state practice for purposes of determining the content of CIL. This makes sense; it is at least reasonable to expect that a state's future behavior will conform to its treaty obligations as to form a similar expectation based on a non-binding statement. However, if a states behavior does not conform to its treaty obligations, it would seem that it would be unreasonable to rely on the fact of treaty adherence in forming expectations as to future actions; hence, the weight to be given to treaty adherence as an item of state practice supporting a rule of cil would appear to depend on a states actual performance under the treaty 18 D' Amato CONCEPT OF CUSTOM. 113-38: Richard Baxter. Multilateral Treaties a Evidence of customary International Law, 41 BRIT. Y.B. INT'L L 275, 275-76(1968) 19 These points are only the beginning of the consideration of treaties and the relationship to CIL. First, it often happens that at least some provisions of a treaty are intended to codify previously existing CIL, though an assertion to that effect in the treaty itself can hardly be taken at face value. Second, even a treaty clearly representing a departure from Cil may give rise to a new Cil rule, at least if states behave as the treaty would require in situations not covered by the treaty -for example, in transactions in which all participating states are not parties to a given treaty, D'Amato, CONCEPT OF CUSTOM, supra n 8, at 104-66 and authorities therein cited. Professor D'Amato, indeed, takes the position that generalizable rules in any treaty give rise to rules of Cil binding upon all states, even if the treaty in question is bilateral The foregoing statement is therefore more conservative than Professor D'Amato's position; ably, he would disagree with it only in that, in his view, it does not go far enough. Third all other things equal, the more states who are parties to a treaty, the easier it is to make the case that practice embodied in a treaty is CIL. Fourth, a treaty may itself negate the argument that it has any effect on CIL. Leaving aside the question of the effect of specific statements in the treaty context. For example, a treaty may limit states parties to particular types of remedies gainst one another for violating the treaty; if those remedies are more limited than thos available for violations of CIL obligations, the implication of the treaty limitation is that n.5,$901, rptr'sn8, for a discussion regarding the practice of the United States regarding 9 bligations imposed by the treaty are not duplicated by Cil obligations( See Restatement, sup claims by and against foreign states); otherwise, the treatys limitation on remedies could be circumvented simply by making a claim based on Cil rather than on the treat An example of a treaty, the remedial limits of which have implications for CIL, is provided by the International Covenant on Civil and Political Rights(the Covenant), Dec. 19. 1966, 999UNTS. 171(1976). That treaty imposes numerous substantive obligations on parties
9 18 D’Amato, CONCEPT OF CUSTOM, 113-38; Richard Baxter, Multilateral Treaties as Evidence of Customary International Law, 41 BRIT. Y.B. INT’L L. 275, 275-76 (1968). 19 These points are only the beginning of the consideration of treaties and their relationship to CIL. First, it often happens that at least some provisions of a treaty are intended to codify previously existing CIL, though an assertion to that effect in the treaty itself can hardly be taken at face value. Second, even a treaty clearly representing a departure from CIL may give rise to a new CIL rule, at least if states behave as the treaty would require in situations not covered by the treaty - for example, in transactions in which all participating states are not parties to a given treaty, D’Amato, CONCEPT OF CUSTOM, supra n. 8, at 104-66 and authorities therein cited. Professor D’Amato, indeed, takes the position that generalizable rules in any treaty give rise to rules of CIL binding upon all states, even if the treaty in question is bilateral. The foregoing statement is therefore more conservative than Professor D’Amato’s position; presumably, he would disagree with it only in that, in his view, it does not go far enough. Third, all other things equal, the more states who are parties to a treaty, the easier it is to make the case that practice embodied in a treaty is CIL. Fourth, a treaty may itself negate the argument that it has any effect on CIL. Leaving aside the question of the effect of specific statements in the treaty denying that it is intended to embody CIL, id. at 150-62, treaties may include provisions inconsistent with the argument that the rules of the treaty would bind the parties outside the treaty context. For example, a treaty may limit states parties to particular types of remedies against one another for violating the treaty; if those remedies are more limited than those available for violations of CIL obligations, the implication of the treaty limitation is that obligations imposed by the treaty are not duplicated by CIL obligations (See Restatement, supra n. 5, § 901, rptr’s n. 8, for a discussion regarding the practice of the United States regarding