The Yale Law Journal


Withdrawing from Customary International Law: Some Lessons from History

17 Dec 2010

The modern view is that “[c]ustomary international law results from a general and consistent practice of states followed by them from a sense of legal obligation.” General and consistent practice can generate a rule of customary international law (CIL) that is binding on all nations even if that practice is not universal. New states are bound by existing rules, and no state may unilaterally withdraw from a rule of CIL. As the International Court of Justice put it, CIL rules, “by their very nature, must have equal force for all members of the international community, and cannot therefore be the subject of any right of unilateral exclusion exercisable at will by any one of them in its own favour.” In short, customary international law today binds all nations irrespective of their individual consent.

In a recent article, Professors Curtis Bradley and Mitu Gulati point out that this has not always been true and ask whether it should continue to be. They contrast what they call the “Default View” of international law, which permitted nations to withdraw unilaterally from some international law rules upon proper notice, with what they call the “Mandatory View,” which denies nations a right of withdrawal. The authors date the shift from the Default View to the Mandatory View to the late nineteenth and early twentieth centuries, suggesting that it “may have evolved as part of an effort to bind new nations and former colonies to international law rules that had already been worked out by a handful of powerful states.” Having tarred the Mandatory View with the brush of imperialism, they propose a limited return to the Default View, under which some CIL rules—like sovereign immunity—would again be subject to a right of withdrawal.

Professors Bradley and Gulati have done a service by reminding us of the Default View. Unless one understands that some CIL rules were once optional, it is easy to misread a number of important early U.S. Supreme Court cases like Ware v. Hylton, Brown v. United States, and The Schooner Exchange v. McFaddon. But Bradley and Gulati have misread the history of customary international law in three ways that cast doubt upon their proposal for change.

First, Bradley and Gulati significantly overstate the importance of the Default View during the late eighteenth and early nineteenth centuries. As explained in Part I, most CIL rules at that time were subject to a Mandatory View—call it the “Old Mandatory View”—which was based on natural law and permitted no unilateral withdrawal. In fact, as will become clear from Part II, the Default View has never been the predominant understanding of customary international law.

Second, Bradley and Gulati are mistaken about when and why the idea developed that general consent could produce universally binding rules. The decline of natural law during the nineteenth century made the Old Mandatory View more difficult to sustain. As Part II explains, the shift to international-law positivism led some scholars to the Default View, but it led others to develop a new mandatory view based on general consent. This “Modern Mandatory View” developed during the mid-nineteenth century—much earlier than the authors recognize. It was used to bind only nations within the Western community, and a number of its advocates expressly stated that it could not be used to bind non-Western nations without their consent. Contrary to Bradley and Gulati’s assertions, the Modern Mandatory View did not develop as a tool of imperialism.

Third, while arguing for greater flexibility in international rules, Bradley and Gulati ignore the comity doctrines that have historically allowed the kind of discretion they seek. As Part III explains, the United States has long treated foreign sovereign immunity—Bradley and Gulati’s principal candidate for opt-out rights—as a matter of comity. Creating a new category between comity and universally binding CIL seems unlikely to increase the flexibility of immunity doctrines, but it does seem likely to complicate conversations about other CIL rules. Historically, the existence of multiple categories of international law created opportunities for confusion and manipulation.

I. The Old Mandatory View

The eighteenth-century understanding of the law of nations was, as Professors Bradley and Gulati point out, different from our own. Both Vattel’s treatise and early American cases recognized several categories within the law of nations. But these early sources also show that the Default View applied to a relatively small and unimportant set of rules and that the larger and more important category was considered binding regardless of a nation’s individual consent.

A. Vattel

Americans in the late eighteenth century drew their conceptions about the law of nations primarily from the Swiss writer Emmerich de Vattel. Bradley and Gulati quote at length from Vattel’s discussion of the “customary law of nations,” noting that its rules were based on individual consent, which nations were free to withdraw by giving sufficient notice. But their account paints an incomplete picture. In Vattel’s system, the “customary law of nations” was just one of several categories. More important for Vattel was the “voluntary law of nations”—rulesbased on natural law, which were not optional. Bradley and Gulati mention the voluntary law of nations only briefly, failing to note that its nonderogable rules vastly overshadowed the derogable customary law.

