the inductive and deductive methods in customary ... - Georgetown Law

THE INDUCTIVE AND DEDUCTIVE METHODS IN CUSTOMARY INTERNATIONAL LAW ANALYSIS: TRADITIONAL AND MODERN APPROACHES WILLIAM THOMAS WORSTER* ABSTRACT Contemporary customary international law analysis is currently understood to be a struggle between traditional and modern approaches, implicating significant normative outcomes. These opposing approaches are supposed to reflect the use of the inductive and deductive methods respectively, with the former excusing the freedom of action of the state and the latter limiting the freedom of the state. This image of two schools, possibly in struggle, however, does not fully capture the ways in which the inductive and deductive methods are actually intertwined in customary international law analysis. The methods are not two opposing monolithic techniques. Instead, in practice, the methods are intermixed, combining a variety of choices. This Article will suggest the complex ways inductive and deductive analyses are layered in the assessment of customary international law and refute the notion that the inductive and deductive approaches can be so easily separated. Customary international law is based on a deductive foundation that that the binding nature of the law can be based on consent, that custom is a source of law, and that custom is established by state practice and opinio juris. In determining which rules are binding under customary international law, we use either induction or deduction to construct hypotheses about the law, and determine that those hypotheses can be tested through a sampling study, amassing evidence. The conclusion that we can test for the law through an inductive process, is itself a deduction, and a deduction that avoids the question of whether the issue to be tested is truly a question of law or one of fact. After all, questions of fact are more easily understood as inductively verifiable. Once we determine that we can test for the law, we must construct most frequently through deduction, the types, forms and qualities of evidence, and the nature of the actors, that will be admissible for this examination. We reach conclusions on the form of the data pool and how it

* Lecturer, International Law, The Hague University of Applied Sciences, The Hague, The Netherlands; LL.M. (Adv.) in Public International Law, cum laude, Leiden University, Faculty of Law, Leiden, The Netherlands; J.D., Chicago-Kent College of Law, Illinois Institute of Technology, Chicago, Illinois; B.A., Modern European History, University of Kansas, Lawrence, Kansas. The author would like to express his appreciation to his research assistant, Veronika Georgieva, for her unfailing efforts to find many of the sources herein. All errors are the author’s own. © 2014, William Thomas Worster.

445

GEORGETOWN JOURNAL OF INTERNATIONAL LAW

is to be assembled and we assess the sufficiency of the evidence to prove the hypothesis being tested, again most likely through deduction. Once we are convinced that the threshold of proof for the hypothesis has been met, we need to take the final step of determining whether that convincingly descriptive study can be realized as a prescription. What we see is that fundamental, foundational questions about the methodology, process of analysis, and forms of evidence are largely deduced, whereas questions about the empirical verification of practice and opinio juris are largely induced. This conclusion means that no analysis of customary international law can be completely classified as deductive or inductive. Instead, each act of customary international legal analysis employs a blend of approaches across a handful of questions. In doing so, it balances the tension between apology and utopia, and realizes the benefit of the differing approaches. We can only critique the balance struck in each case and argue whether it is a just balance. I. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. DEDUCTION, INDUCTION, AND THE SCIENTIFIC METHOD . . . . . . . A. Deduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Induction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Types of Induction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. The Hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV. THE INTERPLAY OF DEDUCTION AND INDUCTION IN CUSTOMARY INTERNATIONAL LAW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. The Foundations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Custom as a Source of Law. . . . . . . . . . . . . . . . . . . 2. The Elements of Customary International Law . . . B. Articulating the Hypothesis . . . . . . . . . . . . . . . . . . . . . . . . C. Testing the Hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Proving Domestic Customary Law . . . . . . . . . . . . . 2. Proving Customary International Law . . . . . . . . . . D. Establishing the Types and Forms of Evidence . . . . . . . . . . . 1. Qualifying Categories of Actors . . . . . . . . . . . . . . . 2. Qualifying Categories of Acts . . . . . . . . . . . . . . . . . 3. Quality of the Acts: Density, Uniformity, Consistency, Extensitivity, Representativity . . . . . . E. Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Assessing the Sufficiency of the Evidence . . . . . . . . . . . . . . . 1. Inductive Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . 2. Deductive Analysis . . . . . . . . . . . . . . . . . . . . . . . . . 3. Burden of Proof . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Essentially Inductive Step in the Analysis . . . . . . . . 446

447 448 455 455 456 459 461 464 465 465 468 471 476 477 480 483 484 486 494 496 501 503 505 509 516

[Vol. 45

INDUCTIVE AND DEDUCTIVE METHODS IN CIL ANALYSIS

G. Conclusion on Whether a Prescriptive Rule has been Established or Not . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I.

519 520

INTRODUCTION

Contemporary customary international law analysis is currently understood to be a struggle between traditional and modern approaches, with those approaches having significant normative outcomes. These opposing approaches are supposed to reflect the use of the inductive and deductive methods respectively, with the former excusing the freedom of action of the state and the latter limiting the freedom of the state.1 This image of two schools, possibly in struggle, however, does not fully capture the ways in which the inductive and deductive methods are actually intertwined in customary international law analysis. The methods are not two opposing monolithic techniques. Instead, in practice, the methods are intermixed, combining a variety of choices. Deductive steps are taken regarding the use of induction, inductive steps are taken in reaching patterns from practice, deductive steps are taken in identifying sample pools, and so on. Each use of deduction or induction may indeed have certain normative effects, but the intermixing of the two methods provides some checks and balances, and the modern and traditional schools largely agree on many aspects of this approach. This Article seeks to examine the application of the inductive and deductive methods in customary international legal analysis and criticize the current characterization of two competing analytical methods. It will substantiate the claim that induction and deduction are both used in the analysis of customary international law in a delicate, yet valuable, balance where two trends exist in corrective tension with each other. Furthermore, this situation is neither traditional, nor modern—in a chronological sense— but is consistent throughout the history of customary international law analysis. As a preliminary matter, the author must first clarify the terms “deduction” and “induction.” Deductive reasoning is often described as

1. See, e.g., MARTTI KOSKENNIEMI, FROM APOLOGY TO UTOPIA: THE STRUCTURE OF INTERNATIONAL LEGAL ARGUMENT 2 (1989) (arguing that the assessor prejudices the outcome of the customary international law analysis because of a predisposition for an “apology” for power politics or “utopian”); W. Michael Reisman, The Cult of Custom in the Late 20th Century, 17 CAL. W. INT’L L.J. 133, 135 (1987) (arguing that any increased use of custom is a “great leap backwards” that necessarily apologizes for the behavior of more powerful states).

2014]

447


going from “the general to the specific” or “truth-preserving.” In essence, a valid deductive argument is one in which the premises—if true—must lead to a true conclusion. Induction, on the other hand, is often described as drawing inferences from specific observable phenomena to general rules, or “knowledge expanding.” Here evidence is collected about observable events and a premise is constructed based on the collected data. The degree to which the conclusion is probably true is based on the quality of the evidence used to support it. Following an introductory first section reviewing the background to the debate over approaches, the Article will consider deduction and induction in more detail, attempting to understand these two methods of logical reasoning. The next section will then examine the way customary international law is analyzed by international tribunals by deconstructing the analytical process into smaller logical steps. For each step, an analysis will be undertaken of whether the conclusion is reached through induction or deduction, and whether this approach is consistent across tribunals and historical eras. Through this approach, the Article will take a first step to suggest the complex ways inductive and deductive analyses are layered in the assessment of customary international law and refute the notion that the inductive and deductive approaches can be so easily separated. II.

BACKGROUND

As is well known, customary international law is established when there is a widespread and consistent state practice and opinio juris.2 State practice consists of the highly consistent acts of a widespread number of states manifesting a certain behavior.3 Opinio juris is the belief on the part of the states manifesting the relevant behavior that

2. See Statute of the International Court of Justice art. 38, June 26, 1945, 59 Stat. 1055, 33 U.N.T.S. 993; Statute of the Permanent Court of Justice, Dec. 16, 1920; S.S. Lotus (Fr. v. Turk.), 1927 P.C.I.J. (ser. A) No. 10, at 18 (Sept. 7); Asylum (Colom. v. Peru), Judgment, 1950 I.C.J. 266, 276-77 (Nov. 20); North Sea Continental Shelf (Ger. v. Den.; Ger.v. Neth.), Judgment, 1969 I.C.J. 3, 44 (Feb. 20); Military and Paramilitary Activities in and Against Nicaragua (Nicar. v U.S.), Judgment, 1986 I.C.J. 14, ¶ 207 (June 27); Gregory Shaffer & Tom Ginsburg, The Empirical Turn in International Legal Scholarship, 106 AM. J. INT’L L. 1 (2012); Lassa Oppenheim, The Science of International Law: Its Task and Method, 2 AM. J. INT’L L. 313, 315 (1908) (“The rules of the present international law are to a great extent not written rules, but based on custom.”). 3. See Asylum, 1950 I.C.J. at 276 (“The Colombian Government must prove that the rule invoked by it is in accordance with a constant and uniform usage practised by the States in question . . . .”); id. at 276-77 (holding that state practices were lacking in the consistency and certainty required to constitute “constant and uniform usage”); IAN BROWNLIE, PRINCIPLES OF PUBLIC INTERNATIONAL LAW 4-11 (5th ed. 1998).

448

[Vol. 45


they are compelled to act as they do.4 According to the International Court of Justice (ICJ), the existence of state practice and opinio juris “can be tested by induction based on the analysis of a sufficiently extensive and convincing practice, and not by deduction from preconceived ideas.”5 The deductive process has been appreciated for its objectivity, consistency and formalism, and thus has proven more resistant to the trend toward state apology.6 However, that appreciation of its degree of detachment from state practice means that the laws it discovers also lack an element of having been deliberately created by states.7 Many scholars have identified a shift in customary international legal analysis from the “traditional” to the “modern” approach.8 What

4. See Military and Paramilitary Activities in and Against Nicaragua, 1986 I.C.J. ¶ 207 (“[F]or a new customary rule to be formed, not only must the acts concerned ‘amount to a settled practice’, but they must be accompanied by the opinio juris sive necessitatis. Either the States taking such action or other States in a position to react to it, must have behaved so that their conduct is ‘evidence of a belief that this practice is rendered obligatory by the existence of a rule of law requiring it.’”); North Sea Continental Shelf (Ger. v. Den.; Ger.v. Neth.), Judgment, 1969 I.C.J. 3, 44 (Feb. 20) (“[A]n indispensable requirement would be that within the period in question, short though it might be, state practice, including that of states whose interests are specially affected, should have been both extensive and virtually uniform in the sense of the provision invoked, and should moreover have occurred in such a way as to show a general recognition that a rule of law or legal obligation is involved.”); Right of Passage Over Indian Territory (Port. v. India), Judgment, 1960 I.C.J. 6, 42-43 (Apr. 12); Asylum, 1950 I.C.J. at 276-77; Lotus (Fr. v. Turk.), 1927 P.C.I.J. (ser. A) No. 10, at 28 (Sept. 7) (“[O]nly if such abstention were based on [states] being conscious of a duty to abstain would it be possible to speak of an international custom.”). 5. Delimitation of the Maritime Boundary in the Gulf of Maine Area (Can. v. U.S.), Judgment, 1984 I.C.J. 246, ¶ 111 (Jan. 20). The use of “can” rather than “is” in the quoted passage above does imply that customary international law could be discovered deductively. However, the Court went on to explicitly reject the norms proffered by the parties on the basis that they were tested deductively. See also Continental Shelf (Libya v. Malta), Judgment, 1985 I.C.J. 13, ¶ 27 (June 3). 6. See Jo¨rg Kammerhofer, Uncertainty in the Formal Sources of International Law: Customary International Law and Some of Its Problems, 15 EUR. J. INT’L L. 523, 542 (2004) (“The advantage of this approach is that it keeps the norm apart from the facts, this approach does not mix the ideal with the real. It can be an internally consistent system.” (citations omitted)). 7. See id. (“The problem is that a deduction of norms means that these are not the result of a human act of will, a human legislation in the widest sense.”); id. (citing HANS KELSEN, ALLGEMEINE THEORIE DER NORMEN 5-6 (1979) (translated by Kammerhofer)) (“[T]he norms of the Vernunftrecht [are] the sense of an act of cognition, [they are] not willed, but imagined norms . . . . I can imagine such a norm only as the sense of a presupposed act of will. I can imagine a norm, as if it had been enacted by an authority, even though it has not been, in fact, enacted, [even though] there is in fact no act of will whose sense [the norm] is.”)). 8. See, e.g., ANDREW T. GUZMAN, HOW INTERNATIONAL LAW WORKS 218 (2008); OSCAR SCHACHTER, INTERNATIONAL LAW IN THEORY AND PRACTICE 35-36 (1991); JAMES BRIERLY, THE LAW OF NATIONS 62

2014]

449


is meant is that the relative importance and roles of state practice and opinio juris have shifted over time within the methodology.9 Whereas once the assessment of customary international law, primarily by international tribunals, began with practice and confirmed the practice with opinio juris, scholars now see the reverse.10 Moreover, under the traditional methods, the assessor of customary international law would have used an inductive approach, but under the modern approach, the assessor will apply a deductive approach.11 In fact, some scholars have even argued that the change in approach has gone unnoticed, with practitioners of the “modern” method believing that they were still applying the “traditional” approach.12 As evidence of this change in approach, commentators look to the special reliance of international

(6th ed. 1963); Jan Wouters & Cedric Ryngaert, The Impact on the Process of the Formation of Customary International Law, in THE IMPACT OF HUMAN RIGHTS LAW ON GENERAL INTERNATIONAL LAW 111 (Menno T. Kamminga & Martin Scheinin eds., 2009); George Norman & Joel P. Trachtman, The Customary International Law Game, 99 AM. J. INT’L L. 541, 542 (2005); Kammerhofer, supra note 6; Anthea Elizabeth Roberts, Traditional and Modern Approaches to Customary International Law: A Reconciliation, 95 AM. J. INT’L L. 757 (2001); Jack L. Goldsmith & Eric A. Posner, Understanding the Resemblance between Modern and Traditional Customary International Law, 40 VA. J. INT’L L. 639, 640 (2000); Bruno Simma & Philip Alston, The Sources of Human Rights Law: Custom, Jus Cogens, and General Principles, 12 AUSTL. Y.B. INT’L L. 82 (1988-1989); Anthony A. D’Amato, Trashing Customary International Law, 81 AM. J. INT’L L. 101 (1987); Bin Cheng, United Nations Resolutions on Outer Space: ‘Instant’ International Customary Law?, 5 INDIAN J. INT’L L. 23 (1965), reprinted in INTERNATIONAL LAW: TEACHING AND PRACTICE 237, 249 (Bin Cheng ed., 1982). 9. See id. 10. See, e.g., Roberts, supra note 8, at 758-59 (citing, inter alia, North Sea Continental Shelf (Ger. v. Den.; Ger.v. Neth.), Judgment, 1969 I.C.J. 3, 44 (Feb. 20); Right of Passage Over Indian Territory (Port v. India), Judgment, 1960 I.C.J. 6, 42-43 (Apr. 12); Asylum (Colom. v. Peru), Judgment, 1950 I.C.J. 266, 276-77 (Nov. 20); Lotus (Fr. v. Turk.), 1927 P.C.I.J. (ser. A) No. 10, at 28 (Sept. 7); RESTATEMENT (THIRD) OF THE FOREIGN RELATIONS LAW OF THE UNITED STATES § 102(2) (1987)) (“[T]raditional custom results from general and consistent practice followed by states from a sense of legal obligation . . . . [modern custom] approach emphasizes opinio juris rather than state practice because it relies primarily on statements rather than actions.”); see also Eduardo Jimeńez de Arećhaga, Remarks, in CHANGE AND STABILITY IN INTERNATIONAL LAW-MAKING 47, 48-50 (Antonio Cassese & Joseph H. H. Weiler eds., 1988); Jonathan Charney, International Agreements and the Development of International Law, 61 WASH. L REV 971, 990-96 (1986); Cheng, United Nations Resolutions, supra note 8. 11. Roberts, supra note 8, at 758 (“Traditional custom is evolutionary and is identified through an inductive process in which a general custom is derived from specific instances of state practice . . . . By contrast, modern custom is derived by a deductive process that begins with general statements of rules rather than particular instances of practice.” (citations omitted)). 12. See Wouters & Ryngaert, supra note 8, at 111 (“As already noted, State practice in the field of human rights is often inconsistent, especially at a universal level. This has prompted the courts and the doctrine to emphasize opinio juris over State practice, and verbal over physical State practice. In so doing, they may believe that they still apply the classical positivist paradigm.”).

450

[Vol. 45


tribunals on statements of the U.N. General Assembly declaring customary international law.13 Despite the criticisms of Frederic Kirgis’ sliding scale analysis, it is still one of the better attempts to understand the fluctuating relationship between approaches emphasizing state practice over opinio juris and vice versa.14 To all of these observers, there is now a split in methodology used to find customary international law. The discussion over the relative value of the inductive and deductive approaches has largely reached a standstill with scholars arguing over the value and appropriateness of the perceived change in method.15 Many allege that the choice of methodology results in differing outcomes, and that the choice of methodology might be deliberately made to produce desired results,16 an argument with considerable merit. Koskenniemi argues that favoring the inductive process naturally favors

13. See Jonathan I. Charney, Universal International Law, 87 AM. J. INT’L L. 529, 544-45 (1993); Jimeńez de Arećhaga, supra note 10, at 48-50; Michael Akehurst, Custom as a Source of International Law, 47 BRIT. Y.B. INT’L L. 1, 6-7 (1975); see also Roberts, supra note 8, at 758 (citing, inter alia, Declaration on Principles of International Law Concerning Friendly Relations and Co-operation Among States in Accordance with the Charter of the United Nations, GA Res. 2625, U.N.G.A.O.R. 25th Sess., Supp. No. 28, at 121, U.N. Doc. A/8028 (1970); Final Act of the Conference on Security and Co-operation in, Aug. 1, 1975, 14 I.L.M. 1292 (1975)) (“The Court paid lip service [in the Nicaragua case] to the traditional test for custom but derived customs of non-use of force and nonintervention from statements such as General Assembly resolutions.”). 14. See Frederic L. Kirgis, Jr., Custom on a Sliding Scale, 81 AM. J. INT’L L. 146 (1987). Although Kirgis articulated this view most clearly, other authors prior to and following him have either ¨ suggested it or endorsed it. See, e.g., REIN MULLERSON , ORDERING ANARCHY: INTERNATIONAL LAW IN INTERNATIONAL SOCIETY 229 (2000); LOUIS HENKIN, INTERNATIONAL LAW: POLITICS AND VALUES 31-32 (1995); G. M. DANILENKO, LAW-MAKING IN THE INTERNATIONAL COMMUNITY 107 (1993); Maurice Mendelson, The Formation of Customary International Law, 272 REC. DES COURS HAGUE ACAD. INT’L L. 155 (1998); John Tasioulas, In Defence of Relative Normativity: Communitarian Values and the Nicaragua Case, 16 OX. J. LEG. STUD. 85 (1996); Int’l L. Ass’n, Final Report of the Committee on Formation of Customary (General) International Law: Statement of Principles Applicable to the Formation of General Customary International Law, princ. 19 cmt. (a) (2000); Oscar Schachter, Entangling Treaty and Custom, in INTERNATIONAL LAW IN A TIME OF PERPLEXITY: ESSAYS IN HONOUR OF SHABTAI ROSENNE 731 (Y. Dinstein ed., 1989). For an attempt to give Kirgis’ paradigm more theoretical backing, see generally Roberts, supra note 8. 15. See, e.g., David Fidler, Challenging the Classical Concept of Custom, 1996 GERM. Y.B. INT’L L. 198, 216-31 (1996) (arguing that traditional customary international law is outdated in modern society); Charney, supra note 13, at 543 (arguing the same); see also Richard B. Lillich, The Growing Importance of Customary International Human Rights Law, 25 GA. J. INT’L & COMP. L. 1, 8 (1995) (presenting a similar argument but focusing on human rights). 16. See Wouters & Ryngaert, supra note 8, at 111-31; Martti Koskenniemi, The Pull of the Mainstream, 88 MICH. L. REV. 1946, 1953 (1990) (critiquing the “modern/human rights” approach of emphasizing opinio juris evidenced through statements as being pre-determined by an “anterior— though in some respects largely shared— criterion of what is right and good for human life”).