claims by and against foreign states); otherwise, the treaty’s limitation on remedies could be circumvented simply by making a claim based on CIL rather than on the treaty. An example of a treaty, the remedial limits of which have implications for CIL, is provided by the International Covenant on Civil and Political Rights (the Covenant), Dec. 19, 1966, 999 U.N.T.S. 171 (1976). That treaty imposes numerous substantive obligations on parties state practice for purposes of determining the content of CIL.18 This makes sense; it is at least as reasonable to expect that a state’s future behavior will conform to its treaty obligations as to form a similar expectation based on a non-binding statement. However, if a state’s behavior does not conform to its treaty obligations, it would seem that it would be unreasonable to rely on the fact of treaty adherence in forming expectations as to future actions; hence, the weight to be given to treaty adherence as an item of state practice supporting a rule of CIL would appear to depend on a state’s actual performance under the treaty.19
If the foregoing discussion does not make clear that dealing with CIL is difficult, perhaps that point can be made if one recalls that the Restatements definition of CIL calls, not only for a general and consistent practice of states, but for practice"followed.. from a sense of legal obligation. That is, it is not enough to know what states have done, it is also necessary to to it to protect the human rights of individuals, id, arts. 1, 6-27. It establishes a Human Rights Committee to perform various functions, id, art. 28. but permits a state party to the Covenant to complain to this committee of another state party's violations only if that state party has consented to the Committees hearing such complaints, and only if the complaining state has agreed to the Committee's authority to hear similar complaints against itself, id, art. 41 Further, the only actions the Committee may take in response to such a complaint are making its good offices available to the states in question with a view toward resolving the dispute, or, with the consent of the parties, appointing a conciliation commission for the same purpose, id, arts 41, 42. Further, Article 44 of the Covenant provides The provisions for the implementation of the present Covenant shall apply without prejudice to the procedures prescribed in the field of human rights by or under the constituent instruments and the conventions of the United Nations and of the specialized agencies and shall not prevent the States Parties to the present Covenant from having recourse to other procedures for settling a dispute in accordance with general or special international agreements in force between The clear implication of these provisions is that states would have no right under the Covenant to make claims against one another for violating its provisions outside the framework provided by arts. 41 and 42. Otherwise, the elaborate provisions of those articles would be pointless; if one state party may make a claim against another regardless of the second states willingness to be subjected to such ould it make to require such a consent for claims made to the Human Rights Committee, especially when claims made under the Article 44 preserves any rights states have accorded one another under other internationa urse Covenants dispute settlement provisions can lead only to non-binding procedures? of ce agreements: but that language underlines the lack of recourse absent such an agreement All of this is significant because it implies that the parties to the Covenant did not see that instrument as merely codifying CIL. If CIL imposed on states an obligations to other states to refrain from any violations of human rights, breaches of that CIL obligation would entitle other states to claim reparation from the offender 20 Restatement, supra n5,$ 102(2)
10 to it to protect the human rights of individuals, id., arts. 1, 6-27. It establishes a Human Rights Committee to perform various functions, id., art. 28. but permits a state party to the Covenant to complain to this committee of another state party’s violations only if that state party has consented to the Committee’s hearing such complaints, and only if the complaining state has agreed to the Committee’s authority to hear similar complaints against itself, id., art. 41. Further, the only actions the Committee may take in response to such a complaint are making its good offices available to the states in question with a view toward resolving the dispute, or, with the consent of the parties, appointing a conciliation commission for the same purpose, id., arts. 41, 42. Further, Article 44 of the Covenant provides: The provisions for the implementation of the present Covenant shall apply without prejudice to the procedures prescribed in the field of human rights by or under the constituent instruments and the conventions of the United Nations and of the specialized agencies and shall not prevent the States Parties to the present Covenant from having recourse to other procedures for settling a dispute in accordance with general or special international agreements