions in treaties.

Although the ex ante benefits of exit are considerable, treaties that permit easy denunciation may also create impediments to future cooperation. One concern is that a State will invoke a denunciation or withdrawal clause (or credibly threaten to do so) whenever economic, political, or other pressures make compliance costly or inconvenient. Seen from this vantage point, an exit provision enables a State to quit a treaty and, after the withdrawal takes effect, engage in conduct that would have been a violation had it remained a member of the agreement. But the risks of exit extend beyond such opportunistic behaviour. States that prefer to cooperate but fear that their treaty partners may withdraw from the agreement also have less incentive to invest in treaty compliance. These deterrents to cooperation favour making treaties more durable and binding by eliminating or restricting exit opportunities—a position directly contrary to the ex ante perspective that favours broad exit rights.

These competing perspectives on the benefits and costs of exit suggest that a key challenge that negotiators face is not to close exit options but rather to set optimal conditions on exit ex ante so as to deter opportunistic invocations of exit ex post. Exit clauses that are too capacious will encourage self-serving denunciations and lead to a breakdown in cooperation. Exit provisions that are too onerous will reduce such behaviour, but may prevent the parties from reaching agreement in the (p. 649) first instance or trigger widespread violations if the costs of compliance rise unexpectedly. These alternative vantage points help to explain the diversity of exit clauses and the different uses of those clauses reviewed in Part II above. Such variation reflects the efforts of negotiators to calibrate the costs of exit in light of the often divergent preferences of States and the myriad transborder cooperation problems they seek to resolve.


This chapter has analysed the different mechanisms that States invoke to end their treaty-based relationships, including express termination, denunciation, and withdrawal clauses and the default rules provided by the VCLT. The chapter has argued that these ‘exit’ provisions help States to mitigate the uncertainties that are endemic to international affairs.

In closing, it is important to stress that treaty exit clauses do not exist in a vacuum. Rather, they operate in tandem with other flexibility devices—such as reservations, amendment rules, escape clauses, and renegotiation provisions—that treaty-makers use to manage risk. The relationship among these flexibility tools has long been a concern of government officials and commentators interested in improving the treaty-making process.83 It would be useful to link these studies to recent scholarship analysing the form and substance of international agreements.84 Such research might consider how States select from among a diverse array of flexibility mechanisms, and how they actually exercise the mechanisms available to them. The findings of these studies could also aid negotiators in designing treaties that more effectively address the diverse array of legal issues that are subject to international regulation.

Recommended Reading

  • M Akehurst, ‘Withdrawal from International Organisations’ (1979) 32 Current Legal Problems 143
  • RB Bilder, Managing the Risks of International Agreement (University of Wisconsin Press, Madison 1981)
  • E Evatt, ‘Democratic People’s Republic of Korea and the ICCPR: Denunciation as an Exercise of the Right of Self-defence?’ (1998) 5 Australia J Human Rts 215
  • N Feinberg, ‘Unilateral Withdrawal from an International Organization’ (1963) 39 BYBIL 189
  • (p. 650) LR Helfer, ‘Exiting Custom: Analogies to Treaty Withdrawals’ (2010) 21 Duke J Comp & Intl L 65
  • LR Helfer, ‘Exiting Treaties’ (2005) 91 Virginia L R 157
  • LR Helfer, ‘Flexibility in International Agreements’ in J Dunoff and M Pollack (eds), International Law and International Relations: Insights from Interdisciplinary Scholarship (CUP, Cambridge 2012)
  • LR Helfer, ‘Overlegalizing Human Rights: International Relations Theory and the Commonwealth Caribbean Backlash Against Human Rights Regimes’ (2002) 102 Columbia L Rev 1832
  • MF Imber, The USA, ILO, UNESCO and IAEA: Politicization and Withdrawal in the Specialized Agencies (Palgrave Macmillan, New York 1989)
  • N Kontou, The Termination and Revision of Treaties in the Light of New Customary International Law (Clarendon Press, Oxford 1994)
  • B Koremenos and A Nau, ‘Exit, No Exit’ (2010) 21 Duke J Comp & Intl L 81
  • T Meyer, ‘Power, Exit Costs, and Renegotiation in International Law’ (2010) 51 HILJ 379
  • B Ress, ‘Ex Ante Safeguards Against Ex Post Opportunism in International Treaties: Theory and Practice of International Public Law’ (1994) 150 J Institutional and Theoretical Economics 279
  • E Schwelb, ‘Withdrawal from the United Nations: The Indonesian Intermezzo’ (1967) 61 AJIL 661
  • H Tobin, The Termination of Multipartite Treaties (Columbia University Press, New York 1933)
  • Y Tyagi, ‘The Denunciation of Human Rights Treaties’ (2008) 79 BYBIL 86
  • K Widdows, ‘The Unilateral Denunciation of Treaties Containing No Denunciation Clause’ (1982) 53 BYBIL 83