Vattel divided the law of nations into four categories that varied in their jurisprudential foundations and in their obligatory force: (1) the necessary, (2) the voluntary, (3) the conventional, and (4) the customary. Both the necessary law of nations and the voluntary law of nations were based on natural law. The necessary law of nations flowed from the direct application of natural law to nation-states. It was immutable and absolutely binding but only internally upon the conscience of the sovereign. The voluntary law of nations was also based upon natural law but mediated through a principle of sovereign equality. In contrast to the necessary law, the voluntary law created externally binding obligations that one nation could enforce against another, by war if need be.

Vattel borrowed the term “voluntary” from the writer Christian Wolf. It was an unfortunate choice because it suggested an optional character that Vattel expressly denied:

  [W]hat we call the voluntary Law of Nations consists in the rules of conduct, of external law, to which the natural law obliges Nations to consent; so that we rightly presume their consent, without seeking any record of it; for even if they had not given their consent, the Law of Nature supplies it, and gives it for them. Nations are not free in this matter to consent or not; the Nation which would refuse to consent would violate the common rights of all Nations.  

In Bradley and Gulati’s terms, Vattel’s voluntary law of nations represented a Mandatory View of international law.

Vattel’s last two categories—the conventional and the customary—were founded not upon natural law but upon individual consent and bound only those nations that actually agreed to them. The conventional law of nations was based on express consent, set forth in treaties, and bound “only the contracting parties.” The customary law of nations consisted of rules “consecrated by long usage and observed by Nations as a sort of law.” It was based on “tacit consent” and bound “only those Nations which have adopted it.” Nations were presumed to have adopted an established custom, but their consent could be withdrawn. Nations were bound to observe the customary law of nations only “so long as they have not expressly declared their unwillingness to follow it any longer.” In Bradley and Gulati’s terms, the customary law of nations represented the Default View of international law.

For Vattel, the mandatory rules of international law completely overshadowed the default rules in both scope and importance. Natural law’s leading role is clear from his title—The Law of Nations or the Principles of Natural Law Applied to the Conduct and to the Affairs of Sovereigns. Vattel wrote in his preface that the necessary and voluntary law “based upon fixed and permanent principles . . . will form the principal subject of my work.” Although he sometimes mentioned customary law for completeness, Vattel doubted that it belonged “within a systematic treatise on the Law of Nations.” The customary law of nations was merely interstitial. It was restricted to questions on which natural law was “[i]ndifferent.” Vattel explained, moreover, that “if there be anything unjust or unlawful in such a custom it is of no force, and indeed every Nation is bound to abandon it, since there can be neither obligation or authorization to violate the Law of Nature.”

Vattel’s discussion of ambassadors’ rights illustrates the relative importance of the voluntary law’s mandatory rules and the customary law’s default rules. Bradley and Gulati quote a passage in which Vattel reiterates the right of a nation to withdraw from certain customs respecting ministers. But Vattel is explicit, even in this passage, that the right to withdraw applies only to “indifferent” customs and not to rules “essential to an embassy and necessary to its proper success.” The indifferent customs Vattel had in mind were the right of a minister to practice his own religion and the exemption from customs duties, which are discussed in the two sections immediately preceding the passage about withdrawal. Elsewhere, Vattel emphasizes that the various grades of public ministers and the honors due to them are matters of custom. But the overwhelming majority of the rules concerning ambassadors—rules to which Vattel devotes five full chapters—are rules of voluntary law from which unilateral withdrawal was not permitted. These include the rights to send and receive public ministers, to their “perfect safety and inviolability,” to independence from “both the civil and the criminal jurisdiction of the country,” and to immunity from search and seizure of an ambassador’s house, carriages, and retinue. None of these rules was optional. As Vattel explained in a passage relating to immunity from jurisdiction, “natural law imposes upon all sovereigns the obligation of consenting to those conditions without which Nations could not cultivate the society which nature has established among them, could not have mutual intercourse, negotiate their affairs, or adjust their differences.”

In sum, the Old Mandatory View dominated Vattel’s system. Most rules—and all of the most important rules—were rules of voluntary law, based on natural law, from which no withdrawal was allowed. The Default View was limited to less important questions, on which natural law was indifferent and concerning which nations might establish customary rules through tacit consent. As the next Section describes, early American courts adopted Vattel’s system, treating only rules of the customary law of nations as subject to a right of unilateral withdrawal.

