COUNTERINTUITING COUNTERMEASURES By David J. Bederman* The adoption in August 2001 by the International Law Commission (ILC) of its articles on responsibility of states for internationally wrongful acts1 well and truly brings to a close the twentieth century’s engagement with international law as (in Martti Koskenniemi’s mem- orable refashioning of George Kennan’s savage critique) a “gentle civilizer of nations.”2 Including the entry into force of the Rome Statute of the International Criminal Court,3 the seven pillars of international legal codification have been completed with some form of as- sistance by the ILC: diplomatic immunities, the law of the sea, a comprehensive law of trea- ties, the Nuremberg Principles, and jurisdictional immunities of states. Indeed, the articles on state responsibility may represent an even greater methodological challenge for interna- tional law codification because they pose fundamental questions regarding the identity and nature of states. Like the Montevideo Convention on Rights and Duties of States4 and the ILC’s own somewhat obtuse efforts on the international law of state succession,5 the articles on state responsibility go to the intellectual core of public international law by delimiting the character of states and the nature of their obligations when they interact with other in- ternational actors. Perhaps, then, it is no surprise that the ILC’s journey into that doctrinal realm took over half a century, and consumed the attention of five special rapporteurs and countless Commission members. The topic of countermeasures always occupied a central place in the ILC’s deliberations on state responsibility, and proved—along with the notion of international crimes—to be one of the lightning rods of criticism and controversy for the articles.6 As Professor James Crawford, the special rapporteur for the final stages of the ILC’s effort, observed: “Concerns * Of the Board of Editors. The writer serves as Chair of the American Society of International Law’s Panel on State Responsibility, which reviewed and commented on earlier versions of the International Law Commission’s draft articles on state responsibility. The writer is grateful for the helpful suggestions of David Caron and James Crawford, but—as befits the topic—remains accountable for the ideas presented here. 1 Draft Articles on Responsibility of States for Internationally Wrongful Acts, in Report of the International Law Commission on the Work of Its Fifty-third Session, UN GAOR, 56th Sess., Supp. No. 10, at 43, UN Doc. A/56/10 (2001), available at <http://www.un.org/law/ilc>, reprinted in JAMES CRAWFORD, THE INTERNATIONAL LAW COMMIS- SION’S ARTICLES ON STATE RESPONSIBILITY: INTRODUCTION, TEXT AND COMMENTARIES (2002). References to, and quotations of, the articles, as well as the official ILC commentaries to the articles, which appear in the Commis- sion’s Fifty-third Report and Crawford’s volume, supra, will be identified below by article and paragraph number. Crawford’s comments in his individual capacity, set forth in the introduction to his compendium, will be specially noted as “Crawford Comments.” For more on the ILC’s progress in considering the articles, see James Crawford, Pierre Bodeau, & Jacqueline Peel, The ILC’s Draft Articles on State Responsibility: Toward Completion of a Second Reading, 94 AJIL 660 (2000). 2 MARTTI KOSKENNIEMI, THE GENTLE CIVILIZER OF NATIONS: THE RISE AND FALL OF INTERNATIONAL LAW 1870– 1960, at v (2002) (quoting GEORGE KENNAN, AMERICAN DIPLOMACY 54 (expanded ed. 1984) (“[W]e would confine these concepts [of international law and morality] to the unobtrusive, almost feminine, function of the gentle civi- lizer of national self-interest . . .”)). 3 Rome Statute of the International Criminal Court, July 17, 1998, UN Doc. A/CONF.183/9* (1998), 37 ILM 999 (1998), corrected through May 8, 2000, by UN Doc. CN.177.2000.TREATIES–5, available at <http://www.un.org/law/ icc> (entered into force July 1, 2002). 4 Convention on Rights and Duties of States, Dec. 26, 1933, 165 LNTS 19. 5 Vienna Convention on Succession of States in Respect of Treaties, Aug. 23, 1978, 17 ILM 1488 (1978); Vienna Convention on Succession of States in Respect of State Property, Archives and Debts, Apr. 8, 1983, 22 ILM 306 (1983), both available at <http://www.un.org/law/ilc/texts/tresutoc.htm>. 6 Crawford Comments, supra note 1, at 47–49. 817 818 THE AMERICAN JOURNAL OF INTERNATIONAL LAW [Vol. 96:817 [regarding the provisions on countermeasures] were expressed at various levels. The most fundamental related to the very principle of including countermeasures in the text, either at all or in the context of the implementation of State responsibility.”7 Indeed, this appeared to be an issue that many states were passionate about. Unlike the arid academic debates raised in the construction of some of the articles—the rules of attribution of state conduct and the relationship in timing between obligations and breaches come to mind—the discussions on the countermeasure provisions implicated the reality of the rough-and-tumble of interna- tional politics, offering a muscular vision of what is practically required to enforce interna- tional norms. Perhaps even more significantly, this subject created interesting alliances of states, as well as of nonstate actors, all anticipating future disputes where countermeasures might be deployed. If, as has been consistently observed in relation to the articles, the entire project was prem- ised on elaborating “secondary” rules of responsibility for states in international relations— the modalities and structures of state responsibility, and not the content of the primary obli- gations themselves—then countermeasures always had to be regarded as a crucial subject. Countermeasures, or “self-help,” are a necessary part of any legal system, like the international system, that lacks strong “vertical” enforcement.8 There is no world policeman to command or coerce obedience to international law rules; instead, states and other actors rely on a com- bination of other mechanisms such as countermeasures to win respect and compliance for these duties.9 But how are countermeasures to be codified in a coherent system of law? How does one write a rule book for future conflicts involving various circumstances and stakes? How can one divorce the contours of a struggle for law from the underlying rules and values being con- tested? Perhaps of even greater moment to those interested in the way that international law gets codified are the normative predilections and political compromises involved in the process. The countermeasures section of the articles10 reflects a few of these dynamics. This essay seeks to explore the wider dimensions of the codification of rules for countermeasures and self-help in the face of internationally wrongful acts. Nevertheless, a short contribution cannot provide a complete and detailed analysis of the articles’ complex provisions on coun- termeasures. Nor can it fully explore the synergies between the countermeasures section and the bundles of provisions on such matters as attribution of state conduct, the timing of an international law breach, circumstances precluding wrongfulness, the remedies available for injuries, and standing to invoke responsibility. Instead, this essay will examine a few key intellectual underpinnings of the countermea- sures articles. First, it considers what the Commission regarded as the relevant sources for coun- termeasures practice, particularly the role played by decisions of the International Court of Justice (ICJ). An intriguing “feedback loop” between the ILC and the ICJ helped to shape the key provision on proportionality (Art. 51). Second, the essay considers the way the Com- mission adopted rules that have the effect of influencing the behavior of states as they vindi- cate their rights by use of countermeasures. This approach partially reflected the “primary/ secondary rule” distinction consistently made by the ILC drafters as a means to justify and 7 Id. at 48. 8 See discussion in DAVID J. BEDERMAN, THE SPIRIT OF INTERNATIONAL LAW 187–94 (2002). For more on the relation between countermeasures and self-help, see OMER ELGAB, THE LEGALITY OF NON-FORCIBLE COUNTER-MEASURES IN INTERNATIONAL LAW 4 (1988); Richard B. Lillich, Forcible Self-Help by States to Protect Human Rights, 53 IOWA L. REV. 325, 347–51 (1967); John Norton Moore, Enhancing Compliance with International Law: A Neglected Remedy, 39 VA. J. INT’L L. 881 (1999); Mary Ellen O’Connell, Enforcing the New International Law of the Environment, 1992 GER. Y.B. INT’L L. 293, 318. 9 See Commentary to pt. 3, ch. II, para. 1 (“Countermeasures are a feature of a decentralised system by which injured States may seek to vindicate their rights and to restore the legal relationship with the responsible State which has been ruptured by the internationally wrongful act.”). 10 Arts. 22, 49–54. 