1  VCLT Arts 42–5, 54–6, 65–8, 70–1. The VCLT applies only to treaties between States. Agreements involving international organizations are governed by the Vienna Convention on the Law of Treaties between States and International Organizations or between International Organizations (adopted 21 March 1986, not yet in force) [1986] 25 ILM 543 (‘1986 VLCT’). The first seventy-two articles of the 1986 VCLT—which is widely regarded as reflecting customary international law—address the same subjects as Arts 1 through 72 of the original VCLT. Anthony Aust, Modern Treaty Law and Practice (2nd edn CUP, Cambridge 2007) 7–8 and n7. Inasmuch as the 1986 VCLT’s provisions relating to termination, denunciation, or withdrawal are materially indistinguishable from those of the VCLT, this chapter focuses only on the VCLT.

2  LR Helfer, ‘Exiting Treaties’ (2005) 91 Virginia L R 1579, 1582 (explaining that ‘exit clauses create a lawful, public mechanism for a state to terminate its treaty obligations or withdraw from membership in an intergovernmental organization’).

3  Eg MM Gomaa, Suspension or Termination of Treaties on Grounds of Breach (Martinus Nijhoff, The Hague 1996) 167–8; S Rosenne, Breach of Treaty (Grotius, Cambridge 1985) 117–25; AE David, The Strategy of Treaty Termination: Lawful Breaches and Retaliations (Yale University Press, New Haven 1975) 159–202. For a discussion of treaty breach, see Chapter 23. It is also important to distinguish denunciation, withdrawal, and termination of a treaty pursuant to its terms from the termination or suspension of a treaty due to supervening impossibility or fundamental change of circumstances. For further discussion of those topics, see Chapter 24, Parts I and II, 606 et seq. For a review of the literature on the design and use of treaty suspension and derogation clauses, see LR Helfer, ‘Flexibility in International Agreements’ in J Dunoff and M Pollack (eds), International Law and International Relations: Taking Stock: Insights from Interdisciplinary Scholarship (CUP, Cambridge 2012).

4  Eg A McNair, The Law of Treaties (Clarendon Press, Oxford 1961) 510 (stating that treaty clauses permitting unilateral denunciation ‘occur[] so frequently that [they] hardly require[] illustration’ or discussion).

5  RB Bilder, Managing the Risks of International Agreement (University of Wisconsin Press, Madison 1981), is an early and influential analysis of treaty flexibility mechanisms as risk management tools.

6  UN Office of Legal Affairs, Final Clauses of Multilateral Treaties Handbook (UN Sales No E04V3 2003) 109 (‘Final Clauses Handbook’) (‘The words denunciation and withdrawal express the same legal concept’). Anthony Aust asserts that ‘although the term denunciation is sometimes used in relation to a multilateral treaty, the better term is withdrawal, since if a party leaves a multilateral treaty that will not normally result in its termination’. A Aust, Handbook of International Law (CUP, Cambridge 2010) 93. Although there is much to recommend this view, in fact multilateral agreements use both terms interchangeably.

7  There are a number of exceptions. If a multilateral agreement, such as the Comprehensive Nuclear-Test-Ban Treaty (adopted 10 September 1996, not yet in force) [1996] 35 ILM 1439, Art XIV(1), requires a particular State to join the agreement as a condition of its entry into force and that State subsequently withdraws from the treaty, ‘it can be assumed that … the treaty would be terminated’. ME Villiger, Commentary on the 1969 Vienna Convention on the Law of Treaties (Martinus Nijhoff, Leiden 2009) 694. Termination also occurs when a multilateral treaty specifies that it shall no longer be in force if denunciations reduce the parties to below a specified number. Eg Convention on the Political Rights of Women (adopted 20 December 1952, entered into force 7 July 1954) 193 UNTS 135, Art 8(2) (providing that the convention ‘shall cease to be in force as from the date when the denunciation which reduces the number of Parties to less than six becomes effective’). However, the default rule in VCLT Art 55 allows the treaty to continue in force unless it specifies a minimum number of required parties.

