法学资料集（英文）：美国法官和国际法

University of North Carolina – Chapel Hill School of Law Public Law & Legal Theory Research Paper No. 02-16 AMERICAN JUDGES AND INTERNATIONAL LAW A. Mark Weisburd Working paper, to be submitted for publication (2002). Full index of research papers available at http://www.law.unc.edu/faculty/researchpapers/. Visit UNC School of Law at www.law.unc.edu. http://ssrn.com/abstract_id=338440 AMERICAN JUDGES AND INTERNATIONAL LAW Introduction ............................................................1 I. The Nature of Customary International Law.................................3 II. The Traditional American Approach to Customary International Law...........12 III. Current American Approaches to Determining the Content of Customary International Law.................................................17 A. The Concept of Jus Cogens ......................................17 B. The Courts of Appeals and Jus Cogens .............................25 C. Critique of the Jus Cogens Decisions of the Courts of Appeals ..........34 1. Articles Basing Legal Rules on Sources Other than State Practice . . 40 2. The Restatement .........................................51 3. Decisions of American Courts ..............................57 4. Decisions of International Courts ............................60 IV. A Suggested Alternative ..............................................72 V. Conclusion .........................................................74 1 * Professor of Law, University of North Carolina at Chapel Hill. The author gratefully acknowledges the support of the North Carolina Law Foundation for this paper. 1 E.g., Siderman de Blake v. Repub. of Argentina, 965 F.2d 699 (9 th Cir. 1992) 2 E.g., Hilao v. Estate of Marcos, 103 F.3d 767 (9 th Cir. 1996). 3 E.g., Doe v. Unocal Corp., 110 F.Supp.2d 1294 (C.D. Cal. 2000). 4 Actually, they raise a whole host of problems. Perhaps the most basic is, what is the place of customary international law in American law? While there is considerable judicial and scholarly authority for the proposition that customary international law is part of “the law of the United States” as that phrase is used in Article III of the Constitution, see Filartiga v. Pena-Irala, 630 F.2d 876, 886-87 (2d Cir. 1980); In re Estate of Marcos Human Rights Litigation, 978 F.2d 493, 502 (9 th Cir. 1992), cert. denied sub nom. Marcos-Manotoc v. Trajano, 508 U.S. 972 (1993); Harold H. Koh, Is International Law Really State Law?, 111 HARV. L. REV. 1824 (1998); Gerald L. Neuman, Sense and Nonsense About Customary International Law: A Response to Professors Bradley and Goldsmith, 66 FORDHAM L. REV. 371 (1997); Beth Stephens, The Law of Our Land: Customary International Law as Federal Law after Erie, 66 FORDHAM L. REV. 393 (1997), a number of writers have challenged this position, see Ernest A. Young, Sorting Out the Debate Over Customary International Law, 42 VA. J. INT’L L. 365, 462-63 (2002); Daniel J. Meltzer, Customary International Law, Foreign Affairs, and Federal Common Law, 42 VA. J. INT’L L. 513, 519 (2002); Curtis A. Bradley & Jack L. Goldsmith, Customary International Law as Federal Common Law: A Critique of the Modern Position, 110 HARV. L. REV. 815 (1997); A.M. Weisburd, State Courts, Federal Courts and International Cases, 20 YALE J. INT’L L. 1 (1995); Arthur M. Weisburd, The Executive Branch and International Law, 41 VAND. L. REV. 1205 AMERICAN JUDGES AND INTERNATIONAL LAW A. Mark Weisburd * Introduction More and more frequently, American courts find themselves dealing with cases that raise issues under public international law. These cases may involve claims against foreign governments, 1 claims based on acts by foreign individuals, 2 or claims against corporations alleged to have cooperated with foreign governments. 3 While such claims may depend substantively on treaties or on federal statutes, very frequently they also rely on customary international law (CIL). And claims so based raise a problem. 4 2 (1988); Phillip R. Trimble, A Revisionist View of Customary International Law, 3 UCLA L. REV. 665 (1986), and at least one court has taken note of the controversy, Sampson v. Fed. Repub. of Germany & Claims Conf., 250 F.3d 1145, 1153 n.4 (7th Cir. 2001). In any event, this dispute does not bear on the matters discussed in this article, and is therefore not addressed. 5 RESTATEMENT (THIRD) OF THE FOREIGN RELATIONS LAW OF THE UNITED STATES § 102(2) (1987) (hereafter cited as RESTATEMENT). To understand the difficulty it is helpful to start with the concept of CIL. The Restatement (Third) of Foreign Relations Law (“Restatement”) describes customary international law as resulting “from a general and consistent practice of states followed by them from a sense of legal obligation.” 5 [In this article, the term “state” will be applied in the sense in which it is normally employed in international law, that is, as referring to independent countries.] Although this article will take issue with a number of assertions made in the Restatement, this definition raises little controversy. It does, however, illustrate the considerable difficulty facing a court forced to address an issue of CIL. How does the court determine, at the most basic level, what the various governments of the world have done regarding a particular matter? What counts as “practice”? How does one determine whether a practice is “general”? Federal courts have sought to escape this morass by relying primarily on academic writings, the Restatement, and decisions by American and international courts - and herein lies the difficulty. For, with respect to some areas of CIL - particularly the law of human rights, the aspect of CIL most frequently considered in American courts - neither modern academic writing nor the Restatement nor most judicial decisions purport to derive CIL from evidence of what governments actually do. Rather, they rely on other academic writings, other decisions of international courts, non-binding resolutions of international bodies, and hazy notions of natural law to justify their assertions regarding this CIL. 3 This article will seek to demonstrate that the approach the American courts have taken to determining the content of international law is fundamentally flawed. It leads courts to treat as law norms whose legal basis is either more circumscribed than the courts assert or, in some cases, non-existent. More fundamentally, it essentially converts law professors into philosopher kings, imposing their ideas of what the law should be under the guise of describing the law’s content. The discussion which follows will first explain just how strange CIL is when viewed from the perspective of the American legal system, and discuss as well the difficulties of determining the content of CIL. The article will then describe the traditional approach taken by American courts to deal with these difficulties. It will illustrate the contemporary approach to such matters by discussing the treatment of the concept of jus cogens by the federal courts of appeals. As will be shown, these courts have relied on doubtful authorities when forced to deal with this concept, and some doubtful results have, not surprisingly, followed. The final substantive section will suggest an alternative approach for the element of customary international law most frequently before American courts, that is, international human rights law. I. The Nature of Customary International Law The more a lawyer trained in American law reflects on the concept of CIL, the more peculiar that concept appears. CIL differs from domestic law in a number of important respects. First, in the CIL system, there is no sovereign with authority to control the actions of the independent states who are the subjects of the law; while states are free to subordinate themselves to such an authority, they have rarely done so. Likewise, there is no court with compulsory jurisdiction over states. Instead of proceeding from a sovereign, the law which 4 6 Standford v. Ky., 492 U.S. 361 (1989). controls the actions of states proceeds from the group of entities who are the law’s primary subjects - that is, governments. Not only are the subjects of the law also the law makers; they are also the law enforcers. In the American system, in contrast, law proceeds either from specific basic documents (constitutions) or more commonly from the actions of relatively small groups of designated individuals, be they legislators, administrators, or judges, who have the legal capacity to constrain every one in society. Likewise, law enforcement is a specialized function carried out by a limited number of people. A second difference between CIL and domestic law lies in the formality of the domestic law-making process. A legislature acts only when it is formally in session and according to the limitations of the relevant constitution and its own rules. Informal actions by legislators acting as individuals thus have no impact on the law - if, by coincidence, every member of a state’s legislature happened to be simultaneously exceeding the speed limit, no one would argue that the state’s traffic laws had been altered. CIL, however, is derived from the individual actions of governments, which actions may be undertaken in any type of setting and for reasons having little to do with the impact of those actions on international law. For example, when the U.S. Supreme Court upheld the constitutionality of sentencing a person to death for a crime committed prior to that person’s seventeenth birthday, 6 that action amounted to an instance of state practice which weakened any argument that CIL forbids governments to impose the death penalty for crimes committed when 5 7 Both the opinion of the court and the dissent in Standford take note of the practice of other countries with respect to execution of juveniles, with the court characterizing such practice as irrelevant, id. at 369, n.1, and the dissent giving weight to that practice, id. at 389-90 (Brennan, J., dissenting). Neither opinion, however, speaks in terms of CIL. 8 Professor D’Amato has also discussed this phenomenon, see Anthony D’Amato, The Concept of Human Rights in International Law, 82 COLUM. L. REV. 1110 (1982); Anthony D’Amato, THE CONCEPT OF CUSTOM IN INTERNATIONAL LAW 93-94 (1971) (hereinafter cited as D”Amato, CONCEPT OF CUSTOM). 9 Filartiga, supra n. 4, at 884 n. 15; Louis Sohn, The International Law of Human Rights: A Reply to Recent Criticism, 9 HOFSTRA L. REV. 347, 350 (1981). the perpetrator was as young as 16. However, nothing in the opinion of the Court suggested any concern with or even awareness of the impact of its decision on CIL. 7 A third, and especially important, difference between CIL and domestic law in the United States is that violations of CIL may lead, not only to some form of legal sanction, but also to a change in the content of the law. 8 Such a result is so contrary to that which would obtain under a domestic legal system that it requires careful explication. It must be stressed that this outcome is a necessary consequence of the way in which CIL is made in the first place. As pointed out above, CIL derives from the practice of states, even when that practice is not, in the first instance, undertaken because of its legal implications. Hence, any act by a government may simultaneously be analyzed under existing CIL and as what amounts to a legislative act. Thus, acts conforming to existing rules are not simply unremarkable instances of obedience to law; they are examples of practice reinforcing that law. Similarly, acts contrary to existing law can be characterized as violations of that law, but can also be seen as what amount to votes either to “repeal” the existing rule, or to modify that rule in some way. To be sure, it is frequently asserted that violations of CIL have no effect on the content of that law, just as domestic law is unaffected by acts violating it. 9 Such assertions, however, rely on a false analogy. Private 6 10 Convention on the High Seas, opened for signature Apr. 29, 1958, 13 U.S.T. 2312, T.I.A.S. No. 5200, 450 U.N.T.S. 82. 11 Id., preamble. 12 Id., arts. 1, 2. 13 For a fuller discussion of changes in the law of the sea regarding fishing, see Arthur M. Weisburd, Customary International Law: The Problem of Treaties, 21 VAND. J. INT’L L. 1, 17-19. (1988). persons violating domestic law do not act as legislators when they act; with regard to CIL, however, states are never not acting as legislators. Hence, acts contrary to law at the time they are done may, if emulated by other states, lead to a change in the law. An example of this phenomenon is provided by the law of the sea. The 1958 Convention on the High Seas 10 described itself as codifying CIL. 11 It defined the “high seas” as including all waters seaward of a state’s territorial sea, and guaranteed freedom of fishing to all states in this area. 12 Yet this rule of CIL, even though codified in a treaty, was altered by state practice within about thirty years. By the end of that period, so many states had proclaimed their right to deny freedom of fishing in zones extending far beyond their territorial seas that it became impossible to deny that CIL had been altered. That is, actions violating CIL (and, for that matter, a treaty) cumulated to change the law. 13 In addition to these differences, and as noted above, CIL is supposed to derive from a general and consistent practice of states followed by them from a sense of legal obligation. This concept of state practice raises its own problems. For example, what counts as state practice? What is the consequence if some types of behavior engaged in by one state are inconsistent with other behavior of that same state? 7 14 J.G. Starke, INTRODUCTION TO INTERNATIONAL LAW 38-39 (10 th ed. 1989). I suggest that one way to address these issues is to ask why a customary practice ought to be law. That is, is there some non-arbitrary justification for ascribing binding effect to a general and consistent practice? Perhaps the most plausible answer to this question is Professor Starke’s: Recurrence of the . . . practice tends to develop an expectation that, in similar future situations, the same conduct or the abstention therefrom will be repeated. When this expectation evolves further into a general acknowledgment by states that the conduct or the abstention therefrom is a matter both of right and of obligation, the transition . . . to custom may be regarded as consummated. 14 If, then, the rationale for treating custom as law is that states ought to be able to rely on the assumption that other states will behave in the future as they have in the past, it would seem to follow that behavior would count as practice if it is of a sort as would give rise to reasonable expectations that it would be followed in future similar situations. Further, if an act which would, other things equal, give rise to such expectations, is contradicted by behavior which is a better predictor of future action, it would seem that it would be unreasonable to expect that future conduct would conform to the act rather than to the contradictory behavior. These considerations facilitate addressing a particularly important issue in connection with CIL: what effect on that body of law should be accorded to actions which may be seen as proxies for behavior? For example, suppose State A announces that it feels constrained by CIL to behave in accord with Rule X. Standing alone, that announcement may well support a reasonable expectation that State A will abide by Rule X in the future, and should count as an item of practice supporting the existence of Rule X. However, if it is clear that State A in fact systematically violates Rule X, it would seem that the net effect of State A’s actions is to 8 15 1986 I.C.J. 14. 16 Id. at 98, ? 186. 17 See, discussion infra at nn. 207-232. undermine the rule. That is, if State A’s policy is in fact to violate Rule X, one could not reasonably expect it to conform to the rule simply because it made a dishonest statement. To be sure, the International Court of Justice (ICJ) in the Case Concerning Military and Paramilitary Activities in and Against Nicaragua (Nicaragua v. United States), 15 (“Nicaragua v. United States”) stated that If a State acts in a way prima facie incompatible with a recognized rule, but defends its conduct by appealing to exceptions or justifications contained within the rule itself, then whether or not the State’s conduct is in fact justifiable on that basis, the significance of that attitude is to confirm rather than weaken the rule. 16 As will discussed below, the ICJ’s assertions regarding legal principles in its judgments are not binding on states generally, 17 but even if the quoted language is considered simply on its merits, it is hard to defend. Why do hypocritical statements confirm a legal principle if the law-making process is one in which legal principles derive from conduct creating reasonable expectations? Surely, if it is known that a given state says one thing but does another, it would hardly be reasonable to rely on what the states says when forming expectations. Suppose, however, that the proxy for behavior is not simply a statement, but adherence to a treaty requiring certain behavior - how should this affect CIL? (Of course, regardless of its status as CIL, a rule established in a treaty is binding on states parties to the treaty. The question here is whether such a rule’s inclusion in a treaty should count as practice establishing a CIL obligation). In fact, it has long been established that states often rely on treaties as evidence of 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 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 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 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 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 14 27 Id. at 161 and 163, note a. 28 Id. at 162-63. 