2014]

451


description over normativity and thus produces an apology for the exercise of state power; but favoring the deductive process does the reverse and naturally produces idealized, utopian laws detached from reality.17 Kennedy makes the point similarly, submitting that this argument is an endless, and apparently pointless, discussion of the relationship between international law and international politics.18 In fact, many authorities have gone so far as to argue that “modern” customary international law is so radical that it is no longer even customary international law.19 It might simply be the application of general principles of law using the language of custom, a wolf in sheep’s clothing as it were, although this implies that being a general principle of law is a negative thing being snuck past unaware guardians of the law.20 Even more radical is the thought that “modern” customary

17. KOSKENNIEMI, FROM APOLOGY TO UTOPIA, supra note 1, at 41-42; Roberts, supra note 8. See also KOSKENNIEMI, FROM APOLOGY TO UTOPIA, supra note 1, at 2; J. Patrick Kelly, The Twilight of Customary International Law, 40 VA. J. INT’L L. 451 (2000) (describing custom as a “matter of taste.”); Reisman, supra note 1, at 135 (focusing on the opportunities for abuse of modern customary international law). 18. David Kennedy, When Renewal Repeats: Thinking Against the Box, 32 N.Y.U. J. INT’L L. & POL. 335, 355, 407 (2000). 19. See Robert Y. Jennings, The Identification of International Law, in INTERNATIONAL LAW, supra note 8, at 3 (“[M]ost of what we perversely persist in calling customary international law is not only not customary law: it does not even faintly resemble a customary law.”). See also North Sea Continental Shelf (Ger. v. Den.; Ger.v. Neth.), 1969 I.C.J. 3, 44 (Feb. 20) (dissenting opinion of Judge Sørensen); Kelly, Twilight, supra note 17, at 454 n.20, 484; Curtis A. Bradley & Jack L. Goldsmith, Customary International Law as Federal Common Law: A Critique of the Modern Position, 110 HARV. L. REV. 815, 838 (1997); Hiram Chodosh, Neither Treaty nor Custom: The Emergence of Declarative International Law, 26 TEX. INT’L L.J. 87, 102 n.70 (1991); Hilary C. M. Charlesworth, Customary International Law and the Nicaragua Case, 11 AUSTL. Y.B. INT’L L. 1 (1984); Cheng, supra note 8. 20. See Simma & Alston, supra note 8, at 88, 96, 102-06 (arguing that the modern approach is in actuality the identification and application of general principles of law). Indeed there is some intermingling of general principles and customary international law. See, e.g., Prosecutor v. Furundzˇjia, Case No. IT-95-17/1-T, Judgment, ¶¶ 168-82 (Int’l Crim. Trib. Former Yugoslavia Dec. 10, 1998). In the Furundzˇjia trial judgment, the tribunal, on the one hand, held that rape in armed conflict was established as a crime under customary international law, and, on the other hand, held that the definition of the elements of rape could be found through the application of general principles of law. However, the methodologies the tribunal applied for finding the content from the different sources were quite similar: customary international law was proved by examining the provisions of the Lieber Code, Hague Regulations, Control Council Law No. 10, the Charter of the International Military Tribunal for the Far East, and the convictions of Toyoda, Matsui, Hirota and Yamashita; the relevant general principles of law were proved by examining provisions of various domestic criminal codes and case law, as well as the U.S. submissions at the U.N. Diplomatic Conference of Plenipotentiaries on the Establishment of an International Criminal Court.

452

[Vol. 45


international law might be the dawn of a new source of international law.21 Schwarzenberger has written the most detailed and emphatic works advocating for the use of the inductive approach to international legal analysis.22 Citing remarkably little practice himself in support of such an approach,23 and instead relying on deductive techniques concerning the value of induction, he concluded that induction reduced the risk of the subjective role of the assessor (and the assessor’s views on lex ferenda) and placed customary international law on a more firm footing.24 He did concede that absolute objectivity was not possible in any legal analysis, and he admittedly only sought to achieve the maximum objectivity possible.25 The principal target of his complaint was what he termed the “eclectic” school.26 His critique of the “eclectic” method was that it selected evidence that prejudiced the outcome (i.e., anything can be proved)27 yet he did not offer a sampling methodology

21. See Daniel Bodansky, Customary (and Not So Customary) International Environmental Law, 3 IND. J. GLOBAL LEG. STUD. 105, 116 –19 (1995); Charney, supra note 13, at 543, 546-47; Chodosh, supra note 19, at 102. 22. See generally GEORG SCHWARZENBERGER, THE INDUCTIVE APPROACH TO INTERNATIONAL LAW (1965); Georg Schwarzenberger, The Inductive Approach to International Law, 60 HARV. L. REV. 539 (1947). 23. See, e.g., In re Piracy Jure Gentium, [1934] A.C. 586, 588 (Viscount Sankey, L.C.) (“It is a process of inductive reasoning.”); 1 CHARLES CHENEY HYDE, INTERNATIONAL LAW CHIEFLY AS INTERPRETED AND APPLIED BY THE UNITED STATES 13 n.12 (2d ed. 1945); ARNOLD DUNCAN MCNAIR, THE LAW OF TREATIES, at viii (1938); 1 H.A. SMITH, GREAT BRITAIN AND THE LAW OF NATIONS (1932); 2 H.A. SMITH, GREAT BRITAIN AND THE LAW OF NATIONS (1935). Schwarzenberger also positively acknowledges the Harvard Research Drafts as exemplars of the inductive method. 24. See SCHWARZENBERGER, supra note 22, at 4; see also id. at 47, 51, 65-66, 71. It might have been an interesting exercise for Schwarzenberger to attempt to show that the inductive method was inductively verifiable as more objective. Also see criticism of the objective/subjective charge in Wouters & Ryngaert, supra note 8. Schwarzenberger’s concern should probably be placed in the larger trend of the international legal system to move away from auto-interpretation to increased judicial interpretation. See, e.g., Thomas Skouteris, The New Tribunalism: Strategies of (De)Legitimation in the Era of Adjudication, 17 FINNISH Y.B. INT’L L. 307 (2006). 25. See SCHWARZENBERGER, supra note 22, at 56, 65. In particular, he argues that lawyers do a better job than diplomats in being objective about the law. See id. at 24, 28. However, lawyers are also expressly charged with vigorously advocating for their clients, a position not so dissimilar to diplomats. 26. See SCHWARZENBERGER, supra note 22, at 564 (“It is suggested, therefore, that the most that can be done at this stage is a systematic analysis, country by country, of the attitudes of the subjects of international law to the rules of international customary law.”). This is a plea that has only partly been answered, many states understandably prioritizing other matters in their limited foreign ministry budgets. 27. See id. at 13-19.

2014]

453


in substitution that would ensure objective selection of evidence.28 In fact, he appears to denigrate randomized selection as arbitrary.29 More urgently, he felt, international law needed more extensive reporting on judicial decisions that could evidence state practice more completely,30 a project which has, now, partly been achieved.31 At the time he wrote, Schwarzenberger acknowledged that the application of the inductive method was only at a beginning, yet, in a sense, Schwarzenberger won the argument. The impact of the inductive movement continues in an increased drive to apply empirical research techniques to the study of law, not only the study of the sociological causes and impacts of the law, but also the study of what the law is.32 However, that is not the whole story. Of course, if Schwarzenberger was writing in the 1940s to 1960s and arguing against what he saw as the bankruptcy of contemporary international legal analysis, which in his view was deductive, his assessment then suggests that the “traditional” method is probably better characterized as deductive and the “modern” method better seen as inductive. Strangely, this situation is just the reverse of the view of most contemporary scholars. Yet if we compare an older case such as the Wimbledon case with a contemporary one such as the Tadic´ case, we see that the court in the Wimbledon case appeared to apply a much more “modern” approach33 than the tribunal in Tadic´.34 The “modern” approach is not so

28. See id. 29. See id. 30. See id. at 565 (“Only in this way will it be possible to see whether, and on what questions, general consensus exists on the part of municipal courts on questions of international customary law. Provided that this work has been done, annual reports may fulfill a valuable supplementary function. It should, however, be gratefully acknowledged that, without the experiment of the Annual Digest, it would have been much more difficult to become so acutely aware of what appears to be the real— or at least the primary—need.”). 31. In the time since Schwarzenberger wrote his article and book, the legal literature has witnessed the growth of reporting mechanisms from the International Law Reports and International Law in Domestic Courts services, to the profusion of national yearbooks of international law, to websites of supreme courts with case law databases (sometimes even in English). 32. See, e.g., Shaffer & Ginsburg, supra note 2. 33. See S.S. “Wimbledon” (U.K., Fr., It. & Japan v. Ger.), 1923 P.C.I.J. (ser. A) No. 1, at 25 (Aug. 17) (finding customary international law on international waterways by relying only on two examples of practice, the Panama and Suez canals, suggesting that the conclusion reached was being reached deductively from the principles being applied in those cases). 34. See Prosecutor v. Tadic´, Case No. IT-94-1-l, Decision on Defence Motion for Interlocutory Appeal on Jurisdiction, ¶ 99 (Int’l Crim. Trib. for the Former Yugoslavia, October 2, 1995) (finding that customary international law must be discovered through a study of many examples of various states’ practice, suggesting that conclusions must be reached inductively from samples).

454

[Vol. 45


modern after all. The words “traditional” and “modern” are, after all, just descriptive words, so perhaps we do not mind how they are used just so long as we can all agree that they describe the same thing. The problem is that the use of the terms suggests a chronological relationship and thus implies an oppositional change in philosophy. It is this implication that the author of this Article finds problematic since, in his view, “traditional” and “modern” legal analysis have always been applied side-by-side and, importantly, in an intermixed manner. This Article will attempt to take a step back and question whether the two approaches, induction and deduction, can be separated so clearly in international legal analysis, or whether they are intermixed, and have always been so. This author believes that these two competing processes can best be understood as a dialectical process of opposing forces repeatedly and alternatively demanding a synthesis of views, i.e., a hybrid analysis. This dialectical process does not, as some might argue, render international law useless, but expresses a process of building on each other through a competition of opposites. These critiques of international law are, of course, not limited to international law specifically, but could apply to law in general.35 The analysis of customary international law is thus a basket of inductive and deductive conclusions answering a variety of questions. The divergent approaches, such as they are, are collections of certain answers to those questions. In this sense, courts and tribunals apply, and likely have always applied, both the “traditional” and “modern” methods simultaneously. III.

DEDUCTION, INDUCTION, AND THE SCIENTIFIC METHOD A.

Deduction

As mentioned in the Introduction, deductive reasoning begins with premises, which, if true, must lead to a true conclusion. In the case of deductive legal reasoning, the process begins with general legal postulates and the specific norms are derived from them.36 Importantly,

35. See Roberts, supra note 8. Anthea Roberts has argued that these two theories of custom encapsulate Hart’s distinction between descriptive and prescriptive statements of law. See H.L.A. HART, THE CONCEPT OF LAW 183 (1961). See also RONALD DWORKIN, LAW’S EMPIRE 271 (1986); JOHN RAWLS, A THEORY OF JUSTICE 20 (1972). 36. See Kammerhofer, supra note 6, at 542. (“In any case, the origin of the normative order is assumed and a system is deduced from it.”); KELSEN, supra note 7, at 4, 114). However, that analysis is not entirely correct. It draws an arbitrary distinction between legal and extra-legal considerations that is not appropriate for deductive legal reasoning which, by its nature, blurs the lines

2014]

455


deduction does not rely on evidence of instances to prove a conclusion, though it does not suffer the weakness of induction in that any conclusion can be falsified by contradictory evidence. Perhaps it should go without saying, but deduction can result in correct outcomes. The lottery paradox exemplifies this reality where repeatedly buying losing lottery tickets does not logically mean that no person has won the lottery.37 In such an instance, the deduction that there is a winner, even where the winner is not observed, provides the correct conclusion. B.

Induction

Because induction seems to be the preferred theory in contemporary international law, following Schwarzenberger, we will omit a lengthy discussion of deduction and examine induction in far greater detail. Focusing more on induction is necessary here to better analyze the purported practice of the inductive method in customary international law analysis to determine whether indeed induction is being applied. Again, as noted in the Introduction, induction is often described as drawing inferences from specific observable phenomena to general rules. Those inferences cannot be absolutely true, as in the case of deduction, but are only provisionally correct to a certain degree of confidence. The degree to which the conclusion is probably true is based on the quality of the evidence used to support it. Arguments with significant evidence are said to be strong and those without it are said to be weak.38 Thus, no conclusion from induction is certain, instead it is probable to a degree of certainty. For this reason, induction in science must also be done publicly so that the newly discovered knowledge can be tested.39 Strangely, inductive reasoning is quite

between law and non-law. We could just as easily argue that the deductive reasoning process uses extra-legal premises to reach non-legal conclusions, or that it reaches legal conclusions from what must necessarily be legal (or “legalized”) premises. In addition, the general legal premises need not be “assumed” in the sense of being arbitrary or self-serving. Rather certain legal premises might have an extensive pedigree of application and be based on the most noble of aspirations, though their origin remains unclear. 37. See generally ERNEST ADAM, THE LOGIC OF CONDITIONALS (1975); Ernest Adam, A Logic of Conditionals, 8 INQUIRY 166 (1965). 38. See J.H. HOLLAND ET AL., INDUCTION: PROCESSES OF INFERENCE, LEARNING, AND DISCOVERY (1989); John Vickers, The Problem of Induction § 2, in STANFORD ENCYCLOPEDIA OF PHILOSOPHY (rev. ed. Jun. 21, 2010), available at http://plato.stanford.edu/entries/induction-problem/; Karl R. Popper & David W. Miller, A Proof of the Impossibility of Inductive Probability, 302 NATURE 687-88 (1983). 39. See GARY KING, ET AL., DESIGNING SOCIAL INQUIRY: SCIENTIFIC INFERENCE IN QUALITATIVE RESEARCH 7-9 (1994) (arguing that scientific research has four characteristics, one of which is that

456

[Vol. 45


difficult to justify through deduction or induction itself.40 Although we can establish that induction as a technique is potentially reaching true conclusions, we cannot be certain that it works as a method of reasoning without the deductive step that observable phenomena can produce truth to a certain degree of confidence about a situation. The step of inductively generalizing the acts of a given sample to the population as a whole is based on a deductive premise that what is observed in one place is likely to always be the case.41 Situations in which the generalization step is weak are often termed “hasty” generalizations.42 Following the observation, the actor engages in a statistical syllogism where it concludes that if a certain quality exists in a sample, then it likely exists in the entire population under examination.43 Difficulties in this step include the “accident.”44 As befitting a science, the proposed rule must always be open to negation by subsequent

“[s]cientific research uses explicit, codified, and public methods to generate and analyze data whose reliability can therefore be assessed . . . . If the method and logic of a researcher’s observations and inferences are left implicit, the scholarly community has no way of judging the validity of what was done”). See also ROBERT K. MERTON, SOCIAL THEORY AND SOCIAL STRUCTURE 71-72 (1949) (“The sociological analysis of qualitative data often resides in a private world of penetrating but unfathomable insights and ineffable understandings . . . . [However,] science . . . is public, not private.”) 40. See, e.g., 1 DAVID HUME, A TREATISE OF HUMAN NATURE Pt. III, 16 (1739); id. at Pt. III, 89; 2 DAVID HUME, A TREATISE OF HUMAN NATURE Pt. I, 12; id. at Pt. II, 12; Samir Okasha, What Did Hume Really Show About Induction?, 51 PHIL. Q. 307-27 (2001). For responses to the skepticism of Hume and others, see Wesley Salmon, On Vindicating Induction, in INDUCTION: SOME CURRENT ISSUES (Henry Kyburg & Ernest Nagel eds., 1963); Max Black, The Justification of Induction, in LANGUAGE AND PHILOSOPHY: STUDIES IN METHOD (Max Black ed., 1949); Paul Edwards, Russell’s Doubts about Induction, 58 MIND 141 (1949); Hans Reichenbach, On the Justification of Induction, in READINGS IN PHILOSOPHICAL ANALYSIS (H. Feigel & W. Sellars eds., 1949). 41. In turn, this conclusion can be descriptive or causal, see KING, ET AL., supra note 39, at 8 (“[O]ur particular definition of science requires the additional step of attempting to infer beyond the immediate data to something broader that is not directly observed. That something may involve descriptive inference— using observations from the world to learn about other unobserved facts. Or that something may involve causal inference—learning about causal effects from the data observed.”). 42. See KEITH HOLYOAK & ROBERT MORRISON, THE CAMBRIDGE HANDBOOK OF THINKING AND REASONING (2005); Vickers, supra note 38. 43. See KING, ET AL., supra note 39, at 8 (“[O]ur particular definition of science requires the additional step of attempting to infer beyond the immediate data to something broader that is not directly observed. That something may involve descriptive inference— using observations from the world to learn about other unobserved facts. Or that something may involve causal inference— learning about causal effects from the data observed.”). 44. See HOLYOAK & MORRISON, supra note 42; HOLLAND ET AL., supra note 38; Vickers, supra note 38.