B. Early American Cases

Besides Vattel, Professors Bradley and Gulati draw upon three early Supreme Court cases to show that “CIL rules were binding only on nations that continued to accept them.” But each of these decisions involved rules that the Court either acknowledged or assumed to be rules of the customary law of nations. None supports the proposition that nations could withdraw from other rules of the law of nations. Indeed, the early cases followed Vattel by distinguishing rules that allowed a right of unilateral withdrawal from other rules that did not.

In Ware v. Hylton, Justice Chase held that a rule against confiscating debts owed to enemy subjects “was not binding on the state of Virginia, because founded on custom only.”This statement follows Chase’s description of the three categories that Vattel viewed as externally binding—“to wit, general, conventional, or customary.”

  The first [general] is universal, or established by the general consent of mankind, and binds all nations. The second [conventional] is founded on express consent, and is not universal, and only binds those nations that have assented to it. The third [customary] is founded on tacit consent; and is only obligatory on those nations, who have adopted it.  

Had the rule instead been one of general law (Vattel’s voluntary law), Virginia would by implication have had no right to withdraw because, as Chase emphasized, the general law “is universal . . . and binds all nations.”

Brown v. United States similarly involved a rule of the customary law of nations. The question was whether a declaration of war was sufficient to authorize the seizure of enemy property or whether further legislation was required. Chief Justice Marshall acknowledged that the United States had a “right” under the law of nations to confiscate enemy property but found that a “practice” had developed of not doing so immediately, thus allowing enemy subjects time to withdraw their personal property. As with the confiscation of debts in Ware, a sovereign right under the voluntary law was softened by a customary rule of forbearance. Marshall held that only Congress could exercise the United States’ right to confiscate enemy property and that its declaration of war should not be read to depart from custom. His famous statement that “[t]his usage is a guide which the sovereign follows or abandons at his will” referred only to the customary law of nations.

The Schooner Exchange v. McFaddon concerned the immunity of a foreign warship from suit in federal court. Marshall began with the propositions that “[t]he jurisdiction of the nation within its own territory is necessarily exclusive and absolute” and that any exceptions “must be traced up to the consent of the nation itself.” Consent could be implied from “common usage,” and Marshall found that nations had generally not exercised jurisdiction over a foreign sovereign’s person, foreign ministers, foreign troops granted permission to pass through a nation’s territory, or friendly foreign warships in a nation’s ports. He construed the statutes conferring jurisdiction on the federal courts as not applying to “a case[] in which the sovereign power has impliedly consented to wave [sic] its jurisdiction.” By treating the issue of sovereign immunity as one of tacit consent, Marshall treated it as an issue of the customary law of nations. His statements in The Schooner Exchange that the sovereign may withdraw sovereign immunity upon proper notice—like the similar statements in Ware and Brown—were thus limited to the customary law of nations and say nothing about the authority of a nation to depart from other rules of international law.

Other early decisions, on the other hand, said that a nation could not opt out of the voluntary law of nations. In Miller v. The Ship Resolution, the Federal Court of Appeals for Prize Cases declared that an ordinance of the Continental Congress on prize law could not change the law of nations concerning neutral rights—rights that Vattel said were part of the voluntary law. In Page v. Pendleton, Chancellor George Wythe similarly asserted that “the legislature could not retract their consent to observe the praecepts of the law, and conform to the usages, of nations.” Page involved the same Virginia confiscation act that subsequently came before the Supreme Court in Ware, but Wythe reached a different conclusion by characterizing the rule as “depending . . . on the law of nature.” For present purposes, the critical point is that whether a nation could withdraw from the law of nations turned on its classification. The customary law of nations was subject to unilateral withdrawal upon proper notice, but the voluntary law of nations was not.

It was rare, however, for American courts to reach the question of whether a sovereign had properly withdrawn from a law-of-nationsrule. More frequently, courts avoided that question by construing the statute not to violate the law of nations—an interpretative rule that has come to be known as the Charming Betsy canon. American courts applied this canon not only to voluntary rules from which nations could not withdraw but also to customary rules from which they could. In Brown v. United States, the Supreme Court read Congress’s declaration of the War of 1812 not to violate a usage against immediately confiscating enemy property, while in The Schooner Exchange v. McFaddon it read jurisdictional statutes not to violate sovereign immunity. Even where a right of withdrawal existed, the Court presumed that it had not been exercised.