2002] SYMPOSIUM: THE ILC’S STATE RESPONSIBILITY ARTICLES 819 protect their project intellectually.11 During the deliberations on the articles, this distinction had substantial cogency, but some unusual dimensions also emerged in the context of coun- termeasures, especially with respect to the procedural safeguards imposed by some of these provisions (particularly Articles 49 and 52). Indeed, the primary thrust of these provisions is to superimpose procedural values of rectitude and transparency on states’ assessments of countermeasure options, even while incorporating some ambiguities that may constrain such behavior. Ironically, the overall effect on the international legal process of the Commission’s approach may be to permit more aggressive forms of countermeasures. Next, the essay contrasts the roles played by formalism and pragmatism in codification ef- forts. The provisions on countermeasures reflect a desire to perpetuate some formal legal distinctions (such as that between retorsions and reprisals, and the character of “urgent” countermeasures in Article 52), while making some quite pragmatic compromises, such as those with collective countermeasures (in Article 54). Last, this essay highlights the central conceptual mission of the countermeasures provisions in the articles: the search for a polite international society. The perceived need to limit or cabin a “muscular” vision of interna- tional relations as a struggle for law certainly constituted a crucial basis of the formulations of Articles 49 and 50. Taken together, these elements have shaped a notable result—a clear distillation of the norms of self-help in the face of international law violations. In the space of seven well- crafted articles, the ILC has largely encompassed the full range of state experience on the most fundamental question of how international actors can properly vindicate their rights. I. SOURCES AND FEEDBACK LOOPS The ICJ and the ILC have enjoyed a cozy historical relationship. It is not only that ILC members have often been elected to the Court, but also that the Court uniquely privileges some of the drafts, reports, and other work product of the Commission through citations in its decisions. In return, the Commission is careful to incorporate the doctrinal pronounce- ments of the ICJ in its codification efforts. Indeed, sometimes the ILC may have no other prac- tical course, insofar as its work products are reviewed by the Sixth Committee of the United Nations General Assembly and by member states convened at plenipotentiary conferences to consider and conclude treaty texts. The influence of the Court’s Advisory Opinion Reservations to the Genocide Convention12 on the ILC’s subsequent codification of reservations to multilateral instruments in Articles 19 to 23 of the Vienna Convention on the Law of Treaties (Vienna Convention)13 is certainly well-known, as was the effect of the Corfu Channel and Fisheries Judg- ments on the evolution of the ILC’s work leading to the 1958 Geneva Conventions on the Law of the Sea. This dynamic produces a “feedback loop” of jurisprudence, variously reinforcing certain norms previously found in state practice and judicial decisions, or, alternatively, per- petuating certain doctrines of uncertain origin or dubious validity. 11 This emphasis is made clear in the very first paragraph of the official ILC commentaries to the articles: The emphasis [of the articles] is on the secondary rules of State responsibility: that is to say, the general con- ditions under international law for the State to be considered responsible for wrongful actions or omissions, and the legal consequences which flow therefrom. The articles do not attempt to define the content of the international obligations breach of which gives rise to responsibility. This is the function of the primary rules, whose codification would involve restating most of substantive international law, customary and conventional. Commentaries, para. 1. Crawford also considers this distinction in his personal introduction to the articles, see Crawford Comments, supra note 1, at 14–16. 12 Reservations to the Convention on the Prevention and Punishment of the Crime of Genocide, Advisory Opinion, 1951 ICJ REP. 15 (May 28). 13 Vienna Convention on the Law of Treaties, opened for signature May 23, 1969, 1155 UNTS 331 [hereinafter Vienna Convention]; see also IAN SINCLAIR, THE INTERNATIONAL LAW COMMISSION (2d ed. 2000). 820 THE AMERICAN JOURNAL OF INTERNATIONAL LAW [Vol. 96:817 In the area of countermeasures, until recently most of the canonical cases were interna- tional arbitral awards and not World Court decisions. The source of the requirement that responses be made in proportion to offending acts was the Naulilaa arbitration14of 1928, one of the first cases to consider countermeasures, although in the context of forcible reprisals. It stood for the proposition that “[e]ven if . . . the law of nations does not require that the reprisal should be approximately in keeping with the offense, one should certainly consider as excessive and therefore unlawful reprisals out of all proportion to the act motivating them.”15 A more refined rule of proportionality in governing the “punch” and “counter-punch” of self-help can be found in the Air Services Agreement 16 award of 1978. There the question was the propriety of the United States’ cancellation of the Paris–Los Angeles route for Air France in the face of the French refusal to grant a “change of gauge” (or use of a smaller aircraft) for Pan American’s London-Paris flight. Because, in the tribunal’s estimation, the U.S. response was not “clearly disproportionate when compared to those [measures] taken by France,” it was deemed permissible.17 Acknowledging that the proportionality calculus was no exact science, the tribunal ruled that it “can at best be accomplished by approximation.”18 More- over, the appropriate optic of analysis was wide; the tribunal considered that the U.S. reac- tion, when viewed in light of the vast network of air services agreements between many nations in the world, was not “clearly disproportionate.”19 Proportionality thus resided in the eye of the beholder, and was to “be satisfied with a very approximate appreciation.”20 While Paul Reuter’s cogent dissent dealt with this characterization,21 the real insight of the Air Service Agree- ment award was that there had to be a permissible level of escalation in response to illegal acts, or else the malefactor would simply not regard the threats made by the injured state as credible. The key, as the tribunal indicated, was that the response not be disproportionate, and that any escalation proceed in relatively carefully measured increments. The ICJ’s engagement with this key question on the scope of legitimate countermeasures has not been entirely consistent or helpful. In the Gab!íkovo-Nagymaros Project Judgment,22 the Court considered whether Czechoslovakia’s diversion of the waters of the Danube was a proper countermeasure in response to Hungary’s lack of cooperation on the joint hydroelec- tric project. “In order to be justifiable, a countermeasure must meet certain conditions,” the Court noted, citing the Air Services Agreement arbitration and an earlier version of the ILC ar- ticles.23 But without much discussion or elaboration, the Court reframed the proportionality inquiry potentially to limit the responding state’s latitude of action: “In the view of the Court, 14 Responsibility of Germany for Damage Caused in the Portuguese Colonies in the South of Africa (Port. v. Ger.), 2 R.I.A.A. 1011 (1928)[hereinafter Naulilaa Arbitration]; see also Cysene Arbitration (Port. v. Ger.), 2 R.I.A.A. 1035 (1930) (responsibility of Germany for acts committed subsequent to July 31, 1914, and before Portugal en- tered war). 15 Naulilaa Arbitration, supra note 14, 2 R.I.A.A. at 1028. 16 Air Services Agreement of 27 March 1946 (U.S. v. Fr.), 18 R.I.A.A. 417 (1978) [hereinafter Air Services Agree- ment Award]. 17 Id. at 444, para. 83. 18 Id. at 443. 19 Id. at 444 (emphasis added). 20 Id. 21 Id. at 448 (Reuter, arb., dissenting) (“the Tribunal has been unable to assess definitely” the proportionality of the U.S. response); see also ELISABETH ZOLLER, PEACETIME UNILATERAL REMEDIES: AN ANALYSIS OF COUNTER- MEASURES (1984); DENIS ALLAND, JUSTICE PRIVÉE ET ORDRE JURIDIQUE INTERNATIONAL; ETUDE THÉORIQUE DES CON- TRE-MESURES EN DROIT INTERNATIONAL PUBLIC (1994). 22 Gab!íkovo-Nagymaros Project (Hung. v. Slovk.), 1997 ICJ REP. 7 (Sept. 25). 23 Id. at 55, para. 83 (citing Military and Paramilitary Activities in and Against Nicaragua (Nicar. v. U.S.), Merits, 1986 ICJ REP. 14, 127, para. 249 ( June 27) [hereinafter Nicaragua Judgment] (“The acts of which Nicaragua is accused, even assuming them to have been established and imputable to that State, could only have justified pro- portionate counter-measures on the part of the State which had been the victim of these acts . . .”); also citing Air Services Agreement Award, supra note 16, 18 R.I.A.A. at 443; Draft Articles on State Responsibility Adopted by the International Law Commission on First Reading [hereinafter First Reading Report], Arts. 47–50, in Report of the International Law Commission on the Work of Its Forty-eighth Session,  2 Y.B. Int’l L. Comm’n, pt. 1, at 1, 68–71, UN Doc. A/CN.4/SER.A/1996/Add.1). 