Breaking the Window: Withdrawals from Humanitarian Disarmament Treaties

There has been a recent wave of withdrawal from humanitarian disarmament treaties. On 6 March 2025, Lithuania’s withdrawal from the Convention on Cluster Munitions (CCM) finally took effect after the 6-month notification period. The withdrawal of Lithuania from the CCM sounds an alarm for the humanitarian disarmament regimes and the international community, as it is the first State in history that formally withdraw from a treaty of such kind (if we exclude the context of nuclear non-proliferation from this category). Similarly, it has been reported that the Baltic States, Poland, and Finland are also preparing for or contemplating withdrawing from the Ottawa Convention, which prohibits anti-personnel landmines. The urges of these States to withdraw are deemed to be spurred by the increasing security concerns related to Russia. Lithuania’s departure from the humanitarian disarmament cause is most likely not the last, and potentially more humanitarian disarmament regimes may be under the same danger.

In light of the wave of withdrawal (and the intention to withdraw), this post will briefly discuss the procedural questions around the withdrawal process in these treaties, not limited to the context of cluster munitions and landmines. This blog also argues that the wave of withdrawal reveals, through a realistic lens, the inherent weakness of these regimes and may lead to a broken-window effect.

Special withdrawal procedures: Any scrutiny on the “reason”?

Humanitarian disarmament treaties typically contain a withdrawal clause that provides special notification procedures. Take Article 20(2) of the CCM as an example: the withdrawing State is required to

“[g]ive notice of such withdrawal to all other States Parties, to the Depositary and to the United Nations Security Council. Such instrument of withdrawal shall include a full explanation of the reasons motivating withdrawal.”

An identical withdrawal provision is also found in the Ottawa Convention (Article 20(2)). With slight differences but obvious similarities, the procedure to notify all other State Parties and the UN Security Council and to explain its withdrawal to them is also legally required in many other major disarmament treaties, such as the Convention on Biological Weapons (CBW) (Article XIII(2)) and the Chemical Weapon Convention (CWC) (Article XVI (2)).

These withdrawal clauses contain a delicate balance: on one hand, they respect sovereignty by acknowledging the State’s right to withdraw; and on the other hand, with the provision on unlimited duration and these elaborate procedures, they intend to keep the regime relatively stable. But do these extra procedures and the obligation to provide reasons offer any safeguard against arbitrary withdrawal? Some questions relating to the procedure may be worth noting: Are there any requirements for the reasons provided under these clauses? And is there any scrutiny or oversight role invested in these notified actors over the reasons?

Regarding the reasons for withdrawal, considerable ambiguity lies behind the required explanation. For the “full explanation of the reasons motivating withdrawal” required under CCM and the Ottawa Convention, no further qualifications were given in these treaties. In academic works, it is commonly understood that there is no requirement for the soundness of these reasons (Casey-Maslen, p. 372). This would also mean that there is no statutory requirement on the nature of the reasons given, whether legal or political. Although lacking abundant practice, what can certainly be observed in comparative terms is that such “full explanation of the reasons” is likely to be drafted intentionally loose, as it deliberately deviates from the pre-existing formulation of the stricter wording “a statement of the extraordinary events it regards as having jeopardised its supreme interests”, as provided in CBW, CWC, and the earlier nuclear disarmament treaties.

For the humanitarian disarmament treaties that contains the requirement of “supreme interests”, as listed above, a higher bar is set for the explanation provided, at least technically so. Under the regimes where the jeopardize of “supreme interests” is required, as Sims summarizes, the reasons given by the withdrawing State can be assessed by three elements: (1) the extraordinariness of the event; (2) relevance to the treaty in questions; and (3) magnitude of threat to its national security. In theory, the requirement of “supreme interests” may imply restrictions on the State’s right to withdraw. However, as academic works have pointed out, the exact meaning of “extraordinary events” that “jeopardize[s] its supreme interests” remains uncertain.

There are some limited examples of the reasons provided by the withdrawing State in practice. In the case of Lithuania’s withdrawal from CCM, its official letter sent to the depositary is frank about the fact that the decision to leave is mainly due to the “evolving regional security dynamics and geopolitical threats” caused by Russia, especially after its invasion of Ukraine. Previously, Eritrea had also informed the depository of its decision to withdraw (which was rescinded later) from the Ottawa Convention for the alleged misconduct of the relevant experts in gathering unauthorized information, thus threatening its national security. The latter appears to be less relevant to the subject-matter of the treaty and less of an imminent threat.