in force between them. The clear implication of these provisions is that states would have no right under the Covenant to make claims against one another for violating its provisions outside the framework provided by arts. 41 and 42. Otherwise, the elaborate provisions of those articles would be pointless; if one state party may make a claim against another regardless of the second state’s willingness to be subjected to such claims, what sense would it make to require such a consent for claims made to the Human Rights Committee, especially when claims made under the Covenant’s dispute settlement provisions can lead only to non-binding procedures? Of course, Article 44 preserves any rights states have accorded one another under other international agreements; but that language underlines the lack of recourse absent such an agreement. All of this is significant because it implies that the parties to the Covenant did not see that instrument as merely codifying CIL. If CIL imposed on states an obligations to other states to refrain from any violations of human rights, breaches of that CIL obligation would entitle other states to claim reparation from the offender. 20 Restatement, supra n.5, § 102(2). If the foregoing discussion does not make clear that dealing with CIL is difficult, perhaps that point can be made if one recalls that the Restatement’s definition of CIL calls, not only for a general and consistent practice of states, but for practice “followed . . . from a sense of legal obligation.”20 That is, it is not enough to know what states have done; it is also necessary to
know why they have acted. More precisely, even if a state's acts suggest that it sees itself constrained in some fashion, evidence of the states motive for restraining itself is very mportant. If the state does not see itself as legally obliged to take the action it has taken, the practice in question simply does not support the argument that CIL imposes an obligation to behave as the state in question has done. It is important to note, however, that it is constraint that must be explained, not the absence of constraint. As in domestic law, there is no need to explain why an actor is free to act, since there is no presumption that every act whatever must be authorized. Rather, explanation is required for limitations on freedom, whether they take the form of prohibitions or of duties to act An American judge facing a case in which Cil plays a role, then, faces a complex task He must seek to determine whether there exists a general and consistent state practice on the matter, while keeping in mind the practice he is investigating can take many forms, and that proof that a practice was general ten years earlier cannot eliminate the possibility that any rule established by that practice has subsequently been undermined by contrary practice. The judg must, furthermore, not only investigate the behavior of governments, but must also seek to determine the motives for that behavior. The question thus becomes, how have judges sought to carry out this task? The next section will seek to answer this question This discussion assumes that the concept of cil has some meaning. Of course, if it does not then judges would never have a basis for deciding a case in reliance on CIL, whether derived from the types of authorities criticized in this article or in some other manner. It should be noted, however, that some recent scholarship questions the concept of CIL itself. Professors Goldsmith and Posner have argued that any behavioral regularities in state practice can be explained as the result of coincidence of interest among states, coercion of weaker states by stronger, bilateral cooperation resulting from a repeating prisoner's dilemma, or bilateral coordination, Jack L. Goldsmith and Eric A. Posner, A Theory of customary International Law, 66 U CHI. L REV. 1113, 1120-1131(1999). Behavioral regularities coming
11 21 This discussion assumes that the concept of CIL has some meaning. Of course, if it does not then judges would never have a basis for deciding a case in reliance on CIL, whether derived from the types of authorities criticized in this article or in some other manner. It should be noted, however, that some recent scholarship questions the concept of CIL itself. Professors Goldsmith and Posner have argued that any behavioral regularities in state practice can be explained as the result of coincidence of interest among states, coercion of weaker states by stronger, bilateral cooperation resulting from a repeating prisoner’s dilemma, or bilateral coordination, Jack L. Goldsmith and Eric A. Posner, A Theory of Customary International Law, 66 U. CHI. L. REV. 1113, 1120-1131 (1999). Behavioral regularities coming know why they have acted. More precisely, even if a state’s acts suggest that it sees itself as constrained in some fashion, evidence of the state’s motive for restraining itself is very important. If the state does not see itself as legally obliged to take the action it has taken, the practice in question simply does not support the argument that CIL imposes an obligation to behave as the state in question has done. It is important to