29 The Paquete Habana, 175 U.S. 677 (1900). (including Grotius, Bynkershoek and Bacon), 27 charges to juries by English judges, and the practice of states in punishing all persons committing the offense. 28 A few comments on Smith. First, it should be remembered that there were no international courts in 1820; the court’s reference to “judicial decisions” thus necessarily referred to decisions of domestic courts. Second, in context, judicial decisions were a form of state practice; the question, after all, was how “piracy” was defined, and one type of circumstance in which states would necessarily engage in practice on the subject would have been court proceedings against alleged pirates. Finally, it must be stressed in this case that all of the sources on which the court relied in Smith were consistent with one another; this was not a situation, in other words, in which state practice on a subject differed from jurists’ opinions as to the proper interpretation of the law. Indeed, it seems doubtful that there was at that time much controversy as to the proper definition of piracy in international law. The Supreme Court’s next discussion of the method of determining the content of CIL came in The Paquete Habana. 29 This famous case arose when, during the American blockade of Cuba during the Spanish-American War, United States naval vessels captured two small coastal fishing boats operating out of Havana. The captured boats were adjudged by a lower federal court, sitting in admiralty, to be prizes of war; they were sold, and the subsequent appeal concerned the disposition of the sale proceeds. The original owners sought to recover the proceeds, arguing that small coastal fishing vessels could not lawfully be treated as prizes of war 15 30 Id. at 678-79, 686. 31 Id. at 700. 32 Id. at 686-700, 701-708. under CIL 30 . In addressing this argument, the court was obliged to explain how it determined the content of CIL on this subject. The court stated: For this purpose, where there is no treaty, and no controlling executive or legislative act or judicial decision, resort must be had to the customs and usages of civilized nations; and, as evidence of these, to the works of jurists and commentators, who by years of labor, research and experience, have made themselves peculiarly well acquainted with the subjects of which they treat. Such works are resorted to by judicial tribunals, not for the speculations of their authors concerning what the law ought to be, but for trustworthy evidence of what the law really is. 31 The court then provided a 14 page review of state practice, followed by an 8 page discussion of the views of prominent writers, 32 concluding that, as a matter of international law, vessels of the type in issue could not lawfully be taken as prizes. There are several points to emphasize about this case. First, the heart of the court’s opinion as to the content of international law was its recital of state practice. It appears that writers were quoted primarily to demonstrate the correctness of the court’s narrative. The language quoted above reinforces this conclusion. Jurists and commentators, according to the court, serve the function of providing evidence of the customs and usages of nations, being able to do so because of their knowledge of their subject. In other words, writers are important in addressing matters of international law, not because their views as to what the law should be carry any weight, but because they provide a convenient catalog of the crucial element of the analysis, that is, state practice. 16 33 United States v. Smith, 18 U.S. ( 5 Wheat.) 153, 157-58 (1820). 34 Paquete Habana, supra n. 29, Brief for Appellants at 8-10. 35 Id., Brief for Captors at 7-8. 36 Id., Brief for the United States at 19. Second, it seems that the method of determining the content of CIL employed in The Paquete Habana is consistent with that employed in Smith. To be sure, the later case stressed that the works of publicists were relied upon only as evidence of state practice, while the earlier case appears to treat scholarly opinion and state practice as equally authoritative sources of law. But the Smith court was not faced with any divergence between the parties regarding the weight to be given to the opinions of writers in determining the content of CIL (though the parties did disagree as to the substance of scholarly opinion). 33 It was not, therefore, obliged to address the question of the relative importance of the writers’ views in determining the content of CIL. In contrast, in Paquete Habana, appellants took the position that the writings of publicists weighed equally with the practice of states in determining the relevant rule of CIL. 34 The captors of the vessels in question, however, asserted that the views of writers were entitled to no weight in the matter. 35 The United States, arguing for the legality of the capture, asserted as follows: Reference has been made in the brief to the fact that the writers on international law, and especially the continental writers, are far in advance of the law as determined by legislation or decisions, or by the plain consensus of agreement on the part of the nations, and that they indulge in speculations which are not justified. We are contending that the court will regard principles fairly settled, but will not be influenced by hypothetical views or considerations of what the law ought to be or may be in the future rather than what it is. 36 It would appear therefore that the language from the court’s opinion quoted above was in response to disagreement among the parties as to the weight to be given to the opinions of 17 37 See discussion at pp. 3-11, supra. 38 See, e.g., Vienna Convention on the Law of Treaties, May 23, 1969, art. 53, 1155 U.N.T.S. 331, 344; 8 I.L.M. 679, 698-99 (1969) (hereinafter Vienna Convention). writers. In essence, the court split the difference - writers’ opinions will be given weight, says the court, not as sources equivalent to the practice of states in determining the content of law, but because they describe that practice. Finally, and most important, it seems that the approach taken in Paquete Habana to determining the content of CIL fits easily with the nature of that body of law as discussed above. 37 That is, it insisted that the source of international law was “custom”, that is, actual practice which, it could reasonably be assumed, would be followed in the future. American courts therefore have two reasons for focusing on the actual practice of states in determining the content of CIL. First, such an approach best comports with the nature of that body of law. Second, the Supreme Court has held that approach to be correct. III. Current American Approaches to Determining the Content of Customary International Law In order to provide an example of the problem this article seeks to address, the following discussion will recount the approach American federal courts of appeals have taken to a particular international legal problem: dealing with the concept of jus cogens. The discussion will first explain that concept and spell out some of the difficulties it presents, and then describe the treatment that courts have given it. A. The Concept of Jus Cogens There is authority for the proposition that there is a class of rules of international law from which states are not permitted to derogate. 38 The term jus cogens is applied to this class of 18 39 Id. 40 CONGRESSIONAL RESEARCH SERVICE, LIBRARY OF CONGRESS, TREATIES AND OTHER INTERNATIONAL AGREEMENTS: THE ROLE OF THE UNITED STATES SENATE 43 n. 2 (Sen. Comm. Print 106-71, 106 th Cong., 2d Sess. 2001) (hereinafter cited as ROLE OF THE SENATE). 41 THE STATESMAN’S YEARBOOK: THE POLITICS, CULTURES AND ECONOMIES OF THE WORLD 2002 6 (Barry Turner ed. 2001). rules. The idea that such a class raises a number of problems. One way to evaluate the current approach taken by American courts to CIL is to examine the approach taken by the courts in addressing these problems. This section of the article provides some background on the subject; the sections following described and critique the decisions of the federal appellate courts dealing with jus cogens. One difficulty is basic: what reason is there for an American court to accept the existence of such a category of rules? To be sure, the concept is expressly embodied in the Vienna Convention on the Law of Treaties, 39 (“Vienna Convention”) but the United States is not a party to that treaty. Further, only 91 other states are parties; 40 by comparison, there are 189 states are members of the United Nations. 41 That is, fewer than half the states in the world have bound themselves to the Vienna Convention, and thus to a treaty obligation to accept the jus cogens concept. If American courts are to treat the concept as a part of international law, then, they must justify their action by relying on some source of law other than either a treaty obligation of the United States or near universal acceptance by states through a multilateral treaty. One might hope to gain some insight into the legal basis for jus cogens by examining the history of the idea; unfortunately, that inquiry only complicates the question. The proposition that such a class of norms existed began to receive systematic attention from scholars of international law following World War I. Initially, it was discussed in connection with the law 19 42 A. Mark Weisburd, The Emptiness of the Concept of Jus Cogens, As Illustrated by the War in Bosnia-Herzegovina, 17 MICH. J. INT’L L.1, 10-14. (1995) 43 Report of the International Law Commission on the Work of Its Eighteenth Session, U.N. GAOR, 21st Sess., Supp. No. 9, U.N. Doc. A/6309/Rev. 1 (1966), reprinted in [1966] 2 Y.B. Int'l L. Comm'n 169, 248, U.N. Doc. A/CN.4/SER.A/1966/Add.1. of treaties; proponents of the concept argued that treaties purporting to achieve certain forbidden objectives were void and unenforceable by international tribunals, analogizing to the rule that, in domestic legal systems, contracts seeking to attain objectives contrary to public policy are similarly unenforceable. Those taking this position appeared to identify forbidden objectives by reference to moral principles, rather than by focusing on positive acts of states. In 1953, the concept was relied upon when the International Law Commission of the United Nations (ILC) began to consider the codification of the law of treaties. The first special rapporteur of the ILC on this subject, then-Professor Hersch Lauterpacht, prepared a draft convention on the law of treaties which included an article embodying the concept of jus cogens. He and his successors as special rapporteur continued the approach of relying on moral principles as at least a partial basis for the claim for the existence of jus cogens principles. 42 By the time of the Vienna Conference on the law of treaties in 1968 and 1969, however, this concept had been modified. The ILC, in its comments to its draft convention, took the position that a rule of jus cogens could be modified by a general multilateral treaty - that is, that governments had the authority to change the rules. 43 This view was reinforced at the Conference itself. The ILC’s draft article on jus cogens provided: A treaty is void if it conflicts with a peremptory norm of general international law from which no derogation is permitted and which can be 20 44 Id. at 247. 45 Vienna Convention on the Law of Treaties, May 23, 1969, part V, section 2, art. 53, 1155 U.N.T.S. 311. 46 See, e.g., INGRID DETTER, THE INTERNATIONAL LEGAL ORDER 174-176 (1994); Mark W. Janis, The Nature of Jus Cogens, 3 CONN. J. INT'L L. 359, 360-63 (1988); Karen Parker and Lyn Beth Neylon, Jus Cogens: Compelling the Law of Human Rights, 12 HASTINGS INT'L & COMP. L. REV. 411, 419-22 (1989). modified only by a subsequent norm of general international law having the same character. 44 As finally adopted by the representatives of the governments involved, however, the article on jus cogens read as follows: A treaty is void if, at the time of its conclusion, it conflicts with a peremptory norm of general international law. For the purposes of the present Convention, a peremptory norm of general international law is a norm accepted and recognized by the international community of States as a whole as a norm from which no derogation is permitted and which can be modified only by a subsequent norm of general international law having the same character. 45 As modified, the article clearly sees the test of a rule’s jus cogens status as its acceptance as such by states, not its derivation from moral rules. In other words, a concept that originated in the belief that moral principles imposed legal limits on state authority - in effect, applying a natural law approach - was codified in a form that grounded limitations on states’ freedom solely on acceptance of those limits by states, that is, in a form shaped to satisfy positivist conceptions of the nature of law. Since 1969, this doctrinal confusion has not abated. A number of writers since that time have continued to argue that jus cogens should be understood as a natural law doctrine. 46 This group even includes scholars who normally do not rely on natural law concepts in their 21 47 Compare MYRES S. MCDOUGAL, HAROLD D. LASSWELL, and LUNG-CHU CHEN, HUMAN RIGHTS AND WORLD PUBLIC ORDER: THE BASIC POLICIES OF AN INTERNATIONAL LAW OF HUMAN DIGNITY 68-71 (1980) with id. at 339-342. 48 W. Michael Reisman, Unilateral Action and the Transformations of the World Constitutive Process: The Special Problem of Humanitarian Intervention, 11 EUR. J. INT’L L. 3, 15 (2000). 49 Id., n. 29. 50 Resolution 34/65 B, ? 4, A/RES/34/65 (Nov. 29, 1979). expositions of international law. 47 Professor Reisman has observed that the jus cogens concept has come to be used, in the human rights context, in a way quite different from its use in the Vienna Convention. 48 As he has observed: In the [Vienna] Convention, a jus cogens deprives of putative legal effect other, inconsistent treaty obligations. In human rights discourse, jus cogens has acquired a much more radical meaning, evolving into a type of super-custom, based on trans-empirical sources and hence not requiring demonstration of practice as proof of its validity. 49 In short, the phrase jus cogens is currently applied to two radically different concepts, the legal bases and implications of which differ fundamentally. Nor has state practice since 1969 made the matter any clearer. The only examples of state practice apparently embodying the jus cogens concept are rhetorical. Delegates to international organizations and conferences occasionally label particular norms as jus cogens, and the General Assembly adopted a resolution declaring the 1978 Camp David accords between Israel and Egypt to “have no effect in so far as they purport to determine the future of the Palestinian people and the Palestinian territories occupied by Israel since 1967,” 50 apparently applying the concept of jus cogens if not the term itself. The General Assembly also adopted numerous resolutions labeling “void” various unilateral acts of states that arguably violated the 22 51 LAURI HANNIKAINEN, PEREMPTORY NORMS (JUS COGENS) IN INTERNATIONAL LAW: HISTORICAL DEVELOPMENT, CRITERIA, PRESENT STATUS 302-03 (1988). 52 CHESTER A. CROCKER, HIGH NOON IN SOUTHERN AFRICA 392-446 (1992). 53 Hannikainen, supra n. 51, at 302. 54 ROLE OF THE SENATE, supra n. 40, at 45. principle of self-determination of peoples, particularly acts by South Africa with respect to its treatment of Namibia and the institution of apartheid. 51 But these declarations had little effect on actual behavior of states with respect to the issues in question; for example, despite condemnations of its occupation of Namibia by U.N. organs, South Africa gave up its control of Namibia only when it received a quid pro quo in the form of Cuba’s agreement to withdraw its troops from Angola. 52 That is, South Africa’s occupation was ended by an arrangement in which its interests were taken into account rather than treated as illegitimate, even in connection with a situation in which it had, if the implications of the General Assembly’s resolutions are taken seriously, derogated from a norm from which no derogation was permitted. Indeed, in 1988, Hannikainen concluded, after an exhaustive survey of state practice, that with the exception of punitive action against pirates and with the possible exception of enforcement action against the gravest forms of racial discrimination, the international community of States as a whole has not engaged with any consistency in enforcement or punitive action against grave violations of the basic norms of the international legal order. 53 Nothing has happened since 1988 to alter the conclusion Hannikainen reached in that year. While American courts might hope for some guidance in dealing with jus cogens from the political branches of the federal government, the executive and the Congress have taken different views. The Vienna Convention, including, of course, its provision regarding jus cogens, was submitted to the Senate for consent to ratification in 1971. 54 The Senate last held 23 55 Id. at 45, 48. 56 Id. at 45-49. 57 Id. at 20-21. 58 Id. at 21. 59 Id. 60 Reports of the Commission to the General Assembly, [1966] 2 Y.B. Int’l L. Comm’n 169, 247-248, U.N. Doc. A/CN.4/SER.A/1966/Add.1 (hereinafter cited as ILC Reports). hearings on the Convention in 1986, and it remains in Committee. 