2014]

457


evidence proving the contrary.45 Since induction is non-monotonic, it is always subject to falsification by a new premise and as such it is conditioned by the principle of total evidence, i.e., that all relevant evidence must be considered.46 The probability of the validity of the inductive conclusion is then, in turn, an entirely different question.47 The challenge is to determine what probability is sufficient to support the inference.48 In order to determine whether the hypothesis should be changed in light of new evidence is usually governed by Bayesian logic.49 Bayes’ theorem of

45. See Popper & Miller, supra note 38. 46. See KARL R. POPPER, THE LOGIC OF SCIENTIFIC DISCOVERY (Karl R. Popper, Julius Freed & Jan Freed trans., 1959). 47. See generally IAN HACKING, AN INTRODUCTION TO PROBABILITY AND INDUCTIVE LOGIC (2001); ALFRED RENYI, FOUNDATIONS OF PROBABILITY (1970); A.N. KOLMOGOROV, FOUNDATIONS OF THE THEORY OF PROBABILITY (2d ed. 1956); RUDOLF CARNAP, LOGICAL FOUNDATIONS OF PROBABILITY (1950); HAROLD JEFFREYS, THEORY OF PROBABILITY (1939); JOHN MAYNARD KEYNES, A TREATISE ON PROBABILITY (1921); James M. Joyce, A Nonpragmatic Vindication of Probabilism, 65 PHIL. SCI. 575 (1998). 48. See KING, ET AL., supra note 39, at 9 (“[U]ncertainty is a central aspect of all research and all knowledge about the world. Without a reasonable estimate of uncertainty, a description of the real world or an inference about a causal effect in the real world is uninterpretable. A researcher who fails to face the issue of uncertainty directly is either asserting that he or she knows everything perfectly or that he or she has no idea how certain or uncertain the results are.”); see also RICHARD JEFFREY, SUBJECTIVE PROBABILITY: THE REAL THING (2004); HACKING, supra note 47; DONALD GILLIES, PHILOSOPHICAL THEORIES OF PROBABILITY (2000); RICHARD JEFFREY, PROBABILITY AND THE ART OF JUDGMENT (1992); GEORGE N.G SCHLESINGER, THE SWEEP OF PROBABILITY (1991); PAUL HORWICH, PROBABILITY AND EVIDENCE (1982); IAN HACKING, THE EMERGENCE OF PROBABILITY: A PHILOSOPHICAL STUDY (1975); A.W.F. EDWARDS, LIKELIHOOD (1972); RENYI, supra note 47; KOLMOGOROV, supra note 47; CARNAP, supra note 47; JEFFREYS, supra note 47; KEYNES, supra note 47; Bernard Koopman, The Bases of Probability, 46 BULLETIN AM. MATH. SOC’Y 763 (1940), reprinted in STUDIES IN SUBJECTIVE PROBABILITY (Henry Kyburg, Jr. & Howard Smokler eds., 2d ed. 1980); David Lewis, A Subjectivist’s Guide to Objective Chance, in 2 STUDIES IN INDUCTIVE LOGIC AND PROBABILITY 263 (Richard Jeffrey ed., 1980). 49. See generally COLIN HOWSON & PETER URBACH, SCIENTIFIC REASONING: THE BAYESIAN APPROACH (3d ed. 2005); LUC BOVENS & STEPHAN HARTMANN, BAYESIAN EPISTEMOLOGY (2003); JOHN EARMAN, BAYES OR BUST? A CRITICAL EXAMINATION OF BAYESIAN CONFIRMATION THEORY (1992); LEONARD SAVAGE, THE FOUNDATIONS OF STATISTICS (1954); David Dowe, Steve Gardner & Graham Oppy, Bayes, Not Bust: Why Simplicity is No Problem for Bayesians, 58 BRIT. J. FOR THE PHIL. OF SCI. 709 (2007); Branden Fitelson, Likelihoodism, Bayesianism, and Relational Confirmation, 156 SYNTHESE 473 (2007); James M. Joyce, Bayesianism, in THE OXFORD HANDBOOK OF RATIONALITY (Alfred Mele & Piers Rawling eds., 2004); Michael Strevens, Bayesian Confirmation Theory: Inductive Logic, or Mere Inductive Framework?, 141 SYNTHESE 365-79 (2004); Elliott Sober, Bayesianism—Its Scope and Limits, in BAYES’S THEOREM 21-38 (Richard Swinburne ed., 2002); Colin Howson, Bayesianism in Statistics, in BAYES’S THEOREM, supra note 49, at 39-71; Branden Fitelson, A Bayesian Account of Independent Evidence with Applications, 68 PHIL. OF SCI. 123 (2001); Branden Fitelson, The Plurality of Bayesian Measures of Confirmation and the Problem of Measure Sensitivity, 66 PHIL. OF SCI. 362 (1999); Geoffrey

458

[Vol. 45


probability is the most commonly applied measure for the adequacy of the evidence for proof of a fact.50 Bayesian confirmation theory holds that we need to evaluate opposing hypotheses by comparing their relative probabilities of being correct.51 C.

Types of Induction

There are, however, several types of induction.52 The first type of induction is direct inference by enumeration of cases. In this type, we examine samples from the data pool and determine the proportion that has a certain quality. Then we reach a conclusion on the likelihood of the remainder of the population having the same distribution of qualities as we observed in the sample.53 This type of induction is limited to logical scenarios where either a universal quality or a

Hellman, Bayes and Beyond, 64 PHIL. OF SCI. 191 (1997); Hawthorne, Degree-of-Belief and Degree-ofSupport: Why Bayesians Need Both Notions, 114 MIND 277 (2005); Charles Chihara, Some Problems for Bayesian Confirmation Theory, 38 BRIT. J. PHIL. OF SCI. 551 (1987); W. Edwards, H. Lindman & Leonard Savage, Bayesian Statistical Inference for Psychological Research, 70 PSYCH. REV. 193-242 (1963). 50. See HOWSON & URBACH, SCIENTIFIC REASONING, supra note 49; BOVENS & HARTMANN, supra note 49; EARMAN, BAYES OR BUST, supra note 49; SAVAGE, supra note 49; Dowe, Gardner & Oppy, supra note 49; Fitelson, supra note 49; Hawthorne, supra note 49; J.; Joyce, supra note 49; Howson, supra note 49; Strevens, supra note 49; Sober, supra note 49; Fitelson, Bayesian Account, supra note 49; Fitelson, Plurality, supra note 49; Hellman, Bayes, supra note 49; Chihara, supra note 49; Edwards, Lindman & Savage, supra note 49. 51. See R.D. Rosenkrantz, Does the Philosophy of Induction Rest on a Mistake?, 79 J. PHIL. 78, 78-97 (1982); Paul Teller, Goodman’s Theory of Projection, 20 BRIT. J. FOR THE PHIL. OF SCI. 219, 219-38 (1969); Haskell Fain, The Very Thought of Grue, 76 PHIL. REV. 61, 61 (1967). 52. Note that the descriptive classification of the types of induction is not uncontroversial in itself. This author generally follows Carnap for the classification. See CARNAP, supra note 47, at 207. 53. See generally HACKING, supra note 48; NELSON GOODMAN, FACT, FICTION, AND FORECAST (4th ed. 1983); CLARK GLYMOUR, THEORY AND EVIDENCE (1980); RUDOLF CARNAP, THE CONTINUUM OF INDUCTIVE METHODS (1952); Jon Williamson, Inductive Influence, 58 BRIT. J. FOR THE PHIL. OF SCI. 689 (2007); Henry E. Kyburg, Jr., Belief, Evidence, and Conditioning, 73 PHIL. SCI. 42 (2006); Patrick Maher, A Conception of Inductive Logic, 73 PHI. SCI. 513 (2006); Sherrilyn Roush, Problem of Induction, in THE PHILOSOPHY OF SCIENCE: AN ENCYCLOPEDIA (Jennifer Pfeifer & Sahotra Sarkar eds., 2005); Sherrilyn Roush, Positive Relevance Defended, 71 PHIL. SCI. 110 (2004); Norton, A Material Theory of Induction, 70 PHI. SCI. 647 (2003); Jonathan Hawthorne & Luc Bovens, The Preface, the Lottery, and the Logic of Belief, 108 MIND 241 (1999); Patrick Maher, Inductive Logic and the Ravens Paradox, 66 PHI. SCI. 50 (1999); Patrick Maher, Depragmatized Dutch Book Arguments, 64 PHIL. SCI. 291-305 (1997); Patrick Maher, Subjective and Objective Confirmation, 63 PHIL. SCI. 149-174 (1996); Isaac Levi, Confirmational Conditionalization, 75 J. PHIL. 730-7 (1978); T. Seidenfeld, Direct Inference and Inverse Inference, 75 J. PHIL. 709-30 (1978); Henry E. Kyburg, Jr., Randomness and the Right Reference Class, 74 J. PHIL. 501 (1977); Isaac Levi, Direct Inference, 74 J. PHIL. 5 (1977); Patrick Maher, The Concept of Inductive Probability, 65 ERKENNTNIS 185-206 (1976); Rudolf Carnap, Replies and Systematic Exposi-

2014]

459


proportion of a quality in a sample can be discovered, and all samples will either have the quality or not.54 The second type of induction is predictive inference. In this type, the investigator uses the knowledge gained from the sample, and the inferences inductively drawn from the sample, to make a prediction that a subsequent sample—a specific case—will reflect the same universal characteristic or will have a probability of having a less than universal characteristic. This is the singular predictive inference.55 The third type of induction is inference by analogy. This type resembles the first two but uses certain commonly occurring characteristics to make, supposedly, better predictions. In this case, the investigator uses secondary characteristics that appear to coincide with the qualities in question, to make a prediction that the quality in question exists in a new case where the secondary characteristics can be found.56 The fourth and fifth types of induction reach conclusions about a population as a whole, not individual cases. The fourth type is inverse inference. In this method, the investigator reaches a prediction about a population, not just a singular case, after examining a sample of that population.57 The fifth type is universal inference, or hypothetical reasoning, where the investigator uses the sample of the population to infer the existence of a general rule that applies to the entire population. Various specific events are examined to determine the underlying commonality in all of them. The investigator supposes an explanation that could account for the various phenomena and then examines each

tions, in THE PHILOSOPHY OF RUDOLF CARNAP (P.A. Schilpp ed.,1963); Carl Hempel, Studies in the Logic of Confirmation, 54 MIND 1-26, 97 (1945). 54. See GLYMOUR, supra note 53; COLIN HOWSON, HUME’S PROBLEM: INDUCTION AND THE JUSTIFICATION OF BELIEF (2000); PATRICK MAHER, BETTING ON THEORIES (1993); GOODMAN, supra note 53; HACKING, supra note 48; PAUL HORWICH, PROBABILITY AND EVIDENCE (1982); R.D. ROSENKRANTZ, FOUNDATIONS AND APPLICATIONS OF INDUCTIVE PROBABILITY (1981); ISAAC LEVI, THE ENTERPRISE OF KNOWLEDGE (1980); HENRY E. KYBURG, JR., THE LOGICAL FOUNDATIONS OF STATISTICAL INFERENCE (1974); ISAAC LEVI, GAMBLING WITH TRUTH (1967); HANS REICHENBACH, EXPERIENCE AND PREDICTION: AN ANALYSIS OF THE FOUNDATIONS AND THE STRUCTURE OF KNOWLEDGE (1938); Carnap, supra note 53; Hawthorne & Bovens, supra note 53; Hempel, supra note 53; Kyburg, Belief, supra note 53; Levi, Direct Inference, supra note 53; Maher, Subjective and Objective, supra note 53; Maher, Depragmatized, supra note 54; Patrick Maher, Ravens Paradox, supra note 53; Maher, Concept, supra note 53; Maher, Conception, supra note 53; Norton, supra note 53; Roush, Positive Relevance, supra note 53; Roush, Problem of Induction, supra note 53; Seidenfeld, supra note 53; Williamson, supra note 53. 55. See generally CARNAP, supra note 53; CARNAP, supra note 47, at 207. 56. See HOWSON, supra note 54; CARNAP, supra note 53; CARNAP, supra note 47; HUME supra note 40, at I.III.16, I.I.12, II.II.12; Okasha, supra note 40. 57. See CARNAP, supra note 53; CARNAP, supra note 47; Seidenfeld, supra note 53.

460

[Vol. 45


phenomenon to determine if it supports or refutes the hypothesis. If the commonality occurs in all of them, then we can infer that there is a rule of general application.58 D.

The Hypothesis

The first step in a verifiable scientific inquiry is forming a hypothesis. It has been argued that hypothetical reasoning is an inductive universal inference.59 However a hypothesis might be based not on evidence, but rather an intelligent, though not infallible, estimation of the content of the rule, open to invalidation. Thus, the precise type of inferential step in forming the hypothesis is not completely clear. In addition, there is no reason why a hypothesis could not be reached through deduction. The construction of the hypothesis is, however, a loaded step. The hypothesis already frames the question and thus the direction of the inquiry.60 It tends to construct a risk of confirmation bias, i.e., the bias in favor of confirmation rather than refutation of hypotheses.61 Once the hypothesis is articulated, the scientific inquiry proceeds to testing the hypothesis to confirm or refute it.62 As for means of testing the hypothesis, the assessor could rely on ethnography, i.e., an observation or interviewing technique for assessing practice.63 Courts do not usually use this technique, although in domestic jurisdictions experts

58. See CARNAP, supra note 53; CARNAP, supra note 47; Ellery Eells & Branden Fitelson, Measuring Confirmation and Evidence, 97 J. OF PHIL. 663 (2000); David Christensen, Measuring Confirmation, 96 J. PHIL. 437 (1999); Kyburg, Belief, supra note 53, at 42-65; Ellery Eells, Problems of Old Evidence, 66 PAC. PHIL. Q. 283 (1985). 59. Sometimes hypotheses are explained as a method of retroduction, sometimes called abduction, following Peirce. In this form of reasoning, the actor explains an outcome by inferring the existence of a mechanism that could create the observed outcome. This process then sets up a post hoc ergo propter hoc problem, which can be tested inductively or deductively. See Charles Sanders Peirce, Prolegomena to an Apology for Pragmatism, 4 NEW ELEMENTS OF MATH 319 (1906); Charles Sanders Peirce, Deduction, Induction, and Hypothesis, 13 POP. SCI. MONTHLY, 470, 472 (1878). See also ANDREW SAYER, METHOD IN SOCIAL SCIENCE: A REALIST APPROACH 107 (2d ed. 1992); Jonathan Koehler, The Base Rate Fallacy Reconsidered: Descriptive, Normative and Methodological Challenges, 19 BEHAV. & BRAIN SCI. 1 (1996). 60. See HOLLAND ET AL., supra note 38; HOLYOAK & MORRISON, supra note 42. 61. See id. 62. See ANTHONY GIDDENS ET AL., ESSENTIALS OF SOCIOLOGY 27 (2010) (describing the three main methods used in sociological research: ethnography, survey and experiment); id. at 24-27 (describing and defining the seven steps of the research process: (1) define the research problem; (2) review the evidence; (3) make the problem precise, i.e., form a hypothesis; (4) work out a design; (5) carry out the research; (6) interpret the results; (7) report the findings). 63. See id. at 28 (describing ethnography as a research method involving public participant observation or interviewing).

2014]

461


are sometimes consulted for evidence on foreign law. Of course, in this view the question of the foreign law is a question of fact. In any event, ethnography runs the risk of not being objective,64 and is also difficult to generalize to the population at large.65 The second way to test a hypothesis is an experiment. The dangers with an experiment can be its very artificial nature and inability to account for all variables at times.66 Thus, the investigator of customary international law is usually left with the survey as the most viable option. The third option for testing a hypothesis is a sampling survey, i.e., acquiring samples of data from a pool67 and then generalizing the findings to the population as a whole.68 In customary international law analysis, this method appears to be the most commonly applied so we will concentrate on this method. In order to produce a strong induction, the sampling of data must be random, but there are various meanings of “random.”69 The sampling technique might be a simple, random sampling (each sample is chosen randomly so that each sample has an equal probability of being selected), stratified sampling (each sample is chosen randomly but from constructed sub-groups of the overall population), cluster sampling (each sample is one of a group of samples taken at the same time from the same population), and so on.70 Only after determining the hypothesis and the most appropriate

64. See id. (critiquing the view that embedded ethnographic study could be truly objective). 65. See id. (observing the limitations of ethnography being the weak viability when studying large or diverse groups, the difficulty of identification with the subjects of study, and problematic generalizations to other communities). 66. See, e.g., Richard H. Franke & James D. Kaul, The Hawthorne Experiments: First Statistical Interpretation, 43 AM. SOC. REV. 623 (1978)); William Thomas Worster, Between a Treaty and Not: A Case Study of the Legal Value of Diplomatic Assurances in Expulsion Cases, 21 MINN. J. INT’L L. 253, 312 n.191 (2012) (citing Lyle Yorks & David A. Whitsett, Hawthorne, Topeka, and the Issue of Science Versus Advocacy in Organizational Behavior, 10 ACAD. MGMT. REV. 21 (1985) (discussing the “observer effect,” i.e., the effect of the external researcher observing the data on the content of the data); see also Int’l L. Ass’n, Statement of Principles, supra note 14, at 5 (“In studying the customary process, it is necessary to be aware of the issue of the observational standpoint. By this reference is not mainly made to the (obvious) need to identify for oneself and for others one’s own assumptions and goals. Rather, the suggestion is that different functions may lead the persons performing them to adopt a somewhat different attitude to the sources . . . .” (footnote omitted)). 67. See GIDDENS, ET AL., supra note 62, at 28 (describing the research method of survey as seeking answers from randomly selected members of a population to structured questionnaires). 68. See id. (observing that results from surveys are generalizable). 69. See CHARLES SANDERS PEIRCE, PHILOSOPHICAL WRITINGS OF PEIRCE 152 (Justus Buchler ed., 1955). 70. See THE PRACTICE OF STATISTICS (Daniel S. Yates et al. eds., 3d ed. 2008).

462

[Vol. 45


method for sampling to confirm or refute the hypothesis does the investigator proceed to actually drawing and analyzing samples, i.e., selecting evidence submitted to it that is or is not probative. Representativity in drawing samples is usually understood to be achieved by a randomized selection process71 (even if the technique is stratified or clustered, the drawing of items of data is still randomized within certain parameters). By “random,” we understand the sampling technique to be one where all instances have an equal chance of selection.72 No sample should be chosen because it is perceived to be valuable or representative by the investigator. Most frequently investigators permit computer randomization to select samples to eliminate the risk of a self-selected, or outcome-driven, sampling.73 With the sample pool assembled, the investigator will then apply one of the forms of the inductive method to reach an inference about the universal pool or a particular instance. Rarely does all the data point in a conclusive direction, so the investigator must apply some notion of a degree of certainty about the inference. The investigator must implicitly have in mind a tipping point where the evidence of the rule arising from the sample pool is sufficient. Where that tipping point is in terms of the sufficiency of evidence, however, is inherently a value judgment.74 In arguing that value judgments inform the location of the tipping point, we do not mean to say that we consider desirability of the conclusion but rather factors of the case that make the situation more or less likely to tip. Isaac Levi makes the argument that we can distinguish between moral

71. See GIDDENS, supra note 62, at 28 (noting that sampling must be representative, i.e., typical of the population as a whole, and that the best procedure for attaining representative samples is by selected them at random). 72. See id. at 29-30 (defining random selection of samples as a sampling technique where all members of a population have an equal probability of being selected). 73. See id. (recommending use of computer randomization for selecting samples from a population). 74. See Richard Rudner, The Scientist qua Scientist Makes Value Judgments, 20 PHIL. SCI. 1, 2 (1953) (“[S]ince no hypothesis is ever completely verified, in accepting a hypothesis the scientist must make the decision that the evidence is sufficiently strong or that the probability is sufficiently high to warrant the acceptance of the hypothesis”); see also William Thomas Worster, The Transformation of Quantity into Quality: Critical Mass in the Formation of Customary International Law, 31 B.U. INT’L L.J. 2 (2013) (regarding the notion of a tipping point in customary international law analysis and how our understanding of customary international law might be informed by the study of tipping points in other sciences, especially that of the sociological study of collective decision-making).

2014]

463


values and what he terms “cognitive value[s].”75 For conclusions that have more significant consequences, the actor making the assessment should expect greater volume and probative force of evidence, and may set the standard of proof higher.76 Throughout this process, we often apply lex parsimoniae (sometimes called Ockham’s razor: “Pluralitas non est ponenda sine necessitate”77) or its cousin, inference to the best possible explanation.78 In essence, these rules dictate that, in analyzing the evidence, we should seek the inference that produces the simplest explanation for the data. That being said, simplicity itself—its definition (elegance and parsimony), application in differing fields, and rational justification—is admittedly problematic in itself.79 Nonetheless, with the simplest explanation we can then confirm or refute the hypothesis. Keeping the context of scientific process in mind, we will now turn to examine how induction and deduction are applied in the assessment of customary international law. IV.