8  Villiger (n 7) 685.

9  VCLT Art 59 (identifying those situations as occurring when ‘(a) it appears from the later treaty or is otherwise established that the parties intended that the matter should be governed by that treaty; or (b) the provisions of the later treaty are so far incompatible with those of the earlier one that the two treaties are not capable of being applied at the same time’).

10  Ibid Art 64. For additional discussion, see N Kontou, The Termination and Revision of Treaties in the Light of New Customary International Law (Clarendon Press, Oxford 1994).

11  VCLT Art 61. For a discussion of impossibility, see Chapter 24, Part I, 606 et seq.

12  VCLT Art 42(2) (‘The termination of a treaty, its denunciation or the withdrawal of a party, may take place only as a result of the application of the provisions of the treaty or of the present Convention’).

13  Aust (n 1) 305–7.

14  VCLT Art 54(a) (‘The termination of a treaty or the withdrawal of a party may take place … in conformity with the provisions of the treaty’); Villiger (n 7) 685 (characterizing Art 54(a) as ‘independent of the will of the parties in a particular situation’). Reservations to withdrawal, denunciation, or termination clauses are extremely rare.

15  Helfer (n 2) 1589 n23.

16  VCLT Art 54(b).

17  Ibid Art 56(1). Another default rule is the presumption that exit rights ‘may be exercised only with respect to the whole treaty’. Villiger (n 7) 564; VCLT Art 44(1) (‘A right of a party, provided for in a treaty or arising under article 56, to denounce, withdraw from or suspend the operation of the treaty may be exercised only with respect to the whole treaty unless the treaty otherwise provides or the parties otherwise agree’).

18  VCLT Art 56(2). Whether these VCLT rules constitute customary international law is an open question, but at least one scholar insists they have such status. See eg Villiger (n 7) 689 (discussing customary law basis of VCLT Art 54(b)); ibid 705 (noting it was ‘doubtful’ if Art 56 reflected customary international law at the time of the VCLT’s adoption, but contending that it has since ‘generated a new rule of customary law’).

19  GG Fitzmaurice, ‘Second Report on the Law of Treaties’ [1957] YBILC, vol II, 16, 22.

20  Ibid.

21  H Waldock, ‘Second Report on the Law of Treaties’ [1963] YBILC, vol II, 36 (draft Art 17).

22  Ibid draft Art 17(3)(a) and (b).

23  Ibid draft Art 17(4).

24  Ibid draft Art 17(5).

25  Eg T Christakis, ‘Article 56’ in O Corten and P Klein (eds), The Vienna Convention on the Law of Treaties: A Commentary (OUP, Oxford 2011) vol II, 1257–66; M Fitzmaurice and O Elias, Contemporary Issues in the Law of Treaties (Eleven International Publishing, Utrecht 2005) 357.

26  Eg K Widdows, ‘The Unilateral Denunciation of Treaties Containing No Denunciation Clause’ (1982) 53 BYBIL 83 (summarizing these debates).

27  Prominent examples of denunciations of multilateral organizations included the withdrawal of Indonesia from the UN in 1965; of Czechoslovakia, Hungary, and Poland from UNESCO in the 1950s; and of the Soviet Union and eight Eastern European States from WHO in the same period. N Feinberg, ‘Unilateral Withdrawal from an International Organization’ (1963) 39 BYBIL 189, 204–11; E Schwelb, ‘Withdrawal from the United Nations: The Indonesian Intermezzo’ (1967) 61 AJIL 661, 666–71; Widdows (n 26) 99–102. For an overview of State practice prior to the Second World War, see Christakis (n 25) 1262–3.