2002] SYMPOSIUM: THE ILC’S STATE RESPONSIBILITY ARTICLES 821 an important consideration is that the effects of a countermeasure must be commensurate with the injury suffered, taking account of the rights in question.”24 Relying on materials bearing on the non-navigational uses of international watercourses, the Court simply con- cluded that Czechoslovakia, by unilaterally assuming control of a shared resource, and thereby de- priving Hungary of its right to an equitable and reasonable share of the natural resources of the Danube—with the continuing effects of the diversion of these waters on the ecol- ogy of the riparian area of the Szigetköz—failed to respect the proportionality which is required by international law.25 Although its analysis failed to indicate whether the Court intended its discussion of propor- tionality solely to be an adjunct to its broader consideration of principles of international resource management, many authorities assumed that the ICJ was sending a signal of the ne- cessity of narrowing states’ freedom of action in taking certain forms of countermeasures. The ILC clearly drew this conclusion, since, from the Court’s rather conclusory analysis arose the ILC’s provision in Article 51 that “[c]ountermeasures must be commensurate with the injury suffered, taking into account the gravity of the internationally wrongful act and the rights in question.”26 Echoing the Court’s thumbnail sketch of the doctrine—although with the qualifier that the “gravity” of the offense must be weighed with the “rights in ques- tion”—this provision also connects with the important qualification contained in Article 49 that a countermeasure be taken only “in order to induce [a] State to comply with its obliga- tions.”27 The article thus would limit the scope of state conduct as to both the “quantitative” element (the impact or effect of a countermeasure) and the “qualitative” inquiry (the pur- pose of the action), but subject to a significant ambiguity: the “broad meaning” to be given to the formulation of the “gravity” of the wrong and the “rights in question.”28 The article also inverts the negative formulation used in the Air Services Agreement award; namely, that coun- termeasures be “not disproportionate,” to a positive requirement that countermeasures be “commensurate” with a wrongful act.29 This change may be a semantic subtlety, but it was not lost on the Commission. The commentaries note that “[a] negative formulation [‘not dis- proportionate’] might allow too much latitude, in a context where there is concern as to the possible abuse of countermeasures.”30 While this seems a minor detail, the Commission’s adoption of the ICJ’s proportionality language in Gab!íkovo-Nagymaros raises some troubling concerns. The first is whether the sub- stitution of the relative neologism “commensurate” can at all be regarded as preferable to 24 1997 ICJ REP. at 55, para. 85. 25 Id. (citing and quoting Territorial Jurisdiction of the International Commission of the River Oder, 1929 PCIJ (ser. A) No. 23, at 27). 26 Art. 51. 27 Art. 49, para. 1. 28 See Commentaries, Art. 51, para. 6. 29 The negative formulation was also apparently adopted by Roberto Ago in his Third Report on State Respon- sibility,  2 Y.B. Int’l L. Comm’n, pt. 1, at 69, UN Doc. A/CN.4/SER.A/1971/Add.1 [hereinafter Ago, 1971 Commentary], and was quoted by Judge Stephen Schwebel in his dissent in the Nicaragua Judgment, 1986 ICJ REP. at 368: There must of course be some proportion between the wrongful infringement by one State of the right of another State and the infringement by the latter of a right of the former through reprisals. In the case of conduct adopted for punitive purposes, of specifically retributive action taken against the perpetrator of a particular wrong, it is self-evident that the punitive action and the wrong should be commensurate with each other. But in the case of action taken for the specific purpose of halting and repelling an armed attack, this does not mean that the action should be more or less commensurate with the attack. Its lawfulness cannot be measured except by its capacity for achieving the desired result. 30 Commentaries, Art. 51, para. 5; see also ENZO CANNIZZARO, IL PRINCIPIO DELLA PROPORZIONALITÀ NELL’ORDINA- MENTO INTERNAZIONALE (2000); Enzo Cannizzaro, The Role of Proportionality in the Law of International Countermea- sures, 12 EUR. J. INT’L L. 889 (2001) [hereinafter Cannizzaro, Role]. 822 THE AMERICAN JOURNAL OF INTERNATIONAL LAW [Vol. 96:817 the continued elaboration of the concept of proportionality. Although state practice on pro- portionality is admittedly sparse, it often does lie in the interstices of states’ reactions to international incidents.31 Even while acknowledging the difficulty of assembling and analyzing such evidence, the Commission might well be faulted for failing to cite any source other than a published arbitral or judicial decision.32 Through its change in terminology, the Commission has raised questions regarding the weight of a century’s worth of state practice on proportion- ality. Second, “commensurate” is nowhere defined in Article 51, nor in the Gab!íkovo-Nagymaros decision itself, exposing the articles to the criticism, made by some national governments,33 that “commensurate” was intended to have a meaning rather narrower than “proportional.” This conclusion, however, does not necessarily follow either logically or semantically from the Commission’s text.34 Moreover, no discussion by the Commission gives any hint that the commensurability analysis should focus on the level of response required to induce a respon- sible state to conform its conduct to international obligations. This was an important aspect of earlier ILC considerations of countermeasures35 but appears to have been collapsed into the threshold requisites of Article 49(1) for the imposition of countermeasures—that they be taken only against a state responsible for an internationally wrongful act as a way to im- plement the remedies set forth in part 2 of the articles. The use of countermeasures for purely punitive reasons—without any hope or expectation that the malefactor state will actually back down from its offensive conduct36—appears to be now precluded by the articles. This change in emphasis on the nature and purpose of countermeasures likewise places in some doubt the Commission’s use of earlier international law sources. Thus, what we appear to have in Article 51 is a provision whose key term is based exclusively on isolated language in a recent World Court decision. Perhaps even more disconcerting is that, to the extent that “commensurability” can be traced to the 1971 report of Roberto Ago, he took a strikingly different approach to the topic. Unlike the final draft of the article, Ago’s version recognized that limited forms of escalation were consistent with a proportional response, especially in light of the need to “induce” a malefactor state to abide by its interna- tional obligations.37 In short, Article 51 indicates the hazards of relying on ostensibly neutral and superior legal authority (ICJ decisions) to reach codification outcomes, rather than on state practice, the weight of previous arbitral decisions, and the Commission’s own doctrine. Absent such traditional indicia of authority in international law, which go beyond the con- clusory pronouncements of the ICJ, the articles’ intellectual support and coherence must come from other sources or policy preferences. These considerations are not meant to suggest that many of the articles suffer from this kind of authority deficit, but the surprising lack of support for the crucial clause on proportionality in the use of countermeasures does give one pause about what other difficulties may be lurking in the construction of the articles. II. STRUCTURING INTERNATIONAL LEGAL PROCESS Aside from the technique seen in the articles’ use of international law sources, the coun- termeasures provisions are largely framed in terms of what might be called “international 31 An excellent assemblage of state practice on proportionality can be found in Cannizzaro, Role, supra note 30, at 900–13 (drawing in large measure on statements made by governments in diplomatic correspondence or in sub- missions before the UN Security Council). 32 See Commentaries, Art. 29, paras. 1–4. 33 See, e.g., Draft Articles on State Responsibility: Comments of the Government of the United States of America 4 (Mar. 1, 2001), excerpted in 95 AJIL 626, 627 (2001) [hereinafter U.S. Comments]. 34 Indeed, Article 51 is entitled “Proportionality,” perhaps suggesting that the terms are meant to be synonymous. 35 See Ago, 1971 Commentary, supra note 29. 36 See id. This appears to be precisely the point made by Judge Schwebel in his Nicaragua dissent, invoking Ago’s earlier commentary. See note 29 supra. 