THE INTERPLAY OF DEDUCTION AND INDUCTION IN CUSTOMARY INTERNATIONAL LAW

Having discussed deduction and induction generally, and the usual process of scientific inquiry, we can now turn to the way international courts and tribunals analyze customary international law to determine whether and how that method is inductive, deductive or both. We will generally look at how international courts80 are assessing the existence

75. See RICHARD C. JEFFREY, THE LOGIC OF DECISION (2d ed. 1983); LEVI, ENTERPRISE, supra note 54; LEVI, GAMBLING, supra note 54; ISAAC LEVI, HARD CHOICES: DECISION MAKING UNDER UNRESOLVED CONFLICT 43-46 (1986); FRANK PLUMPTON RAMSEY, Truth and Probability, in THE FOUNDATIONS OF MATHEMATICS AND OTHER LOGICAL ESSAYS 156 (R.B. Braithwaite ed., 1931); SAVAGE, supra note 49; LEVI, GAMBLING, supra note 54; Levi, Direct Inference, supra note 53; Levi, Confirmational, supra note 53. 76. See Rudner, supra note 74. 77. See IMMANUEL KANT, THE CRITIQUE OF JUDGMENT 14 (J.H. Bernard trans., 1914); see also 2 THOMAS AQUINAS, BASIC WRITINGS OF SAINT THOMAS AQUINAS 129 (Anton C. Pegis trans., 1945) (“If a thing can be done adequately by means of one, it is superfluous to do it by means of several . . . .”). 78. See PETER LIPTON, INFERENCE TO THE BEST EXPLANATION (2d ed. 2004); Gilbert H. Harman, The Inference to the Best Explanation, 74 PHIL. REV. 88, 88-95 (1965). 79. See Elliott Sober, What is the Problem of Simplicity?, in SIMPLICITY, INFERENCE AND MODELLING: KEEPING IT SOPHISTICATEDLY SIMPLE 13 (Arnold Zellner et al. eds., 2001); Elliott Sober, The Principle of Parsimony, 32 BRIT. J. PHIL. SCI. 145 (1981). 80. For purposes of this Article, international courts will include regional courts such as the European Court of Justice (“ECJ”). This is appropriate because it is not the nature of the court

464

[Vol. 45


of customary international law because the papers of foreign ministries and other sources are usually not so clear about— or omit entirely—the legal reasoning method employed to reach the conclusion. This Article will suffer this risk of not being inclusive and the reader is invited to find this a fault. One overriding difficulty with the analytical method of courts, however, is the vague nature of the analytical method being employed, sometimes only marginally more clear than a foreign ministry. Courts rarely, if ever, explicitly state when they will use the deductive or inductive method.81 Furthermore, domestic courts, as opposed to international courts, are often significantly constrained by constitutional limitations on their ability to receive and/or analyze international law that might prejudice its application of the inductive or deductive methods. Thus, we must reach our own inferences about the method being employed by each body on a case-by-case basis and keep their limitations in mind, but by looking primarily at courts we have a better ability to analyze the logical steps. A. 1.

The Foundations

Custom as a Source of Law

First, courts acknowledge that customary international law is a source of international law.82 There is an initial difficulty here. Courts exam-

assessing the norm that is at issue but rather the assessment process. The ECJ has made its own assessment of the existence of rules of customary international law. Firstly, the European Court of Justice has observed that the European legal order must comply with international law. See Case C-366/10, Air Transp. Ass’n of Am. v. Sec’y of State for Energy & Climate Change, ¶ 101 (Dec. 21, 2011), http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri⫽CELEX:62010CJ0366:EN: HTML; Case C-162/96, Racke v. Hauptzollamt Mainz (1998), E.C.R. I-3655, ¶ 45. In particular, the ECJ has held that international law includes customary international law, see Case Air Transp. Ass’n, ¶¶ 101-02; Racke, at ¶ 46, even though it “does not have the same degree of precision as a provision of an international agreement,” Racke, ¶ 52. Since the ECJ is engaging in customary international law analysis, we can include it in this study. 81. Cf. KING, ET AL., DESIGNING, supra note 39, at 8 (“[M]any [qualitative researchers] proceed as if they had no method—sometimes as if the use of explicit methods would diminish their creativity. Nevertheless they cannot help but use some method.”). Schwarzenberger’s quote to In re Piracy Jure Gentium is a glaring exception and perhaps the reason why he used it. See In re Piracy Jure Gentium, [1934] A. C. 586, 588 (per Viscount Sankey, L. C.) (“It is a process of inductive reasoning.”). 82. See Lotus (Fr. v. Turk.), 1927 P.C.I.J. (ser. A) No. 10 (Sept. 7); North Sea Continental Shelf (Ger. v. Den.; Ger.v. Neth.), Judgment, 1969 I.C.J. 3, 42, 44 (Feb. 20); Continental Shelf (Libya v. Malta), Judgment, 1985 I.C.J. 13, ¶¶ 27-30 (June 3); Jurisdictional Immunities of the State (Ger. v. It.), Judgement, 2012 I.C.J. 99, ¶ 55 (Feb. 3) (citing North Sea Continental Shelf (Ger. v. Den.; Ger.v. Neth.), Judgment, 1969 I.C.J. 3, 44 (Feb. 20)), available at http://www.icj-

2014]

465


ine the text of the Statute of the ICJ and Permanent Court of International Justice (PCIJ), which provide for the application of customary international law by the court. The early findings of the PCIJ, which discovered the elements, e.g., in the Lotus case, based their reasoning on the PCIJ Statute only.83 However, the Statute alone does not make international law. The first conclusion we can draw from the Court’s analysis is that the Statute points the Court to a source of law that does indeed exist. Subsequent practice of courts and states demonstrates that the provision in the Statute has been accepted as a correct statement of the law. In this author’s view, this conclusion is correct, though it is a deductive step. The Court does not examine practice to reach the conclusion, instead it applies reason based on certain premises about international law. Therefore, in the final analysis, both traditional and modern customary international law analyses agree that the fundamental basis for custom as a source of law is reached through a deductive process. At this point, we should observe that it could be problematic to rely on state practice itself to analyze the rules on customary international law, although perhaps the practice of the Court is a different consideration. The International Law Association (ILA), in its principles on the formation of customary international law, suggested that looking to the practice of states—an inductive approach—to determine the rules on the formation of general customary international law was not a problem.84 The ILA committee thought that this was not circular reasoning based on Hart’s rule of recognition, i.e., the rules of any legal system are based, primarily, on what the actors in the system recognize as

cij.org/docket/files/143/16883.pdf; Int’l L. Ass’n, Statement of Principles, supra note 14, ¶ 9; see also SCHWARZENBERGER, supra note 22, at 19. 83. See Lotus, 1927 P.C.I.J. at 28-30 (remarking that identifying practice, without opinio juris, is not customary international law). In paragraphs 77 and 81, the PCIJ did look to examples of case law, but that inquiry was directed at the content of the law, not the deduction of the elements of customary international law. Id. Oddly enough, the STL stated that the Lotus case impliedly criticized the P.C.I.J. for its fully consent-based customary international law analysis. See also Prosecutor v. Ayyash, Case No. STL-11-01/I, Interlocutory Decision on the Applicable Law: Terrorism, Conspiracy, Homicide, Perpetration, Cumulative Charging, ¶ 99 (Spec. Trib. for Lebanon Feb. 16, 2011) (citing Lotus, 1927 P.C.I.J. at 28-30) (“[I]n the celebrated Lotus case . . . [the P.C.I.J.] still took a voluntarist view of custom.”). 84. See INT’L LAW ASS’N, supra note 14, ¶ 6 n.6 (“Essentially, this Statement’s approach is that, in the absence of a proper written constitution, the test of what constitute the “sources” of international law, and of the specific rules about the operation of each of them, depends on what the key participants in the system—States—recognize as the processes by which law is made.”).

466

[Vol. 45


such.85 The problem with the ILA’s conclusion is that state practice is assumed to be the act of recognition of the applicable rules. Kammerhofer cites the supposedly inductive approach of Schwarzenberger, Van Hoof,86 Walden,87 and Mendelson88 (and ultimately Hart’s “rule of recognition”89), and rightly argues that “[t]he problem with this and similar approaches is that they mix the description of reality and prescription.”90 True, the ILA states that it is not simply using customary international law to prove customary international law,91 but it is, more or less, using the state practice element of customary international law to prove the rules of customary international law. There is no necessary reason to believe that Hartian recognition is manifested by the very same type of act also qualifying as state practice for purposes of customary international law. The scope of acts could be wider, narrower, or of a different character altogether. In essence the ILA is finding the rules on the formation of customary international law by using customary international law minus opinio juris.92 In a later section of the committee’s report, the ILA concluded that opinio juris can be dispensed with where there is significant consonant practice93 which then subjects the committee’s own analysis to the circular critique once more. This discussion is not intended to be a harsh criticism of the excellent work of the ILA, but rather to bring to light the efforts of the ILA to prove induction through induction, an impossible task. There must come at some point a deduction to provide the foundation. This foundational deduction that underpins subsequent inductive steps

85. See also id. ¶ 6 n.6 (“These ‘secondary’, systemic rules of recognition etc. are not the same as the substantive rules created by one of the processes so recognized— custom.”). Cf. HART, supra note 35, at 91-94. 86. See G.J.H. VAN HOOF, RETHINKING THE SOURCES OF INTERNATIONAL LAW 53 (1983). 87. See Raphael M. Walden, Customary International Law: A Jurisprudential Analysis, 13 ISR. L. REV. 86 (1978). 88. See id.; Mendelson, supra note 14, at 172-76. 89. See HART, supra note 35, at 235. 90. See Kammerhofer, supra note 6, at 544 (citing KELSEN, supra note 7) (“Kelsen would not be able to stop at the level of ‘material constitutional norms’, because one must not base a legal system’s validity on empirical facts, as Hart purports to do.” (citations omitted)). 91. See also Int’l L. Ass’n, Statement of Principles, supra note 14, at ¶ 6 n.6 (“This is not, as it might at first seem, circular: one is not simply using customary law to prove customary law.”) 92. See Kammerhofer, supra note 6, at 545 (“The difficulty with this empirical approach to determining the rules of law is that ‘If we were to apply the contentions of this method to customary international law we would be trying to prove the existence of custom-creating norms by the ‘practice of states’ alone . . . . How could one distinguish between fact and law when every fact is made law, every application law-making?”). 93. See also Int’l L. Ass’n, Statement of Principles, supra note 14, princ. 19.

2014]

467


could be that state practice is conclusive proof of state consent, that Hart’s rule of recognition applies to the rules on the formation of customary international law, or even that induction can be proved by induction. In any event, the conclusion rests on a deduction. Of course customary international law, and the consistent practice and opinio juris that underlie it, only evidences state consent to a rule, a presumption that appears to be irrebuttable.94 But an irrebutable presumption would render true state consent illusory.95 If we are examining objective consent (i.e., the consent evidenced by the actions or statement of consent), then we must deduce consent from behavior. This author has elsewhere argued that explaining customary international law purely by reference to each individual state’s consent to the rule misses the nuanced interaction of consent and non-consent in bottom-up, collective rule formation processes.96 However, for purposes of this inquiry, we can still identify the presumption that customary international law evidences the consent of states as a deductive step. 2.

The Elements of Customary International Law

The next step of finding the precise elements of customary international law is also a deductive one. The Court could have engaged in an empirical assessment of how this source of international law is discovered, but, given the lack of jurisprudence on the Statute, the Court probably had little choice but to use a deduction. Thus, the court already began with an assumption that the statutory provision could be divided into smaller parts.97 It is important to note that this step is already part of the customary international law analysis, for without this step we would not know what to test to determine if there was a custom. Of course, as observed above, all inductions or deductions must, in the end, be ultimately based on a deduction. Thus, any customary international law analysis is inherently, at some ultimate point, based on a deduction.

94. See KOSKENNIEMI, FROM APOLOGY TO UTOPIA, supra note 1, at 430; ANTHONY A. D’AMATO, THE CONCEPT OF CUSTOM IN INTERNATIONAL LAW 252-54, 258-62 (1971); C.H.M. Waldcock, General Course on Public International Law, 106-II REC. DES COURS HAGUE ACAD. INT’L L. 51 (1962). 95. See KOSKENNIEM, FROM APOLOGY TO UTOPIA, supra note 1, at 430. 96. See Worster, supra note 74. 97. See, e.g., Lotus (Fr. v. Turk.), 1927 P.C.I.J. (ser. A) No. 10 (Sept. 7); North Sea Continental Shelf (Ger. v. Den.; Ger.v. Neth.), Judgment, 1969 I.C.J. 3, 44 (Feb. 20); Continental Shelf (Libya v. Malta), Judgment, 1985 I.C.J. 13, ¶¶ 27-30 (June 3); Jurisdictional Immunities of the State (Ger. v. It.), Judgement, 2012 I.C.J. 99, ¶ 55 (Feb. 3), available at http://www.icj-cij.org/ docket/files/143/16883.pdf.

468

[Vol. 45


Following the determination that there are two aspects, state practice and opinio juris, both of which must be proved, the Court must determine whether the Statute sets up a regime of elements, factors, or evidence. The Statute is ambiguous on this point.98 It could be that customary international law is formed by the concurrence of widespread and consistent practice with opinio juris, and so we prove it by submitting evidence of the elements.99 Or it could be that customary international law exists on some other basis, and it is merely proved by demonstrating practice and opinio juris, with the degree to which it is widespread and consistent going to supporting the probability that the custom exists.100 Lastly, the two could be factors, meaning that custom would exist when there are both items or at least a great deal of one. The Statute does not specify that these are distinct legal elements. Instead manifestations of behavior, whether they are statements of belief or actions, could very well be factors rather than elements, and moreover factors that influence each other.101 Both traditional and

98. The precise text of the I.C.J. Statute, which has been repeated so many times that we often omit reading it again, is as follows: “b. international custom, as evidence of a general practice accepted as law.” Statute of the I.C.J., art. 38(1)(b). It is interesting to note that the provision does not state “general practice accepted as law, as evidence of an international custom.” This text could suggest that custom is the evidence of the legally obligatory practice, not the reverse, as we usually understand it. 99. Compare Statute of the I.C.J., art. 38 (1)(b) (describing “international custom, as evidence of a general practice accepted as law” and suggesting that international custom might be formed by some unknown process, but merely evidenced by practice and opinio juris), with European Union Guidelines on Promoting Compliance with International Humanitarian Law (IHL), [2005] O.J. C327/04, ¶ 7 (Dec. 23, 2005) (“Customary international law is formed by the practice of States which they accept as binding upon them.”). Interestingly, the PCIJ and ICJ Statutes call for the application of “international custom, as evidence of a practice accepted as law,” see Statute of the I.C.J., art 38(1)(b) (emphasis added), suggesting the reverse of the usual interpretation: that custom itself is evidentiary of legal practice, not that legal practice is evidentiary of custom. 100. See id. 101. KOSKENNIEMI, FROM APOLOGY TO UTOPIA, supra note 1, at 434 (“If the sole criterion for something to count as custom-forming practice is whether it ‘reflects’ the opinio juris, then we in fact lack the criterion altogether. For in order to be able to make use of such criterion, we should already know the opinio juris.”) If fact, some scholars have indicated that the two elements are so intermixed as to really only constitute one question. See, e.g., Peter Haggenmacher, La Doctrine des ´ . DE DROIT INT’L deux e´le´ments du droit coutumier dans la pratique de la Cour international, 90 REV. GEN PUBL. 5, 114 (1986). En ve´rite´, aucun des deux e´le´ments n’existe comme tel dans les faits historiques cense´s eˆtre a` la base d’une re`gle coutumie´re concrete . . . Les deux pre´tendus e´le´ments n’ont en reálite´ aucune individualite´ propre; ils se trouvent inextricablement meˆle´s au sein d’une “pratique” unitaire. Cette pratique forme pour ainsi dire un seul “e´le´ment” complexe, fait d’aspects “mate´riels” et “psychologiques.”

2014]

469


modern customary international law analysis agree that there is a deduction here and appear to agree that the elements are not really equal, but disagree on the deduction regarding the interplay of the elements.102 Traditional custom focuses on state practice; modern custom focuses on opinio juris. This conclusion points in the direction that we are not dealing with elements, but factors whose relative weight can compensate for the other. However, the objective of this Article is not to recharacterize the elements as factors, but to argue that there is a deduction at the heart of the rule of customary international law. This conclusion that these aspects are elements of existence and not an evidentiary standard is not particularly problematic, but we should acknowledge that it is deductive. On the other hand, the PCIJ and ICJ have suggested that some rules of customary international law may exist for which the state practice and opinio juris elements are not required. In the Gulf of Maine case, the court held that customary international law also includes “a limited set of norms for ensuring the coexistence and vital cooperation of the members of the international community.”103 This view on alternate means for discovering norms of customary international law is not confined to the Gulf of Maine case, as the Court has articulated or applied it subsequently.104 We could understand this view to mean that there are two classes of customary international laws, arising from two separate processes. This interpretation would mean that the element

Int’l L. Ass’n, Statement of Principles, supra note 14, ¶ 10(c) (“[U]ndoubtedly it is often difficult or impossible to separate the two elements.”). In addition, some scholars have even suggested that there is, in fact, only one element. See Cheng, supra note 8, at 36 (“[I]nternational customary law has in reality only one constitutive element, the opinio juris.”); Mendelson, supra note 14, at 292 (arguing for the opposite: that opinio juris is not necessary in obvious cases). 102. See, e.g., Roberts, supra note 8. 103. See Delimitation of the Maritime Boundary in the Gulf of Maine Area (Can. v. U.S.), Judgment, 1984 I.C.J. 246, ¶ 111 (Jan. 20). 104. See, e.g., Armed Activities on the Territory of the Congo (D.R. Congo v. Uganda), Judgment, 2005 I.C.J. 168, ¶¶ 161-62, 213-14, 244 (Dec. 19) (finding customary international humanitarian law by relying on the role of the Hague Convention Respecting the Laws and Customs of Land Warfare, and customary international law on the “principle of permanent sovereignty over natural resources” by relying on UNGA resolutions); Frontier Dispute (Burkina Faso v. Mali), Judgment 1986 I.C.J. 554, ¶ 20 (Dec. 22) (concluding that uti possidetis is customary international law because “it is logically connected with the phenomenon of the obtaining of independence”); Military and Paramilitary Activities in and Against Nicaragua (Nicar. v U.S.), Judgment, 1986 I.C.J. 14, ¶¶ 215, 218 (June 27) (beginning with “elementary considerations of humanity” to arrive at the conclusion that Common Article 3 of the Geneva Conventions was customary international law); Corfu Channel (U.K. v Alb.), Judgment, 1949 I.C.J. 4, ¶ 219 (Apr. 9) (also beginning with “elementary considerations of humanity”).