The real problem, however, does not lie in the exact qualification of these reasons, but the fact that there is no separate mechanism, whether conducted by the notified actors or any other bodies, to review and scrutinize these reasons. Reading the text of these disarmament treaties, regardless of their different requirements for the reasons provided, there is no expression of investing the authority to in any sense examine the withdrawal reasons. The existing practice with regard to the withdrawal of Lithuania and the rescinded withdrawal of Eritrea also proves that the notified bodies asserted nothing but a procedural role. For the withdrawal clauses that contain the “supreme interest” tests, all of them explicitly mention that it is the State that “decides” whether such an “extraordinary event” exists. The “sole prerogative” of the withdrawing State. It is yet to be seen, but unlikely, that these reasons are judiciable either. It means that, even if there are requirements for the reasons, the withdrawing State is its own and only “judge”, which makes every reason valid.

These withdrawal provisions are certainly realistic political compromises and may have served a “safety-net” function. At the same time, it must be clear that these disarmament regimes do not provide any effective scrutiny or control over withdrawal, but a procedure with some shades of “public shaming”. Despite the trend of making withdrawals complicated in formal clauses (Casey-Maslen and Vestner, p. 468), the barriers they bring are primarily moral discouragements and political pressures.

Inherent weakness revealed: the prisoner’s dilemma and the broken-window effects

On Lithuania’s withdrawal, the ICRC commented that “[i]nternational humanitarian law is not made for the hopeful days of peacetime. It is made for humanity’s darkest days, when armed conflict rages and people are in grave danger”. It is both true and ironic that it is during armed conflicts that these weapon-ban treaties make the most difference, yet States consider withdrawing from them precisely because of armed conflict. In technical terms, these humanitarian disarmament regimes are attentive to such situations and balance them out by including a notification period during which the withdrawal does not enter into effect instantly. And if the State engages in an armed conflict within the period, the treaty will stay in force until the end of the armed conflict. One may also consider whether the principle of good faith may, in some extreme cases, restrict the right to withdraw from these treaties if the State is taking the escape route too easily (some inspirations, see here and here). But at a more general level, the States’ increasing will to leave at the moment most needed reveals some serious weaknesses of the disarmament regimes.

First, these regimes create a “prisoner’s dilemma” situation, a constant question in disarmament. Russia is not party to either CCM or the Ottawa Convention, and if Lithuania (or another neighboring State) is bound by the treaties, they would face a disadvantaged asymmetrical situation. Certainly, both sides would benefit from the ban if both were to abide by it; yet if only one of them is a State Party, realistically, it would be particularly damaging for the State that respected the ban; and therefore, they would likely rush to a lose-lose situation. It is an understandable concern. The prisoners would not be in the dilemma if they had the means to communicate faithfully with each other, and analogously, it also explains the reason why disarmament regimes need trust and solidarity among States, which are naturally nowhere to be found between States at risk of armed conflicts against each other (for more insights and critiques on the prisoner’s dilemma and IHL, see here).

Second, the withdrawal of Lithuania may trigger broken-window effects on disarmament regimes. Borrowed from criminology, the broken window metaphor refers to the idea that signs as small as broken windows usually invite more and greater offenses over time. Withdrawing from these treaties is definitely not a crime, but Lithuania’s withdrawal may have thrown the first rock at the window of disarmament regimes. It is likely that States that intend to withdraw in the future would feel less moral burden and political toll with their decision to leave, in preparation for armed conflict or other less serious reasons. (The limited influence of North Korea’s withdrawal from the NPT is a different case, as it faced serious sanctions later due to nuclear activities.) The detrimental effects arising therefrom link back to the problem that these regimes provide no effective scrutiny over withdrawal, but only moral and political pressure, which is likely to be weakened in the future. These undesirable effects could be particularly damaging to the regimes on weapons whose prohibition is not yet crystallized as customary norms.

Conclusion

The withdrawal of Lithuania in no way cancels the efforts spent on humanitarian disarmament. This post also has no intention to place any blame on Lithuania or other States that are considering their withdrawal from these treaties, as security concerns play a role in every policy-making. The increasing willingness to withdraw is a realistic lesson from which we learn about the inherent weakness of the humanitarian disarmament design, including the lack of scrutiny over withdrawal and the difficult prisoner’s dilemma. The breaking of the first window should ring an alarm, and its damaging effects should be brought to awareness. Efforts need to be made to prevent, or at least mitigate, further compromises to the system.