37 Ago, 1971 Commentary, supra note 29. 2002] SYMPOSIUM: THE ILC’S STATE RESPONSIBILITY ARTICLES 823 legal process.” Given the professed objective of the articles to articulate the “secondary rules” of state responsibility—including the modalities by which states can legitimately claim redress for violations of “primary” international norms—this focus is quite understandable. But even beyond adherence to this significant intellectual (if not also pragmatic and politi- cal) limit on the Commission’s work, the articles also express a serious concern to structure rules that channel state conduct. Indeed, the Commission often sought to deflect criticism from the state responsibility project by asserting that, essentially, the articles prescribe only “the legal consequences which flow” from breaches of international obligations, and do “not attempt to define the content of the international obligations breach of which gives rise to responsibility.”38 The extent to which this approach is combined in the countermeasure ar- ticles with the objective of creating a polite international society will be discussed further below. What needs to be considered here is how international law codification projects such as the articles have come to be regarded as a legitimate forum for the creation of international legal process values and mechanisms.39 The International Law Commission made its first foray into inculcating process values in international society with its preparatory work for what in 1969 became the Vienna Conven- tion on the Law of Treaties (Vienna Convention), although earlier hints were certainly evident in the Commission’s work on diplomatic and consular immunities. Not content with restat- ing the rules of treaty formation, the application and interpretation of those instruments, and their subsequent amendment and modification, the ILC also articulated the manner in which states would terminate or suspend treaty obligations. Some of these provisions were specifically denominated as “procedural.”40 But sometimes the Commission embedded an important procedural safeguard into what would otherwise be regarded as a substantive pro- vision. A good example was Article 60, which largely codified the “material breach” rule pre- viously found in customary international law.41 The material breach rule prevents states from pretextually terminating international obligations merely because of trivial violations by their treaty partners.42 In practice,43 the rule obliges states to exercise substantial caution in sus- pending or terminating treaty obligations absent clear and significant misconduct by a treaty partner. When in doubt, states will be obliged to continue their treaty obligations unless the threshold of material breach has been crossed. A treaty termination in response to an action that is later found not to be “material” would place the first mover in the position of having made an anticipatory breach, and thus of subjecting itself to sanctions.44 The material breach rule thus operates as a sort of safety valve in the sometimes volatile world of international relations. In operation, Article 60 of the Vienna Convention is like the ICJ’s ruling in the Nicaragua case,45 that the first nation found to cross the tripwire of the “armed attack” stan- dard of UN Charter Article 51 will be found to be the aggressor. Both Article 60 and the Nicaragua gloss on the threshold level of force rely on ambiguity to condition state action— to the extent that a standard of conduct is uncertain ex ante (How can states know for sure 38 Commentaries, para. 1. 39 For more on this connection between codification and legal process, one might examine ABRAM CHAYES, THOMAS EHRLICH, & ANDREAS F. LOWENFELD, INTERNATIONAL LEGAL PROCESS (1968–69); ROSALYN HIGGINS, PROBLEMS AND PROCESS: INTERNATIONAL LAW AND HOW WE USE IT (1994); Harold Hongju Koh, Why Do Nations Obey International Law? 106 YALE L.J. 2599 (1997); Harold Hongju Koh, Transnational Legal Process, 75 NEB. L. REV. 181 (1996). 40 Vienna Convention, supra note 13, Arts. 65–68. 41 See, e.g., Tacna-Arica Question (Chile v. Peru), 2 R.I.A.A. 921, 926 (1925). 42 Vienna Convention, supra note 13, Art. 60, para. 3 (defining material breach as an actual “repudiation of the treaty” or a “violation of a provision essential to the accomplishment of the object or purpose of the treaty”). 43 For U.S. practice, see Unilateral Abrogation or Material Breach, 1978 DIGEST, ch. 5, §3, at 767. 44 For more on this phenomenon, see Moore, supra note 8; John K. Setear, Responses to Breach of a Treaty and Rationalist International Relations Theory: The Rules of Release and Remediation in the Law of Treaties and the Law of State Responsibility, 83 VA. L. REV. 1 (1997); John K. Setear, An Iterative Perspective on Treaties: A Synthesis of International Relations Theory and International Law, 37 HARV. INT’L L.J. 139 (1996). 45 Nicaragua Judgment, 1986 ICJ REP. 14. 824 THE AMERICAN JOURNAL OF INTERNATIONAL LAW [Vol. 96:817 that a breach will later be regarded as “material”?), states should act to restrain responses, at least in marginal cases. Several provisions in the countermeasures section of the articles are also intended to mar- ginally restrain atavistic state behavior.46 The requirement in Article 49 that countermeasures be limited to suspensions of obligations that can be resumed once the offending nation’s conduct again conforms with international norms47—what might be called the “reversibility” principle—is one of those provisions. Article 49 does not directly prescribe a substantive limit to the application of countermeasures but achieves precisely the same result by a pro- cedural default. Whatever countermeasure a state selects has to be capable of being reversed. Interestingly, the commentaries specifically acknowledge the state responsibility project’s debt to the process rules contained in the Vienna Convention.48 Even more striking is Article 52, which governs “conditions relating to resort to counter- measures.” In its earlier form after the Commission’s first reading, the provision essentially required that states not apply countermeasures for as long as good faith negotiations were under way between the responsible state and the injured state.49 This restriction led to the bizarre result that a malefactor government could control the application of countermea- sures against it by maintaining the mere appearance of negotiations, a result particularly re- jected by the arbitral tribunal in the Air Services Agreement award in assessing French intransi- gence prior to the U.S. application of countermeasures.50 After the Commission’s final reading, the provision was altered to require merely that injured states are to “call on the responsible State . . . to fulfil its obligations” and to “notify the responsible State of any de- cision to take countermeasures and offer to negotiate with that State.”51 In any event, these notification requirements—sometimes called sommation—enjoy ample support, as the ILC commentaries properly suggest, in state practice and judicial decisions.52 Thus, after flirting with a modification of the Air Services Agreement rule on the timing of countermeasures, the Commission wisely chose to adhere to that expression of custom. So far, so good. Articles 49 and 52 manage to structure some explicit expectations into countermeasures practice. The principle of reversibility, especially read in conjunction with the commensurateness requirement of Article 51, demands that a victim state take no action that cannot be undone later. The notification requirement of Article 52 at least provides some transparency by putting a state on notice that another state considers it to be in breach of an international obligation and is contemplating responsive actions. Consequently, once in- formed, a state that unwittingly finds itself in violation of an international law norm can quickly remedy the situation. The very structure of these provisions allows states to stand down from potential disputes with face-saving diplomatic maneuvers. 46 Needless to say, analogues to the Vienna Convention on the Law of Treaties were significant in the construc- tion of the countermeasures section of the articles on state responsibility. See Commentary to pt. 3, ch. II, para. 4 & n.780; Art. 49, para. 9; Art. 50, para. 8 & n.809. 47 “Countermeasures are limited to the non-performance for the time being of international obligations of the State taking the measures towards the responsible State. . . . Countermeasures shall, as far as possible, be taken in such a way as to permit resumption of performance of the obligations in question.” Art. 49, paras. 2, 3. 48 Commentaries, Art. 49, para. 9: Paragraph 3 of article 49 is inspired by article 72(2) of the Vienna Convention on the Law of Treaties, which provides that when a State suspends a treaty it must not, during the suspension, do anything to preclude the treaty from being brought back into force. By analogy, States should as far as possible choose countermea- sures that are reversible. 