470

[Vol. 45


nature of state practice and opinio juris would be thrown into doubt. We could alternatively understand this statement to mean that there is a single class of customary international law, possibly arising from the same underlying process but discovered by two distinct methods of proof. Alternatively, this statement might suggest that there are not two distinct methods of proof, but that the burden of proof of the existence of a customary rule is affected by the utility of the rule to be proved. This possibility will be examined in more detail later. If this is not a burden of proof issue, but rather distinct methods of proof, then the ICJ has apparently established two different methods. The first method of proof is, apparently, an inductive analysis of the existence of the elements, and the second is, apparently, a deductive analysis beginning with the vital needs of the international community. In this latter interpretation, state practice and opinio juris might still be elements, but the proof of them could be deductive. In this author’s view, the Gulf of Maine case, among others, perfectly demonstrates the intermingling of inductive and deductive methods that have always been applied to discover rules of customary international law. B. Articulating the Hypothesis It might seem that the next step is looking at the content of the practice and attempting to derive a rule, but that is not the case. In the Court’s examination, the actor making the assessment is either making or challenging a hypothesis about the content of the rule. It could be competing hypotheses from parties or the single hypothesis of an examiner where the competing hypothesis is implicit and unarticulated.105 However, the Court might also entertain a third hypothesis: that neither hypothesis was correct. This possibility is linked to the curia nova juris concept, i.e., the general principle that the court always knows the law and does not necessarily need it to be proved. While the ICJ explicitly endorses and applies this concept, suggesting that it could contemplate a non-proffered hypothesis,106 in reality it seems to some-

105. See Jurisdictional Immunities of the State (Ger. v. It.), Judgment, 2012 I.C.J. 99, ¶¶ 62-63 (Feb. 3), available at http://www.icj-cij.org/docket/files/143/16883.pdf (“The essence of the first Italian argument is that customary international law has developed to the point where a State is no longer entitled to immunity in respect of acts occasioning death, personal injury or damage to property on the territory of the forum State, even if the act in question was performed jure imperii . . . Germany maintains that, in so far as they deny a State immunity in respect of acta jure imperii, neither Article 11 of the European Convention, nor Article 12 of the United Nations Convention reflects customary international law.”). 106. Military and Paramilitary Activities in and Against Nicaragua, 1986 I.C.J. 14.

2014]

471


times decide based only on the proffered hypotheses.107 This practice undermines the curia nova juris principle and makes the competing proffered hypotheses all the more significant. As noted above, forming a hypothesis is usually understood to be a universal inference, where the investigator uses a sample of a population to infer the existence of a general rule covering the entire population.108 This view seems to be an accurate description of how customary international law hypotheses are formed in cases of litigation at the ICJ. In the Jurisdictional Immunities case, Italy defended its courts’ assessment of the state of customary international law by proposing a hypothesis on the state of the law and submitting the evidence that its own courts used to reach their conclusion on the law.109 Traditional and modern customary international law do not disagree on this point. That being said, the formation of the hypothesis need not be an inductive universal inference and we cannot definitively say that hypotheses submitted to the Court have been formed by universal inference. States do not need to be clear to the Court how they formed their hypothesis in order to lodge a suit.110 Those hypotheses that were presented could have been arrived at from a variety of analytical processes. The Court could examine state practice and opinio juris to determine whether there is a norm requiring an international tribunal to accept and test any hypothesis as formulated by the parties.111 On

107. See, e.g., Certain Questions of Mutual Assistance in Criminal Matters (Djib. v. Fr.), Judgment, 2008 I.C.J. 177 (June 4) (applying the Vienna Convention on the Law of Treaties even though neither France nor Djibouti is a party to the Convention, but neither raised it as an objection). 108. See CARNAP, supra note 53; CARNAP, supra note 47; Christensen, supra note 58; Eells, supra note 58; Eells & Fitelson, supra note 58; Kyburg, Belief, supra note 53, at 42-65. 109. See Jurisdictional Immunities of the State (Ger. v. It.), Judgement, 2012 I.C.J. 99, ¶¶ 62-63 (Feb. 3), available at http://www.icj-cij.org/docket/files/143/16883.pdf. 110. Compare President’s Signing Statement Objects to Detention Provisions in Defense Legislation, 106 AM. J. INT’L L. 361 (2012) (discussing the position of the administration in relation to the 2012 Defense Authorization Act and although considering matters as diverse as the use of force, nuclear arms, Guantanamo detainees and other topics of diplomacy and foreign affairs, referring only to U.S. constitutional law and never to international law, thus providing no good evidence of opinio juris regarding those matters), with Jurisdictional Immunities of the State (Ger. v. It.), Judgment, 2012 I.C.J. 99 (Feb. 3), available at http://www.icj-cij.org/docket/files/143/16883.pdf; Ferrini v. Ger., Cass. 11 Mar. 2004, n.5044/2004, 2004 R.D.I. 540 (It.), reprinted in 128 INT’L L. REPS. 658 (2006) (articulating the state practice as legally justified by a court). 111. In this case, the evolving practice of certain quasi-judicial bodies from administrative assessment of claims to judicial assessment of cases is interesting, and the changing rules on procedure. See generally C.L. Lim, On the Law, Procedures and Politics of the United Nations Gulf War

472

[Vol. 45


the other hand, the Court could view its role deductively from certain premises, i.e., the ICJ Statute commands that the ICJ receive and hear arguments from the parties and test their hypotheses on the law, and so it concludes that it must. Thus, the hypotheses to be tested might also be derived deductively from the hypotheses presented.112 The Court has not made it clear how it reaches the conclusion that it must receive hypotheses. This vague conclusion then opens the door for hypotheses that are developed through induction or deduction. The importance of the hypothesis step for assessing a rule of customary international law is that the act of forming the hypothesis itself can prejudice the outcome, and thus have normative effects. One difficulty is the problem of confirmation bias.113 This bias presents different and reinforcing obstacles, and already prejudices the instances of state practice that will be sampled as falling into either one of two competing theories. It provides limitations for which items of potential evidence are probative of the rule or not. To some degree the prejudice is compensated by the competing hypothesis, but it does not entirely eliminate confirmation bias. Firstly, the hypothesis could be framed in a way that is more in favor of the opposing party, i.e., the hypothesis that the “considerations of humanity” should always prevail.114 It is difficult to imagine a situation in which a court will adjudge that considerations of humanity should not prevail,115 which then affects

Reparations, 4 SING. J. INT’L & COMP. L. 435 (2000); Hans Wassgren, The U.N. Compensation Commission: Lessons of Legitimacy, State Responsibility, and War Reparations, 11 LEIDEN J. INT’L L. 473 (1998). 112. Cf. Accordance with Int’l L. of the Unilat. Decl. of Indep. in Resp. of Kosovo, 2010 I.C.J. Reps. ___ (July 22) (where the Court had to interpret the question asked by the UNGA, which was not an inductive process but a deductive one). Admittedly, asking an advisory question is a different process than proposing a hypothesis. 113. For a discussion of confirmation bias, see also FENNA POLETIEK, HYPOTHESIS-TESTING BEHAVIOUR 73 (2001); HOLLAND ET AL., supra note 38; JONATHAN BARON, THINKING AND DECIDING 195 (3d ed. 2000); Jane L. Risen & Thomas Gilovich, Informal Logical Fallacies, in CRITICAL THINKING IN PSYCHOLOGY 113 (Robert Sternberg et al. eds., 2007); Margit Oswald & Stefan Grosjean, Confirmation Bias, in COGNITIVE ILLUSIONS: A HANDBOOK ON FALLACIES AND BIASES IN THINKING, JUDGEMENT AND MEMORY 79 (Ru¨diger Pohl ed., 2004); Raymond Nickerson, Confirmation Bias: A Ubiquitous Phenomenon in Many Guises, 2 REV. GEN. PSYCH. 175 (1998). 114. See Corfu Channel (U.K. v Alb.), Judgment, 1949 I.C.J. 4, ¶ 219 (Apr. 9). 115. The ICJ has admittedly exempted rules of procedure and immunity from jus cogens norms. See Jurisdictional Immunities of the State (Ger. v. It.), Judgment, 2012 I.C.J. 99, ¶ 95 (Feb. 3), available at http://www.icj-cij.org/docket/files/143/16883.pdf; Case Concerning the Arrest Warrant of 11 April 2000 (D.R. Congo v. Belg.), 2002 I.C.J. Reps. 1 [hereinafter “Arrest Warrant case”], available at http://www.icj-cij.org/docket/files/121/8126.pdf. In those cases, the Court was able to re-frame the issue as not one where considerations of humanity must prevail.

2014]

473


the inductive process that follows. Secondly, even if neither hypothesis could be said to be unfairly advantaged by the framing of issues, any evidence that does not comfortably fall into one of those two competing theories will be, presumably, forced into one theory or discarded as irrelevant in the process of assembling the data pool. Consider the way in which the hypotheses were differently articulated in the Jurisdictional Immunities case. Where Germany proposed a hypothesis that this particular derogation was never permitted, Italy proposed a hypothesis that the rule was increasingly derogable. Firstly, Germany stated: “No general practice, supported by opinio juris, exists as to any enlargement of the derogation from the principle of state immunity in respect of violations of humanitarian law committed by military forces during an armed conflict.”116 Then Italy articulated an opposing hypothesis: Italy will show that, while recognizing sovereign immunity as a fundamental principle of international law, States have also accepted the fact that sovereign immunity is not an absolute principle. The rule of sovereign immunity has been subjected over time to a process of progressive redefinition of its content in the direction of a restriction of the scope of State activities covered by immunity. The evolution towards a restrictive immunity has been fuelled by the increasing concern not to unjustly limit the possibility for individuals to have access to courts against a foreign State.117 If the German hypothesis was tested, then the Court would only consider historical evidence where the rule was derogated from for violations of humanitarian law. If the Italian hypothesis was tested, then the Court would only regard evidence as relevant where it showed an increasing practice of derogation generally, perhaps with more value placed on recent practice. Italy noticed this problem in its Rejoinder: However, Germany’s position appears to be founded on an error of perspective. According to Germany, reference to practice would serve only to prove the existence vel non of a new

116. Jurisdictional Immunities of the State (Ger. v. It.), Memorial of the Fed. Rep. of Germany, 2012 I.C.J. 99, ¶ 55 (June 12), available at http://www.icj-cij.org/docket/files/143/ 16644.pdf. 117. Jurisdictional Immunities of the State (Ger. v. It.), Counter-Memorial of Italy, ¶ 4.3 (Dec. 22), available at http://www.icj-cij.org/docket/files/143/16648.pdf.

474

[Vol. 45


customary rule allowing States to deny immunity with regard to each and every case arising from a breach of a jus cogens rule. In Italy’s view, practice must not be looked at having in mind only the problem of determining whether a new rule providing for a general exception to immunity has crystallized. Statements of States as well as decisions of domestic and international courts are of major significance insofar as they show how Germany errs in pretending to deny the existence of any relations between the substance of jus cogens rules and their enforcement. Contrary to Germany’s contention, States are increasingly aware of the overriding impact of the existence of jus cogens rules on the law of State immunity . . . .118 It is not clear which hypothesis the Court tested (or whether both were tested in parallel), but the outcome of the case suggests an answer.119 Thus, the very act of making a hypothesis about the state of the law—which could be produced through a deductive process—already conditions the selection of data that will be used in reaching the conclusion. We could attempt to assess customary international law without hypotheses. This strategy would avoid the influence from the hypothesizing action itself on finding the norm. In its study on customary international humanitarian law, the ICRC took the approach of sampling state practice and opinio juris on certain issues rather than hypotheses on the existence or lack of certain articulated norms.120 This approach, however, may have had hypotheses in the shadows, for

118. I Jurisdictional Immunities of the State (Ger. v. It.), Rejoinder of Italy, ¶ 4.19, (Jan. 10), available at http://www.icj-cij.org/docket/files/143/16652.pdf. 119. See Riccardo Pavoni, An American Anomaly? On the ICJ’s Selective Reading of United States Practice in Jurisdictional Immunities of the State, 21 ITAL. Y.B. INT’L L. 143 (2011) (citing Jurisdictional Immunities of the State, 2012 I.C.J. ¶ 71, available at http://www.sidi-isil.org/wp-content/uploads/ 2012/05/Pavoni_an-american-anomaly.pdf.) (arguing that, while the ICJ stated it could not find any case in which the relevant practice was upheld, it omitted to consider practice where the courts rejected the claims of the law, and their reasoning: “The Court did not refer to US cases where claims based on the tort exception and relating to wartime acts of foreign States were rejected. It instead professed its unawareness of any US court having even been called upon to apply the tort exception to such situations”). 120. See Tullio Treves, Customary International Law, in MAX PLANCK ENCYCLOPEDIA OF PUBLIC INTERNATIONAL LAW ¶ 92 (2006), available at http://www.mpepil.com/sample_article?id⫽/epil/ entries/law-9780199231690-e1393&recno⫽34&. See also Continental Shelf (Libya v. Malta), Judgment, 1985 I.C.J. 13, ¶ 34 (June 3); Continenetal Shelf Case (Libya/Malta), 1982 I.C.J. Reps. 18, 74, ¶ 100 (Feb. 24) (finding that there was sufficient consistency of the principle of an EEZ to find a rule of customary international law, even though the precise allocations were still contested); see

2014]

475


example, seeking to confirm or refute the hypothesis that a particular practice during armed conflict was prohibited. States could document all state practice without paying any attention to questions of action, but such an approach would be intensely wasteful, since the state would be unable to address questions about the legality of future behavior. States might act, sometimes with and sometimes without the benefit of counsel (or sophisticated counsel), and sometimes believing the act to be pragmatic, solution-oriented, in line with past practice, or not, with the true legality of the act being evaluated ex post facto.121 It could be that practice without a clearly articulated hypothesis is simply the normal formation of customary international law.122 States are, after all, approaching their relationship with customary international law from a very different legal position than an international court. C.

Testing the Hypothesis

Next, the Court must test the hypothesis that has been articulated. First, courts have determined, implicitly, that the existence of state practice and opinio juris as elements can be proved by evidence. This step of determining the make-up of customary international law, apparently only based on a reading of the ICJ Statute and partly an assumption inherent in any legal system about answering a legal question, is primarily a deductive step that evidence of a rule can prove

generally 2 INT’L COMM. RED CROSS, CUSTOMARY INTERNATIONAL HUMANITARIAN LAW (Jean-Marie Henckaerts & Louise Doswald-Beck eds., 2005) [hereinafter “2 ICRC CUSTOMARY IHL”]. 121. Int’l L. Ass’n., Statement of Principles, supra note 14, ¶ 7. Rather, the suggestion is that different functions may lead the persons performing them to adopt a somewhat different attitude to the sources, including this particular one. For example, whereas a judge has essentially to determine the law impartially, as it is at the time of judgment, a government legal adviser may well feel called upon also to assess how matters are likely to develop in the future, and even to try to assist or retard that development in accordance with the national interest. (citing Int’l L. Ass’n., 1st Interim Report of the Committee, Ann., in INT’L L. ASS’N, REPORT OF 63RD CONFERENCE 938, 941 (Warsaw 1988); Int’l L. Ass’n, 1st Report of the Rapporteur (Mendelson), Formation of International Law and the Observational Standpoint, App. (1986)). See also Wouters & Ryngaert, supra note 8, at 111 (citing Int’l L. Ass’n, supra note 14) (“A choice of methodology is often not value- or politically neutral. This also holds true in law and international law. Scholar/lawyer A may have a preference for method 1 because it best serves his or her moral or political proclivities, while scholar/lawyer B may opt for method 2 because that method best transmits his or her value preferences. Quite conceivably, both A and B will term their own method ‘objective’, and another method ‘subjective.’”). 122. See, e.g., BRIAN D. LEPARD, CUSTOMARY INTERNATIONAL LAW: A NEW THEORY WITH PRACTICAL APPLICATIONS 103, 158-59 (2010).

476

[Vol. 45


a rule.123 Usually when we answer questions of law, we do not employ “evidence” in the strict sense. However, in the proof of state practice and opinio juris, it appears that we might. We could answer this conclusion inductively by examining the practice of states in reaching conclusions on questions of international law. Whether the practice of tribunals could be considered in such an inquiry is an open question. But it is not proposed that this is a mandatory step and early tribunals dispensed with any inductive analysis.124 In fact, where cases are cited they are not cited as a factual sampling of practice but in a quasiprecedential manner, confirming the validity of the initial deduction. In any event, there is a deductive conclusion that state practice and opinio juris must be established. 1.

Proving Domestic Customary Law

One way to determine how a court should assess the existence of customary international law, is to examine the way domestic courts assess the existence of domestic customary law. Domestic legal systems sometimes have to determine whether customary law exists on a certain question125 and those courts have developed some methodology. Generally, the existence of a customary law must be proved by the party

123. See Kammerhofer, supra note 6 (“In effect he [Georg Schwarzenberger] simply assumes Article 38 to be the relevant provision on which to build his theory; this is a deductive step. The inductive approach needs a referent by which to check the results of the induction or, rather, to clarify the ‘arbitrary origin’, as engineers would call it, of the induction. In order to do this he postulates Article 38 as ‘having its sheet-anchor firmly embedded in the near universally expressed ¨ berbau of ‘law-creating processes and law-determining will of the organised world society’. This U agencies’ is deduced, not induced—it is the dogma Schwarzenberger does not question or ¨ berbau is not objectionable from a strictly inductive viewpoint, validate. While the content of this U the method employed to know it is.” (footnotes omitted)). 124. See, e.g., Corfu Channel (U.K. v Alb.), Judgment, 1949 I.C.J. 4 (Apr. 9). 125. See, e.g., Mmusi v. Ramantele, Case No. MAHLB-000836-10, (High Court, Bots. 2012) (on appeal from the Customary Court of Appeal); Bhe & Others v. Magistrate, Khayelitsha and Others, 2005 (1) SA 580 (CC) (S. Afr.); Navajo Nation v. Rodriguez, No. SC-CR-03-04 (Navajo. Sup. Ct. Dec. 16, 2004); Ephraim v. Pastory, (2001) AHRLR 236 (High Ct. Tanz.); Crow Tribe of Indians v. Big Man, No. 00-410, 2000 Crow 7, 2000.NACT.0000007 (Crow Ct. App. 2000); Vill. of Mishongnovi v. Humeyestewa , No. 96AP000008, 1998.NAHT.0000017 (Hopi Ct. App. 1998); Mojekwu v. Mojekwu, [1997] 7 NWLR 283 (C.A. Nigeria); Rave v. Reynolds, 23 Indian L. Rep. 6150 (Winnebago Tribe of Nebraska Supreme Court 1996)); Hamlet of Baker Lake v. Min. Ind. Aff’rs & N. Develop., [1980] 1 F.C. 518 (T.D.) (Can.); D.P.P. v. Camplin, [1978] AC 705 (H.L.); Mofja v. R., (1977) 13 A.L.R. 225 (Austr.); Calder v. A.-G. Brit. Col., [1973] S.C.R. 313, [1973] 4 W.W.R. 1 (Sup. Ct., Can., 1973); R. v. Pilimapitjimiri, ex parte Gananggu, (1965) N.T.J. 776 (Austl.); Akrofi v. Akrofi, [1965] GLR 17 (High Ct. Ghana); Bedder v. DPP, [1954] 2 All E.R. 801 (Austl.); Angu v. Attah, (Gold Coast 1916), in JUDGMENTS OF THE JUDICIAL COMMITTEE OF THE PRIVY

2014]