55 While the Executive Branch has consistently favored the United States’s becoming a party to the Vienna Convention, 56 concerns in the Senate have blocked further action. The primary disagreement involves the effect of the Vienna Convention on executive agreements into which the President enters without Senate action. 57 However, concerns about jus cogens are also important. Specifically, the Senate appears to be concerned about “by whom and how” jus cogens norms would be established 58 and about the Vienna Convention’s requirement of compulsory adjudication by the ICJ of unresolved disputes over jus cogens. 59 Thus, concerns about the scope of the obligation which would be assumed in respect of jus cogens are one of the reasons why the United States has never become a party to the Vienna Convention. This last point illustrates a further problem with the concept of jus cogens: what are the peremptory norms from which no derogation is permitted? In its commentary to its draft Convention on the Law of Treaties respecting the article setting out the jus cogens concept, the ILC gave only one example of a rule of that class: the provision of the United Nations Charter prohibiting the use of force between states. 60 At the subsequent conference on the law of treaties, however, the various states’ representatives offered widely differing lists of rules 24 61 JERZY SZTUCKI, JUS COGENS AND THE VIENNA CONVENTION ON THE LAW OF TREATIES: A CRITICAL APPRAISAL 119-20 (1974). 62 Id. at 119. 63 ILC Reports, supra n. 60, at 248. 64 See Opinion no. 2, Conference on Yugoslavia Arbitration Commission, Jan. 11, 1992, 31 INT'L LEGAL MATERIALS 1497, 1498 (1992). 65 Id. 66 MICHAEL J. GLENNON, LIMITS OF LAW, PREROGATIVES OF POWER: INTERVENTIONISM AFTER KOSOVO 37- 89 (2001). meeting the requirements of jus cogens; of the 26 delegations that offered examples of peremptory norms, no more than thirteen agreed with respect to any one rule. 61 Moreover the mutability of the lists of such norms is surprising. For example, 6 delegations at the Vienna Conference took the position that respect for the right of self-determination of peoples was a jus cogens norm, 62 as did some members of the ILC. 63 Yet by 1992, an arbitration commission established by the European Community’s Conference on Yugoslavia could assert that “international law as it currently stands does not spell out all the implications of the right to self- determination,” 64 and went on to subordinate that right to the principle that “whatever the circumstances, the right to self-determination must not involve changes to existing frontiers at the time of independence (uti possidetis juris) except where the States concerned agree otherwise.” 65 Again, as noted above, the prohibitions on the use of force provided in the United Nations Charter was the one rule which even the ILC was prepared to label as jus cogens. Yet, as Professor Glennon has demonstrated, it seems impossible now to classify that prohibition even as a matter of CIL, let alone a peremptory norm. 66 And a number of scholars, at least, seem 25 67 See, e.g., sources collected in A. Mark Weisburd, International Law and the Problem of Evil, 34 VAND. J. TRANSNAT’L. L. 225, 233-37 (2001). 68 859 F.2d 929 (D.C. Cir. 1988) (hereinafter Citizens in Nicaragua). 69 Id. at 929, 939-940. prepared to entertain the possibility that international law permits violation of the relevant Charter prohibitions if necessary to prevent massive human rights violations. 67 In short, the status of the concept of jus cogens as an element of international law is quite confused. In the next section of the article, we will see what American federal appellate courts have made of this confusion. B. The Courts of Appeals and Jus Cogens Since the Vienna Convention was signed in 1969, the concept of jus cogens has figured in 16 decisions of the federal courts of appeals. This section describes the approaches to that concept taken in those cases; the following section critiques those decisions. The first of these cases to deal with jus cogens was Committee of United States Citizens Living in Nicaragua v. Reagan 68 (“Citizens in Nicaragua”). Plaintiffs in that case sought injunctive and declaratory relief to prevent the United States from funding the Nicaraguan contras. They based their claim, among other grounds, on the argument that jus cogens obliged states which have submitted to an international court’s jurisdiction to abide by that court’s judgment, and that the failure of the United States to abide by the judgment of the ICJ in Nicaragua v. United States therefore entitled them to relief. 69 In rejecting that argument, the court held that the norm upon which plaintiffs relied did not satisfy any definition of jus cogens - either that in the Vienna Convention or any other. 26 70 Id. at 940, citing Vienna Convention, art. 53. 71 Giorgio Gaja, Jus Cogens Beyond the Vienna Convention, 172 R.C.A.D.I. 271, 283-87 (1982). 72 Citizens in Nicaragua, supra n. 68, at 941. 73 Id. at 941. 74 Kenneth C. Randall, Universal Jurisdiction Under International Law, 66 TEX. L. Rev. 785 (1988). 75 Marjorie M. Whiteman, Jus Cogens in International Law, with a Projected List, 7 GA. J. INT'L & COMP. L. 609 (1977). 76 Id. at 941, citing RESTATEMENT §331 cmt. e. Relying on the language of the Vienna Convention 70 and on one scholarly article 71 , the court asserted that a norm achieved jus cogens status only when a rule of CIL was recognized as jus cogens by the international community as a whole. The court then held that the rule for which plaintiffs contended did not satisfy this definition, citing state practice as collected by various writers for its conclusion. 72 In dictum, the court went on to list rules that “arguably” satisfied the test for jus cogens, mentioning the United Nation Charter’s prohibition on the use of force, and prohibitions on genocide, slavery, murder, torture, prolonged arbitrary detention, and racial discrimination, relying on the Restatement (specifically comment k to section 102, section 702, and comment n to that section) 73 and on law review articles by Professor Randall 74 and Assistant Legal Advisor Whiteman. 75 The court also quoted the Restatement to the effect that the scope of jus cogens is so uncertain that a domestic court should not refuse to enforce an agreement on the ground that it violates jus cogens. 76 27 77 965 F.2d 699 (1992), cert. denied 507 U.S. 1017 (1993). 78 Id. at 702. 79 Id. at 713-14. 80 Id. at 718-19. The court ultimately held that Argentina had waived its immunity defense, id. at 720-22. 81 Id. at 714. 82 Id. at 715. 83 Adam C. Belsky et al., Comment, Implied Waiver Under the FSIA: A Proposed Exception to Immunity for Violations of Peremptory Norms of International Law, 77 CALIF. L. REV. 365 (1989); David F. Klein, Comment, A Theory for the Application of the Customary International Law of Human Rights by Domestic Courts, 13 YALE J. INT’L L. 362 (1988). 84 Note, The Nuremberg Legacy: An Unfulfilled Promise, 66 S. CAL. L. REV. 833 (1990) (hereinafter Legacy). The next case dealing with jus cogens was Siderman de Blake v. Republic of Argentina. 77 Plaintiffs made claims for, among other things, acts of torture carried out by Argentina. 78 Argentina raised the defense of sovereign immunity, and plaintiffs offered several arguments against the availability of that defense. One of these arguments was that state torture was a violation of a jus cogens rule, and that the defense of sovereign immunity was not available in suits alleging jus cogens violations. 79 The court ultimately rejected this argument 80 , but not before discussing jus cogens at length. The court defined jus cogens by referencing the definition in the Vienna Convention. 81 It distinguished jus cogens norms from CIL, however, asserting that, “[w]hereas customary international law derives solely from the consent of states, the fundamental and universal norms constituting jus cogens transcend such consent;” 82 in reaching this conclusion, the court relied in on two student law review comments 83 , on a student law review note 84 , and on a quotation from 28 85 1970 I.C.J. 3. 86 Siderman, supra n.?, at 716. 87 630 F.2d 876 (2d Cir. 1980). 88 672 F.Supp. 1531 (N.D. Cal. 1987). 89 726 F.2d 774, 781 (D.C. Cir. 1984) (opinion of Edwards, J.) cert. denied 470 U.S. 1003 (1985) and id. at 820 (opinion of Bork, J.). 90 December 10, 1984, 1465 U.N.T.S. 85, 23 I.L.M. 1027 (1984). 91 Siderman, supra n. ?, at 716-17. 92 Randall, supra n. 74. 93 Belsky et al., supra n. 83; Klein, supra n. 83. the ICJ decision in the Case Concerning the Barcelona Traction, Power & Light Co. (Belgium v. Spain). 85 The court also quoted comment k to Restatement Section 102 to the effect that jus cogens norms “prevail over and invalidate international agreements and other rules of international law in conflict with them.” 86 The court went on to hold that torture violates CIL, relying on Filartiga v. Pena-Irala, 87 Forti v. Suarez-Mason, 88 two opinions from Tel-Oren v. Libyan Arab Republic, 89 section 702(d) of the Restatement, on the existence of the Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, 90 (Torture Convention) and on the fact of the Senate’s having consented to the ratification of that treaty by the United States. 91 Finally, relying on Filartiga’s characterization of the degree of international condemnation of torture, as well as on the article by Professor Randall also cited in Citizens in Nicaragua, 92 on two of the student comments on which it had previously relied, 93 and on an 29 94 Parker and Neylon, supra n. 46. 95 Siderman, supra n. ?, at 717. 96 In re Estate of Marcos Human Rights Litigation, 978 F.2d 493, 500 (9 th Cir. 1992), (citing Siderman and Citizens in Nicaragua for the concept of jus cogens and Siderman for the proposition that prohibition of torture is a jus cogens norm); Gisbert v. United States Attorney- General, 988 F.2d 1437, 1448 n. 22 (5 th Cir. 1993) (citing Citizens in Nicaragua for the definition of jus cogens); Hilao v. Marcos (In re Estate of Marcos, Human Rights Litigation) 25 F.3d 1467, 1471 n.6, 1475 (citing Siderman for the definition of jus cogens and for the proposition that the prohibition on torture is a jus cogens norm). 97 26 F.3d 1166 (D.C. Cir. 1994). 98 Id. at 1173-74. 99 United States v. Matta-Ballesteros, 71 F.3d 754, 764 n. 5 (9 th Cir. 1995) (citing Citizens in Nicaragua and Siderman for the definition and content of jus cogens); Smith v. Socialist People’s Libyan Arab Jamahiriya, 101 F.3d 238, 242 (2d Cir. 1996) (citing Belsky et al., supra n. 83); Hilao v. Estate of Marcos, 103 F.2d 767, 778 (9 th Cir. 1996) (citing Siderman); Hilao v. Estate of Marcos, 103 F.2d 789, 795 (9 th Cir. 1996) (cites to Siderman for content of jus cogens); Cabiri v. Government of the Republic of Ghana, 165 F.3d 193, 201 (2d Cir. 1999) (using term jus cogens without discussing or defining it). article by Ms. Parker and Ms. Neylon, 94 the Siderman court concluded that the prohibition of torture was a jus cogens norm. 95 The first three appellate decisions dealing with jus cogens after Siderman did not analyze that concept, simply citing to one or both of Citizens in Nicaragua and Siderman. 96 Princz v. Federal Republic of Germany 97 went slightly further, defining jus cogens and specifying its content by reference to comment k to Restatement section 102, Restatement section 702, comment n to section 702, and the law review note cited in Siderman, as well as in reliance on Citizens in Nicaragua and Siderman. 98 Princz was followed by five more decisions that did no more than either cite earlier cases for their treatment of jus cogens or else cite without qualification authorities cited in those earlier cases. 99 The next case that varied its analysis even 30 100 218 F.3d 1004 (9 th Cir. 2000). 101 Hilao, supra n. 96. 102 John Dugard and Christine Van den Wyngaert, Reconciling Extradition with Human Rights, 92 A.J.I.L. 187 (1998). 103 218 F.3d at 1016, n. 15. The court also relied on Filartiga, supra n. 4; Trajano v. Marcos, 978 F.2d 493 (9 th Cir. 1992), cert. denied sub nom. Marcos-Manotoc v. Trajano, 508 U.S. 972 (1993) and Kadic v. Karadzic, 70 F.3d 232 (2d Cir. 1995) for the proposition that official torture violates CIL, 218 F.3d at 1016, n. 15. 104 Sampson v. Federal Republic of Germany, 250 F.3d 1145 (7 th Cir. 2001). 105 Id. at 1146-49. 106 Id. at 1149-50. slightly was Cornejo-Barretto v. Seifert, 100 relying on Restatement Section 702, In re Estate of Marcos, Human Rights Litigation, 101 and an article by Professors Dugard and Van den Wyngaert, 102 characterized the right to be free from torture as a jus cogens norm. 103 Sampson v. Federal Republic of Germany 104 differed from all of the foregoing cases in that it did not accept the concept of jus cogens uncritically. The case was a suit by a Holocaust survivor against the German government, seeking damages for his enslavement in Germany during the Nazi period. Germany relied on the defense of sovereign immunity. The plaintiff sought to defeat that defense by arguing that Germany’s commission of acts violating jus cogens norms amounted to an implicit waiver of sovereign immunity. 105 While the court in that case relied on Siderman and Citizens in Nicaragua for its definition of jus cogens, 106 it also quoted 31 107 Oppenheim’s International Law (7 th ed. 1992). 108 Anthony D’Amato, Human Rights as Part of Customary International Law: A Plea for a Change of Paradigms, 25 GA. J. INT’L & COMP. L. REV. 47 (1995-96). 109 250 F.3d at 1155. 110 Id. at 1150-1156. 111 Alvarez-Machain v. United States, 266 F.3d 1045, 1050 (9 th Cir. 2001) (defines jus cogens by citing RESTATEMENT §102 cmt. k); 112 Buell v. Mitchell, 274 F.2d 337 (6 th Cir. 2001). Oppenheim’s treatise 107 and Professor D’Amato 108 to establish the vast disagreement among scholars as to the content of jus cogens. The court went on to observe: Domestic courts (and for that matter the other two branches of our government), do not determine the content of the jus cogens doctrine. Instead, it emanates from academic commentary and multilateral treaties, even when unsigned by the United States. Only as a last resort should United States courts infer jurisdiction over foreign sovereigns through this loosely woven subject matter. See also Tel- Oren v. Libyan Arab Republic, 233 U.S. App. D.C. 384, 726 F.2d 774, 827 (D.C. Cir. 1984) (Robb, J., concurring) (“Courts ought not to serve as debating clubs for professors willing to argue over what is or what is not an accepted violation of the law of nations.”). Absent congressional direction, such overactive involvement by our judiciary would challenge the consent-based structure of our constitutional system. 109 The court went on to reject plaintiff’s implied waiver argument, and hold his claims against Germany barred by sovereign immunity. 110 Three federal appellate courts have dealt with the concept of jus cogens since Sampson. One of these decisions relied on earlier decided cases or authorities cited therein in its treatment of that subject, 111 but the other two resembled Sampson in considering factors not addressed in the earlier cases. Buell v. Mitchell 112 was an appeal from a denial of a petition for a writ of habeas corpus brought by a person convicted by a state court of aggravated murder and 32 113 Id. at 344. 114 Id. at 370. 115 Id. 116 Id. at 372, quoting RESTATEMENT, supra n. 5, § 102. 117 Buell, supra n. 112 at 372, quoting RESTATEMENT, supra n. 5, § 102 cmt. b. 118 Buell, supra n. 112 at 372, quoting RESTATEMENT, supra n. 5, § 102 cmt. c. 119 Buell, supra n. 112 at 372-73. 120 Id. at 373. sentenced to death. 113 Among many other grounds, the petitioner argued that a prohibition on the death penalty was not simply CIL, but a matter of jus cogens. 114 The court relied on the Vienna Convention, In re Estate of Marcos, Human Rights Litigation, and Citizens in Nicaragua for the definition of jus cogens. 115 However, it stressed language of Restatement section 102 and of comments b and c to that section to the effect that, for a putative rule to be considered a norm of CIL, it must amount to “a general and consistent practice of states followed by them from a sense of legal obligation,” 116 that the practice “should reflect wide acceptance among the states particularly involved in the relevant activity,” 117 and that “a practice which is generally followed but which states feel legally free to disregard does not contribute to customary law.” 118 The court also relied on comment k and reporter’s note n to section 102 for the proposition that jus cogens rules are CIL rules of “higher status.” 119 The court further stressed language from Citizens in Nicaragua underlining that a CIL norm became a jus cogens norm only upon its acceptance as jus cogens by the international community as a whole. 120 The court then examined the problem before it. It noted that a very large number of states retain the death penalty - 33 121 William A. Schabas, International Law and the Abolition of the Death Penalty, 55 WASH. & LEE L. REV. 797 (1998). 122 The Covenant, supra n. 19. 123 Buell, supra n. 112, at 373. 124 Id. The court went on to hold that, even if it was wrong and the prohibition of the death penalty was a jus cogens norm, the fact that the petitioner sought to rely on that norm against an American official rather than as a basis for a civil claim against a foreign official made the matter a question to be answered by the political departments of the federal government, id. at 373-76. 