49 First Reading Report, supra note 23, Art. 53, para. 4. 50 See Air Services Agreement Award, supra note 16, 18 R.I.A.A. at 445 (the tribunal noted that it “does not be- lieve that it is possible, in the present state of international relations, to lay down a rule prohibiting the use of counter- measures during negotiations”); see also James Crawford, Jacqueline Peel, & Simon Olleson, The ILC’s Articles on Re- sponsibility of States for Internationally Wrongful Acts: Completion of the Second Reading, 12 EUR. J. INT’L L. 963, 983 (2001). 51 Art. 52, para. 1(a), (b). 52 Commentaries, Art. 52, para. 3. 2002] SYMPOSIUM: THE ILC’S STATE RESPONSIBILITY ARTICLES 825 Similarly to the “materiality” test of the Vienna Convention, and the ICJ’s “tripwire” test of an armed attack, the articles incorporate elements of ambiguity that may give states further pause in their rush to punch and counterpunch against a malefactor. Article 52 presents two significant areas of ambiguity, although in fairness to the Commission, both of these have always characterized the customary international law of this subject. The first is the right granted in Article 52, paragraph 2, that an injured state “may take such urgent countermea- sures as are necessary to preserve its rights,”53 notwithstanding the notification requirement of Article 52, paragraph 1. There is a pragmatic flavor to this provision, based as it is on the premise that certain kinds of countermeasures (such as the freezing of assets and temporary stay orders) are effective only when the opposing state receives no advance notice of their consideration.54 Nonetheless, the distinction between “urgent countermeasures” and those taken in the normal course (and subject to notification requirements) will likely be trouble- some, and as the special rapporteur noted elsewhere, the distinction may not otherwise find any authority in state practice.55 The second ambiguity presented by Article 52 involves the relation between countermea- sures and dispute settlement mechanisms. The article provides that “[c]ountermeasures may not be taken, and if already taken must be suspended without undue delay if: . . . the inter- nationally wrongful act has ceased, and . . . the dispute is pending before a court or tribunal which has the authority to make decisions binding on the parties.”56 At first glance, this pro- vision seems unexceptional, a rephrasing of the relevant holding from Air Services Agreement that once a dispute is submitted to a tribunal that has the “means to achieve the objectives justifying the counter-measures,” the right to initiate countermeasures is vitiated and those already in force “may” be “eliminated,” but only to the extent that the tribunal can provide equivalent “interim measures of protection.”57 As can be readily seen, the ILC’s language in Article 52 is far more sweeping, and far more hopeful about the effectiveness of dispute set- tlement mechanisms, than the Air Services Agreement award.58 The ILC articles essentially create an absolute bar to the maintenance of countermeasures once the offending conduct has ceased and the matter is submitted “to any third party dispute settlement procedure.”59 That still leaves the problem, though, that the original misconduct could have caused substantial injury to a victim state, damage the responding state sought to redress through the use of coun- termeasures. The permissible uses of countermeasures under Article 49 include inducing a state to meet all of its obligations under part 2 of the articles, consisting of reparation as well as cessation. Despite the fact that the illegal conduct has ceased, its effects may well continue, and the obligation to provide reparations remains. It is thus by no means clear whether arbitral tribunals will be able to enforce provisional measures effectively, especially those similar in character to the countermeasures deployed by victim states in the period preceding the constitution of the arbitration. The fact that a tribunal might ostensibly have the power to indicate provisional measures, even while lack- ing actual jurisdiction over the dispute,60 raises concerns about the reading of Article 52(3). 53 Art. 52, para. 2. 54 Commentaries, Art. 52, para. 6. 55 Crawford, Peel, & Olleson, supra note 50, at 983 n.75 (citing Hungary’s and Japan’s comments criticizing the distinction). 56 Art. 52, para. 3 (emphasis added); see also Commentaries, Art. 52, para. 7. 57 Air Services Agreement Award, supra note 16, 18 R.I.A.A. at 445–46. 58 See id. at 446 (“As the object and scope of the power of the tribunal to decide on interim measures of protec- tion may be defined quite narrowly, however, the power of the Parties to initiate or maintain counter-measures, too, may not disappear completely.”). 59 Commentaries, Art. 52, para. 8. Interestingly, this language—which is significant because it excludes political adjustments (as in the UN Security Council) or recourse to domestic courts of one of the disputing parties—does not appear in the actual text of the article, an unfortunate drafting oversight. 60 See Aegean Sea Continental Shelf (Greece v. Turk.), Jurisdiction, 1978 ICJ REP. 3 (Dec. 19); Aegean Sea Conti- nental Shelf (Greece v. Turk.), Provisional Measures, 1976 ICJ REP. 3 (Sept. 11); Southern Bluefin Tuna Case 826 THE AMERICAN JOURNAL OF INTERNATIONAL LAW [Vol. 96:817 In such a circumstance, will a tribunal’s subsequent announcement of a lack of jurisdic- tion—and cancellation of the earlier imposed provisional measures—mean that counter- measures can resume? The articles are careful to indicate that countermeasures that had been canceled when a dispute was submitted to binding settlement may be reimposed if the tribunal’s provisional measures are ignored by the malefactor government or even if “the responsible State fails to implement the dispute settlement procedures in good faith.”61 But reimposition is not necessarily responsive to the problem caused when a tribunal, having once indicated provisional measures, later decides that it does not have jurisdiction over a dispute. And while the Commission’s commentaries refer to the ICJ’s recent decision in the LaGrand case62 as a hopeful sign confirming the binding effect of provisional measures, gov- ernments may quite properly doubt whether a system of tribunal- or court-imposed provi- sional measures will ever successfully substitute for vigorous countermeasures. That leaves the important question of whether the articles, although deliberately crafted to restrict the potential scope of countermeasures, may actually result in increasing license for their vigorous use. Indeed, various nations explicitly made this point in criticizing earlier versions of the draft articles.63 Codification acts, obviously, to crystallize an area of state con- duct and discretion. Despite the (very) occasional matters to be submitted for judicial settle- ment, the law of countermeasures has been played out in the realm of contextual incidents without an articulated framework of legality, aside from the recognition of a narrow range of restraints (notification and proportionality among them). International legal process the- ory might suggest that even a codification that embeds substantial ambiguity in a legal stan- dard might nonetheless give states a higher degree of confidence in pursuing robust coun- termeasures, which previously, in the complete absence of conventional standards, would have been rejected as too risky. A codified rule (whether or not actually adopted as a convention)64 changes the dynamic of subsequent disputes from consideration of past incidents and prac- tices to exegesis of a written text. The fact that the rule may be ambiguous or problematic may matter less to states contemplating aggressive countermeasures than the availability of some textual defense for the intended action. Such thinking is counterintuitive, for sure, but may well reflect the uncertain realities of state practice regarding an issue as basic as counter- measures. III. FORMALISM AND PRAGMATISM Codification projects offer fertile fields of inquiry for those intrigued by the way that inter- national law, as a discipline, strikes a balance between formal legal distinctions and practical concerns.65 The articles on state responsibility will be especially fruitful in this regard. Close examination of the articles illustrates that international law needs a high degree of formal- ism to be successful as a legal system, and that pragmatism (colored, as it often is, by diplomatic compromise and equivocation) is not always an unalloyed good. Pragmatic codification out- comes may simply hide underlying disagreements that are papered over with “constructive (Austl. & NZ v. Japan), Jurisdiction and Admissibility, 39 ILM 1359 (arb. trib. Aug. 4, 2000); Southern Bluefin Tuna Cases (NZ v. Japan; Austl. v. Japan), Provisional Measures, 38 ILM 1624, 1635 (Int’l Trib. Law of Sea Aug. 27, 1999). 