Publication Date


Publication Title

Duke Journal of Comparative & International Law


The conventional wisdom among international law scholars is that, once a rule of customary international law (“CIL”) becomes established, nations never have the unilateral right to withdraw from it. Instead, if they want to act in a way that is contrary to the rule, they must either violate it and hope that other nations acquiesce in the violation, or they must persuade other nations to enter into a treaty that overrides the CIL rule. In Withdrawing from International Custom, we termed this conventional wisdom the “Mandatory View” of CIL.1

As we explained in Withdrawing, the Mandatory View of CIL can be contrasted with the withdrawal rights that frequently exist under treaties. When nations expressly negotiate the creation of treaty obligations, they often include within the treaty a right of withdrawal, sometimes conditioned upon a period of notice. Even when they do not make such an agreement expressly, the subject matter of the treaty will sometimes itself suggest an implicit right of withdrawal. Moreover, even when there is no general right of withdrawal from a treaty, nations typically will have some right of withdrawal for situations in which there has been a fundamental change of circumstances. Finally, nations often have the ability to remain a party to a treaty while avoiding the application of particular provisions within the treaty through the use of reservations or the invocation of derogation clauses.2

This dichotomy between no exit rights under CIL and frequent and variegated exit rights under treaties is puzzling, for several reasons. Treaties and CIL are the two major sources of international law, and their substantive content frequently overlaps. Moreover, it is in many ways more difficult to create international law through treaty than through custom, since a treaty requires an express act of ratification. As a result, one might expect that, if anything, it would be more difficult to exit from treaties than from CIL, rather than the opposite.

We searched in the literature and found almost no explanation for the Mandatory View, and what little we did find was brief and conclusory. We attempted to trace the intellectual roots of the Mandatory View, but this produced only additional puzzles. We found, for example, that a number of the classic international law commentators of the eighteenth and nineteenth centuries thought that nations could unilaterally exit from at least some CIL rules.3 In addition, we found that the intellectual shift to the Mandatory View began to take place in the late nineteenth and early twentieth centuries and may have been part of an effort to ensure that “uncivilized” nations would be bound to the CIL worked out by a handful of powerful Western countries, something that raises questions about the normative underpinnings of the Mandatory View. Furthermore, we found that the one exception under the Mandatory View to the ban on unilateral exit—the so-called “persistent objector doctrine”—is a modern creation that did not become established until well after World War II and appears to have been in part a response to continuing uncertainties about how the Mandatory View would operate in practice.

After reviewing this history, we considered the Mandatory View from the perspective of institutional design. Because there was so little theoretical defense of the Mandatory View in the literature, we were compelled to speculate about what might be the best arguments in favor of that View. To gain traction on this issue, we drew from theoretical work that has been done concerning exit rights in the areas of contract law, constitutional design, and voting rights. We found that, although there are arguments that can be made to defend the Mandatory View, these arguments at best apply to only a subset of CIL, most notably where CIL is designed to address externality or agency problems. We also found that allowing exit rights under CIL could enhance the usefulness of CIL, by making it more transparent and efficient, and by encouraging broader experimentation. We concluded with some thoughts about how a typology might be developed to allow for variability in exit rights.

In this symposium issue of the Duke Journal of Comparative and International Law, a number of leading scholars engage with our project. Some of these scholars offer critiques of the analysis in Withdrawing, while others raise practical questions about how our ideas might be implemented. At minimum, this symposium fills a gap in the literature in terms of setting forth a sustained assessment of the Mandatory View. We are hopeful that it will also serve as a platform for additional work concerning withdrawal rights under CIL. Regardless, we owe an immense debt to the scholars who took part in the symposium, both for their willingness to consider our ideas and for their insightful comments.4

In this essay, we seek to advance the analysis set forth in Withdrawing by addressing four topics: the current state of CIL; the proper way to conceive of CIL and its relationship to treaties; how a shift away from the Mandatory View might occur in practice; and whether a shift to a Default View would make a meaningful difference in state practice. Most of the criticisms directed at Withdrawing are encompassed within these topics. We conclude the essay with some observations about additional research that might be useful.

Included in

Law Commons