477


alleging the custom.126 At other times, courts simply take judicial notice of sufficiently well-known practices.127 In addition, colonial courts often defer to indigenous courts regarding knowledge of the law, presuming that they would know their law, curia nova juris, and thus substantiating evidence of the law is not necessary. In customary courts (e.g. indigenous or Sharia courts) by contrast, proof of the custom may not be required, since curia nova juris truly applies both de jure and de facto.128 Although appeals within a legal system, not to a foreign colonial court, may consider the question of law de novo,129 it appears that proving customary law, even where forming the applicable law of the jurisdiction and thus not foreign law, is usually thought of as a question of fact, especially where colonial courts were investigating the applicable customary law in indigenous communities.130 This distinction between questions of fact and questions of law then has implications for whether the customary law at issue should be established through the same process as facts, with the same forms of evidence, evidentiary standards, and standards of proof. Some jurisdictions have reversed this practice and determined that the existence of a customary law is a question of law, meaning that the court can receive the law using sources such as reported cases, textbooks, etc. and would not be bound by the evidentiary considerations inherent in a question

COUNCIL ON APPEAL FROM THE GOLD COAST COLONY 1874-JUNE 1928 43 (1978); Wallis v. SolicitorGeneral (1903) NZPCC 23 (N.Z.); R. v. Symonds (1847) NZPCC 387 (N.Z.). 126. See, e.g., Nigerian Evidence Act, 1945, § 14(1); Healy v. Mashantucket Pequot Gaming Cen., No. MTPC-EA-97-132, 1999.NAMP.0000053, ¶ 36 (Mashantucket Pequot Ct, Apr. 8, 1999); Hopi Indian Credit Ass’n v. Thomas, No. AP-001-84, 1996.NAHT.0000007, ¶ 31 (Hopi Ct. App. 1996). 127. See, e.g., Indian Evidence Act, 1872, § 57; Hopi Indian Credit Ass’n, ¶¶ 32-33; Apache v. Rep. Nat’l Life Ins., 3 Nav. R. 250, para. 28 (W.R. Dist. Ct., U.S., Sept. 16, 1982) (applying the standard that “every damn fool knows” the customary practice). See also Tome v. Navajo Nation, 4 Nav. R. 159 (W.R. Dist. Ct., U.S., May 13, 1983) (“The court can ‘find’ the Navajo common law through its own knowledge of it, since it is a matter of common knowledge.”). However, the precise content of the custom might still be open to judicial interpretation. See, e.g., Polingyouma v. Laban, No. AP-006-95, 1997.NAHT.0000018, para. 27 (Hopi Tribal Ct., U.S., Mar. 28, 1997). 128. See Nigerian Evidence Act, 1945, § 14(1); 7 Navajo Nation Code § 354(E) (requiring judges to be fluent in Navajo and have knowledge of the indigenous culture). 129. See, e.g., Native Customs (Recognition) Ordinance (“Papua N. Guinea Customs Recognition Act”), 1963, § 2 (Papua N. Guinea). 130. See, e.g., Native Customs (Recognition) Ordinance, 1963, § 2 (Papua N. Guinea) (noting that the existence of the customary law should be proved as if it were fact, not specifically declaring the issue a question of fact). See also Nigerian Evidence Act, 1945, § 56(1); Angu (Gold Coast 1916), at 43.

478

[Vol. 45


of fact.131 In these systems applying a factual analysis, courts often take testimonial evidence of the practice in the community.132 This evidence can take the form of expert or scientific testimony,133 for example by an anthropologist,134 but it might be considered more reliable to take the testimony of persons who are respected members of the community with more intimate knowledge, and thus experts in a different sense.135 Alternatively, courts can receive evidence in the form of documentary proof of the practice.136 Even where it is a factual consideration, courts are not usually bound by evidentiary rules regarding facts, e.g. limitations on hearsay, etc.,137 and the burden of proof is often unclear.138 Sometimes where a customary law is difficult to establish, a court might lower the evidentiary threshold for the custom if it is supported by strong public policy considerations.139 It would appear that the requirement of widespread practice as we understand it

131. See, e.g., 1969 Customary Law (Application and Ascertainment) Act (Botsw.), § 11. 132. See, e.g., Indian Evidence Act, 1872, § 32(4); Angu (Gold Coast 1916). As is the case with all customary law, it has to be proved in the first instance by calling witnesses acquainted with the native customs until the particular customs have, by frequent proof in the court, become so notorious that the courts will take judicial notice of them. See Nigerian Evidence Act, 1945 § 14(3). 133. See, e.g., In re Estate of Apachee, 4 Nav. R. 178, para. 32 (W.R. Dist. Ct., U.S., 1983). 134. See, e.g., Indian Evidence Act, 1872, § 45; Nigerian Evidence Act 1945, § 56(1); In re Apachee, 4 Nav. R. 178, 180, para. 32. 135. See, e.g., Nigerian Evidence Act 1945 § 58; Indian Evidence Act, 1872, §§ 48-49; Jensen v. Giant Indust., No. SC-CV-51-99, 2002.NANN.0000003, para. 21 (Nav. Sup. Ct., U.S., 2002) (establish custom through an informal conference of experts); Marino v. Mashantucket Pequot Gaming Enter., No. MPTC-CV-97-104, 2000.NAMP.000003, para. 30 (Manshantucket Pequot Ct., U.S., Jan. 6, 2000) (“an expert on the culture, custom, and traditions of the Tribe”); Hoffman v. Colville Confed. Tribes, 24 Ind. L. Rep. 6163, 1997.NACC.0000010, para. 133 (Colv. Confed. Tribes Ct. App., U.S., May 5, 1997) (“Where any doubt arises as to the customs of the Tribes the Court may request the advise of counselors familiar with those customs.”); Hopi Ind. Credit Ass’n v. Thomas, No. AP-001-84, 1996.NAHT.000000, para. 41 (Hopi Tribal Ct., U.S., Mar. 29, 1996) (“expert witnesses will be called to testify as to the existence and scope of traditions”); In re Estate of Bigthumb, 6 Navajo. Rptr. 453, 455, 1989.NANN.0000026, para. 22 (W.R. Dist. Ct., U.S., Jan. 20, 1989); Benally v. Navajo Nation, 5 Navajo. Rptr. 209, 1986.NANN.0000013, para. 18 (W.R. Dist. Ct., U.S., 1986) (“counselors familiar with these customs and usages”). 136. See, e.g., Indian Evidence Act, 1872, § 57. 137. See, e.g., Native Customs (Recognition) Ordinance, 1963, § 2 (Papua New Guinea). 138. See, e.g., Miller v. Mashantucket Pequot Gaming Enter., No. MPTC-EA-98-121, 1998.NAMP.0000041, para. 20 (Manshantucket Pequot Ct., U.S., Aug. 7, 1998) (need to “clearly” demonstrate custom); Watt v. Colville Confed. Tribes, 25 Ind. L. Rep. 6027, 1998.NACC.0000012, para. 19 (Colville Confed. Tribes Ct. App., U.S., Jan. 21, 1998) (mere “suppositions” regarding custom are insufficient). 139. See, e.g., Rave v. Reynolds, 23 Indian L. Rep. 6150, 6158 (Winnebago Tribe of Nebraska Supreme Court 1996) (“In small, close-knit tribal communities, like the Winnebago tribe of Nebraska, denying an opportunity to air and heal grievances . . . could have disruptive ef-

2014]

479


in customary international law, while formally required for domestic customary law, is often dispensed with in terms of proof.140 In essence, a court can “inform itself as it thinks proper.”141 On the other hand, many jurisdictions approach the question of customary international law as a question of law, not fact,142 although what that conclusion means for the proof of the law, e.g. use of expert testimony or not, remains unclear.143 This picture of the method in domestic legal systems for finding domestic customary law sets the stage for our examination of customary international law. 2.

Proving Customary International Law

Turning to customary international law, courts also look for proof of the rule. As a threshold matter, in domestic courts, there is the important role, sometimes determinative, of the executive affirming the existence of a rule of customary international law. Some systems

fects . . . . Accordingly, adopting the narrow standing rules employed in federal courts . . . would not constitute sound public policy.”). 140. See, e.g., Indian Evidence Act, 1872, § 32(4). 141. See, e.g., Native Customs (Recognition) Ordinance, 1963, § 2(2)(a). 142. See, e.g., Nulyarimma v. Thompson, 165 A.L.R. 621 (Austl. 1999); Polyukhovich v.. Australia, (1991) 172 CLR 501 (Austl.); Jumbunna Coal Mine v. Victorian Coal Miners Ass’n, (1908) 6 CLR 309, 363 (Austl.); Republic of “A” Embassy Bank Account (Sup. Ct. Austl., Apr. 3, 1986), reported at 77 INT’L L. REPS. 488; In re Aquarone, 101 REV. GEŃ. DE DROIT INT’L PUBL. 838 (1997) (Con. State, Fr., June 6, 1997); Paul Touvier (Ct. App. Paris, Fr., Apr. 13, 1992), reported at 100 INT’L L. REPS. 338; Bundesgerichtshof [BGH] [Federal Court of Justice] Nov. 3, 1992, 100 INT’L L. REPORTS 364, 1995; Bundesgerichtshof [BGH] [Federal Court of Justice] May 4, 1968, 65 INT’L L. REPORTS 162, 1994; Prefecture of Voitia v. Germany (“Distomo Massacre case”), Case. No. 11/2000 (Ct. Cass (Areios Pagos), Greece, May 4, 2000), reported at 129 INT’L L. REPS. 513; Decision, (Cass. It., Apr. 28, 1994, n. 168) reported at 77 R.D.I. 519 (1994); In re Arafat Salah, Cass., 28 June 1985 (It.), reported at 7 ITAL. YB INT’L L. 189 (1986-87); Russel v. Societa Immobiliare Soblim 1978-79 ITAL. Y.B. INT’L L. 146 (Corte Cost., It.); Publ. Prosecutor v. D.K. 1997 NETH. YB INT’L L. 315 (Sup. Ct. Neth.); Diana A. v. S. Afr., Case 1.293/1990 (Trib. Const., Sp., July 1, 1992) reported at 44-2 Revista Espanola de Derecho internacional (hereinafter “REDI”) 565 (1992), (Dec. 21, 1993, 46 REDI 725 (1994); P. v. Office Fe´d. de la police, 117 lb Arrets du Tribunal Federal 337, c. 5a (340) (Trib. Fed., Switz., 14 October 1991) reported at 48 ANN. SUISSE DROIT INT’L 569-570 (1991); Ferdinand et Imelda Marcos v. Office Fe´de´ral de la police, 115 lb Arrets du Trib. Fed. 496, 498 (Trib. Fed., Switz., Nov 2, 1989) reported at 47-48 ANN. SUISSE DROIT INT’L 533 (1991); R v. Bow St. Metro. Stipendiary Magistrate, ex parte Pinochet Ugarte [Pinochet III], 2 W.L.R. 827 (H.L. 1999); Alcom Ltd. V. Colombia, [1984] 2 W.L.R. 750 (H.L . 1984). 143. Compare In re Bouterse (Sup. Ct., Neths., Sept. 18, 2001), reported at 32 NETH. YB INT’L L. 282-96 (2001) (relying on expert testimony as to the content of international law), with Julius A. Furer case (Ct. Appl., Amsterdam, Neths., June 18, 1987), reported at SCHIP EN SCHADE [Dutch Law Reports on Maritime Law, Insurance Law and Carriage of Goods], No. 55 (1988) (not relying on expert testimony).

480

[Vol. 45


appear to defer to the executive on questions of international law, or at least give the executive’s opinion a presumption of validity.144 Of course, the executive might be correct about the law, but this approach by courts to establishing customary international law does suggest that courts sometimes view the question as one of fact. This approach does not directly apply to international courts, though the parties to a dispute might agree on the rules of customary international law. Of course having the parties in a situation of legal equality stipulate customary international law is different from having the executive branch of a state assert the law in court proceedings, but it does show that international courts might occasionally treat the existence of law as equivalent to a factual matter.145 However, in many cases, the law is contested. A court could use one of various techniques to determine the content of the law, but it remains somewhat unclear whether courts are considering the establishment of the law as a question of law or fact. Sometimes courts accept submissions by counsel to establish the law, suggesting that the question is one of law and need not be proved in the evidentiary sense. On the other hand, some also accept submissions by experts (compare to an ethnography), suggesting the reverse.146 For example, the International Criminal Court called for scholarly comment on the issue of Palestinian capacity to accept the jurisdiction of the Court.147 Note that the scholarly comment received by the Prosecutor was not doctrinal statements of the law, but rather scientific studies of the law. In essence, the court asked scholars to do its job for it. It could be that courts view submissions or advice by experts as more akin to amicus curiae, or as subsidiary sources of international law, not evidence, though it is difficult to find precision in the case law. Of course the weakness in this

144. See R. v. Bottrill, ex parte Kuechenmeister, [1947] 1 K.B. 41 (Ct. Appl., U.K., 1947). 145. See, e.g., Jurisdictional Immunities of the State (Ger. v. It.), Judgment, 2012 I.C.J. 99 (Feb. 3), available at http://www.icj-cij.org/docket/files/143/16883.pdf (Germany did not object to Italy’s characterization of the acts of the German Armed Forces in World War II as constituting violations of international law); Certain Questions of Mutual Assistance in Criminal Matters (Djib. v. Fr.), Judgment, 2008 I.C.J. 177 (June 4) (neither France nor Djibouti objected to the application of the Vienna Convention on the Law of Treaties as a matter of customary international law, even though neither is a party to the convention). 146. See generally Maurice Mendelson, The Effect of Customary International Law on Domestic Law: An Overview, 4 NON-ST. ACTORS & INT’L L. 75 (2004). 147. See Int’l Crim. Ct., Ofc. of the Pros., Situation in Palestine: Summary of submissions on whether the declaration lodged by the Palestinian National Authority meets statutory requirements (May 3, 2010), available at http://www.icc-cpi.int/NR/rdonlyres/D3C77FA6-9DEE-45B1-ACC0B41706BB41E5/282852/PALESTINEFINAL201010272.pdf.

2014]

481


approach is the concern over subjectivity and generalizability mentioned above. In any event, it is important to carefully isolate domestic jurisprudence on this matter since certain constitutional structures might intentionally view international law as analogous to foreign law, and might not be applicable to international courts. The court could rely on an experiment, although it is unclear how an experiment as known in the sciences would work in the context of litigation. In a sense, the rendering of the decision and waiting for comments from states is an experiment, though an incredibly ineffective one. Courts could, for example, render interlocutory decisions on questions of law which would then be referred to states for comment or correction. Though this process resembles an experiment, it would have significant repercussions on the court’s power to state what the law is. Instead, in determining the existence of customary international law, the court will generally use a sampling survey, i.e., acquiring samples of data from a pool and then generalizing the findings to the population as a whole. That sampling is the method that international courts should use is a conclusion that is not clearly inductive, but apparently deductive. Only once we are confident, either through induction or deduction, that sampling is a method that is appropriate for the assessment of the content of the rule, do we proceed to the next step of assessing the content of the rule. Both traditional and modern customary analysis agree that we should sample data of instances to establish the existence of the norm. Perhaps they disagree on what we should sample for, but they do not disagree that we should sample. However in all of the foregoing analysis on the method for proving customary international law, the mechanism for scientific inquiry is examining fact, not law, so it is not clear that this is the correct way to approach questions of law. It is not entirely clear that within the international legal system itself that the existence of international law is a factual inquiry.148 Though courts can, and do, look to state practice

148. See, e.g., W. Rand Centr. Gold Mining Co. v. The King, [1905] 2 K.B. 391, 392, 406-7 (Ct. Appl., U.K., 1905) (per Alverstone, L.C.J.): any doctrine so invoked [as customary international law] must be one really accepted as binding between nations, and the international law sought to be applied must, like anything else, be proved by satisfactory evidence, which must shew either that the particular proposition put forward has been recognised and acted upon by our own country, or that it is of such a nature, and has been so widely and generally accepted, that it can hardly be supposed that any civilized state would repudiate it. From the text and context, it appears that Alverstone viewed the question of customary international law as a question of fact.

482

[Vol. 45


and opinio juris on the proof of international law in order to justify such an approach by induction, it is not clear that such an approach can be supported by induction. In all of the instances of state practice, the court looked at state reception of international law, which, depending on the view that the state held towards international law, could be considered a separate (or foreign) factual question. If domestic courts treat the question as one of fact, then it is difficult to see how this practice can result in international law applicable by international tribunals that should treat the inquiry as one of law. The tribunal would need to distinguish each instance based on the domestic legal system’s distinction between monist and dualist treatment. Perhaps in monist legal systems, where international law is directly applicable in the domestic legal system, international law is also viewed as a question of law, and perhaps in dualist legal systems, where international law must be incorporated into the domestic legal system before being applicable, international law is viewed as a question of fact. It is therefore more likely that an international tribunal, in forming its own inquiry into the existence of rules of customary international law, is applying a sampling technique by deductive analogy, not by inductive empirical analysis. D.

Establishing the Types and Forms of Evidence

Once the tribunal has determined that it may test the hypothesis of the law by a sampling method, the court determines that certain acts or facts can constitute evidence of state practice and opinio juris.149 This is also a deductive step. There is no clear agreement on which volitional manifestations constitute “acts,” and which of those acts qualify as “state practice.”150 There also does not appear to be any distinction between

149. See Jurisdictional Immunities of the State (Ger. v. It.), Judgment, 2012 I.C.J. 99, ¶¶ 55-56 (Feb. 3), available at http://www.icj-cij.org/docket/files/143/16883.pdf. There is actually a step that even precedes this one: the conclusion that external behavior can be indicative (or even conclusive) of internal belief. [W]e cannot automatically infer anything about State wills or beliefs—the presence or absence of custom— by looking at the State’s external behaviour. The normative sense of behaviour can be determined only once we first know the ‘internal aspect’—that is, how the State itself understands its conduct . . . . [D]octrine about customary law is indeterminate because circular. It assumes behaviour to be evidence of the opinio juris and the latter to be evidence of which behaviour is relevant as custom. Kennedy, Renewal, supra note 18, at 355, 407. See also Kammerhofer, supra note 6. 150. Possibilities have been suggested. See, e.g., UNGA Res. 2099(XX); Council of Eur., Comm. Mins., Resolution (68)17, Model Plan for the Classification of Documents concerning State Practice in the Field of Public International Law (June 28, 1968), as amended by Recomm. 97(11)

2014]

483


types of questions and the types of evidence more suitable for each question.151 1.