In sum, the early cases do not show that “CIL rules were binding only on nations that continued to accept them.” American courts applied the Default View only to the customary law of nations. Other CIL rules—those in the category of “general” (that is, voluntary) law—were subject to a Mandatory View based on natural law. Such rules were not “only obligatory on those nations, who have adopted [them];” they were “universal” and bound “all nations.” It was this Old Mandatory View of customary international law that predominated in the early Republic.

II. Evolution of the Modern Mandatory View

Some writers adhered to the Old Mandatory View based on natural law well into the nineteenth century. Robert Phillimore’s 1854 Commentaries, for example, maintained that “[t]he Primary Source . . . of International Jurisprudence is Divine Law” and that “[t]he obligations of Natural and Revealed Law exist independently of consent of men or nations.” But the progress of international-law positivism over the course of the nineteenth century made it difficult to maintain a mandatory view based on natural law. Positivist writers responded in two different ways. Some, like Henry Wheaton, expanded the Default View to cover all customary international law. Wheaton reasoned that, in the absence of natural law, CIL could be based only on individual state consent—consent that a state was free to withdraw. Others, like Richard Wildman, articulated a new Mandatory View based on general consent. Just as custom in a domestic legal system was based on the general consent of a domestic community, so CIL was based on the general consent of the international community. Bradley and Gulati date the emergence of this view to Oppenheim’s 1905 treatise and suggest that it developed in order to bind non-Western states to Western rules. In fact, the Modern Mandatory View developed much earlier, was not used to bind non-Western nations, and was applied in contexts having nothing to do with imperialism.

A. Wheaton and the Default View

The Default View found its broadest application to customary international law in the writings of Henry Wheaton, America’s most celebrated nineteenth-century scholar of international law. Wheaton’s positivism grew over time, and his Default View strengthened as it did. The first edition of Wheaton’s Elements of International Law, published in 1836, repeated Vattel’s taxonomy of the law of nations with apparent approval and listed “[t]he rules of conduct which ought to be observed between nations, as deduced by reason from the nature of the society existing among independent states” first among the sources of international law. Wheaton grew more critical of Vattel in his 1845 History of the Law of Nations, suggesting that the voluntary law of nations should not be considered a separate category but rather a grouping of the conventional and customary law of nations. The 1846 edition of Wheaton’s Elements, published the following year, emphasized that the natural law of nations was only binding internally and “may be more properly called international morality,” while repeating Wheaton’s criticism of Vattel. Although this 1846 edition continued to list “rules deduced by reason” as the first source of international law, two posthumous editions based on changes Wheaton made before his death continued the march towards positivism by omitting such rules as a source of law.

As Wheaton’s writings became more positivist, he began to assert a broader right to withdraw from customary international law. In the first edition of Elements, Wheaton quoted Bynkershoek’s observation that “[t]he law of nations is only a presumption founded upon usage, and every such presumption ceases the moment the will of the party who is affected by it is expressed to the contrary” but without any particular indication of approval. Wheaton’s embrace of this idea in his 1845 History, however, is hard to miss. He criticized the distinction between ambassadorial privileges based on natural law and those based on custom, for example, observing that both “may be disregarded by any state which chooses to incur the risk of retaliation or hostility,” and then quoted Bynkershoek to support his own position. This discussion is repeated in the two posthumous editions of Elements. Wheaton’s expansion of the Default View seems connected to his positivism. Wheaton squeezed Vattel’s voluntary law of nations out of existence, equating it with the conventional and customary law of nations. He struck natural law from the sources of international law and banished it to the realm of morality. The only categories of law that remained were treaties and custom, both of which Vattel had understood to require the individual consent of each nation. For Wheaton, the Default View followed naturally from the shift to positivism.

B. Other Positivists and the Modern Mandatory View

Wheaton’s response to positivism was not the only one possible. Richard Wildman outdid even Wheaton’s positivism, maintaining that “[t]he law of nature forms no part of international law” and that natural law could create legal obligations only when “sanctioned by usage.” But Wildman, writing in 1849, articulated a new Mandatory View based on the idea that general consent could give rise to universally binding rules. He explained the binding force of customary international law by analogizing the community of nations to a community of people: “As the custom of a people forms part of their municipal law, and is binding upon all, so the custom of nations is binding upon each, unless it be distinctly disclaimed, as it may be, before it has been adopted in practice.” Wildman’s opt-out—like the persistent objector rule today—was limited to disclaimers made before a custom became established. He denied that nations could individually withdraw from its rules. Wildman limited Vattel’s Default View so that it applied only to “the peculiar customs of particular states” and not to customs “established by the general practice of nations.” Indeed, Wildman found the notion of a law from which a nation could unilaterally withdraw self-contradictory, “as that which has no binding obligation, can in no sense be properly called law.”