61 Art. 52, para. 4; see also Commentaries, Art. 52, para. 9 (“Under the circumstances of paragraph 4, the limita- tions to the taking of countermeasures under paragraph 3 do not apply.”). 62 LaGrand Case (Ger. v. U.S.), Merits, paras. 99–104 (Int’l Ct. Justice June 27, 2001), 40 ILM 1069, 1092–94 (2001), cited in Commentaries, Art. 52, para. 8 & n.835. 63 State Responsibility, Comments and Observations Received from Governments, UN Doc. A/CN.4/488 (1998) [hereinafter Government Comments]. 64 See Caron’s contribution to this symposium, David D. Caron, The ILC Articles on State Responsibility: The Para- doxical Relationship Between Form and Authority, 96 AJIL 857, 867–68 (2002). 65 See BEDERMAN, supra note 8, at 162–70 (considering the use of formalistic and pragmatic tropes in the Vienna Convention on the Law of Treaties). 2002] SYMPOSIUM: THE ILC’S STATE RESPONSIBILITY ARTICLES 827 ambiguities” or open-ended standards. In contrast, some formalisms establish bright-line rules, which states, for better or worse, are prepared to respect in planning their actions. The articles by their very structure and tone convey a high degree of formalism. This may well be a legacy of Special Rapporteur Ago’s work on the project, although Professor Crawford did much to help persuade the Commission to strip away some superfluous layers of abstrac- tion. It can hardly be said that the articles overtly evidence pragmatic policy preferences. Rather, these preferences are often subtly conveyed in the provisions and commentaries. But for an area as rife with political expediency as the law of countermeasures, formalism may actually be a good thing. One significant formalism dominates this topic. That is the now well accepted tripartite division between reprisals (forcible responses), countermeasures (nonforcible actions), and retorsions (legal, but unfriendly, responses). The articles take care to limit their scope to nonforcible countermeasures. Reprisals are properly relegated to the law of the UN Charter or to international humanitarian law.66 The one narrow exception is the savings clause in Article 50, confirming that “[c]ountermeasures shall not affect . . . obligations for the pro- tection of fundamental human rights [or] . . . obligations of a humanitarian character pro- hibiting reprisals.”67 This exception is the natural corollary of the clause that limits counter- measures to being directed only against states,68and the concerns of humanity implicated in this clause have had a long history in international law doctrine.69 That still leaves the ques- tion, though, of how victim states are to respond when a malefactor transgresses “fundamen- tal human rights,” “obligations of a humanitarian character,” or, indeed, any “other obliga- tions under peremptory norms of general international law.”70 Obviously, such a violation could also trigger the provisions of Articles 40 and 41, the last residue of the concept of “crimes of State” in the project.71 What the articles make clear is that states may not respond in kind to human rights abuses.72 The use of countermeasures in the face of widespread human rights abuses by some nations will be a significant element in the practical implementation of the articles. While the restrictions in Article 50(1) are not likely to have an impact on the devel- opment of human rights sanctions practice, this remains an area that will need to be closely watched. The Commission was also wise to resist the temptation to prescribe rules of state conduct in relation to retorsions. The ILC seemed to eschew suggestions by some nations that retor- sions ought to be subject to the same rules of sommation, relevance, and commensurateness as those imposed for countermeasures. Although stray language in some of the commentaries might be interpreted to lead to this conclusion,73 it does not appear to have been seriously intended. That leaves, as perhaps the most pragmatic turn of this set of provisions, Article 54 on mea- sures taken by others than the injured state, more colloquially known as “collective counter- measures.” This article may also represent the most significant act of indirect progressive 66 See Commentary to pt. 3, ch. II, para. 3. 67 Art. 50, para. 1(b), (c). 68 Art. 49, para. 1 (“An injured State may only take countermeasures against a State . . .” (emphasis added)); see also Commentaries, Art. 49, para. 4. 69 See Naulilaa Arbitration, supra note 14, 2 R.I.A.A. at 1026 (countermeasures must be “limited by the require- ments of humanity and the rules of good faith applicable in relations between States”); 1934 ANNUAIRE DE L’INSTI- TUT DE DROIT INTERNATIONAL 710. 70 Art. 50, para. 1(d). 71 Arts. 40, 41 (governing situations of “gross or systematic failure” of states to follow “peremptory norm[s] of general international law,” entailing such additional consequences as the requirements that other states “coop- erate to bring [the breach] to an end through lawful means” and that states shall not “recognize as lawful a situa- tion created by a serious breach . . . nor render aid or assistance in maintaining that situation”). 72 Commentaries, Art. 50, para. 6. 73 See id., Art. 50, para. 7 (discussing General Comment 8 of the UN’s Committee on Economic, Social and Cultural Rights as limiting the effect of economic sanctions on civilian populations, without seeming to differen- tiate between such sanctions as retorsions and countermeasures). 828 THE AMERICAN JOURNAL OF INTERNATIONAL LAW [Vol. 96:817 development among the countermeasure clauses, with the commentaries frankly admitting that “[p]ractice on this subject is limited and rather embryonic.”74 Textually, Article 54 is a savings clause that purports not to codify a rule of practice regarding collective “lawful mea- sures,” in view of the “sparse” state practice and the “uncertain” posture of the law.75 But, in reality, Article 54 provides the necessary ambit for the significant provisions of Article 48, paragraph 1, granting standing to states other than the injured polity, where the obligation breached “is owed to a group of States . . . and is established for the protection of a collec- tive interest of the group” or “is owed to the international community as a whole.”76 While the import of this clause is examined elsewhere in this symposium,77 the Commis- sion’s choice regarding the extent to engage in codification also deserves mention. To artic- ulate a rule for collective countermeasures prematurely would run the risk of “ ‘freez[ing]’ an area of law still very much in the process of development.”78 But to say nothing on the subject might have raised the (apparently) false impression that collective countermeasures were barred and that only “injured States,” as defined in the articles, were eligible to impose them.79 The pragmatic compromise—and, indeed, the only possible political solution—was to defer debate to another day and to allow customary international lawmaking processes to elabo- rate any conditions on the use of collective countermeasures. Once again, a provision in the articles does not appear to be what it seems. Counterintu- itively, a savings clause that notionally makes no law is actually intended to induce significant state practice and is expected to tend toward the progressive development that will be achieved through time. Pragmatism in the making of international law often works that way. Allowing events to take their course is often the wisest move in codification. Indeed, the most prag- matic turn of the articles on state responsibility was the decision not to present them in the form of a convention to be adopted by the General Assembly, followed by a diplomatic con- ference and the process of entry into force or ultimate rejection.80 The ILC obviously has learned its lesson well in observing the results when it demands that controversial progressive lawmaking projects be subjected to the caustic test of state ac- ceptance through negotiation, adoption, and ratification. One need only look as far as the sad fate of the Commission’s projects on state succession and immunities to realize that the course of codification does not always run smoothly. Nor is it intellectually satisfactory to conclude that the state succession project was doomed because its provisions really reflected progressive development more than a true codification of customary international law. The Commission’s primary successes—including the 1958 law of the sea treaties and the 1961 and 1963 diplomatic and consular relations conventions—were a confection of progressive development and codification. Scholars and practitioners remain in doubt as to how to cate- gorize certain provisions in those instruments, and it hardly matters. The final measure of success was whether enough members of the international community manifested the polit- ical will and desire to bring a law reform project to fruition. There are some topics that states really have little desire to see codified, despite their protestations to the contrary in the UN General Assembly’s Sixth Committee. For all these reasons, the articles on state responsibility—and their provisions on coun- termeasures—may end up being more influential as a statement of principle (much like the 74 Id., Art. 54, para. 3. 