28  For example, the US purported to withdraw from the Optional Protocol to the Vienna Convention on Consular Relations with immediate effect. J Quigley, ‘The United States’ Withdrawal from International Court of Justice Jurisdiction in Consular Cases: Reasons and Consequences’ (2009) 19 Duke J Comp & Intl L 263, 265–6, 292–3 (‘United States, in its communication to the U.N. Secretary-General gave no time period, apparently purporting to make its withdrawal effective immediately’).

29  A notable example occurred in 1971 when Senegal notified the UN Secretary-General of denunciations of the Convention on the Territorial Sea and the Contiguous Zone (adopted 29 April 1958, entered into force 10 September 1964) 526 UNTS 205, and the Convention on Fishing and Conservation of the Living Resources of the High Seas (adopted 29 April 1958, entered into force 20 March 1966) 559 UNTS 285. Senegal justified its actions by asserting that the treaties ‘“profited the wealthier, the better equipped, and not the under-developed, the poorer” who could only attest, powerless, to the over-exploitation of biological resources situated in high seas areas adjacent to their territorial waters’. D Bardonnet, ‘La denunciation par le gouvernement sénégalais de la Convention sur la mer territoriale et la zone continiguë et de la Convention sur la pêche et la conservation des sources biologiques de la haute mer’ (1972) 18 Annuaire Française de Droit International 123, 133 (quoting declaration of Senegalese President). In response, the United Kingdom objected on the ground that the conventions were ‘not susceptible to unilateral denunciation’ and that it ‘therefore cannot accept the validity or effectiveness of the purported denunciation by the Government of Senegal’. UN Treaty Collection, Law of the Sea, Convention on the Territorial Sea and the Contiguous Zone, Multilateral Treaties Deposited with the Secretary General (MTDSG) <http://treaties.un.org/Pages/ViewDetails.aspx?src=TREATY&mtdsg_no=XXI-1&chapter=21&lang=en, Chapter XXI>.

30  Eg M Akehurst, ‘Withdrawal from International Organisations’ (1979) 32 Current Legal Prob 143, 146–49.

31  UNHRC, ‘General Comment 26’ (1997) UN Doc CCPR/C/21/Rev1/Add8/Rev1 [5].

32  Ibid [2].

33  Ibid [4].

34  Ibid [3]; E Evatt, ‘Democratic People’s Republic of Korea and the ICCPR: Denunciation as an Exercise of the Right of Self-defence?’(1998) 5 Australia J Human Rts 215, 219–20. In part in reliance on the Committee’s analysis, most commentators have concluded that human rights treaties that lack an express exit clause—including the International Covenant on Economic, Social and Cultural Rights (adopted 16 December 1966, entered into force 3 January 1976) 993 UNTS 3; Convention on the Elimination of All Forms of Discrimination Against Women (adopted 18 December 1979, entered into force 3 September 1981) 1249 UNTS 13; International Convention for the Protection of All Persons from Enforced Disappearance (adopted 20 December 2006, entered into force 23 December 2010) [2007] 4 IHRR 582, and the Second Optional Protocol to the ICCPR Aiming at the Abolition of the Death Penalty (adopted 15 December 1989, entered into force 11 July 1991) 1642 UNTS 414—are not susceptible to unilateral denunciation or withdrawal. Eg Helfer (n 2) 1642 n172; Villiger (n 7) 703; Y Tyagi, ‘The Denunciation of Human Rights Treaties’ (2008) 79 BYBIL 86, 126–33.

35  Article 54 provides that ‘[t]he termination of a treaty or the withdrawal of a party may take place: (a) in conformity with the provisions of the treaty; or (b) at any time by consent of all the parties after consultation with the other contracting States’. VCLT Art 54.

36  E Bates, ‘Avoiding Legal Obligations Created by Human Rights Treaties’ (2008) 57 ICLQ 751, 755; H Klingenberg, ‘Elements of Nordic Practice 1998: Denmark’ (1999) 68 Nordic J Intl L 163, 164 (indicating that ‘[o]ther states communicated similar responses’).