Wildman in turn influenced Henry Halleck, whose 1861 treatise likewise grounded mandatory CIL rules in “general usage” rather than natural law. Whether a custom became “universal” depended upon “the frequency of its recognition and the numbers who have approved it,” but universal consent was not required. Indeed, Halleck cautioned that we should look to the history of state practice not for “the record of any constant and uninterrupted practice, but because we shall there find what has been generally approved and what has been generally condemned in the variable and contradictory practice of nations.”

Another positivist to take the Modern Mandatory View was William Hall. In his 1880 treatise, Hall looked to state practice as the sole source of customary international law, yet he also held that CIL rules could be established by “general consent.” Hall’s discussion of the Declaration of Paris is illustrative. In 1856, Great Britain, France, Austria, Russia, Prussia, and Turkey agreed to abolish privateering, ensure neutral rights, and establish rules for blockades. Many other states declared their adherence, but the United States did not. This meant that “the provisions of the declaration of Paris cannot in strictness be said to be at present part of international law.” But, Hall continued, if other nations continued to act in accordance with its provisions, “the United States will come under an obligation to conform its practice to them in a time which will depend on the number and importance of the opportunities which other states may possess of manifesting their persistent opinions.”

For positivists who believed that CIL rested upon the general consent of a community, it became important to define the community. William Hall, John Westlake, and Lassa Oppenheim each maintained that a state could be bound by the general consent of the international community irrespective of its individual consent. Each defined the community bound by Western CIL rules to include European nations, American nations that arose from European colonies, and a few others who had joined the existing system. But contrary to Bradley and Gulati’s suggestion, each author expressly disclaimed that Western rules automatically bound those who were not members of the Western community. Hall wrote that “states outside European civilisation” could become bound to the existing system only through “a formal act.” Westlake maintained that only states within European civilization were automatically bound to its system of international law. And Oppenheim said that, for a new state to be admitted “into the circle of the Family of Nations,” it must “expressly or tacitly consent to be bound for its future international conduct by the rules of International Law.” For these authors, the “Family of Nations” idea served not to impose Western rules on non-Western states but rather to limit rules based on the general consent of a community to the members of that community.

C. The Modern Mandatory View in the Supreme Court

The Modern Mandatory View was not confined to the musings of scholars during the nineteenth century. It also found its way into the decisions of the U.S. Supreme Court. The question in The Scotia, decided in 1872, was whether rules for displaying lights adopted by Britain in 1863 and by the U.S. Congress in 1864 applied to the collision of a British ship and an American ship on the high seas. The Court held that they did—not because of their adoption by the British and American governments, but because of their general adoption as part of the law of the sea.

  That law is of universal obligation, and no statute of one or two nations can create obligations for the world. Like all the laws of nations, it rests upon the common consent of civilized communities. It is of force, not because it was prescribed by any superior power, but because it has been generally accepted as a rule of conduct.  

This passage reflects the Modern Mandatory View that CIL rules of “universal obligation” can arise from “common consent.” Common consent did not mean universal consent, and the Supreme Court thrice pointed out that these regulations had become binding on all despite less than universal adherence. Among the thirty states listed as adhering to the rules was not just Turkey (admitted to the European system of international law in 1856) but also Morocco. Non-Western practice was relevant in establishing rules for Great Britain and the United States.

The Paquete Habana followed the path laid by The Scotia. The Supreme Court repeatedly declared that customary international law rests upon “the general consent of civilized nations.” Indeed, Justice Gray quoted The Scotia at length on this very point. In finding that coastal fishing vessels were not subject to capture as prize under customary international law, the Supreme Court distinguished the practice of Great Britain, while citing that of Japan.

D. The Modern Mandatory View in Historical Perspective

Among the reasons Bradley and Gulati give for reconsidering the current system is that it “may not have persisted for as long as many assume and that its foundations may be rooted in normatively unattractive propositions.” The history recounted above undercuts both halves of this claim.