125 Hain v. Gibson, 287 F.3d 1224 (10 th Cir. 2002). 126 Id. at 1226-27. 127 Id. at 1242-43. relying for this information on an article by Professor Schabas in which the data was presented 121 - and on the fact that a great many states were parties to the International Covenant on Civil and Political Rights, 122 which permits the death penalty. 123 The court further observed that there was no indication that states abolishing the death penalty had done so because they felt they were legally obliged to take that step, and concluded that, since CIL could not be understood as requiring the end of the death penalty, such a requirement could not have risen to the level of jus cogens. 124 The last case in this group of sixteen was Hain v. Gibson. 125 Like Buell, Hain was an appeal from a denial of a petition for a writ of habeas corpus filed by a person sentenced to death. 126 The petitioner raised, among other grounds for his petition, the argument that, since he was 17 at the time he committed the murders for which he had received the death sentence, executing him would violate a jus cogens norm. 127 The court cited Buell for the proposition that a jus cogens norm was one “that has ‘risen to the level that the international community as a 34 128 Id. at 1243, citing Buell, supra n. 112, at 373. 129 Hain, supra n. 125, at 1243-44. This court also relied on Buell for the proposition that, even if the rule against executing persons for crimes committed prior to the eighteenth birthday was jus cogens, the issue of how to respond to that rule was one properly left to the political branches of the federal government, id. at 1244. whole recognizes it as . . . a norm from which no derogation is permitted.’” 128 Without reviewing state practice, but relying on Buell, the court concluded that the rule against the imposition of the death penalty on a person for crimes committed prior to the person’s eighteenth birthday had not been adopted from a sense of legal obligation. For this reason, the court held that the rule was not a rule of CIL and therefore not a rule of jus cogens. 129 These cases, then, illustrate the federal appellate courts’ approach to issues of international law not turning on treaties. In the next section, we turn to the question of how well this approach withstands analysis. C. Critique of the Jus Cogens Decisions of the Courts of Appeals Of the foregoing cases, only five appear to have engaged in any significant analysis of the concept of jus cogens: Citzens in Nicaragua, Siderman, Sampson, Buell, and Hain. The other cases simply cite to one or more of these five, or to authorities cited in those cases, with no real independent consideration of the question. Each of these five courts was obliged to address two related issues: first, it was necessary to decide whether the concept of jus cogens is a part of international law; assuming an affirmative answer to the first question, the court then had to address whether the particular activity involved in the case before it violated a rule with jus cogens status. What would be a fair evaluation of the treatment of the jus cogens issue in these decisons? 35 130 Citizens in Nicaragua, supra n. 68, at 939-41. 131 Whiteman, supra n. 75, at 609-626; Randall, supra n. 74, at 830. 132 Siderman, supra n. ?, at 715. 133 Id. 134 Belsky et al., supra n. 83, at 385-86; Klein, supra n. 83 at 350-52; Legacy, supra n. 84, at 868. First, it will be helpful to recall briefly how each court dealt with these two issues. Citizens in Nicaragua relied on the Vienna Convention and one writer to define jus cogens and on state practice as reported in scholarly articles to conclude that the activity in issue was not a violation of a jus cogens norm. The court in that case went on in dictum to list certain norms which it suggested would have jus cogens status, relying on the Restatement and two articles, 130 neither of which purported to base its conclusions on state practice. 131 Though the court in Citizens Living in Nicaragua did not acknowledge the confusion as to the manner by which a particular rule becomes jus cogens, its reliance on the Vienna Convention and its focus on the actual practice of states amounts to an acceptance of the idea of jus cogens as a product of state acceptance, rather than as deriving simply from moral principles. Siderman took an approach somewhat different from that of Citizens in Nicaragua. Siderman purported to rely on the Vienna Convention to define jus cogens, but its description of that body of norms as transcending state consent 132 is difficult to reconcile with the treaty’s focus on state acceptance as the hallmark of jus cogens norms. The Siderman court derived its view of a transcendental jus cogens from the Restatement, from three items of student law review writing and by reference to an ICJ opinion. 133 The student pieces on which the court relied all based their “jus cogens as moral principles” approach on scholarly writing, 134 though two of the three 36 135 Klein, supra n. 83, at 352; Legacy, supra n. 84, at 868. 136 Siderman, supra n. ?, at 716. 137 Id. at 716-17. 138 Randall, supra n. 74, at 830. 139 Belsky et al., supra n. 83 at 393-94. 140 Legacy, supra n. 84, at 354 n. 111. 141 Parker and Neylon, supra n. 46, at 437-439. relied on the implications of the Nuremberg trials as well. 135 With respect to the issue before it, the court concluded that state-sponsored torture was a violation of CIL in reliance on three decisions by other federal courts, on the Restatement, and on the existence of and Senate consent to ratification of the Torture Convention. 136 Its conclusion that torture was also a violation of a jus cogens norm was based on one of the federal cases that it believed supported the CIL status of the prohibition on torture, on Citizens in Nicaragua, on the Restatement, on Professor Randall’s article, on two of the student pieces on which it had earlier relied, and on an article by Ms. Neylon and Ms. Parker. 137 Of the articles, Professor Randall supported his claim that torture violates a jus cogens norm by reference to academic writing, 138 one of the two student pieces offered no support for the assertion, 139 the other student piece relied on unnamed international conventions, 140 and Ms. Parker and Ms. Neylon relied on the existence of various conventions and on resolutions by the United Nations General Assembly, on a statement by a United Nations special rapporteur, and on scholarly opinion. 141 The Sampson court differed from all the others discussed here in acknowledging the confused character of the concept of jus cogens, relying on academic commentary for the 37 142 Sampson, supra n. 104, at 1155. 143 Buell, supra n. 112, at 372-74. 144 Hain, supra n. 125, at 1243-44. proposition that academics disagreed on the issue. The character of slavery as a violation of jus cogens was not contested in that case. 142 The Buell court relied on familiar authorities to define jus cogens, but put particular weight on its characterization as a higher degree of CIL. It relied on state practice, collected in a scholarly article, as a basis for concluding that the practice under consideration - imposition of the death penalty - did not violate CIL, and therefore could not violate jus cogens. It likewise relied on the absence of evidence that those states which had abolished the death penalty had done so from a sense of legal obligation. 143 It thus appears to follow Citizens in Nicaragua rather than Siderman in treating jus cogens as a product of state acceptance rather than of the application of transcendental values. Finally, Hain essentially relied on Buell for its jus cogens analysis, even concluding in reliance on Buell that states which had abolished the death penalty for crimes committed prior to the offender’s eighteenth birthday had not acted out of a sense of legal obligation. 144 This analysis is questionable, since Buell dealt with the very different issue of whether CIL required the complete abolition of the death penalty; the analysis may reflect the court’s ultimate conclusion that the international law issues did not really matter in any event, since the question was one that should be left to the political branches whatever the state of international law. What do these cases tell us about the methods American courts currently use to determine the content of CIL? First, what is perhaps most striking about these decisions is their generally 38 145 See discussion at nn. 38-67, supra. 146 Sampson, supra n. 104, at 1149-55. uncritical acceptance of the jus cogens concept itself. As noted above, the idea that such a doctrine is part of international law seems mired in confusion and is certainly not supported by the actual practice of states. 145 Further, the fact that the Vienna Convention includes the concept is part of the reason the Senate rejected that treaty. Yet none of the decisions discussed simply rejected the existence of the concept. Only Sampson hesitated to take it at face value, andthe court’s concern was apparently more a matter of the difficulty of determining which rules had jus cogens status thanof the status of the concept itself. 146 In other words, according to the standards set out in Paquete Habana for assessing the content of CIL, there is at least considerable doubt as to whether the jus cogens concept is anything other than a club with which academics beat each other. Yet most of the decisions discussed here accept that doctrine without even hinting that its standing is at all doubtful. To be sure, the decisions are much less questionable in their outcomes. Citizens in Nicaragua, Buell and Hain all held that the international norms alleged to have been violated in those cases were not of jus cogens status. While Siderman accepted the jus cogens status of the rule against torture and the status of the prohibition on slavery was not challenged in Sampson, both cases rejected the argument that violations of jus cogens operated to eliminate the protections of sovereign immunity. Nonetheless, the decisions discussed here treat jus cogens as a category of international legal norms despite the absence of any evidence from which they could infer that states’ behavior would lead to the reasonable expectation that this doctrine would be applied by 39 governments in their relations with one another. Since these decisions reach their conclusions regarding the existence of jus cogens without such evidence, they are doubtful as a matter of international law. Further, and for the same reason, they simply flout the teaching of Paquete Habana as to the proper method of determining international law in the absence of a treaty. How then do we account for this fundamental mistake by a number of courts? I would suggest that their problems developed because of the way in which they set out to determine the content of the law. Rather than looking to state practice to determine the status of jus cogens as a doctrine of international law, they based their legal conclusions on sources that in fact cannot generate legal rules and which, in this case, misstated the content of state practice. More specifically, the sources upon which the courts relied included: 1) scholarly articles cataloging state practice; 2) scholarly articles, including student pieces, basing assertions as to the content of law on moral principles derived from the writings of other scholars and from international treaties and General Assembly resolutions; 3) the Restatement; 4) decisions by other federal courts; and 5) decisions by the ICJ. Recall that the focus in Paquete Habana was on the practice of states, with the court relying in that case on scholarly writings as convenient collections of practice. The reliance of the courts in Siderman and Buell on similar articles thus is consistent with the traditional American approach on this subject. But what about the other types of authority on which these courts relied? It will be recalled that the rationale for relying on state practice as the source for CIL is expectations based. Consistent practice can give rise to rules of law, according to this theory, because states may reasonably expect that other states will in the future continue to behave as they consistently done in the past. Courts determining the content of non-treaty international law 40 by relying on articles which collect state practice, then, can justify their approach, not merely as sanctioned by the authority of Paquete Habana, but as consistent with the basis for treating CIL as law in the first place. The only function of the writers of the articles is to spare the judges the necessity of doing the drudge work of collecting state practice regarding a particular subject. And in fact, the courts in Citizens in Nicaragua and Buell rejected very weak arguments that certain rules had attained jus cogens status precisely because these collections of state practice showed just how weak the arguments were. Each of the other sources on which the courts discussed in this article relied, however, seems doubtful. This is not because reliance on these sources is not expressly permitted by Paquete Habana, but because they are unreliable guides to state practice. Certainly, in these cases, they did not alert the courts to the uncertain status of the doctrine of jus cogens. But the problem with these sources was not that, somehow, they misunderstood state practice. The problem is much more fundamental - they simply treated state practice as irrelevant. That is, they purported to find law in sources which could not give rise to reasonable expectations as to the course of future behavior by governments. The following discussions in this section examine non-practice based scholarly articles, the Restatement, and domestic and international judicial decisions, explaining why reliance on these sources amounts to conferring legislative authority on entities that simply have no claim to it. 1. Articles Basing Legal Rules on Sources Other than State Practice When a legal writer asserts a proposition to be law, one of the obvious responses to the assertion is, why do you think so? Such a response necessarily assumes that the proposition is not law simply because the writer makes the assertion; rather, it is assumed that a writer looks to 41 147 Whiteman, supra n. 75, at 625-26; Randall, supra n. 74, at 830. 148 Whiteman, supra n. 75, at 625-626. 149 Randall, supra n. 74, at 830. 150 Egon Schwelb, Some Aspects of International Jus Cogens as Formulated by the International Law Commission, 61 A.J.I.L. 946, 955 (hereinafter cited as Aspects of Jus Cogens); Egon Schwelb, The Actio Popularis and International Law, 2 ISRAEL Y.B. ON HUMAN RIGHTS 46, 56 (hereinafter cited as Actio Popularis). some source authorized to generate legal rules, and explains how the proposition the writer puts forward flows from the authorized source. The dilemma posed by articles purporting to describe CIL without relying on state practice is thus obvious. If CIL derives from state practice, but a writer claims to be able to discern a rule of CIL without considering state practice, the claim would appear to be a contradiction in terms. Examining some of the articles on which the courts relied demonstrates the weak bases for the conclusions the articles’ authors reached. Consider the articles by Assistant Legal Advisor Whiteman and Professor Randall, which were relied upon by the court in Citizens in Nicaragua to support the conclusion that acts such as genocide and torture were violations of jus cogens norms. In both articles, the authors did indeed list norms which might have jus cogens status, including those mentioned by the court. 147 Yet Assistant Legal Advisor Whiteman apparently relied solely on her own judgment as to what rules ought to fall within the jus cogens category, citing no authority to support her conclusions, 148 while Professor Randall simply cited two articles written by Professor Schwelb, 149 neither of which even purported to rely on state practice, 150 - and neither of which, for that matter, seems to support the propositions for which 42 151 Professor Randall in the text of his article characterizes “norms against hijacking, hostage-taking, crimes against internationally protected persons, apartheid, and torture” as likely to be included in any list of jus cogens norms, Randall, supra n. 74. His only support for this characterization is a footnote in which he claims that Schwelb in Actio Popularis, supra n. 150, “assert[s] that states have an obligation to outlaw acts of aggression, genocide, slavery and racial discrimination,” Randall, supra n. 74, at 830, n. 255. In fact, the page Randall cites from Actio Popularis consists solely of language Schwelb quotes from the ICJ’s opinion in Barcelona Traction, supra n. 85, in which the court characterized aggression, genocide, slavery and racial discrimination as outlawed by international law, and as acts to which any state is permitted to object; the cited language contains no assertion by Schwelb at all and no assertion that individual states have the obligation to outlaw the acts which Randall mentions, Schwelb, Actio Popularis, at 56. Randall also asserts that Schwelb in Aspects of Jus Cogens, supra n. 150, “recogniz[es] [the] principles underlying the Genocide Convention as among those that civilized nations regard as binding, Randall, supra n. 74, at 830, n. 255. In fact, at the page Randall cites, Schwelb is characterizing a holding of the ICJ, rather than recognizing anything himself, Schwelb, Aspects of Jus Cogens, at 955. No where in the language Randall cites from Schwelb’s articles is there any reference to hijacking, hostage-taking, crimes against internationally protected persons, or torture - acts which he asserts are “likely” violations of jus cogens norms in purported reliance on those articles. 