37  Aust (n 1) 291.

38  Bates (n 36) 755–6.

39  VCLT Art 70(1)(a).

40  Ibid Art 70(1)(b).

41  Ibid Art 43.

42  Ibid Arts 43, 70(2). A few multilateral human rights and humanitarian law conventions reiterate that an exiting State’s obligations continue until the date that a denunciation or withdrawal takes effect. Eg American Convention on Human Rights (adopted 22 November 1969, entered into force 18 July 1978) 1144 UNTS 123, Art 78(2). The drafters of other multilateral agreements, accepting Art 70’s invitation to contract around the VCLT default rules, expressly indicate which obligations survive a State’s unilateral exit. Eg UN Convention on the Law of the Sea (10 December 1982, entered into force 16 November 1994) 1833 UNTS 397, Art 317(2) (providing that a denunciation does not affect the ‘financial and contractual obligations’ accrued while a State was a party).

43  VCLT Art 65(1) (‘A party which, under the provisions of the present Convention, invokes … a ground for … terminating [a treaty], withdrawing from it or suspending its operation, must notify the other parties of its claim’); see also ibid Art 67 (requiring that notices of withdrawal, denunciation, or termination be in writing and be made by officials with actual treaty-making powers or those possessing full powers); ibid Art 68 (providing that notice of withdrawal, denunciation or termination may be revoked at any time before taking effect).

44  Eg Roodal v Trinidad and Tobago, Case 12.342, Inter-Am Comm’n HR 89, OEA/ser L/V/II114, doc 5 rev (2001) <http://cidh.org/annualrep/2001eng/TT12342.htm> (concluding that ‘[n]otwithstanding Trinidad and Tobago’s denunciation of the Convention [on 26 May 1999], the Commission will retain jurisdiction over complaints of violations of the Convention by Trinidad and Tobago in respect of acts taken by that State prior to’ the date the denunciation became effective as well as over ‘acts taken by the State prior to [that date] even if the effects of those acts continue or are not manifested until after that date’).

45  LR Helfer, ‘Exiting Custom: Analogies to Treaty Withdrawals’ (2010) 21 Duke J Comp & Intl L 65, 78–9.

46  B Koremenos and A Nau, ‘Exit, No Exit’ (2010) 21 Duke J Comp & Intl L 81, 106. The study’s findings may be influenced by the fact that a large majority of treaties in the random sample are bilateral. Ibid 112–19. An earlier survey found that just nearly 90 per cent of bilateral and multilateral treaties registered with the UN between 1967 and 1971 contained denunciation or withdrawal clauses. Widdows (n 26) 95.

47  In 1951, 1957, and 2003, the UN Office of Legal Affairs published a Handbook of Final Clauses. The Handbook is a reference tool of examples from existing treaties intended to assist State representatives who draft international agreements. Eg Final Clauses Handbook (n 6). See also Committee of Ministers, Council of Europe, Model Final Clauses for Conventions and Agreements Concluded within the Council of Europe (February 1980) <http://conventions.coe.int/Treaty/EN/Treaties/Html/ClausesFinales.htm>; H Blix and JH Emerson (eds), The Treaty Maker’s Handbook (Oceana Publications, New York 1973) (‘Treaty Maker’s Handbook’) (collecting examples of final clauses).

48  Helfer (n 2) 1597 (reviewing handbooks of final clauses).

49  Common Art 63 of the four Geneva Conventions of 1949 provides that a denunciation takes effect one year after notification. However, a notice of denunciation ‘made at a time when the denouncing Power is involved in a conflict shall not take effect until peace has been concluded, and until after operations connected with the release and repatriation of the persons protected by the present Convention have been terminated’. Eg Geneva Convention for the Amelioration of the Condition of the Wounded and Sick in Armed Forces in the Field (opened for signature 12 August 1949, entered into force 21 October 1950) 75 UNTS 31, Art 63.

50  JW Salacuse, ‘The Emerging Global Regime for Investment’ (2010) 51 HILJ 427, 471–2.

51  The date that a notice of denunciation, withdrawal, or termination takes effect is calculated differently depending on whether the treaty lists the notice period in days or months. For further discussion, see Chapter 7, Part IV, 204.

52  Eg Koremenos and Nau (n 46) 106–7; G Haraszti, Some Fundamental Problems of the Law of Treaties (Akadémiai Kiadó, Budapest 1973) 264.

53  Treaty Maker’s Handbook (n 47) 114–16.

54  Eg Convention on the Prohibition of the Use, Stockpiling, Production and Transfer of Anti-Personnel Mines and on their Destruction (adopted 18 September 1997, entered into force 1 March 1999) 2056 UNTS 211, Art 20 (‘Land Mines Convention’) (requiring a State to provide ‘a full explanation of the reasons motivating [its] withdrawal’). For additional analysis, see A Chayes, ‘An Inquiry into the Workings of Arms Control Agreements’ (1972) 85 Harvard L Rev 905, 957–8.