75 Id., Art. 54, para. 6. Ironically, though, the commentaries provide more evidence of state practice on this par- ticular point than they do in relation to any other issue having to do with countermeasures. See id., paras. 3, 4 (citing no fewer than eight incidents). 76 Art. 48, para. 1. 77 See the contribution to this symposium by Edith Brown Weiss, Invoking State Responsibility in the Twenty-first Cen- tury, 96 AJIL 798 (2002). 78 Crawford, Peel, & Olleson, supra note 50, at 981. 79 Id. at 981–82. 80 See Crawford Comments, supra note 1, at 58–60. 2002] SYMPOSIUM: THE ILC’S STATE RESPONSIBILITY ARTICLES 829 Nuremberg Principles adopted by the Commission in 1950 and the Model Rules on Arbitral Procedure in 1958) than as a legally binding treaty instrument.81 While Caron, in his contri- bution to this symposium, expresses some concerns on this point,82 the previous practice of standing international courts and tribunals of regularly citing the draft articles,83 even when the provisions were in a fairly inchoate (some might say, incoherent) form, should dispel such concerns. Without the possibility of repudiation or rejection, the only real challenge to the authority of the articles will be posed by the ongoing process of customary international law formation, which either may render portions of the clauses irrelevant over time, or, if there is a seismic shift in international values and processes (such as acceptance of “crimes of State” as a valid principle of state responsibility) may subject the entire project to the indignities of desuetude. I think there is little risk of that, however, because the articles at least have the virtue of being pitched at a level of abstraction that will prevent them from being superseded too quickly by changes in international life. Still, one cringes now to contemplate the fate of García- Amador’s draft project,84 which emphasized injuries to alien persons and property as a sub- set of wider state responsibility, had it actually been adopted by the Commission and the Gen- eral Assembly in the 1960s. As technically accomplished as it was, that draft overemphasized primary rules of state conduct, norms that were quickly evolving over time. It would have quickly been relegated to the dustbin of history. The stripping of primary rules from the suc- cessive Ago, Riphagen, Arangio-Ruiz, and Crawford versions of the draft articles fixed that problem, as did the elimination of the insuperably difficult provisions on crimes of state. And while the Commission flirted with the potentially fatal attraction of severely restricting the scope of countermeasures, it ultimately distanced itself from that position and accepted, instead, the set of modest procedural limitations we now see. These adjustments do not mean that the articles are completely immune from the criticism, intimated in the introduction to this symposium,85 that they reflect the law of the twentieth century more than they prescribe rules for the next. No resort to abstraction, or elimination of contentious issues, can obscure the fact that the articles concern, ultimately, the affairs of states. Only to the extent that a positivist vision of twentieth-century international law placed states at the center of the doctrinal universe does it make sense, I suppose, to develop a set of rules for state responsibility. Ironically, in the fifty years it took the ILC to draft the articles, states have progressively lost ground in their competition with other international actors. The exclusion of individuals and international organizations from the scope of the articles86 was a relatively uncontroversial move, but it could be criticized for a lack of vision. Does it really make sense to hermetically seal state responsibility from that of other international actors, particularly at a time when nonstate actors are gaining in strength, power, prestige, and legiti- macy? Ultimately, all this may well matter the most at the intersection of state power, individ- ual rights, and countermeasures. 81 See Caron, supra note 64, at 861–66. 82 Id. at 868, 868–73. 83 My admittedly imperfect survey of online databases for international tribunal decisions and awards indicates about twenty-two World Court decisions (opinions, judgments, or orders) that cite the draft articles on state respon- sibility (twelve, if the iterative opinions in the Legality of the Use of Force cases are combined), about fifteen awards of the Iran Claims Tribunal, about the same for WTO dispute panels, and one case decided in the International Criminal Tribunals for the Former Yugoslavia and for Rwanda. Westlaw search (INT-ICJ, INT-IRAN, INT-ICTY, INT- ICTR, WTO-DEC databases) (performed Aug. 6, 2002). 84 See F. V. GARCÍA-AMADOR, LOUIS B. SOHN, & R. R. BAXTER, RECENT CODIFICATION OF THE LAW OF STATE RESPON- SIBILITY FOR INJURIES TO ALIENS (1974). 85 Daniel Bodansky & John R. Crook, Introduction and Overview, 96 AJIL 773, 790 (2002). 86 Arts. 57, 58 (“These articles are without prejudice to any question of the responsibility under international law of an international organization, or of any State for the conduct of an international organization.” “These ar- ticles are without prejudice to any question of the individual responsibility under international law of any person acting on behalf of a State.”). 830 THE AMERICAN JOURNAL OF INTERNATIONAL LAW [Vol. 96:817 IV. POLITE INTERNATIONAL SOCIETY As already suggested, the overarching thrust of the articles on countermeasures is to codify rules to limit the risk of dangerous and reflexive acts by governments faced with a challenge to their legal rights. The overall tone of the ILC commentaries is that countermeasures are a necessary evil, a regrettable byproduct of the imperfect, premillenarian condition of inter- national law and relations, which is a “decentralised system by which injured States may seek to vindicate their rights.”87 In short, “countermeasures [may be] justified under certain circum- stances” but, at the same time, international law needs “to ensure, by appropriate conditions and limitations, that countermeasures are kept within generally acceptable bounds.”88 The grudging wariness with which the Commission viewed countermeasures permeated the entire project. This attitude can be seen most readily in the articulation in Article 49 of the objects and limits of countermeasures, particularly the antecedent requirement that coun- termeasures be taken only to induce wrongdoers to comply with their duties under interna- tional law and to make reparation for their wrongful acts, and that the action taken must be reversible. Article 49 thus speaks to the concern that countermeasures may often be used pretextually to coerce innocent states or actors. The rule of proportionality/commensurate- ness in Article 51 is assuredly another aspect of the Commission’s desire to cabin and restrain any tendency for the use of countermeasures to escalate. Lastly, the requirement of Article 53 that countermeasures “be terminated as soon as the responsible State has complied with its obligations”89 alleviates the fear that countermeasures could become a permanent feature of relations between states. In short, the articles attempt to circumscribe the legitimate con- ception, objectives, escalation, and continuance of self-help among states. And if pretext, escalation, and permanence were not enough to worry about, the Commis- sion also sought to create a “safe zone” of international life that would be totally unaffected by unilateral, or (under Article 54) collective, countermeasures. As it now reads, Article 50 is a veritable mélange of fundamental norms, nonderogable rules, and procedural mechanisms that are immunized from self-help practice. There appears to be little rhyme or reason to the structure of this clause, although this deficiency may have to do with its long and tortured drafting history. Nevertheless, the prohibitions of Article 50 are not without their logic. The Commission viewed uses of force in violation of the Charter and reprisals contravening international humanitarian law as the two clear examples in Article 50, paragraph 1. But exempting from the scope of countermeasures “obligations for the protection of fundamental human rights”90 drew objections from some governments,91 and even the special rapporteur noted that the provision was “problematic.”92 Part of the problem is that the phrase “fundamental human rights” has no established meaning, aside from suggesting that certain human rights, while not rising to the level of peremptory norms, may nonetheless deserve special protection. While the commentaries attempt to connect this formulation with conventional rules of nondero- gation of human rights norms,93 this linkage is nowhere made express. Article 50 is far more successful when it specifically makes the “inviolability of diplomatic or consular agents, 87 Commentary to pt. 3, ch. II, para. 1. 