152 Parker and Neylon, supra n. 20. 153 Siderman, supra n. ?, at 716. Professor Randall cites them. 151 In short, the analyses of both Assistant Legal Advisor Whiteman and Professor Randall with respect to the points for which they were cited in Citizens in Nicaragua are rather doubtful, and neither, in any event examined state practice to support her or his conclusions regarding jus cogens. Siderman provides other examples. That case relied on the article by Professor Randall, just discussed, and also on an article by Ms. Parker and Ms. Neylon 152 for the proposition that torture violates a jus cogens rule. 153 That article addressed the question as follows: Torture is widely recognized as contravening jus cogens. All major human rights agreements and instruments contain a prohibition against torture. In the relevant treaties, the prohibition is non-derogable. Torture in time of war is a grave breach of humanitarian law. To reinforce the prohibitions against torture, the United Nations General Assembly promulgated the Torture Convention. Because of the universal concern about the widespread occurrence of torture, the United 43 154 Parker and Neylon, supra n. 20, at 437-39 (footnotes omitted.) 155 Rosalyn Higgins, Derogation Under Human Rights Treaties, 48 BRIT. Y.B. INT’L. L. 281, 282 (1977). 156 Id. and id. at n. 3. 157 Parker and Neylon, supra n. 20, nn. 165-170 at 437-39. Nations Commission on Human Rights appointed a special rapporteur on torture, Peter Kooijmans, to 'promote the full implementation of the prohibition under international and national law of the practice of torture and other cruel, inhuman, or degrading treatment or punishment.' In Mr. Kooijmans' 1986 report, he emphasized the jus cogens nature of the prohibition against torture: Torture is now absolutely and without any reservation prohibited under international law whether in time of peace or of war. In all human rights instruments the prohibition of torture belongs to the group of rights from which no derogation can be made. The International Court of Justice has qualified the obligation to respect the basic human rights, to which the right not to be tortured belongs beyond any doubt, as obligations erga omnes . . . which every State has a legal interest [to implement]. The International Law Commission . . . has labelled serious violations of these basic human rights as 'international crimes,' giving rise to the specific responsibility of the States concerned. In view of these qualifications the prohibition of torture can be considered to belong to the rules of jus cogens. If ever a phenomenon was outlawed unreservedly and unequivocally it is torture. 154 The first sentence in the foregoing quotation is supported by a citation to an article in which then-Professor Higgins, asserts that, with respect to human rights conventions, “[t]here certainly exists a consensus that certain rights - - the right to life, to freedom from slavery or torture - - are so fundamental that no derogation may be made.” 155 However, this assertion is supported only in part, and then only by reference to the opinions of two other scholars and to a decision by the ICJ. 156 All of the other assertions made by Parker and Neylon are supported solely by reference to treaties and non-binding resolutions of various international organizations. 157 44 158 Kelly, supra n. 21, at 478, 492. 159 Louis B. Sohn, Sources of International Law, 25 GA. J. INT’L. & COMP. L. 399 (1995/96). 160 See discussion at n. 8, supra. This point raises the issue of the value to be accorded such sources in determining the content of CIL. First, consider the reliance on the authority of a noted scholar to support the assertion that a particular proposition is a legal rule. This would seem justified only so long as the scholar’s opinion flows reasonably from a consideration of some rule whose legal character is undisputed. If we ignore the source of the scholar’s view, and rely on that opinion solely on the basis of the scholar’s reputation, we effectively treat the scholar as a legislator. In fact, it is not uncommon for international law scholars to view themselves in this fashion, as Professor Kelly has noted. 158 Indeed, Professor Sohn has asserted, with respect to the CIL of human rights, that “states really never make international law on the subject of human rights. It is made by the people that care; the professors, the writers of textbooks and casebooks, and the authors of articles in leading international law journals.” 159 But it is by no means clear that, because law professors are pleased to see themselves as legislators, anyone else should take the same view. Yet the views of then-Professor Higgins upon which Parker and Neylon relied seem to have been only weakly supported. It would appear, that is, that they are assuming that her assertion is true simply because she said it. It is also necessary to consider the reliance by Parker and Neylon on the language of various treaties and non-binding declarations to support their conclusion regarding the jus cogens status of torture. As noted above, 160 it would appear that treaties and non-binding declarations are useful evidence of state intentions, absent better evidence. That is, if a 45 161 Id. at 437 n. 165. 162 A.M. Weisburd, Implications of International Relations theory for the International Law of Human Rights, 38 COLUM. J. TRANSNAT’L. L. 45, 55-60 (1999). government binds itself legally to behave in a particular way, it is hardly unreasonable to assume that the government’s future behavior will conform to its legal obligation, all other things being equal. Likewise, if a government simply announces that it will adhere to a certain pattern of behavior, without assuming any legal obligation to do so, one’s most likely assumption upon hearing of the announcement is to assume that the government will behave as it has said it would behave, all other things being equal. But now suppose that all other things are not equal. Suppose that it quickly becomes apparent that the government is not conforming its behavior to its treaty commitment or to its non-binding announcement, as the case may be. Since the crucial issue is how the government will actually behave, its behavior is the best evidence on that issue. And if the behavior contradicts the government’s undertakings, it seems unreasonable to insist that the undertakings are the only relevant basis upon which to form expectations, and to argue further that the behavior can be disregarded. This last situation seems to be the case for the instruments on which Parker and Neylon rely. For example, one of the treaties on which they base their assertion is the Covenant. 161 But a very large minority of parties to that treaty violate its provisions relating to torture. 162 More generally, Professor Hathaway’s careful research suggests that not only is treaty ratification not associated with better human rights practices than otherwise expected, but it is often associated with worse practices. Countries that ratify human rights treaties often appear less likely, rather than more likely, 46 163 Oona Hathaway, Do Human Rights Treaties Make a Difference?, 111 YALE L.J. 1935, 1989 (2002). to conform to the requirements of the treaties than countries that do not ratify these treaties. 163 That is, treating a state’s ratification of human rights treaties as evidence that the state protects human rights is simply not justified by the evidence. Therefore, whatever positive expectations regarding protection of human rights might be generated by the fact that many states have ratified human rights treaties, if that fact is considered in isolation, an account of the actual behavior of states in this area makes it unreasonable to maintain such expectations. However doubtful a given treaty may be as a source of CIL, it does at least create legal obligations in its own right. This may offer some excuse for the reliance of Parker and Neylon on widely-ignored human rights treaties as sources of jus cogens. But neither the ILC, nor the United Nations Commission on Human Rights, nor any special rapporteurs of that Commission, have any capacity whatever to create law. Why then do Parker and Neylon rely on such sources to support their arguments? Either those entities correctly state the content of international law or they do not. In neither case are the entities’ conclusions themselves particularly helpful. This is perhaps obvious in the case where they misstate the content of the law. But even if their views of the law are correct, presumably what ought to count are the sources from which they were able to form their opinions - and if those sources are what is crucial, why not cite to them directly? One is forced to the conclusion that Parker and Neylon simply do not care about state practice, that is, about behavior which gives rise to reasonable expectations regarding future behavior. They cite treaties whose language contains the rule which they wish to label as legally 47 164 Paquete Habana, supra n. 29, at 700. 165 Klein, supra n. 83. 166 Siderman, supra n. ?, at 715. 167 Id. at 717. 168 Klein, supra n. 83, at 351 n. 98. 169 “[T]he specific content of norms of [jus cogens] involves the irrelevance of protest, recognition, and acquiescence: prescription cannot purge this type of illegality,” IAN BROWNLIE, PRINCIPLES OF PUBLIC INTERNATIONAL LAW 514 (3d ed. 1979). binding, ignoring completely the question of compliance with those treaties. They cite statements from entities without law-creating authority which agree with their views, but do not indicate any basis for concluding that those entities’ views of the law are correct. They are simply citing authorities that lend whatever color they have to the conclusion which they wish to reach. Their result is therefore suspect. Siderman also relied for its conclusion on three comments by law students. One may question whether law students fall within Paquete Habana’s reference to “jurists and commentators, who by years of labor, research and experience, have made themselves peculiarly well acquainted with the subjects of which they treat,” 164 but these articles have their flaws in any event. One 165 was cited for the proposition that jus cogens rules bind states regardless of their consent, 166 and also for the proposition that “there is widespread agreement among scholars that the prohibition against official torture has achieved the status of a jus cogens norm.” 167 The student author bases his assertion that state consent is irrelevant to jus cogens on a scholarly article and on a treatise. 168 The treatise contains language, not supported by references to state practice, that may support the student author’s point if taken in isolation, 169 but which comes 48 170 See discussion at nn. 43-45, supra. 171 Id. at 512-15. 172 Alan Brudner, The Domestic Enforcement of International Covenants on Human Rights: A Theoretical Framework, 35 U. TORONTO L.J. 219, 231 (1985). 173 Klein, supra n. 83, at 351. 174 See Hathaway, supra n. 163, at 1989. 175 See Hannikainen, supra n. 51, at 302. 176 Klein, supra n. 83, at 354, n. 111. from a discussion which notes that the firmest support for the concept of jus cogens is provided by the Vienna Convention - which treaty, of course, grounds that concept in state consent 170 - and is in any event quite cautious in its treatment of the issue. 171 The article cited by the student author does not purport to base its conclusions on state practice, but instead offers a theoretical justification for the doubtful assertion that human rights treaties are inherently self-executing. 172 To be sure, the student author asserts that “nations do observe jus cogens, seek to enforce it upon each other, and deny their own violations of it,” 173 showing that the author understood the importance of grounding his argument in state practice. However, he supports his assertions regarding state behavior only by mentioning the fact that states have entered into numerous treaties purporting to protect human rights, without addressing the poor human rights performance even of parties to those treaties; 174 he offers no examples of, for instance, an effort to enforce jus cogens, presumably because he could find none. 175 His assertion that the prohibition against torture is a jus cogens norm is entirely unsupported; 176 by citing him for this proposition, the Siderman court was essentially treating his assertion as true simply because he made it. 49 177 Legacy, supra n. 84. 178 Siderman, supra n. ?, at 715. 179 Legacy, supra n. 84, at 868 n. 261. 180 See discussion at nn. 43-45, supra. 181 Belsky et al., supra n. 27. 182 Id. at 382. 183 Id. at 386. A second student piece, a note, 177 is also cited in Siderman for the proposition that the obligations created by jus cogens norms transcend state consent. 178 That note, however, bases its argument that such a transcendental concept of jus cogens has entered international law solely by citing to the Vienna Convention and the Restatement. 179 As noted above, the Vienna Convention in fact rejects this approach to jus cogens; 180 in any event, the student writer does not support his argument by references to state practice. Siderman also relied on yet another student piece to support its conclusion that the concept of jus cogens is part of international law. 181 In their discussion of the basis for the concept, the student authors of that article observed that, during the sixteenth century, natural law was an important basis for international law. 182 They then assert that jus cogens “has revived the natural law idea of a law binding irrespective of the will of the sovereign states.” 183 No where in their article do the students betray any awareness of the tremendous difficulties presented by the concept of natural law, let alone seek to justify their reliance on that concept. Surely an article that deals so superficially with an idea fundamental to its thesis is a weak basis for the opinion of a court. 50 184 Siderman, supra n. ?, at 716. 185 Belsky et al., supra n. 27 at 393-94.. 186 Id., n. 152 at 393. 187 MYRES S. MCDOUGAL, ET AL., HUMAN RIGHTS AND WORLD PUBLIC ORDER 274 n.366 (1980). 188 Id. at 320-332. It should be noted that the argument regarding the CIL status of the Universal Declaration is based solely on statements by scholars, resolutions of the General Assembly, and assertions that that Declaration has been frequently “invoked” - not on assertions, The Siderman court relied on this last student article, among others, for the proposition that torture is a violation of a jus cogens norm. 184 The article addresses that issue by asserting, “The most fundamental individual rights are embodied in the concept of jus cogens. For example, a state policy of genocide, torture, or slavery, some of the worst violations of individual rights, is generally accepted as violating jus cogens norms.” 185 The only support given for this assertion is a footnote, inserted after the first of the two sentences just quoted, which read, omitting only a cross-reference, as follows: The suggestion has been made that the Universal Declaration of Human Rights, G.A. Res. 217A, U.N. Doc. A/180 at 71 (1948), reprinted in INTERNATIONAL HUMAN RIGHTS INSTRUMENTS, . . . at § 440.1, has 'the attributes of jus cogens.' M. MCDOUGAL, H. LASSWELL & L. CHEN, HUMAN RIGHTS AND WORLD PUBLIC ORDER 274 (1980). 186 In essence, what this article offers is ipse dixit from a group of law students. To be sure, they cite to a famous treatise on international law, but that work is characterized only as making a suggestion. The language to which reference is made in the cited work is itself supported only by an internal citation to another part of the same work; 187 that other part argues that the Universal Declaration of Human Rights has become CIL, but does not support the jus cogens characterization to which reference is made in the quoted footnote. 188 All in all, these students’ 51 let alone on evidence, that it has been put into practice. writing seems a weak basis for a court’s assertion that a particular rule amounts to non-derogable international law. The foregoing discussion, excessively lengthy though it may be, shows that at least some courts have decided questions of international law by relying on articles that appear to be doubtful in their reasoning and weakly supported by authority. This is not to say, of course, that courts should ignore articles in dealing with CIL, but only to say that the mere fact that an assertion about international law appears between the covers of a respected legal periodical does not necessarily mean that the assertion should be given any weight. Indeed, given the tendency of some scholars of international law to see themselves as legislators, it would seem that a particularly high degree of caution is appropriate in relying on academic writing in this field. 2. The Restatement Several of the cases discussed herein relied heavily on the Restatement in their discussions of jus cogens. While this might be seen as another instance of reliance upon academic authority, that is a bit of an over-simplification. Indeed, given the eminence of the American Law Institute, taking its views seriously is understandable. Unfortunately, however, a number of the assertions the Restatement makes regarding the content of international law seem weakly supported and, indeed, not to “restate” anything. For example, several courts relied on comment k to section 102 in their discussions of jus cogens. That comment reads: Peremptory norms of international law (jus cogens). Some rules of international law are recognized by the international community of states as peremptory, permitting no derogation. These rules prevail over and invalidate international 52 189 RESTATEMENT, supra n.5, §102 cmt. k (emphasis supplied). 