First, the Modern Mandatory View developed, not at the turn of the twentieth century, but during the middle of the nineteenth, at roughly the same time that Phillimore continued to advance the Old Mandatory View and Wheaton attempted to expand the Default View. Although the Modern Mandatory View was just one of several theories in the 1850s, by the end of the nineteenth century it was widely accepted among international-law scholars and by the U.S. Supreme Court. The Default View, by contrast, was never the dominant understanding of international law. Wheaton was an important voice to be sure but only one of many at mid-century, and his views did not prevail.

Second, the Modern Mandatory View was not developed “to bind new nations and former colonies to international law rules that had already been worked out by a handful of powerful states.” Because this view based CIL rules on the general consent of a community, it bound only nations that were part of that community. Writers from Hall to Westlake to Oppenheim expressly stated that Western rules did not bind non-Western nations without their consent. To the extent that the Modern Mandatory View looked beyond the Western community, the bonds of obligation tended to run in the opposite direction. Thus, the U.S. Supreme Court in The Scotia looked to the practices of Turkey and Morocco among others to find a rule binding on Great Britain and the United States, while in The Paquete Habana the Court cited Japan’s practice as it found that the U.S. Navy had violated customary international law. Bradley and Gulati’s assertion that the Modern Mandatory View is the offspring of imperialism is simply false.

To be clear, I do not claim that the Modern Mandatory View is free of racism or that it could not be used for imperialist ends. But these same charges may be laid against other doctrines of international law including the Default View. Wheaton, for example, remarked approvingly of the First Opium War that “the Chinese Empire . . . has been compelled to abandon its inveterate anti-commercial and anti-social principles, and to acknowledge the independence and equality of other nations in the mutual intercourse of war and peace.” Wheaton may have believed in individual consent, but he apparently felt no qualms about compelling such consent by force of arms. My claim here is simply that the Modern Mandatory View did not develop as “another form of colonialism” and should not be rejected as “normatively unattractive” on that basis.

III. Between Comity and Law

Bradley and Gulati’s affirmative case for the Default View is based on a perceived need for greater flexibility in international rules. They argue that the Modern Mandatory View makes customary international law too “sticky,” creates “holdout” problems, and is insufficiently adaptable. The authors are careful to note that some CIL rules should remain mandatory. They suggest that human rights and background principles like pacta sunt servanda, on which the international system of treaties is based, should not be subject to unilateral withdrawal. Their “prime candidates” for opt-out rights are “certain immunity doctrines.”

Given their focus on flexibility and their acknowledgment that different degrees of flexibility may be appropriate for different international rules, it is odd that Bradley and Gulati ignore international comity. Historically, comity has spawned a number of doctrines that, although not binding as a matter of customary international law, have played a critical role in mediating international legal relations. In the United States, comity has provided the basis for the conflict of laws and the enforcement of foreign judgments. Comity has limited the extraterritorial reach of American laws and the jurisdiction of American courts. Comity has allowed foreign governments to bring suit in the United States and immunized them from suits by others. Comity doctrines are inherently discretionary: “Every nation must be the final judge for itself, not only of the nature and extent of the duty, but of the occasions on which its exercise may be justly demanded.”

In the United States, foreign sovereign immunity has long been understood as a matter of comity. In the early nineteenth century, the dividing line between comity and the derogable customary law of nations was unclear and perhaps unimportant since both allowed for discretion. The Schooner Exchange v. McFaddon treated foreign sovereign immunity as part of the customary law. But the doctrine’s discretionary aspect soon led the Court to recharacterize it as a matter of “public comity and convenience.” The characterization stuck and profoundly shaped the doctrine. During the first half of the twentieth century, foreign sovereign immunity’s grounding in comity led the Supreme Court to defer to executive-branch determinations of immunity. Even after its codification in the Foreign Sovereign Immunities Act of 1976 (FSIA), the Supreme Court has continued to describe foreign sovereign immunity as “a matter of grace and comity on the part of the United States.” In recent years, the Court has seldom looked to customary international law to construe the FSIA, while it has repeatedly invoked to comity to resolve interpretative questions.

Since Bradley and Gulati offer foreign sovereign immunity as a prime candidate for the Default View, it is worth asking what their proposal would add to the doctrine’s current treatment as one of comity. Under the Default View, foreign sovereign immunity would formally bind the United States as customary international law. But practically, the authors’ proposal would make no difference because the United States would be bound under international law only so long as it wanted to be bound.