88 Id., para. 2. 89 Art. 53. 90 Art. 50, para. 1(b). 91 See U.S. Comments, supra note 33, at 2–3. 92 Crawford Comments, supra note 1, at 50. In light of the fact that the first reading of the draft articles in 1996 referred merely to “basic human rights,” the revised formulation might have even been regarded as an improve- ment. First Reading Report, supra note 23, Art. 50, para. (d). 93 Commentaries, Art. 50, para. 6 n.803 (citing the International Covenant on Civil and Political Rights, Dec. 16, 1966, Art. 4, 999 UNTS 171; European Convention for the Protection of Human Rights and Fundamental Free- doms, Nov. 4, 1950, Art. 15, 213 UNTS 221; American Convention on Human Rights, Nov. 22, 1969, Art. 27, 1144 UNTS 123). 2002] SYMPOSIUM: THE ILC’S STATE RESPONSIBILITY ARTICLES 831 premises, archives and documents” not subject to any form of countermeasure, since the contours of inviolability are clearly established in law and practice.94 Most of the attention devoted to Article 50 after the first reading in 1996 focused on whether to retain the provision banning countermeasures that constitute “extreme economic or polit- ical coercion designed to endanger the territorial integrity or political independence of the State which has committed the internationally wrongful act.”95 Special Rapporteur Crawford, in his private comments about the first-reading draft, was savagely critical of this clause: By definition countermeasures are coercive . . . . To say that countermeasures may not involve “extreme” coercion is to say one of two things: either the countermeasures are disproportionate, in which case they are excluded by the principle of proportionality, or they are not, in which case they will necessarily be a response to extreme wrongdoing causing injuries which have not been redressed. Article  had accordingly to be reconsidered.96 Thus, it came as no surprise that the “extreme economic or political coercion” restriction was excised from the articles. But that nod to political reality should not be confused with a ringing endorsement of an energetic system of self-help. Quite the contrary, as Crawford’s comments indicate,97 the requirement of commensurateness in Article 51 may well serve as an even stronger restraint on particularly coercive forms of countermeasures. The real question, of course, is whether it is even a desirable goal for international society to become too polite. This inquiry should transcend the somewhat rhetorical debate that has been waged between “liberal” internationalists, who see the virtues of muscular self-help ini- tiatives by governments, and “restrictionists,” who view countermeasures as little more than a cover for cowboy diplomacy by rich, powerful states. Posited in the way that I would prefer, the discussion must turn to whether international society has sufficient “vertical” enforcement power—whether manifested through collective security regimes, effective sanctions by re- sponsible institutions, or credible mobilizations of shame against bad governments—to war- rant a departure from vigorous “horizontal” enforcement through self-help measures. I cer- tainly do not wish to endorse what has been called the “Heinlein effect” for international relations: an armed world society has not historically proven to be a polite one.98 Yet one must still reckon that a calibrated recourse to nonforcible countermeasures may be preferable to other outlets for enforcement, even police actions by legitimate international institutions. Given the cumbersome procedure to initiate actions under Chapter VII of the UN Charter, and their sometimes striking lack of effectiveness, having in place a robust and parallel coun- termeasures regime may prove to be a saving grace for international law. The countermeasure clauses of the articles on state responsibility, more than any other set of their provisions, feature a profound impulse toward social engineering for international relations. In this respect, the articles are forward-looking, imagining a time in international life when unilateral and horizontal means of enforcement through robust self-help will be a thing of the past. The drafters of the articles certainly attempted to reject a nineteenth- century vision of reprisals and great-power gunboat diplomacy in creating a twenty-first-cen- tury regime for countermeasures. That they were at least partially foiled in this process— 94 Art. 50, para. 2(b). 95 First Reading Report, supra note 23, Art. 50, para. (b). 96 Crawford Comments, supra note 1, at 50–51. 97 See id. 98 The Robert Heinlein effect is named after the science fiction author who coined the aphorism “an armed so- ciety is a polite society.” See James P. Pinkerton, Nullification: Wrong in 1832 and in 1995, L.A. TIMES, Oct. 12, 1995, at B9 (quoting Robert Heinlein). For sociological literature on this point, see ROGER D. MCGRATH, GUNFIGHTERS, HIGHWAYMEN, AND VIGILANTES (1984); John Umbeck, Might Makes Rights: A Theory of the Foundation and Initial Distribution of Property Rights, 19 ECON. INQUIRY 38 (1981). 832 THE AMERICAN JOURNAL OF INTERNATIONAL LAW [Vol. 96:817 usually through the interventions and comments99 of states still quite desirous of preserving freedom of action with respect to self-help—is illustrative of the real dynamic of codification and progressive development on a topic as sensitive as state responsibility. The articles may well have a postmodern structure and texture, but, at least for the countermeasure provisions, they reflect as much the imperative of realist power politics. The articles pretend to be a rule book of clearly articulated limits on state action, pithily presented in the form of legal terms of obligation, but their greatest impact will be in not prescribing clear courses of conduct for situations in that middle ground of gray uncertain- ties—where law matters most, and yet is really helpless. Some may rightly speak of the “con- structive ambiguities” of the articles, and compliment the drafters on avoiding and resolving the difficulties of some of the provisions. But the ambiguous standards embedded in the countermeasure clauses (commensurateness in Article 51, reversibility in Article 49, urgency in Article 52 (2), and fundamental human rights in Article 50(1), to name a few) will assuredly give rise to significant disputes, though not by design, and at the same time will cause some states to restrain their behavior (when they are otherwise disposed to be restrained or fear the consequences of making false moves), while others may well test the limits of the regime. The problem with ambiguity is that it will tend to restrain those states that may have suffered the most grievous wrongs or that are pursuing the most wholesome values and objectives that international law can conceive. On the other hand, it may allow atavistic states to game the system in ways that we now cannot fully appreciate. At stake in this vision of the future is nothing less than a structure for states to vindicate their rights and responsibilities in international law. A cynic doubtful of the effect of legally binding instruments in conditioning state behavior—much less the “soft law” articulation of principles that the articles now aspire to—would be quick to dismiss the likely impact of the countermeasure articles, whether positive or negative, on states’ future recourse to self- help. The articles, like all international law, would be epiphenomenal.100 But I am actually quite confident that the articles will be immensely influential. And that may be precisely the problem. Of all the counterintuitions noted here—that am- biguous standards may help restrain states, that savings clauses may actually spur the progres- sive development they disclaim, and that a declaration of principles may be more authoritative than a formal treaty—the one that is most troubling is that writing a rule book for self-help may actually encourage governments to play a game of punch and counterpunch that they had previously avoided. Alternatively, it may be feared that observance of the articles’ restric- tive provisions may hamstring effective responses to truly atavistic state behavior. Then again, they may not lead to either result and instead will achieve their purpose in channeling and civilizing state behavior and ushering in this century’s superior regime of international law enforcement. The good news is that within a generation we should have solid empirical data to test the practical effects of the countermeasure provisions of the articles. In the meantime, we are left to speculate about the role of good intentions in international law codification. 99 See Government Comments, supra note 63; U.S. Comments, supra note 33. 100 See KENNAN, supra note 2, at 95; see also HANS J. MORGENTHAU, POLITICS AMONG NATIONS: THE STRUGGLE FOR POWER AND PEACE 3–17 (Kenneth W. Thompson ed., 6th ed. 1985).