note, however, that it is constraint that must be explained, not the absence of constraint. As in domestic law, there is no need to explain why an actor is free to act, since there is no presumption that every act whatever must be authorized. Rather, explanation is required for limitations on freedom, whether they take the form of prohibitions or of duties to act. An American judge facing a case in which CIL plays a role, then, faces a complex task. He must seek to determine whether there exists a “general and consistent” state practice on the matter, while keeping in mind the practice he is investigating can take many forms, and that proof that a practice was general ten years earlier cannot eliminate the possibility that any rule established by that practice has subsequently been undermined by contrary practice. The judge must, furthermore, not only investigate the behavior of governments, but must also seek to determine the motives for that behavior. The question thus becomes, how have judges sought to carry out this task? The next section will seek to answer this question.21
IL. The Traditional american Approach to Customary International law The Supreme Court has seldom had occasion to pronounce upon the proper method for determining the content of CIL, and last did so more than a century ago. Still, with caveats to be addressed below, surely that Courts approach to the subject should carry weight with American about for such reasons do not fit the traditional definition of CIL, they argue, because states are in such cases shaping their actions either out of self-interest or fear, rather than from a sense of legal obligation. id at 1131-33 This view assumes that states motivated either by self-interest or by fear are by definition not acting from a sense of legal obligation. This assumption seems to reflect a very narrow concept of what it means to act from a sense of legal obligation. Apparently, actions taken from any motive other than a disinterested desire to comply with the law do not count. But surel common practice, many individuals obey legal rules out of fear( they refrain from speeding to void being ticketed )or out of self-interest( they make sure that their contracts are in writing so that they can be enforced). At least with respect to coercion, further, it would seem that Goldsmith and posner would need to show that states who conform their actions to the demands of more powerful states do so without regard to the legal basis for such demands. It is possible that demands seen as having a plausible legal basis are less costly to the more powerful state than are demands with no such basis, for example, because they attract less negative reaction from other powerful states, or because the weaker state is less inclined to offer such resistance as it could. In short, the argument of Goldsmith and Posner is provocative but leaves some questions unanswered Professor Kelly offers an even more fundamental criticism of the concept of CIL. He asserts that most norms characterized as rules of CIl derive, not from the general practice of states, but from the practice of a small number of states and/or deductions by writers from sources other than the practice of states, J. Patrick Kelly, The Twilight of customary International Law, 40 VA.J. INT'L L 449, 472-73, 475-76. While he acknowledges that there considerable empirical support for certain structural norms of CIL, he insists that there is no such not in states'motives for ad posing liabilities on states, id at 479-84. That is, the problem lies ering to certain apparent behavioral regularities, but in doubt as to the very existence of such regularities The only way to refute Professor Kelly's argument is to demonstrate that particular CIL rules purporting to limit states freedom of action have strong empirical support. Rather than attempt to make such a demonstration in a footnote, it is enough here to note that, if his argument is accepted, it is at least clear that American courts ought not base decisions on purported cIL rules derived from any source other than an actual showing of the existence of a general practice conforming to the alleged rule
12 about for such reasons do not fit the traditional definition of CIL, they argue, because states are in such cases shaping their actions either out of self-interest or fear, rather than from a sense of legal obligation, id. at 1131-33. This view assumes that states motivated either by self-interest or by fear are by definition not acting from a sense of legal obligation. This assumption seems to reflect a very narrow concept of what it means to act from a sense of legal obligation. Apparently, actions taken from any motive other than a disinterested desire to comply with the law do not count. But surely, in common practice, many individuals obey legal rules out of fear (they refrain from speeding to avoid being ticketed) or out of self-interest (they make sure that their contracts are in writing so that they can be enforced). At least with respect to coercion, further, it would seem that Goldsmith and Posner would need to show that states who conform their actions to the demands of more powerful states