190 Reisman, supra n. 48. agreements and other rules of international law in conflict with them. Such a peremptory norm is subject to modification only by a subsequent norm of international law having the same character. It is generally accepted that the principles of the United Nations Charter prohibiting the use of force (Comment h) have the character of jus cogens. See §331(2) and Comment e to that section. 189 The emphasized language in this quotation goes beyond the Vienna Convention in its approach to jus cogens, since the comment asserts that jus cogens can invalidate “other rules of international law,” presumably referring to rules of CIL, while the Convention deals only with international agreements. The Restatement’s formulation appears inconsistent with the Convention’s deriving jus cogens from state acceptance. This follows when one considers the possible forms of a conflict between a jus cogens rule and a rule of CIL. If a CIL rule was derived from state practice, but state practice subsequently changed, such that the new practice was accepted as jus cogens, there would be no conflict between the rules. Rather, the old CIL rule would simply not be a rule any longer - it would no longer correspond to state practice. Indeed, this would be true whether or not a rule embodying the new state practice was labeled jus cogens. But if a rule was accepted by the international community as a whole as a rule from which no derogation was permitted, it is difficult to see how such acceptance could co-exist with a general practice of states that was in conflict with it. If, however, jus cogens is seen as deriving from trans-empirical sources, as Professor Reisman suggests some have argued, 190 such a conflict is much easier to envision. Obviously, actual practice can differ from some rule based on something other than actual practice. But it hardly seems reasonable to characterize a rule based on something other than actual practice as a rule accepted by the international community 53 191 RESTATEMENT, §102, rptr’s. n. 6. as a whole as one from which no derogation is permitted, which is the Vienna Convention’s definition of jus cogens. How then does the Restatement explain its characterizing jus cogens in a way inconsistent with the Vienna Convention? Its explanation for comment k is set out in reporter’s note 6 to section 102: Peremptory norms (jus cogens). The concept of jus cogens is of relatively recent origin. See Schwelb, "Some Aspects of International Jus Cogens as Formulated by the International Law Commission," 61 [American Journal of International Law] 946 (1967). It is now widely accepted, however, as a principle of customary law (albeit of higher status). It is incorporated in the Vienna Convention on the Law of Treaties, Articles 53 and 64. See § 331(2) and Comment e to that section. Comment k to this section adopts the definition of jus cogens found in Article 53 of the Vienna Convention. The Vienna Convention requires that the norm (and its peremptory character) must be "accepted and recognized by the international community of States as a whole" (Art. 53). Apparently that means by "a very large majority" of states, even if over dissent by "a very small number" of states. See Report of the Proceedings of the Committee of the Whole, May 21, 1968, U.N. Doc. A/Conf. 39/11 at 471-72. 191 Not only does this discussion fail to explain why comment k’s definition of jus cogens departs from that of the Vienna Convention, it even asserts, inaccurately, that comment k adopts the definition of the Vienna Convention. Further, that comment characterizes the concept of jus cogens as “widely accepted” without identifying the entities that accept it. In particular, it offers no evidence of acceptance by states generally or by the United States in particular. In essence, readers of the Restatement are to accept its assertions on faith. Reporter’s note 6 cross-references to comment n of section 702. That comment provides that, “Not all human rights norms are peremptory norms (jus cogens), but those in clauses (a) to 54 192 Id., §702, cmt. n. 193 Id., §702(a)-(f). 194 Id. rprtrs. n. 11. 195 MCDOUGAL ET AL., supra n. 187, at 338-50. (f) of this section are, and an international agreement that violates them is void. See § 331(2).” 192 This comment is supported by reporter’s note 11, which provides: Human rights law and jus cogens. Not all human rights norms are jus cogens, but [the prohibitions on genocide, slavery or slave trade, the murder or causing the disappearance of individuals, torture or other cruel, inhuman, or degrading treatment or punishment, prolonged arbitrary detention, and systematic racial discrimination] 193 , have that quality. It has been suggested that a human rights norm cannot be deemed jus cogens if it is subject to derogation in time of public emergency; see, for example, [Article] 4 of the Covenant on Civil and Political Rights, § 701, Reporters' Note 6. Nonderogability in emergency and jus cogens are different principles, responding to different concerns, and they are not necessarily congruent. In any event, the rights recognized in clauses (a) to (f) of this section are not subject to derogation in emergency under the Covenant. Article 4 of the Covenant explicitly excludes from derogation the right to life and freedom from slavery and from torture, as well as from racial discrimination. Freedom from arbitrary detention is not included among the nonderogable provisions, but since derogation is permitted only "in time of public emergency which threatens the life of the nation," and only "to the extent strictly required by the exigencies of the situation," detentions that meet those standards presumably would not be arbitrary. See, generally, McDougal, Lasswell and Chen,[Human Rights and World Public Order] 338-50 (1980). 194 The Restatement thus offers no support from state practice for its assertions. To be sure, it cites to a treatise, but that treatise justifies its characterization of certain rules as jus cogens norms solely by reference to statements made by delegates at the conference which produced the Vienna Convention, to scholarly opinion, and to certain decisions of the ICJ. 195 It too makes no reference to state practice. 55 196 RESTATEMENT, supra n. 5, § 701, rptr.’s n. 1. In short, the Restatement’s sections on which the courts in the cases discussed above relied are not reliable guides to state practice regarding the subjects they address. To be fair, the Restatement does not claim otherwise. It states, in reporter’s note 1 to section 701, [T]he universal acceptance of human rights in principle, and active international concern with human rights, has led to some readiness to conclude that states have assumed human rights obligations. There is a disposition to find legal obligation in indeterminate language about human rights in international agreements, e.g., the United Nations Charter . . . . There is some willingness to find that the practice of states, perhaps under constitutional, political, or moral impetus, is practice with a sense of international legal obligation creating a customary international law of human rights, even though many states sometimes violate these rights . . . . Absorption into international law of principles common to national legal systems generally is only a secondary source of international law . . . , but there is a willingness to conclude that prohibitions common to the constitutions or laws of many states are general principles that have been absorbed into international law. 196 The Restatement, that is, does not pretend to rely on the actual practice of states regarding human rights in its discussion of the CIL or, we may assume, the jus cogens norms of human rights. It justifies this approach by reference to “a disposition to find legal obligation in indeterminate language about human rights in international agreements,” without identifying the entities so disposed, and to “some willingness to find that the practice of states, perhaps under constitutional, political, or moral impetus, is practice with a sense of international legal obligation creating a customary international law of human rights,” without indicating who is willing to make this leap. More fundamentally, the Restatement does not explain why such an approach is justified. It is as though the drafters of the Restatement have determined that human rights rules ought to be treated as exceptions to the normal processes of formation of 56 197 Accounting for the approach the RESTATEMENT takes on these matters is difficult. Professor Henkin, Chief Reporter, rejected the argument that scholars’ opinions should be considered a source of international law, 57 A.L.I. PROC. 79 (1980). Nonetheless, the American Law Institute’s debates on these subjects make little reference to state practice, as distinct from actions by international organizations and the language of treaties, see id. at 123-26, 128-30; 59 A.L.I. PROC. 204-26 (1982); 62 A.L.I. PROC. 395-400, 539-44 (1985). Some comments made in the course of the debates support the argument that actual state practice was not really controlling in drafting the RESTATEMENT. In one of the debates on the RESTATEMENT, Professor Henkin characterized section 702 as including in its lists of prohibited behaviors “that which no state admits it practices, brags about practicing, or asserts the right to practice,” 59 A.L.I. PROC. 207 (1982). That is, the list consisted, not of actions from which states in fact refrained, but of actions states were unwilling to claim the right to take. But it is not clear why a state’s refusal to admit to a practice would create an expectation that the state would refrain from the practice. Further, with respect to the content of jus cogens, Professor McDougal, one of the advisers to the Reporters, defended his conclusion that certain human rights protections were jus cogens norms by reference to the views of one hundred writers, 57 A.L.I. PROC. 128 (1980), not by reference to the practice of states. Further, in debate Professor Lillich characterized section 702 as “very revolutionary,” as going “very, very far” and as “rather radical,” 59 A.L.I. PROC. 219-20 (1982). Such statements suggest some awareness that the positions taken in the RESTATEMENT were not uncontroversial. It should also be noted that a note-writer in the YALE JOURNAL OF INTERNATIONAL LAW pointed out a few years ago that, with respect to another element of the RESTATEMENT, the reporters appear to have chosen to “restate” the law according to their notions of good policy, aware of the fact that their views did not, in fact, reflect the state of the law as of the time of their drafting efforts, David B. Massey, Note, How the American Law Institute Influences Customary Law: The Reasonableness Requirement of the Restatement of Foreign Relations Law, 22 YALE J. INT’L. L. 419, 441-42 (1997). Such a thing would not seem impossible with respect to the issues addressed in this article, especially given the strong views on the subject of the Chief Reporter, Professor Henkin., see Louis Henkin, Human Rights and State “Sovereignty”, 25 GA. J. INT’L. & COMP. L. 31, 37-39 (1995/96). international law formation, even though nothing in those normal processes would indicate that such rules have any special status. 197 To sum up, the Restatement of Foreign Relations Law should be treated with great caution. It does not even purport to justify its assertions regarding the subjects discussed in this article, except by inaccurate citations to the Vienna Convention and by relying on scholarly opinion. If CIL is supposed to be created by the practice of states rather than by law professors, 57 198 See quotation at n. 109, supra. 199 Tel-Oren v. Libyan Arab Republic, 726 F.2d 774, 827 (D.C. Cir.1984) (Robb, J., concurring). 200 See n. 4, supra. it would seem the Restatement is a poor guide to CIL. As noted above, 198 Judge Robb once asserted that, “[c]ourts ought not to serve as debating clubs for professors willing to argue over what is or what is not an accepted violation of the law of nations.” 199 Unfortunately, the Restatement would appear to be a collection of only side’s arguments in a professorial debate. 3. Decisions of American Courts Several of the decisions described above relied on the decisions of other American courts when addressing the content of CIL. While it might seem unremarkable for a court to rely on a sister court’s decision when addressing a question of law, this approach is in fact problematic in the situation addressed in this article. This is so for two reasons. First, decisions by American federal courts regarding CIL are necessarily less authoritative than are their decisions regarding either the Constitution or federal statutes. Those courts are the only judicial bodies whose constructions of the Constitution and federal statutes can be considered binding. When the federal courts construe the Constitution or an act of Congress, they are addressing legal instruments adopted by the very sovereignty of which they are themselves an arm. In essence, when federal courts construe federal law, the United States government is announcing what its own law means. In contrast, whatever the status of CIL vis à vis federal law, 200 no one maintains that CIL is created solely by actions of the various parts of the government of the United States. When federal courts construe CIL, their task is more akin to that of addressing the law of a foreign 58 country than to that of interpreting a federal statute, and their conclusions should be treated with appropriate caution. After all, if a federal court found itself forced to interpret, for example, a Canadian statute, it would presumably rely, in the first instance, on Canadian readings of the statute, whether those of a Canadian court or of some other entity considered authoritative in Canada. This would be true, I suggest, even if another federal court had recently addressed the same statute. Canada’s law is determined by Canada, and an American court’s understanding of that law cannot be controlling in the same way such a court’s view of American law can be controlling. Similarly with respect to CIL, the focus for American courts should be on CIL itself, which is to say on the state practice which creates that body of law. Those courts are in essence relying on what amounts to secondary sources of CIL if, when faced with a CIL issue, they rely on American judicial interpretations of CIL rather than on the substance of CIL itself. The second reason for caution in relying on American court decisions to determine the content of CIL is that such a determination is heavily fact specific. In addressing questions of American law, an American court can look to the language of the Constitution, of statutes, or of administrative regulations, and can read relevant judicial opinions. While interpreting these resources may be difficult, there is typically no difficulty in determining their content. It is seldom a problem to determine what words the legislature uttered, even if it may be difficult to determine what those words mean. With respect to CIL, however, a court must engage in a factual inquiry to determine the content of the relevant state practice. The problem is not simply that an earlier decision may have found the facts of state practice incorrectly, although that is of course a concern. It is also that the interactions between governments which create CIL are never-ending. Even if a court’s 59 201 Buell v. Mitchell, 274 F.2d 337 (6 th Cir. 2001). 202 Id. at 371-72. 203 Hain v. Gibson, 287 F.3d 1224 (10 th Cir. 2002). 204 Id. at 1243-44. 205 The Covenant, supra n. 19, art. 6, ? 5. conclusion regarding a particular CIL issue in 1999 was defensible, in light of state practice at that time, a court addressing the same issue in 2002 cannot simply assume that state practice in the intervening three years has had no effect on the law. Nor is it necessarily safe to infer that, because state practice on one subject takes one form, it necessarily takes the same form with respect to a related subject. The cases discussed above offer an example of the problems that can arise when one American court relies on another to support a conclusion regarding the content of CIL. Buell v. Mitchell 201 found that the death penalty was not contrary to the general practice of states, relying in part on the fact that the Covenant permits the death penalty. 202 Hain v. Gibson 203 relied on Buell to support the conclusion that states which had abolished the death penalty for persons whose crimes were committed before their eighteenth birthdays had not necessarily done so from a sense of legal obligation. 204 But the issues in the two cases are not the same; one dealt with the complete elimination of the death penalty, the other with partial abolition. Further, to the extent that the result in Buell depended on the Covenant’s approach to the death penalty, reliance on Buell’s method of analysis should have led the Hain court to a result opposite to that which it reached. This follows because the Covenant forbids imposition of the death penalty for crimes committed prior to a defendant’s eighteenth birthday. 