Bradley and Gulati emphasize that “the Default View would allow only prospective withdrawal” and that “a reasonable notice period might be imposed in situations in which reliance interests are at stake.” But comity doctrines can also act as a precommitment strategy. Some applications of comity, such as the public policy exceptions in the conflict of laws and the enforcement of judgments, are determined ex post and case-by-case. But others—in particular, foreign sovereign immunity—are codified by statute, assuring foreign nations some notice if the United States changes its rules. It is true that changes in the FSIA may be retroactive, whereas changes in customary international law are not. But Bradley and Gulati’s proposal would not alter this, since it is well established that an act of Congress can supersede a CIL rule as law of the United States. The only difference the Default View would seem to make in foreign sovereign immunity is to give a foreign country the right to complain diplomatically about a retroactive (though not a prospective) change in the rules.

My point is not to quarrel with the proposition that flexibility is often useful in international legal relations. It is not clear, however, that the Default View meaningfully adds to that flexibility. What Bradley and Gulati propose is the creation of a new category of rules between discretionary comity and universally binding customary international law. But the difference between their new category and comity seems vanishingly thin.

If the Default View would not add much, would it do any harm? We cannot know for sure, but history suggests some reasons to worry. We know that the existence of multiple categories of international law created opportunities for confusion and manipulation. In the 1790s, courts faced the question whether Virginia’s debt-confiscation statute was valid under the law of nations. The Virginia High Court of Chancery held the act invalid because the rule against confiscating debts depended “on the law of nature,” while the U.S. Supreme Court held that the statute was valid under the law of nations (though not under the peace treaty with Great Britain) because the rule was “founded on custom only.” The Supreme Court had the better of this argument, because Vattel clearly characterized the rule against confiscating debts as one of “general custom.” In The Schooner Exchange v. McFaddon, on the other hand, the Supreme Court engaged in its own manipulation. It took rules of foreign sovereign immunity that Vattel clearly viewed as part of the nonderogable voluntary law of nations and reconceived them as customary rules based on tacit consent. Similar manipulation by U.S. courts would be less likely today because the stakes are lower. Even without a Default View, it is well-established that Congress may, as a matter of domestic law, override rules of customary international law.

Manipulation seems more likely at the international level. Bradley and Gulati suggest that “[a]ny withdrawal rights would need to be invoked publicly, presumably with reference to the purported CIL rule at issue and also perhaps to the reasons for withdrawal, thereby potentially increasing transparency and dialogue.” They assume that only “bad” actors would “attempt to violate CIL rules quietly . . . in an effort to avoid the reputational costs of withdrawal.” But if withdrawal entails reputational costs, it seems likely that “good” actors would seek to avoid them too. Under the current system, states alleged to have violated customary international law typically respond by denying the conduct, arguing that it does not fall within the CIL prohibition or that it is excused by some exception, all of which tend “to confirm rather than to weaken the rule.” Under Bradley and Gulati’s proposal, by contrast, the process of claim and response might additionally involve the proper classification of a rule as mandatory or optional, the amount of notice required, and whether such notice should be inferred from past conduct. Responses of this sort would do nothing to strengthen existing rules but would gradually chip away at them, diminishing the force of customary international law.


History can tell us several things about the Default View of customary international law. First, history can tell us that this view was never the dominant understanding. During the late eighteenth and early nineteenth centuries, the Default View applied only to interstitial rules not covered by natural law under the Old Mandatory View. When the Old Mandatory View declined, a Modern Mandatory View based on general consent developed almost immediately to take its place. Second, history can tell us that under the Modern Mandatory View, the general consent of the community of nations bound only nations that belonged to that community. It was not a tool of imperialism designed to bind non-Western nations to Western rules. And third, history can tell us that multiplying CIL’s categories increases opportunities for confusion and manipulation. The international system has discretionary doctrines based on comity, and it is far from clear that creating a new category between comity and binding law would add anything useful. The Default View is an interesting and important part of international law’s past, but history suggests that it should not be part of international law’s future.

William S. Dodge is Professor of Law at the University of California, Hastings College of the Law. The author would like to thank Tony Anghie, Oona Hathaway, and Chimène Keitner for comments on an earlier draft.

Preferred citation: William S. Dodge, Withdrawing from Customary International Law: Some Lessons from History, 120 Yale L.J. Online 169 (2010),