do so without regard to the legal basis for such demands. It is possible that demands seen as having a plausible legal basis are less costly to the more powerful state than are demands with no such basis, for example, because they attract less negative reaction from other powerful states, or because the weaker state is less inclined to offer such resistance as it could. In short, the argument of Goldsmith and Posner is provocative but leaves some questions unanswered. Professor Kelly offers an even more fundamental criticism of the concept of CIL. He asserts that most norms characterized as rules of CIL derive, not from the general practice of states, but from the practice of a small number of states and/or deductions by writers from sources other than the practice of states, J. Patrick Kelly, The Twilight of Customary International Law, 40 VA. J. INT’L L. 449, 472-73, 475-76. While he acknowledges that there is considerable empirical support for certain structural norms of CIL, he insists that there is no such empirical support for norms imposing liabilities on states, id. at 479-84. That is, the problem lies not in states’ motives for adhering to certain apparent behavioral regularities, but in doubt as to the very existence of such regularities. The only way to refute Professor Kelly’s argument is to demonstrate that particular CIL rules purporting to limit states’ freedom of action have strong empirical support. Rather than attempt to make such a demonstration in a footnote, it is enough here to note that, if his argument is accepted, it is at least clear that American courts ought not base decisions on purported CIL rules derived from any source other than an actual showing of the existence of a general practice conforming to the alleged rule. II. The Traditional American Approach to Customary International Law The Supreme Court has seldom had occasion to pronounce upon the proper method for determining the content of CIL, and last did so more than a century ago. Still, with caveats to be addressed below, surely that Court’s approach to the subject should carry weight with American judges
13 The earliest case in which the Supreme Court appears to have addressed the question of the manner of determining the content of CIl was United States v. Smith. The defendant in that case had been tried under a federal statute which imposed the death penalty upon persons onvicted of"the crime of piracy, as defined by the law of nations. "Defendant argued that the statute was unconstitutional in that it failed to define the offense of piracy other than by reference to the law of nations. 4 The court held that the statute was not unconstitutional simply because its meaning depended upon the interpretation of a term not itself defined in the statute, stating that Congress may as well define by using a term of a known and determinate meaning, as by an express enumeration of al the particulars included in that term. The court then addressed the question whether the law of nations in fact provided a reasonably certain meaning for the term"piracy. In this connection the court stated that, What the law of nations on this subject is, may be ascertained by consulting the works of jurists, writing professedly on public law, or by the general usage and practice of nations, or by judicial decisions recognising [sic] and enforcing that law. 26 The court proceeded to demonstrate that the listed sources provided a clear definition of piracy. That is, the court relied on the opinions of well known jurists 22 United States v Smith, 18 U.S.( 5 Wheat )153(1820) 23ldat153-54.n.a Id. at 158 Id at 159 ld.at160-61
13 22 United States v. Smith, 18 U.S. ( 5 Wheat.) 153 (1820). 23 Id. at 153-54, n. a. 24 Id. at 158. 25 Id. at 159. 26 Id. at 160-61. The earliest case in which the Supreme Court appears to have addressed the question of the manner of determining the content of CIL was United States v. Smith.22 The defendant in that case had been tried under a federal statute which imposed the death penalty upon persons convicted of “the crime of piracy, as defined by the law of nations.”23 Defendant argued that the statute was unconstitutional in that it failed to define the offense of piracy other than by reference to the law of nations.24 The court held that the statute was not unconstitutional simply because its meaning depended upon the interpretation of a term not itself defined in the statute, stating that “Congress may as well define by using a term of a known and determinate meaning, as by an express enumeration of al the particulars included in that term.”25 The court then addressed the question whether the law of nations in fact provided a reasonably certain meaning for the term “piracy.” In this connection the court stated that, “What the law of nations on this subject is, may be ascertained by consulting the works of jurists, writing professedly on public law; or by the general usage and practice of nations; or by judicial decisions recognising [sic] and enforcing that law.”26 The court proceeded to demonstrate that the listed sources provided a clear definition of piracy. That is, the court relied on the opinions of well known jurists