205 60 206 It might be objected that, if this argument is correct, then this article’s stress on the method of determining CIL employed in Paquete Habana makes no sense; there is no reason to assume that the Supreme Court is any more authoritative regarding CIL than is any other American court, so arguments criticizing contemporary methods of determining CIL for departing from the approach of Paquete Habana make no sense. While this argument has some plausibility, it depends on the assumption that Paquete Habana is either controlling or irrelevant. Surely, however, the views of the Supreme Court ought to carry some weight with lower federal courts, even if those views are not controlling. At minimum, the lower courts ought to acknowledge any departures they make from an approach taken by the Supreme Court, and explain why they believe the Supreme Court was wrong. The point in citations to Paquete Habana is not that the lower courts should have adhered slavishly to that decision’s method of determining the content of CIL, but rather that those courts’ decisions show no awareness that they are departing from the Supreme Court’s approach to this issue, and provide no justification for their actions. The foregoing is not meant to suggest that a careful discussion of a point of CIL should be ignored simply because it is found in the opinion of an American court. It is rather to say that, however helpful a court decision may be in providing insight into a question of CIL, it cannot substitute for an examination of the fundamental sources of that body of law. 206 4. Decisions of International Courts If a judge seeking to determine the content of a rule of international law can rely on the decisions of American courts only with great caution, it would not be surprising for that judge to assume that more confidence could be reposed in the decisions of international tribunals. Such tribunals deal only with questions of international law; their judges are, in many cases, expert in that field. Reliance on such tribunals, therefore, would seem to be a reasonable way for a busy judge to deal with a complex body of law with which he or she might well be unfamiliar. Nonetheless, the decisions of international tribunals must also be treated with caution by American judges faced with questions of international law. At least two factors support this conclusion. First, the legal instruments establishing the court in question may indicate questions 61 207 STATUTE OF THE INTERNATIONAL COURT OF JUSTICE, June 26, 1945, 59 Stat. 1055, T.S. No. 993 (hereinafter STATUTE). 208 CHARTER OF THE UNITED NATIONS, art. 94, June 26, 1945, 59 Stat. 1055, T.S. No. 993 (hereinafter CHARTER). as to the scope of authority conferred upon it. Second, even if a court’s authority seems unambiguous, the inevitable delay in states’correcting a judicial decision they see as incorrect means that too quick an embrace of controversial decisions by an international tribunal may result in an American judge treating as established a judicial construction of a legal rule which states ultimately reject. An example of a tribunal of limited authority is provided by the best-known of these bodies, the ICJ. Its basic document indicates that its decisions are not to be treated as having precedential value. That document, read with the United Nations Charter, also shows that states even limited the ICJ’s capacity to finally resolve individual disputes. The Statute of the ICJ 207 (the Statute) includes three articles which, taken together with Article 94 of the Charter of the United Nations 208 (the Charter), indicate both that the decisions of the ICJ were not intended to create law and that states were quite conservative in the scope of authority they were willing to grant to the Court. A consideration of Articles 38(1)(d) and 59 of the Statute demonstrates the first point. Article 38(1)(d) provides: 1. The Court, whose function is to decide in accordance with international law such disputes as are submitted to it, shall apply: . . . d. subject to the provisions of Article 59, judicial decisions and the teachings of the most highly qualified publicists of the various 62 209 Id., art. 38(1). 210 Id., art. 59. 211 The portions of the Statute here in question are all are drawn from the correspondingly numbered articles of the Statute of the Permanent Court of International Justice (PCIJ), Statute of the Permanent Court of International Justice, December 16, 1920, 6 L.N.T.S. 391, which was established in 1921. The initial draft of that statute was prepared by an Advisory Committee of Jurists. As finally adopted, that Committee’s version of the language which became article 38(1)(d) read, “The Court shall . . . apply . . . 4. judicial decisions and the teachings of the most highly qualified publicists of the various nations, as subsidiary means for the determination of rules of law.” PERMANENT COURT OF INTERNATIONAL JUSTICE, ADVISORY COMMITTEE OF JURISTS, Annex 1, 32d Meeting, PROCèS-VERBAUX OF THE PROCEEDINGS OF THE COMMITTEE JUNE 16 TH -JULY 24 th 1920 WITH ANNEXES 673, 680. The Committee’s intention to make clear that the PCIJ would lack law-making authority is shown by its continual adjustment of the language that became Article 38(1)(d) in the direction of emphasizing that meaning. Its first version provided, “the following rules are to be applied by the judge in the solution of international disputes; . . .4. International jurisprudence as a means for the application and development of law,” Annex No. 3, 13 th Meeting, id. at 281, 306. This language nations, as subsidiary means for the determination of rules of law. 209 Article 59 provides that “[t]he decision of the Court has no binding force except between the parties and in respect of that particular case.” 210 The most natural reading of these two articles is that judicial decisions are solely a subsidiary means for the determination of the content of international law. Article 38(1)(d) states this position expressly. Article 59 negatives the existence of any legal effect of a judgment of the ICJ except as between the parties to a case in respect of that case; the phrase in Article 38(1)(d) making the application of that paragraph subject to Article 59 can most easily be seen as meaning that the designation of judicial decisions as subsidiary means for determining the content of law is not intended to somehow qualify the limiting effects on the ICJ’s judgments imposed by Article 59. The drafting history of these provisions reinforces this conclusion. 211 63 provoked considerable debate, with its proponent, Baron Descamps, asserting that [Doctrine] could only be of a subsidiary nature; the judge should only use it in a supplementary way to clarify the rules of international law. Doctrine and jurisprudence no doubt do not create law; but they assist in determining rules which exist. A judge should make use of both jurisprudence and doctrine, but they should serve only as elucidation, 15 th Meeting, id. at 336. A drafting committee altered the proposed language to read, “The Court . . . shall . . . apply . . . 4. Rules of law derived from judicial decisions and the teachings of the most highly qualified publicists of the various nations,” Annex 2, 25 th Meeting, id. at 561, 567. This was later modified by the addition of the phrase “, as a subsidiary means for the determination of rules of law” added after the word “nations,” Annex 2, 29 th Meeting, id. at 629, 636. Apparently, controversy continued on the subject, since Baron Descamps subsequently proposed altered language “as a compromise,” 30 th Meeting, id. at 617, 620. The final version is set out in the text. The matter then moved to the Council of the League of Nations. The language ultimately adopted by that body for Articles 38(4) (now Article 38(1)(d)) and 59 is identical to that of those articles in their current form, Compare Statute for the Permanent Court of International Justice Provided for by Article 13 of the Covenant of the League of Nations, arts. 38(4), 59, LEAGUE OF NATIONS, PERMANENT COURT OF INTERNATIONAL JUSTICE, DOCUMENTS CONCERNING THE ACTION TAKEN BY THE COUNCIL OF THE LEAGUE OF NATIONS UNDER ARTICLE 14 OF THE COVENANT AND THE ADOPTION BY THE ASSEMBLY OF THE STATUTE OF THE PERMANENT COURT 258, 264, 266 (hereinafter PCIJ DOCUMENTS) with Statute, arts. 38(1)(d), 59. The adoption of the wording that became these articles, PCIJ DOCUMENTS at 42, 44 was preceded by the submission of a memorandum by the British delegate observing that the PCIJ’s decisions must necessarily have the effect of “moulding and modifying international law,” Note on the Permanent Court of International Justice, id. at 38, expressing the opinion that this development was not contemplated by the Covenant of the League of Nations. The memorandum went on to suggest that there ought to be a mechanism whereby a state could make a protest against changes in international law that might flow from a particular decision, id. The memorandum described such a mechanism as particularly necessary because non-members of the League of Nations, including the United States, Germany, and the Soviet Union, could “not be expected to take their views on international law from the court’s decision,” id. A report, prepared by the French representative to the Council of the League and adopted by that body, id. at 45, makes clear that the adoption of what became Article 59 was seen as a response to the concern that states not parties to a case could nonetheless be affected by the decision of that case if the PCIJ’s decisions were seen as a source of law generally. More specifically, that report indicates that Article 59 was intended to make explicit that states not parties to a given case were not bound by the PCIJ’s legal pronouncements in that case, id. at 50. It is true that a subcommittee of the Third Committee of the Assembly of the League rejected an Argentine amendment to what became Article 38 that would have limited “the power of the court to attribute the character of precedents 64 to judicial decisions” because the subcommittee “considered that it would be one of the Court’s important tasks to contribute, through its jurisprudence, to the development of international law,” Report and Draft Scheme Presented to the Assembly by the Third Committee, id. at 206, 211. But it seems to go too far to say that the rejection of this amendment should be seen as indicating that the PCIJ was intended to have the capacity, through its decisions, to make law. As indicated above, the language of articles 38(4) and 59 was clearly drafted to preclude that result, and that language was not altered by the subcommittee. Further, the Argentine amendment would have would have limited the PCIJ to applying only national judicial decisions, and those only against the state of the court delivering the decision, id. at 68. That is, the amendment would have had the effect of forbidding the PCIJ to look to its own decisions even as subsidiary means for determining rules of law. Given the highly restrictive character of the amendment, its rejection hardly implies the intention to adopt a very expansive view of the effect of the PCIJ’s decisions. Article 36 of the Statute reinforces the conclusion that the ICJ was not intended to have the capacity to make law through its decisions. Article 36 of the Statute provides: 1. The jurisdiction of the Court comprises all cases which the parties refer to it and all matters specially provided for in the Charter of the United Nations or in treaties and conventions in force. 2. The states parties to the present Statute may at any time declare that they recognize as compulsory ipso facto and without special agreement, in relation to any other state accepting the same obligation, the jurisdiction of the Court in all legal disputes concerning: a. the interpretation of a treaty; b. any question of international law; c. the existence of any fact which, if established, would constitute a breach of an international obligation; d. the nature or extent of the reparation to be made for the breach of an international obligation. 3. The declarations referred to above may be made unconditionally or on condition of reciprocity on the part of several or certain states, or for a certain time. . . . 65 212 STATUTE, supra n. 207, art. 36. 213 CHARTER, supra n. 208, art. 94. 6. In the event of a dispute as to whether the Court has jurisdiction, the matter shall be settled by the decision of the Court. 212 The effect of this article is to limit the ICJ’s jurisdiction to states which have voluntarily accepted its authority, either by recognizing its “compulsory” jurisdiction prior to the development of a dispute or by referring a case to the Court after the dispute has arisen. That is, states did not afford the ICJ the authority to compel an unwilling litigant state to appear before it. The court, in other words, is not intended to be able to impose its views on the international community. But if the ICJ’s decisions are seen as sources of law, then states unwilling to appear before the Court are, indirectly, subjected to its authority - strongly curtailing the protection for state autonomy created in Article 36. Article 94 of the Charter supports the same conclusion. That article provides: 1. Each Member of the United Nations undertakes to comply with the decision of the International Court of Justice in any case to which it is a party. 2. If any party to a case fails to perform the obligations incumbent upon it under a judgment rendered by the Court, the other party may have recourse to the Security Council, which may, if it deems necessary, make recommendations or decide upon measures to be taken to give to the judgment. 213 In other words, a state prevailing before the ICJ has no legal right to execution of the judgment in its favor. On the contrary, whether to enforce the judgment is entirely a matter of the discretion of the Security Council. If the ICJ’s decisions were understood to be sources of law, binding on the international community, such discretion would be meaningless. This follows, since the states composing the Security Council would presumably be as obliged to 66 214 See Iain Scobbie, Res Judicata, Precedent and the International Court: A Preliminary Sketch, 20 AUSTL. Y. B. INT’L L. 299 (1999). 215 Mohammed Bedjaoui, Expediency in the Decisions of the International Court of Justice, 71 BRIT. Y.B. INT’L L. 1 (2000). In particular, Judge Shahabuddeen of the ICJ has argued that ICJ decisions may properly be considered a precedent, at least when that body decides cases, MOHAMED SHAHBUDDEEN, PRECEDENT IN THE WORLD COURT 238-39 (1996). However, the heart of his argument is that the members of the Council of the League of Nations who produced the final draft of the PCIJ’s Statute were aware that the Court’s decisions would create precedents, but adopted the Statute anyway. More specifically, he argues that Article 59 means only that a judgment can be enforced only against a state which was a party to the litigation which produced the judgment, id. at 56-63. This conclusion is hard to square with the drafting history of the Statute (which, to be sure, Judge Shahabuddeen describes at length). Article 59 was added after the British note asserted the undesirability of states not parties to litigation before the PCIJ being held bound by rules of law developed by that Court, and was characterized by the French representative as dispositive of that concern. Again, if that article was intended only to indicate that non-parties to a case were not affected by the judgment, it would have been unnecessary to add to Article 59 the phrase “and only in respect of that particular case,” since, according to Judge Shahabuddeen, all the article purports to address is the effect of individual judgments. However, that phrase makes sense if the article is intended to emphasize the non-precedential character of the Court’s decisions, since parties to a case would be as vulnerable to its use as a precedent as would other states. Further, as Judge Shahabuddeen does not note, the phrase “[s]ubject to the provisions of Article 59" was added to what is now accept the view of the law taken by the Court as any other non-parties to a case in which a particular rule was established. Yet if they were so bound, how could they lawfully elect to reject the application of a binding legal rule in the case in which execution was sought? If, however, Article 94 actually confers discretion on the Security Council, it must follow that the Council’s members not parties to a particular case decided by the ICJ are not bound by any legal rules the Court purports to enunciate in that case. And if they are not bound, neither are any other non-parties. In short, Article 94 is inconsistent with the argument that the ICJ’s decisions are sources of law. To be sure, the Court itself puts considerable weight on its own prior decisions, 214 and some of its judges have defended this practice. 215 Scholars, too, frequently cite decisions of the 67 Article 38(1)(d) at the same time that Article 59 was added to the statute, PCIJ DOCUMENTS, supra n. 211, at 42, 44. That phrase makes sense in context only if it is intended to emphasize that reliance on the Court’s judicial decisions as subsidiary means for the determination of rules of law is subject to the limitation that its decisions bind only the parties to a case, and only for that case. Judge Shahabuddeen observes that one might distinguish between the law-making capacity of municipal courts and that of the ICJ on the ground that municipal courts are invested with the sovereignty of the states which establish them, SHAHABUDDEEN at 93. He argues, however, that the states of the world can be seen as having delegated law-making authority to the ICJ as a function of establishing that court, id. That assertion, however, overlooks the implications of the ICJ’s lack of compulsory jurisdiction and the fact that its judgments are enforceable only at the discretion of the Security Council. In contrast, municipal courts can compel the appearance of litigants and executive authorities have no discretion not to execute their judgments. That is, municipal courts themselves possess the essence of sovereignty, the power to coerce obedience. In contrast, the ICJ was denied any coercive powers by governments - surely an indication that they intended that it not exercise anything like sovereign power. 