55  Eg Treaty on the Limitation of Anti-Ballistic Missile Systems (US-USSR) (adopted 26 May 1972, entered into force 3 October 1972) 944 UNTS 13, Art XV (recognizing a right to withdraw if either party ‘decides that extraordinary events related to the subject matter of this Treaty have jeopardized its supreme interests’ and requiring notice ‘of the extraordinary events the notifying Party regards as having jeopardized its supreme interests’). The events surrounding North Korea’s exit from the Nuclear Non-Proliferation Treaty, which contains a similarly worded withdrawal clause, engendered extensive analysis of the ‘extraordinary events’ standard and its self-judging character. Eg M Asada, ‘Arms Control Law in Crisis? A Study of the North Korean Nuclear Issue’ (2004) 9 J Conflict & Sec L 331; AF Perez, ‘Survival of Rights Under the Nuclear Non-Proliferation Treaty: Withdrawal and the Continuing Right of International Atomic Energy Agency Safeguards’ (1994) 34 VJIL 749.

56  K Widdows, ‘The Denunciation of International Labour Conventions’ (1984) 33 ICLQ 1052, 1055.

57  Eg JR Crook, ‘United States, Russia Sign New Strategic Arms Reduction Treaty; Senate Begins Hearings’ (2010) 104 AJIL 514, 515 (explaining that the duration of the New START Treaty between the Russian Federation and the United States ‘will be ten years, unless superseded by a subsequent agreement’ and that the parties ‘may agree to extend the Treaty for a period of no more than five years’); B Koremenos, ‘Can Cooperation Survive Changes in Bargaining Power? The Case of Coffee’ (2002) 31 J Legal Studies 259, 274–6 (‘Koremenos, Coffee Agreements’) (analysing International Coffee Agreements, which had durations of five to seven years with the expectation of renegotiation); B Koremenos, ‘Loosening the Ties that Bind: A Learning Model of Agreement Flexibility’ (2001) 55 Intl Org 289, 305 (‘Koremenos, Loosening the Ties’) (analysing the Nuclear Non-Proliferation Treaty, which ‘entered into force in 1970 for a period of twenty-five years’ and whose parties ‘reconvened [in 1995] and decided to extend the treaty indefinitely’).

58  Land Mines Convention (n 54) Art 20(1) (‘This Convention shall be of unlimited duration’).

59  Eg Aust (n 1) 278–88; Final Clauses Handbook (n 6) 114–17.

60  VCLT Art 54(b).

61  Final Clauses Handbook (n 6) 117.

62  Aust (n 1) 284.

63  B Koremenos, ‘Contracting around International Uncertainty’ (2005) 99 American Political Science Rev 549, 557.

64  This omission is especially striking with regard to treaty terminations. For two notable exceptions, see Koremenos, Coffee Agreements (n 57) (analysing International Coffee Agreements); Koremenos, Loosening the Ties (n 57) (analysing the Nuclear Non-Proliferation Treaty).

65  Helfer (n 2) 1601–7. Of the 5,416 multilateral agreements in the study, 191, or 3.5 per cent, have been denounced at least once. This small percentage suggests that a few multilateral treaties have turned out badly and resulted in withdrawals by multiple States.

66  Ibid 1604–05.

67  Ibid 1602.

68  LR Helfer, ‘Overlegalizing Human Rights: International Relations Theory and the Commonwealth Caribbean Backlash Against Human Rights Regimes’ (2002) 102 Columbia L Rev 1832 (analysing denunciations of the American Convention on Human Rights and the First Optional Protocol to the ICCPR by Guyana, Jamaica, and Trinidad and Tobago).

69  Salacuse (n 50) 469. Bolivia (in 2007) and Ecuador (in 2010) withdrew from the Washington Convention establishing the International Centre for the Settlement of Investment Disputes (ICSID) (18 March 1965, entered into force 14 October 1966) 575 UNTS 159. During the same period, Ecuador denounced nine BITs and Venezuela terminated its BIT with the Netherlands. The States have also announced their intention to renegotiate other BITs to which they are parties. Ibid 469–70.