Qualifying Categories of Actors

The first aspect of this inquiry is the nature of the actor, and whether it qualifies as a state. State practice is generally regarded to be just that, the practice of states, not international organizations or other actors.152 The acts of international organizations have also been suggested as sources of evidence of customary international law.153 Setting aside the question of whether a given entity is or is not a state, which is yet another mixed question of induction and deduction, law and politics, the question that sometimes arises is which actors within a state are capable of manifesting state practice.154 Some authorities have argued

(June 12, 1997); G. Marston, The Evidences of British State Practice in the Field of International Law, in A. CARTY & G. DANILENKO, EDS., PERESTROIKA AND INTERNATIONAL LAW, CURRENT ANGLO-SOVIET APPROACHES TO INTERNATIONAL LAW 35 (1990). 151. See, e.g., THEODOR MERON, HUMAN RIGHTS AND HUMANITARIAN NORMS AS CUSTOMARY INTERNATIONAL LAW (1989) (arguing that customary norms on human rights should especially rely on declarations in international organizations or conferences); Lori Bruun, Beyond the 1948 Convention—Emerging Principles of Genocide in Customary International Law, 17 MD. J. INT’L L. & TRADE 193, 216-17 (1993); Lillich, Growing Importance, supra note 15; Charney, supra note 13, at 543-45 (arguing that common problems facing the world renders “traditional” custom unsuitable). 152. While the ILA easily dismissed the unattributable acts of natural persons and legal persons from contributing to customary international law, its views on the role of international organizations was somewhat less clear. See Int’l L. Ass’n, Statement of Principles, supra note 14, princs. 10, 11 cmt. (a), (b) (addressing the arguments of, inter alia, Profs. Wolfke and Villiger); see also Reservations to the Genocide Convention on the Prevention and Punishment of the Crime of Genocide, 1951 I.C.J. 15 (May 28) (examining the depositary practice of the UN SecretaryGeneral). 153. See Military and Paramilitary Activities in and Against Nicaragua (Nicar. v U.S.), Judgment, 1986 I.C.J. 14, ¶¶ 188, 190-92, 195, 202, 204 (June 27); North Sea Continental Shelf (Ger. v. Den.; Ger.v. Neth.), Judgment, 1969 I.C.J. 3, 33-36 (Feb. 20); Texaco Overseas Petroleum Co. v. Libya, 17 I.L.M. 30, ¶ 78 (Ad Hoc Trib. 1977); Kuwait v. Am. Indep. Oil Co., 21 I.L.M. 976, 1032-33 (Ad Hoc Trib. 1982); Sedco, Inc. v. Nat’l Iranian Oil Co., 25 I.L.M. 633 (Iran-U.S. Cl. Trib., 1986); Int’l L. Comm’n, Report to the United Nations General Assembly, II YB. INT’L L. COMM’N 368-72. 154. There might also be an additional step that could be approached deductively or inductively, the assessment of which actors will qualify as states and thus qualifying contributors to customary international law. For example, the analysis of the ICRC which had to assess customary international humanitarian law as applicable to non-state actors. Some state practice are easy cases, but others, such as either including or excluding the practice of the Republic of China on Taiwan Island, Kosovo or Palestine would be far more controversial. We can add to this consideration the problem of accepting practice as evidence of law by entities that the states in

484

[Vol. 45


that decisions of courts do not qualify as state practice since courts cannot bind the state, e.g. as under a treaty, under international law.155 Others have argued that any act attributable to a state, e.g. under the Draft Articles on State Responsibility, can constitute state practice, and even acts that are not attributable could still constitute state practice where they are adopted by the state.156 This rule is yet another deductive step. An alternative could be drawing an analogy with the

question do not accept as states. For example, could Iran be held to a customary international law (or even more problematic, a regional customary international law) that had been evidenced partly (or even more difficult, mostly) by the practice of Israel, a state Iran refuses to recognize? In addition, the contribution to customary international law of the practice of some entities should be more controversial than it is. An example would be the casual reception of practice of the Holy See, which of course, as opposed to the Vatican city state, is not a state. See, e.g., Alain Pellet, Special Rapp., Int’l L. Comm’n, Third Report on Reservations to Treaties, U.N. Doc. A/CN.4/491, ¶ 6 (Apr. 30, 1998) (discussing the responses to his questionnaires to states on state practice and opinio juris, which includes a footnote that the Special Rapportuer had received a reply from the Holy See). It could be that the Holy See was merely submitting the practice of the Vatican, although in the instance above, that is unlikely since the Holy See generally adheres to treaties in its own name and competence rather than through the personality of the Vatican, so the practice referenced must necessarily be that of the Holy See and not the Vatican. More likely, the Special Rapportuer assimilated the two entities and accepted the statement of practice and opinio juris by a non-state international legal person as evidence of customary international law. Notwithstanding Crawford’s submissions on the matter, see generally JAMES CRAWFORD, THE CREATION OF STATES (2d ed. 2006), this problem of statehood is usually solved by a heavily politically-informed legal analysis. See this author’s conclusions at William Thomas Worster, Law, Politics, and the Conception of the State in State Recognition Theory, 27 BOSTON UNIV. INT’L L. J. 115 (2009). But even where the analysis could be said to be “more legal than political”, the legal analysis tends to be deductive since an inductive approach would, by necessity of the sampling process, suggest the application of the constitutive theory and bring in the consideration of too many decisions that were “more political than legal.” 155. See, e.g., Lotus (Fr. v. Turk.), 1927 P.C.I.J. (ser. A) No. 10, at 23, 26, 28-29 (Sept. 7); Int’l L. Comm’n, Draft articles on Responsibility of States for Internationally Wrongful Acts, with commentaries arts. 5-6, U.N. Doc. A/56/10 (2001) (hereinafter “ILC, Draft articles on Responsibility of States”); Int’l L. Ass’n, Statement of Principles, supra note 14, princ. 9 cmt. (a), (d); 1 DIONISIO ANZILOTTI, COURS DE DROIT INTERNATIONAL 74-75 (G. Gidel trans., 3d ed. 1929) (arguing that the only practice that could be considered state practice for customary international law was the practice of the organs of state capable of binding the state under international law, not, e.g., the courts); K. Strupp, Les re`gles geńe´rales du droit de la paix, 47-I REC. DES COURS HAGUE ACAD. INT’L L. 313 (1934). All of the foregoing sets aside the contemporary argument that the practice and opinio juris of international organizations in their own independent capacity (and not practice of states by and through international organizations) as evidentiary of customary international law. See, e.g., Kenneth S. Gallant, International Criminal Courts and the Making of Public International Law: New Roles for International Organizations and Individuals, 43 J. MARSHALL L. REV. 603 (2010). 156. See Int’l L. Ass’n, Statement of Principles, supra note 14, at princs. 7, 8, 9; ILC, Draft articles on Responsibility of States, supra note 155, arts. 4-11.

2014]

485


Vienna Convention on the Law of Treaties prescribing those actors that have the capacity to bind the state.157 This analogy, however, when compared to the wider scope of actors usually considered in customary international law analysis (e.g. acts of courts), also suffers from the weakness of probably being too narrow. Any analogy drawn, though, would be a deductive exercise. The Draft Articles was largely a codification, and thus presumably an inductive, exercise,158 but drawing an analogy between the Draft Articles rules on attribution (themselves based on customary international law) to determine the rules on the formation of customary international law is a deductive step. The question could be approached inductively, perhaps with the same outcome, but courts usually reach this rule by deducing the requirement from either or both of the ICJ Statute language or postulates about the nature of the international legal system (e.g. “states are the only subjects of international law” etc.). This is not to say that this is incorrect, but merely that it is deductive. 2.

Qualifying Categories of Acts

Second, courts approach the question of which acts can constitute evidence of state practice and opinio juris by identifying categories of acts that are suitable as evidence of customary international law and, implicitly, those that are not. This must be distinguished from the later step where the actual evidence before the court is assessed as reliable and probative of a rule or not.159 This preliminary step involves the building of a threshold classification system for types of evidence that might be submitted. Various courts and tribunals, the International Law Commission (ILC), and certain scholars, notably Brownlie, have reached conclu-

157. See Vienna Convention on the Law of Treaties, May 23, 1969 art. 7, opened for signature May 23, 1969, 1155 U.N.T.S. 331 (entered into force Jan. 27, 1988). 158. The reader is invited to consider whether the draft articles are themselves produced through induction or deduction. See, e.g., ILC, Draft articles on Responsibility of States, supra note 155, arts. 4-11. 159. See Int’l L. Ass’n, Statement of Principles, supra note 14, princ. 3 & cmt. (“What is suggested here is something analogous to (but not the same as) the well-known distinction in the law of evidence between the admissibility of evidence and its weight (convincingness).”). See also Fisheries (U.K. v. Nor.), 1951 I.C.J. 116, 191 (Dec. 18) (dissenting opinion of Judge Reed) (“The only convincing evidence of State practice is to be found in seizures, where the coastal State asserts its sovereignty over the waters in question by arresting a foreign ship and by maintaining its position in the course of diplomatic negotiations and international arbitration.”). However, it is important to note that this observation was made in the context of claims of territorial sovereignty, where special considerations may obtain.

486

[Vol. 45


sions on possible categories of state acts that are suitable for proving the existence of a rule of customary international law. These categories include the existence of a multilateral convention on point where the convention records, defines or develops customary international law;160 the existence of claims to rules of customary international law before national courts and the judgments or other reactions of courts when decided on issues of customary international law;161 domestic legislation of states on issues governed by customary international law;162 public acts or statements made to the international commu-

160. See Jurisdictional Immunities of the State (Ger. v. It.), Judgment, 2012 I.C.J. 99, ¶ 55 (Feb. 3), available at http://www.icj-cij.org/docket/files/143/16883.pdf; Military and Paramilitary Activities in and Against Nicaragua (Nicar. v U.S.), Judgment, 1986 I.C.J. 14, ¶ 212 (June 27); Continental Shelf (Libya v. Malta), Judgment, 1985 I.C.J. 13, ¶¶ 27, 34-35 (June 3); North Sea Continental Shelf (Ger. v. Den.; Ger.v. Neth.), Judgment, 1969 I.C.J. 3, 38-39 (Feb. 20); Nottebohm (Liech. v. Guat.), Judgment, 2d Phase, 1955 I.C.J. 4, 22-23 (Apr. 6); Asylum (Colom. v. Peru), Judgment, 1950 I.C.J. 266, 277 (Nov. 20); Lotus (Fr. V. Turk.), 1927 P.C.I.J. (ser. A) No. 10, at 26 (Sept. 7); Prosecutor v. Furundzˇjia, Case No. IT-95-17/1-T, Judgment, ¶ 168 (Int’l Crim. Trib. Former Yugoslavia Dec. 10, 1998); Case C-366/10, Air Transp. Ass’n of Am. v. Sec’y of State for Energy & Climate Change, ¶ 104 (Dec. 21, 2011), available at http://eur-lex.europa.eu/ LexUriServ/LexUriServ.do?uri⫽CELEX:62010CJ0366:EN:HTML (citing Article 1 of the Chicago Convention); Racke v. Hauptzollamt Mainz, 1998 E.C.R. I-3655, ¶ 53. Draft conventions that have not entered into force can also be used as evidence of rule. See id. ¶ 59 (citing the draft Inter-American Convention on Jurisdictional Immunity of States, 1983, reprinted at 22 I.L.M. 292 (not in force)); but see Barcelona Traction Light and Power Co. (Belg. v. Spain), Judgment, 2d Phase, 1970 I.C.J. 3, ¶ 61 (Feb. 5); Kuwait v. Am. Independent Oil Co. (Ad Hoc Trib. 1982), reprinted in 66 INT’L L. REPS. 518, 606-67. See also II YB Int’l L. Comm’n 368, para. 31. 161. See Jurisdictional Immunities of the State (Ger. v. It.), Judgment, 2012 I.C.J. 99, ¶¶ 55, 72-77 (Feb. 3), available at http://www.icj-cij.org/docket/files/143/16883.pdf; Military and Paramilitary Activities in and Against Nicaragua, 1986 I.C.J. at ¶ 212; Lotus (Fr. v. Turk.), 1927 P.C.I.J. (ser. A) No. 10, at 28, 30 (Sept. 7); Prosecutor v. Furundzˇjia, Case No. IT-95-17/1-T, Judgment, ¶¶ 168-82 (Int’l Crim. Trib. Former Yugoslavia Dec. 10, 1998) (citing The Yamashita Case 1945-1946: Decision of the United State Military Commission at Manila, December 7, 1945, reprinted in 2 THE LAW OF WAR 1597 (Leon Friedman ed., 1972)); Case C-366/10, Air Transport Ass’n of Am., ¶ 104; Racke, 1998 E.C.R. at ¶ 53. See also Lotus (Fr. v. Turk.), 1927 P.C.I.J. (ser. A) No. 10, at 29 (Sept. 7) (looking to states to object or lodge protests to practice not seen as consistent with customary international law). 162. See Jurisdictional Immunities of the State (Ger. v. It.), Judgment, 2012 I.C.J. 99, ¶¶ 55, 70-71 (Feb. 3), available at http://www.icj-cij.org/docket/files/143/16883.pdf; North Sea Continental Shelf (Ger. v. Den.; Ger.v. Neth.), Judgment, 1969 I.C.J. 3, 104-06 (Feb. 20); Nottebohm (Liech. v. Guat.), Judgment, 2d Phase, 1955 I.C.J. 4, 22 (Apr. 6); Fisheries (U.K. v. Nor.), 1951 I.C.J. 116, 134-36 (Dec. 18); Int’l L. Ass’n, Statement of Principles, supra note 14, princ. 9 cmt. (c). But see Prosecutor v. Furundzˇjia, Case No. IT-95-17/1-T, Judgment, ¶ 168 (Int’l Crim. Trib. Former Yugoslavia Dec. 10, 1998) (citing THE TOKYO JUDGEMENT: THE INTERNATIONAL MILITARY TRIBUNAL FOR THE FAR EAST 385 (B.V.A. Ro ¨ ling & C.F. Ru¨ter eds., 1977)). It is unclear whether the decisions of the IMT qualify as court judgments in this sense which would make them evidentiary of customary international law as a primary source or whether they are decisions of an international

2014]

487


nity163 (either in the context of the work of the ILC164 or in treaty negotiations165); the conclusions (be they Draft Articles or otherwise) of the ILC;166 and the teachings of publicists.167 What can be controversial for each category is whether or not that category of acts is evidentiary of the way states actually act168 or the opinions they hold.169 One difficulty with this step is that the process of establishing linguistic categories is always one of necessarily abstracting away from actual phenomena in order to better understand it.170 For example, we

tribunal which would make them subsidiary sources of customary international law or any other source of international law. Judgments of the national military commissions in German under the post-IMT Control Council Law No. 10 are even more difficult to classify as international or domestic. 163. See Gabcˇikovo-Nagymaros Project (Hung. v. Slovk.), Judgment, 1997 I.C.J. 7, 49-54 (Sept. 25); Legality of the Threat or Use of Nuclear Weapons, Advisory Opinion, 1996 I.C.J. 226, ¶¶ 86-88 (July 8); Military and Paramilitary Activities in and Against Nicaragua, 1986 I.C.J. ¶¶ 183207; Delimitation of the Maritime Boundary in the Gulf of Maine Area (Can. v. U.S.), Judgment, 1984 I.C.J. 246, ¶ 34 (Jan. 20); Continental Shelf Case (Tunisia/Libya), Judgment, 1982 I.C.J. Reps. 38, ¶ 24; Fisheries Jurisdiction (U.K. v. Iceland), 1974 I.C.J. 24, 25-26 (Feb. 2); North Sea Continental Shelf (Ger. v. Den.; Ger.v. Neth.), Judgment, 1969 I.C.J. 3, 25-26, 32-33 (Feb. 20); Nottebohm (Liech. v. Guat.), Judgment, 2d Phase, 1955 I.C.J. 4, 21-23 (Apr. 6); Lotus (Fr. v. Turk.), 1927 P.C.I.J. (ser. A) No. 10, at 23, 26-30 (Sept. 7); MARK VILLIGER, CUSTOMARY INTERNATIONAL LAW AND TREATIES 2019-20 (2nd ed. 1997); BROWNLIE, supra note 3, at 5; Int’l L. Ass’n, Statement of Principles, supra note 14, princ. 4 cmt. (a) (“Verbal acts, meaning making statements rather than performing physical acts, are in fact more common forms of State practice than physical conduct . . . . are all forms of speech-act.”). Also possibly including acts by individuals that are adopted by the state. See Hostages case, 1980 I.C.J. Reps. 33-5, ¶¶ 70-74. 164. See Jurisdictional Immunities of the State (Ger. v. It.), Judgment, 2012 I.C.J. 99, ¶ 55 (Feb. 3), available at http://www.icj-cij.org/docket/files/143/16883.pdf. 165. See id. 166. See id. ¶¶ 56-58 (citing 1980-II(2) YB Int’l L. Comm’n 147, ¶ 26). 167. See Lotus (Fr. v. Turk.), 1927 P.C.I.J. (ser. A) No. 10, at 25 (Sept. 7). 168. See Military and Paramilitary Activities in and Against Nicaragua, 1986 I.C.J. ¶¶ 191-92; see also Kirgis, supra note 14. 169. This problem is the basis for D’Amato’s distinction between an act and a statement, being indicative only of state practice and opinio juris respectively and exclusively. See D’AMATO, supra note 94, at 89-90 (arguing that treaties are actions). This categorical and conservative view is rejected by others. See RESTATEMENT (THIRD) OF THE FOREIGN RELATIONS LAW OF THE UNITED STATES § 102(2) (1987); BROWNLIE, supra note 3, at 5-6; Akehurst, supra note 13, at 2-3, 35, 43. See also North Sea Cont’l Shelf (Ger. v. Den.; Ger.v. Neth.), Judgment, 1969 I.C.J. 3, 32-33, 47, 53 (Feb. 20) (suggesting that D’Amato’s distinction is not being accepted). 170. See NOAM CHOMSKY, LANGUAGE AND RESPONSIBILITY 57 (1979) (“Opposition to idealization is simply objection to rationality; it amounts to nothing more than an insistence that we shall not have meaningful intellectual work. Phenomena that are complicated enough to be worth studying generally involve the interaction of several systems. Therefore you must abstract some object of study, you must eliminate those factors which are not pertinent. At least if you want to conduct an investigation which is not trivial. In the natural sciences this isn’t even discussed, it is self-evident.