216 For examples of scholarly reliance on ICJ decisions as establishing rules of CIL see, e.g., J.G. STARKE, INTRODUCTION TO INTERNATIONAL LAW passim (10 th ed. 1989); MALCOLM N. SHAW, INTERNATIONAL LAW passim (4 th ed. 1997); Anthea Elizabeth Roberts, Traditional and Modern Approaches to Customary International Law: A Reconciliation, 95 A.J.I.L. 757 (2001). 217 Siderman de Blake v. Republic of Argentina, 965 F.2d 699, 716 (1992), cert. denied 507 U.S. 1017 (1993). 218 Prosecutor v. Du?ko Tadi?, 38 I.L.M. 1518 (Int’l Crim. Trib. Former Yugo.-App. Chamber 1999). ICJ as though they were in themselves authority for particular propositions of law. 216 And, as noted above, at least one American court has relied on an ICJ judgment as authority supporting its holding. 217 However, not only do the ICJ’s basic documents cast great doubt on the authority of its decisions, but at least one other international tribunal has repeatedly held that it is not bound by the ICJ’s view of international law. It was the International Criminal Tribunal for the Former Yugoslavia (ICTY) which took this position, in the following context. In Prosecutor v. Du?ko Tadi?, 218 the Appeals Chamber of the ICTY was faced with a case in which, among other issues, the prosecution appealed the Trial 68 219 Id., ?? 22, 68-74, 80-82, 86. 220 Id., ? 97. 221 Supra n. 15. 222 Tadi?, supra n. 218, ? 115. 223 Id., ?? 116-45. 224 Prosecutor v. Delali?, 40 I.L.M. 630 (Int’l Crim. Trib. Former Yugo.-App. Chamber 2001). 225 Id. ? 21. Chamber’s apparent conclusion that certain of the crimes of which the defendant was accused were not covered by the relevant treaty because the conflict in Bosnia-Herzegovina was not international. 219 Whether that determination was correct depended on whether the acts of Bosnian Serb forces could be attributed to Yugoslavia, which in turn depended on the degree of control Yugoslavia was required to maintain over those forces in order to satisfy the international legal requirements for such attribution. 220 According to the prosecution, the trial chamber had erred by drawing the legal standards for attribution from the United States v. Nicaragua case. 221 The Appeals Chamber agreed, holding that the test drawn from that case was not persuasive. 222 It based this conclusion on its determination that the result in United States v. Nicaragua was consistent neither with the bases of the law of state responsibility nor with the practice of states and international and national tribunals. 223 In Prosecutor v. Delali?, 224 the Appeals Chamber faced the same legal issue, with the defendants in that case arguing that the Tadi? court had erred in refusing to follow the ICJ’s holding in the Nicaragua case because the Appeals Chamber was “bound by the ICJ’s precedent.” 225 The Appeals Chamber did not accept that argument, stating that 69 226 Id. ? 24 (footnote omitted). 227 Statute of the International Tribunal for the Prosecution of Persons Responsible for Serious Violations of International Humanitarian Law Committed in the Territory of the Former Yugoslavia since 1991, arts. 1, 6, 9, 29, U.N. Doc. S/25704 at 36, annex (1993) and S/25704/Add.1 (1993), adopted by Security Council on 25 May 1993, U.N. Doc. S/RES/827 (1993), as amended by Security Council Resolutions 1166, S/RES/1166 (1998); 1329, S/RES/1329 (2000); and 1411, S/RES/1411 (2002). 228 Id., arts. 23, 24, 27. 229 The ICTY is not the only international tribunal whose judgments are actually enforceable. For example, money judgments against in a state rendered by the Inter-American this Tribunal is an autonomous international judicial body, and although the ICJ is the “principal judicial organ” within the United Nations system to which the Tribunal belongs, there is no hierarchical relationship between the two courts. Although the Appeals Chamber will necessarily take into consideration other decisions of international courts, it may, after careful consideration, come to a different conclusion. 226 The ICTY, in short, has flatly rejected the argument that ICJ decisions give rise to binding legal rules. Rather, it has held that the ICJ is to be followed only if and to the extent its decisions make sense and correspond to the practice of states. This treatment of the ICJ contrasts sharply with the scholarly practice of treating ICJ decisions as establishing legal rules without regard to their coherence and correspondence to practice. The authority of the ICTY is, if anything, less doubtful than that of the ICJ. While it has had limited success in actually obtaining custody over persons it has indicted, its jurisdiction does not depend on consent either from defendants before it or from their home states. Formally, at least, its jurisdiction is compulsory. 227 Furthermore, the ICTY is empowered by its Statute to impose sentences and to select the facility in which the sentences are to be served. 228 That is, the carrying out of its judgments is not subject to the discretion of any political body, and its authority to actually resolve the cases brought before it is thus unquestioned. 229 When such a 70 Court of Human Rights are enforceable in the courts of the state against which the judgment was rendered, American Convention on Human Rights, Nov. 22, 1969, arts. 62, 68, 1144 U.N.T.S. 123. In other words, American judges should not assume that all international tribunals have been kept on a leash as short as that attached to the ICJ. Rather, it is necessary to examine the situation of each such judicial body in order to determine the scope of authority of the entity in question. court explains the law guiding its decisions, it would seem unreasonable simply to ignore that explanation. Yet if these ICTY rulings are taken seriously, they necessarily cast doubt on any reliance on ICJ decisions simply because they are ICJ decisions. Rather, however inconvenient it may be for scholars and domestic courts, the holdings in these ICTY cases would seem to seriously weaken any argument that ICJ decisions on CIL which are poorly reasoned or contrary to state practice are entitled to any deference. We see, then, that limitations on the authority of particular international tribunals cast doubt on arguments that the decisions of all such tribunals are a source of international law. But that is not the only factor counseling caution in dealing with the law-creating effect of decisions of international tribunals. Another is the problem of legal mistakes. This is not to say that judges of international tribunals are somehow more prone to error than are judges of domestic courts. It is, rather, to say that correcting such errors in so much more difficult in the international system than is true in a domestic system that formal action to change a rule relied upon in a doubtful decision may take a long time, even if the decision is understood to be questionable as soon as it is announced. An American judge who seizes upon that decision without examining the reaction of states to it may thus find itself applying an incorrect legal standard. 71 230 Indeed, the reaction of the European Union to a degree of perceived overreaching by the European Court of Justice is a contrary example which proves the point in the text. The judgments of that court interpreting the various legal instruments of the European Union are enforceable in the domestic legal systems of the member states of the Union, Treaty Establishing the European Economic Community, March 25, 1957, arts. 187, 192, 298 U.N.T.S. 3, 78, 79 (1957). Thus, when that court handed down rulings seen as encroaching on national prerogatives, the member states had no alternative to adopting various substantive and procedural measures in order to rein in that court, see Karen J. Alter, The European Union’s Legal system and Domestic Policy: Spillover or Backlash?, 54 INT’L ORGANIZATION 489, 512- 15 (2000). This difficulty follows from the nature of the international legal system. As a practical matter, the only way to clearly alter a rule of international law is to create a treaty regime, the rules of which will supersede any pre-existing rule. But a treaty binds only the states accepting it. Thus, if the rule to be changed is one that affects a large number of states, the group of states which must reach consensus on a rule-changing treaty will be correspondingly large. But convening a diplomatic conference and negotiating agreement among a large number of governments is a very complicated process. It is necessarily more difficult than amending a domestic statute which a court has interpreted in a way contrary to the legislature’s views on the subject, since the legislature can act by majority vote; unanimity is not required. Obviously, before states will undertake the effort of convening such a conference, they must be convinced that the effort is necessary. A decision by an international court, even if perceived as incorrect, will by no means necessarily be seen as a sufficient reason to take on the burden of convening a conference. States may simply ignore the decision. 230 After all, and as noted above, the ICJ has limited means for compelling states to follow the rules it lays down, and in fact cannot even insure the execution of its judgments. So, for example, when the ICJ concluded that it had jurisdiction to hear the claim brought against the United States by 72 231 Case Concerning Military and Paramilitary Activities in and Against Nicaragua (Nicaragua v. United States of America), Jurisdiction of the Court and Admissibility of the Application, 1984 I.C.J. 392. 232 Statement on the U.S. Withdrawal from the Proceedings Initiated by Nicaragua in the International Court of Justice, 24 I.L.M. 246, 247 (1985). Nicaragua in respect of the activities of the contra rebels, 231 the United States simply withdrew from the proceedings, arguing that the Court’s conclusion that it had jurisdiction was entirely wrong. 232 In summary, both because of limitations in their authority imposed by the instruments establishing them, and because any errors they make may not be formally corrected, American courts should react cautiously to suggestions that they rely on the opinions of international tribunals as sources of rules of CIL. IV. A Suggested Alternative The foregoing discussion underlines the problems with the reliance, by American courts, on many of the sources to which they might expect to turn when confronted with a case involving international law. It might reasonably be asked, however, what is a judge supposed to do when faced with such a case? The judge cannot simply refuse to decide, and if much scholarly writing, the Restatement, and domestic and international court decisions are unavailable to instruct the judge in the law to apply, it would seem that the judge is in an impossible position. Fortunately, that need not be true, at least with respect to the international legal issues which most frequently come before American courts, that is, issues involving human rights. Recall that, in matters involving CIL, The Paquete Habana holds that, if an American court must 73 233 Paquete Habana, supra n. 29, at 700. 234 Hathaway, supra n. 163, at 1967. 235 Id. at 1965, 1967. decide a case involving international law but has no treaty, statute, or controlling executive or judicial act upon which to rely, “resort must be had to the customs and usages of civilized nations” 233 as the source of law. We now have available extensive collections of the actual practice of states regarding human rights, published very frequently by trustworthy organizations. For example, Professor Hathaway was able to rely on four sources of information regarding states’ human rights practices in her study of the effectiveness of human rights treaties: “the Center for International Development and Conflict Management at the University of Maryland, College Park, the United States Department of State Country Reports on Human Rights, Freedom House's Annual Survey of Political Rights and Civil Liberties, and the Inter- Parliamentary Union.” 234 From data drawn from these sources, she was able to reach conclusions regarding the performance of 166 states with respect to genocide, torture, civil liberty, fair and public trials, and the political rights of women. 235 In other words, the lack of collections of state practice which forced Justice Gray to look to scholarly writings to determine the customs and usages of civilized nations has, with respect to human rights matters, been remedied. Thus, the data judges need to address questions of state practice regarding human rights is available. This is not to say that using such information is easy; indeed, so much material may be available that sorting through it may be difficult. But judges are able to deal with other types of voluminous material with the assistance of counsel, and there is no reason to think that 74 counsel could not similarly guide a court through, for example, the State Department’s annual report on human rights. That is, judges facing the need to decide questions involving the international law of human rights can ignore the materials which this article suggests are questionable without losing a basis for a reasoned decision. V. Conclusion The foregoing argument has sought to make a number of points. First, that the traditional American approach to the determination of CIL is to focus on the actual behavior of governments, which approach is in any case the one most consistent with the nature of CIL. Second, that, with respect to at least one issue - the nature and content of jus cogens - a number of federal appellate courts have relied in their legal analyses, not on the behavior of governments, but on other types of information, which has led them to reach doubtful conclusions. Third, that these alternatives to a focus on state practice are of doubtful authority. And finally, fourth, that with respect to the class of international legal issues that most commonly come before American courts - those involving the law of human rights - there exist a number of highly regarded compilations of state practice, such that no judge is left without guidance on the law because of any difficulty in determining the nature of state behavior in this area. In short, American judges should, when determining the content of CIL, model themselves on the example of the Supreme Court’s analysis in The Paquete Habana and concentrate on determining the actual behavior of governments with respect to the issue in question. But the question can be raised, suppose they don’t - so what? Would the result of the reliance on the sources criticized in this article really be so bad? From one perspective, perhaps not. The number of cases involved will probably not be great, the plaintiffs in such cases will 75 probably be deserving in some (not necessarily legal) sense and the defendants are likely to be unsavory. But surely a mode of legal analysis should be evaluated on criteria beyond that of, does this way of thinking give us a pretext for reaching a result we like? In this case, the problem with reliance on sources other than state behavior to determine the content of CIL is that such a practice transfers legislative power to groups with little right to claim it - such as judges of international tribunals whose authority is carefully circumscribed in their founding instruments - or no right at all - such as legal academics. The framing of rules of law is necessarily a political act. The ultimate problem with efforts to shift the focus in CIL determinations from state practice to something else is that the something else, whatever it is, will lack any sort of political legitimacy. Surely rules deriving from illegitimate sources are a shakey basis for any body of law, including the law of the international community. 76 ABSTRACT This article addresses an issue with which federal courts have been forced to deal with increasing frequency: how ought a judge go about determining the content of customary international law? The article seeks to demonstrate, using the example of the treatment of the concept of jus cogens by the courts of appeals, that federal courts have come to rely on doubtful sources in addressing questions of international law. More specifically, it sets out to show that courts frequently do not rely on the actual practice of governments to determine the content of customary international law, as would be indicated both by the nature of customary international law and by Supreme Court authority. Rather, they have come to place weight on the works of writers whose conclusions themselves are drawn from dubious bases, on the Restatement, on the views of other domestic courts and on the decisions of international courts. The article explains the problems with relying on such sources, and briefly describes an alternative method of proceeding for cases involving an area of customary international law very frequently before American courts, the law of human rights.