70  A related phenomenon is exit that is associated with the creation of a new treaty regime. In 1992, for example, Iceland denounced the International Convention for the Regulation of Whaling (2 December 1946, entered into force 10 November 1948) 161 UNTS 72, and, together with other pro-whaling States, established the North Atlantic Marine Mammal Commission. DD Caron, ‘The International Whaling Commission and the North Atlantic Marine Mammal Commission: The Institutional Risks of Coercion in Consensual Structures’ (1995) 89 AJIL 154, 155. Iceland rejoined the Convention in 2002.

71  In the Caribbean example, the States that denounced human rights treaties established a new Caribbean Court of Justice to, inter alia, review appeals in death penalty cases. Helfer (n 68) 1882–4. Several of the Latin American nations that withdrew from international investment treaties are advocating for a ‘Bolivarian alternative to free trade’. A Tzanakopoulos, ‘Denunciation of the ICSID Convention under the General International Law of Treaties’ in R Hofmann and CM Tams (eds), International Investment Law and General International Law: From Clinical Isolation to Systemic Integration (Nomos Verlagsgesellschaft Mbh & Co, Baden Baden 2011) 75–93.

72  For additional discussion of these examples and supporting authorities, see Helfer (n 2) 1584.

73  The foundational framework for analysing the relationship between exit and voice in the domestic context is AO Hirschman, Exit, Voice and Loyalty: Responses to Decline in Firms, Organizations, and States (Harvard University Press, Cambridge 1970).

74  C Brummer, ‘The Ties that Bind? Regionalism, Commercial Treaties and the Future of Global Economic Integration’ (2007) 60 Vanderbilt L Rev 1349, 1372. The EU later modified this position somewhat, declaring that all incompatible BIT provisions would have to be removed from the treaties by amendment. The States seeking accession to the EU complied with this demand. Ibid 1379.

75  Ibid 1389.

76  For a recent discussion of the distributional implications of exit costs for powerful and weaker States, see T Meyer, ‘Power, Exit Costs, and Renegotiation in International Law’ (2010) 51 HILJ 379.

77  For additional discussion of these examples and supporting authorities, see Helfer (n 2) 1609–10, 1645–6. Note that even if a later treaty does not expressly provide for the denunciation of an earlier convention, the same result may be achieved by VCLT Art 59, which creates a default rule that allows for the termination of an earlier treaty by implication if all the parties to that agreement conclude a later treaty relating to the same subject matter.

78  Eg CW Jenks, A New World of Law? A Study of the Creative Imagination in International Law (Longmans, Green & Co, London 1969) 180 (deploring treaty withdrawals as ‘a mask for anarchy, a practice which weakens the whole structure of treaty-created international obligations’).

79  B Koremenos and others, ‘The Rational Design of International Institutions’ (2001) 55 Intl Org 761, 773.

80  As Harold Tobin observed nearly eighty years ago, a State’s ability to quit a treaty after it enters into force ‘facilitates the securing of the consent of doubtful states to conventions aiming at universality, by removing the fear that changed conditions will make continued adherence inconvenient or even dangerous’. H Tobin, The Termination of Multipartite Treaties (Columbia University Press, New York 1933) 202.

81  Helfer (n 2) 1599; cf Tobin (n 80) 179–80 (explaining that where ‘conditions are particularly liable to change’, a denunciation or termination clause ‘may materially assist those who are attempting to secure acceptance of a draft’).

82  Helfer (n 2) 1590 (‘A state that … follows the specified procedures [of an exit clause] and explains the basis for its actions projects a real (if somewhat backhanded) respect for international rules, particularly where it is possible to profess adherence in theory but fail to comply in fact’).

83  See generally Bilder (n 5); Treaty Maker’s Handbook (n 47).

84  Eg AT Guzman, ‘The Design of International Agreements’ (2005) 16 EJIL 579; O Hathaway, ‘Between Power and Principle: An Integrated Theory of International Law’ (2005) 72 U Chicago L Rev 469; LR Helfer, ‘Nonconsensual International Lawmaking’ 2008 U Illinois L Rev 71; K Raustiala, ‘Form and Substance in International Agreements’ (2005) 99 AJIL 581.