488

[Vol. 45


might determine that “statements” made during treaty negotiations could qualify as evidence of state practice, opinio juris, or both, but that language category by itself is meaningless when separated from actual instances where communications are made. We must already have a notion of what a statement is, derived from examples of statements made in the past, that serves as the basis for the abstraction of statements generally. Once the category is made based on some predetermined ideal statement, then other communications are tested against the category to determine whether they qualify for inclusion or not. This could be achieved through an inductive step, but the ILC discussion on the forms of customary international law suggests that it is; however, the ILC does not appear to have applied an inductive approach in reaching its conclusion on the categories of evidence.171 Nonetheless, those categories appear to have been adopted by most authorities. A further difficulty with establishing categories is that they are abstractions of various manifestations of behavior formed by identifying certain commonalities amongst them that hold the group together as a category suitable for abstraction, but does not necessarily contemplate all commonalities. Thus these abstractions do not describe outlier events very well.172 For example, it has been argued that a belief in the

That is unfortunate. In the human sciences, people continue to question it. That is unfortunate. When you work within some idealization, perhaps you overlook something which is terribly important. That is a contingency of rational inquiry that has always been understood. One must not be too worried about it. One has to face this problem and try to deal with it, to accommodate oneself to it. It is inevitable.” (emphasis in original)). 171. See e.g. II YB Int’l L. Comm’n 368, para. 31 (“Evidence of the practice of States is to be sought in a variety of materials. The reference in article 24 of the Statute of the Commission to ‘documents concerning State practice’ (documents e´tablissant la pratique des Etats) supplies no criteria for judging the nature of such ‘documents’. Nor is it practicable to list all the numerous types of materials which reveal State practice on each of the many problems arising in international relations.”). 172. KING, ET AL., DESIGNING, supra note 39, at 10 (“One way to understand [complex, and in some sense unique,] events [with enormous ramifications] is by seeking generalizations: conceptualizing each case as a member of a class of events about which meaningful generalizations can be made. This method often works well for ordinary wars or revolutions, but some wars and revolutions, being much more extreme than others, are “outliers” in the statistical distribution.”). For example, the ICTY has cited to Control Council Law No. 10 as evidentiary, not conclusively, of customary international law. See Prosecutor v. Furundzˇjia, No. IT-95-17/1-T, Judgment, ¶ 168 (Int’l Crim. Trib. for the Former Yugoslavia Dec. 10, 1998). What is uncertain is whether this item of evidence qualifies as state practice in the sense of contributing to customary international law. It could be argued that Control Council Law No. 10 merely provides for the further prosecution of additional war criminals under customary international law, and thus is a

2014]

489


compelling nature of a practice must be “articulated” by the acting state in order to qualify as opinio juris.173 The reason for this argument is that the opinio juris we look for is not the state’s subjective beliefs but objective behavior evidencing subjective beliefs, which must be manifested in order to be objectively interpreted.174 This author disagrees with that articulation, in the sense that writing or speech, is necessary. A subjective belief need not be expressed objectively in writing or speech in order for us to attempt to objectively understand it. Writing or speech simply makes it easier to evidence and understand the belief objectively. For example, we have always accepted acquiescence as evidence of the formation of a custom, and acquiescence is, generally, understood to be an omission.175 The ILA has suggested that all of the eligible acts must be “public” in the sense of having been communicated to at least one other state to provide an opportunity for reaction176 and that omissions can constitute practice.177 In any event it is unclear whether there could be any agreement on the descriptive homogenization of all volitional manifestations into

claim to the existence of a rule under customary international law. It could be argued that it furthers the Charter of International Military Tribunal’s work and thus it is a quasi-treaty. It could be argued to be domestic legislation as evidence of a customary norm. These propositions are not necessarily wrong, but they do show that evidence that does not fall into the homogenized categories must be deductively analogized to evidence that does. 173. See, e.g., D’AMATO, supra note 94, at 35-39; Roberts, Traditional and Modern, supra note 8; Akehurst, supra note 13, at 36-37. 174. See, e.g., North Sea Continental Shelf (Ger. v. Den.; Ger.v. Neth.), 1969 I.C.J. 3, 176 (Feb. 20) (dissenting opinion of Judge Tanaka); D’AMATO, supra note 94, at 35-39; Roberts, supra note 8, at 757 (“Thus, actions can form custom only if accompanied by an articulation of the legality of the action.”); Akehurst, supra note 13, at 36-37. 175. We could go farther and distinguish purposive omissions resulting in acquiescence from inadvertent ones, leading us to view acquiescence as either an act (deliberate sitting on one’s hands) or omission, but that distinction does not seem to produce any help. 176. See Int’l L. Ass’n., Statement of Principles, supra note 14, princ. 5 cmt. (a) (“For a verbal act to count as State practice, it must be public - not in the sense that it need necessarily be communicated to all of the world, but that, if it is not publicized generally (e.g., by legislation, press statements, etc.), it must be communicated to at least one other State.”); but see the criticism of this view by this author, William Thomas Worster, The Effect of Leaked Information on the Rules of Customary International Law, 28 AM. UNI. INT’L L. REV. 443 (2013). 177. See Nottebohm (Liech. v. Guat.), Judgment, 2d Phase, 1955 I.C.J. 4, 22 (Apr. 6) (refraining from exercising diplomatic protection); Asylum (Colom. v. Peru), Judgment, 1950 I.C.J. 266, 277-78 (Nov. 20) (refraining from ratifying convention); Lotus (Fr. v. Turk.), 1927 P.C.I.J. (ser. A) No. 10, at 23, 28 (Sept. 7); Int’l L. Ass’n., Statement of Principles, supra note 14, princ. 6 & cmt; Maurice Mendelson, State Acts and Omissions as Explicit or Implicit Claims, in LE DROIT ´ ´ INTERNATIONAL AU SERVICE DE LA PAIX, DE LA JUSTICE ET DU DEVELOPPEMENT : MELANGES MICHEL VIRALLY 373 (1991).

490

[Vol. 45


these prescriptive categories leading to an understanding of the state’s practice.178 This problem is partly due to the differences in the contemporary and historical assessment of the meaning of certain acts,179 the difficulty of the abstraction of the mind of the state180 and the difficulty of the distinction between political positions of state organs and the “beliefs of state.”181 In addition, it is not clear that states always consider existing or forming customary international law when they act or that they truly engage in an offer-acceptance discussion.182

178. See Military and Paramilitary Activities in and Against Nicaragua (Nicar. v U.S.), Judgment, 1986 I.C.J. 14, ¶¶ 190, 207 (June 27); North Sea Continental Shelf (Ger. v. Den.; Ger.v. Neth.), Judgment, 1969 I.C.J. 3, 32-33 (Feb. 20); Asylum, 1950 I.C.J. at 277; Anglo-Norw. Fisheries Case, 1951 I.C.J. Reps. 191 (Read, J., dissenting op.); D’AMATO, supra note 94, at 50-51; HUGH W.A. THIRLWAY, INTERNATIONAL CUSTOMARY LAW AND CODIFICATION; AN EXAMINATION OF THE CONTINUING ROLE OF CUSTOM IN THE PRESENT PERIOD OF CODIFICATION OF INTERNATIONAL LAW 58 (1972); Pavoni, supra note 119 (arguing that the practice that the Court chose to include was only diplomatic practice, not the practice of courts: “Nor should one overlook that here the Court only told the diplomatic side of the story, something which may be taken to express a very controversial preference for the manifestations of practice coming from the executive branch of Governments.”); Strupp, Les re`gles geńe´rales, supra note 155; Akehurst, supra note 13. 179. See ANTHONY CARTY & RICHARD A. SMITH, SIR GERALD FITZMAURICE AND THE WORLD CRISIS: A LEGAL ADVISOR IN THE FOREIGN OFFICE 1932-1945 23-27 (2000) (“It is very difficult to discuss contemporaneous events for a number of reasons. The main one is the fact that those involved are usually still alive and may continue to be engaged in the very same events that are ongoing. Perspectives and opinions about the best course of action will remain openly contested. Furthermore, there will not usually be agreed objective and detached sources from which one can draw to determine the nature of the events. There will be much fresh, firsthand testimony, but it will be conflicting . . . . There is a second, equally important problem with the analysis of contemporary events in which states participate, and that is to know whether one can be sure of the factual circumstances which are supposed to justify the invocation of a norm.”). 180. See Maurice Mendelson, The Subjective Element in Customary International Law, 66 Brit. YB Int’l L. 177 (1995). Related to the problem of the abstraction of the mind of the state is the issue of the volume of practice and opinio juris of a state that remains secret. See Worster, Leaked Information, supra note 176. 181. See Mendelson, supra note 14, at 195-96; Akehurst, supra note 13, at 195 (“We cannot know what states believe. First of all states being abstractions or institutions do not have minds of their own; and in any case since much of the decision-making within governments takes place in secret, we cannot know what states (or those who speak for them) really think, but only what they say they think.”). 182. Int’l L. Ass’n., Statement of Principles, supra note 14, ¶ 2. Similarly, States (and other international actors) tend not to address themselves to the principles of customary law-formation in the abstract; and though politicians and officials sometimes make pronouncements on this subject, it is not always clear that this represents the considered position of their State . . . . An added difficulty is that of finding the material: most States do not publish digests of their practice in international law, and even those that do rarely classify much under the rubric of sources of law, as opposed to substantive topics. See LEPARD, supra note 122, at 103, 158-59 (applying game theory).

2014]

491


By way of example of the often deductive nature of selecting categories of acts, we can again consider the ILC’s draft articles on state responsibility, specifically the portion on attribution. Here we get an example of how the ILC views the evidence of customary international law. Most of the evidence of the rule of attribution constitutes cases of international courts or arbitral tribunals (and in one case a special commission, not a court or tribunal in the usual sense183), which might or might not truly evidence state practice.184 These items of evidence constitute subsidiary or persuasive sources of the law at best. However, if the ILC truly wanted to undertake an inductive approach with these subsidiary sources, it would have had to take the next step, which is to examine the cited cases to determine if they were decided following inductive study. Otherwise, the ILC would be conducting an inductive study of deductive studies. Continuing with the theme of subsidiary sources, the ILC also cited a few jurists, though this was relied on considerably less than case law, though the same concerns as with case law apply here.185 In addition, the ILC referred to several treaty

183. See Tellini case (Ital./Gr. Spec. Comm’n Jurists, 1923), reprinted at L.N.O.J., 4th Yr., No. 11, 1349 (Nov. 1923). 184. See Difference Relating to Immunity from Legal Process of a Special Rapporteur of the Commission on Human Rights, Advisory Opinion, 1999 I.C.J. 62, ¶ 62 (Apr. 29); Elettronica Sicula S.p.A. (U.S. v. It.), Judgment, 1989 I.C.J. 15, ¶ 73 (July 20); Military and Paramilitary Activities in and Against Nicaragua (Nicar. v U.S.), Judgment, 1986 I.C.J. 14, ¶¶ 283, 292 (June 27); Legal Consequences for States of the Continued Presence of South Africa in Namibia (South-West Africa) Nothwithstanding Security Resolution 276 (1970), Advisory Opinion, 1971 I.C.J. 16, 54, ¶ 118 (June 21); Treatment of Polish Nationals & Other Persons of Polish Origin or Speech in the Danzit Territory, 1932 P.C.I.J. (ser. A/B) No. 44, at 4, 24-25 (Feb. 4); German Interests in Polish Upper Silesia (Germ. v. Pol.), 1926 P.C.I.J. (ser. A) No. 7 (May 25); German Settlers in Poland., Adv. Op., 1923 P.C.I.J. (ser. B) No. 6, at 22 (Feb. 3); Report of the Panel, Japan—Trade in Semi-conductors, ¶¶ 110-11, L/6309 (Mar. 24, 1988), GATT B.I.S.D. (35th Supp.) at 116 (1989); Panel Report, Japan—Measures Affecting Consumer Photo Film & Paper, WT/DS44/R (March 31, 1998), ¶¶ 10.12-10.16; Panel Report, Turkey—Restrictions on Imports of Textile & Clothing Products, Case No. WT/DS34/R (May 31, 1999), ¶¶ 9.33-9.44; Xhavara v. Italy & Albania, Appl. No. 39473/98 (Eur. Ct. H.R., Jan. 11, 2001); Swedish Engine Drivers’ Union v. Sweden, 20 Eur. Ct. H.R. (ser. A) 14 (1976); Schmidt & Dahlstro¨m v. Sweden, 21 Eur. Ct. H.R. (ser. A) 15 (1976); X & Y v. Switzerland, Apps. Nos. 7289/75 & 7349/76, 1977 Y.B. Eur. Conv. on H.R. 373 (Eur. Comm’n on H.R.); Prosecutor v. Tadic´, Case No. IT-94-1-l, Decision on Defence Motion for Interlocutory Appeal on Jurisdiction, ¶ 117 (Int’l Crim. Trib. for the Former Yugoslavia, Oct. 2, 1995); Yeager v. Iran, 17 Iran-U.S. Cl. Trib. Rep. 92, 101-02 (1987); Diffe´rend Dame Mosse´, XIII R.I.A.A. 486, 492-93 (Fr.-It. Conciliation Comm’n 1953); Diffe´rend He´ritiers de S.A.R. Mgr. le Duc de Guise, XIII R.I.A.A. 150, 161 (Fr.-It. Conciliation Comm’n 1951); Pellat case (Fr.-Mex. Cl. Comm’n). 185. See INT’L COMM. RED CROSS, Commentary on the Addition of Protocols of 8 June 1977 to the Geneva Convensions of 12 August 1949, 1053-54 (1987); J. A. FROWEIN, DAS DE FACTO-REGIME IM

492

[Vol. 45


provisions when examining the issue of attribution, apparently applying a deductive methodology.186 On the other hand, the ILC did, on a few select occasions, cite to domestic cases, legislation, and statements of states either outside or within international organizations, which would be inductive examination of non-subsidiary evidence of customary international law.187 However, in some of those cases, the ILC admitted to citing them for their conclusions on immunity, not attribution, which could then inform the conclusion on attribution by deductive analogy.188 In sum, this ILC analysis mixes inductive and deductive analysis in establishing its categories of acts that are eligible for study.

¨ VOLKERRECHT 70-71 (1968); H. Atlam, National liberation movements and international responsibility, in UNITED NATIONS CODIFICATION OF STATE RESPONSIBILITY 35 (B. Simma & M. Spinedi eds., 1987). 186. See, e.g., Agreement Relating to Appeals to the High Court of Australia from the Supreme Court of Nauru, Austl.-Nauru, Sep. 6, 1976, 1216 U.N.T.S. 151; Convention for the Protection of the World Cultural and Natural Heritage art. 34, opened for signature Nov. 16, 1972, 1037 U.N.T.S. 151, (entered into force Dec. 17, 1975); Convention (IV) Respecting the Laws and Customs of War on Land, opened for signature Oct. 18, 1907, 36 Stat. 2277, 205 Consol. T.S. 277 (entered into force Jan. 26, 1910); Convention (IV) Respecting the Laws and Customs of War on Land, Regulations art. 2, opened for signature Oct. 18, 1907, 36 Stat. 2277, 205 Consol. T.S. 277 (entered into force Jan. 26, 1910); Convention (II) with Respect to the Laws and Customs of War on Land, opened for signature July 29, 1899, 32 Stat. 1803, 187 Consol. T.S. 429 (entered into force Sept. 4, 1900); Convention (II) with Respect to the Laws and Customs of War on Land, Regulations art. 2, opened for signature, July 29, 1899, 32 Stat. 1803, 187 Consol. T.S. 429 (entered into force Sept. 4, 1900); Convention (III) Relative to the Treatment of Prisoners of War art. 4(A)(6), opened for signature Aug. 12, 1949, 75 U.N.T.S. 135 (entered into force Oct. 21, 1950). ´ ´ ERAL 187. See CONSTITUTION FED [CST] [CONSTITUTION] Apr. 18, 1999, SR 101, arts. 56(3), 172(3); Preparatory Comm. For the Conference for the Codification of Int’l Law, Responsibility of States for Damage caused in their Territory to the Person or Property of Foreigners, League of Nations Doc. C.75.M.69.1929.V, 16, 25, 41, 52, 74, 90 (1926); Replies made by the Governments to the Schedule of Points; Replies of Canada and the United States of America, League of Nations Doc. C.75(a)M.69(a).1929.V, 2-3, 6, 17-18, 20 (1926); LON Doc. C.351(c)M.145(c).1930.V, 90, 92, 108, 116, 118, 237; Duke Almodo´var del Rıó, Note verbale (July 4, 1898), reprinted in ARCHIVIO DEL MINISTERO DEGLI AFFARI ESTERI ITALIANO, ser. politica P, No. 43; Statement of United States Secretary of State Regarding “American Bible Society” Incident (Aug. 17, 1885), in 4 A DIGEST OF INTERNATIONAL LAW 743 (John Bassett Moore & Francis Wharton eds., 1906); Namib. v. Mwandinghi, 1992(2) S. Afr. L. Reps. 355, 360 (Namib. 1991), reprinted at 91 INT’L L. REPS. 341. However, in examining domestic case law, the ILC found inconsistency, see 44123 Ontario Ltd. v. Kiyonga, 11 Kampala L. Reps. 14, 20-21 (High Ct., Uganda, 1992), reprinted at 103 INT’L L. REPS. 259, 266. 188. See, e.g., Church of Scientology case, Case No. VI ZR 267/76, NEUE JURISTICHE WOCHENSCHRIFT [NJW], 21, 1101, 1979 (Ger.) (May 1979) (Fed. Sup. Ct., Germ., Sept. 26, 1978), reprinted at 65 INT’L L. REPS. 193; Propend Fin. Pty Ltd. v. Sing (Ct. Appl., Engl., 1997), reprinted at 111 INT’L L. REPS. 611.

2014]

493


3.

Quality of the Acts: Density, Uniformity, Consistency, Extensitivity, and Representativity

In addition to requiring evidence of certain types, courts also look for evidence of sufficient density,189 uniformity,190 and consistency,191 and that the practice is extensive192 and representative.193 What is unclear for all of these requirements is whether these are questions of “admissibility” of the evidence of custom or assessments of the “weight” of the evidence of custom. That is to say, it is not clear whether certain acts would not be accepted as evidentiary because they were not dense enough or whether the acts would be accepted as evidentiary because they fall within the list of acceptable evidence of custom and the lack of density would go to the weight accorded to them. Furthermore, courts are also not careful to articulate whether these various factors of density, uniformity, etc. are applied as mandatory aspects or contributing factors, though this question will be dealt with in more detail in the section on assessing the weight of the evidence. Lastly, a combination of several types of acts might be necessary. For example in the Interlocutory Decision on the Applicable Law: Terrorism, Conspiracy, Homicide, Perpetration, Cumulative Charging (Terrorism Decision) in the Prosecutor v. Ayyash et al. case,194 the Special Tribunal for Lebanon (STL) Appeals Chamber based its conclusion that there was a definition of terrorism only on the decisions of courts

189. See North Sea Continental Shelf (Ger. v. Den.; Ger.v. Neth.), Judgment, 1969 I.C.J. 3, 42-43 (Feb. 20). 190. See, e.g., Military and Paramilitary Activities in and Against Nicaragua (Nicar. v U.S.), Judgment, 1986 I.C.J. 14, ¶ 186 (June 27); Fisheries (U.K. v. Nor.), 1951 I.C.J. Reps. 116, 131, 138; Asylum (Colom. v. Peru), Judgment, 1950 I.C.J. 266, 277-78 (Nov. 20). 191. See Fisheries (U.K. v. Nor.), 1951 I.C.J. Reps. 116, 131; Asylum, (Colum. V. Peru) 1950 I.C.J. at 277-78. 192. See North Sea Continental Shelf (Ger. v. Den.; Ger.v. Neth.), Judgment, 1969 I.C.J. 3, 32-33, 47, 53 (Feb. 20). See also Mendelson, supra note 14, at 172-76. 193. See North Sea Continental Shelf (Ger. v. Den.; Ger.v. Neth.), Judgment, 1969 I.C.J. 3 (Feb. 20); Nottebohm (Liech. v. Guat.), Judgment, 2d Phase, 1955 I.C.J. 4 (Apr. 6); S.S. “Wimbledon” (U.K., Fr., It. & Japan v. Ger.), 1923 P.C.I.J. (ser. A) No. 1 (Aug. 17); Lotus (Fr. v. Turk.), 1927 P.C.I.J. (ser. A) No. 10 (Sept. 7) (not requiring evidence of the practice of the states in question, but merely representative practice of states generally). Similarly, customary international law will apply to new states when they emerge as such, regardless of consent to the rule. See Right of Passage Over Indian Territory (Port. v. India), Judgment, 1960 I.C.J. 6 (Apr. 12); Certain Ger. Interests in Pol. Upper Silesia, 1926 P.C.I.J. (ser. A) No. 7, at 22, 42 (May 25); Ger. Settlers in Pol. case, Advisory Opinion, 1923 P.C.I.J. (ser. B) No. 6, at 36 (Sept. 10). 194. See Prosecutor v. Ayyash, Case No. STL-11-01/I, Interlocutory Decision on the Applicable Law: Terrorism, Conspiracy, Homicide, Perpetration, Cumulative Charging, ¶ 86 (Spec. Trib. for Lebanon Feb. 16, 2011).

494

[Vol. 45


of Argentina,195 Canada,196 Italy,197 Mexico198 and the U.S.199 It then acknowledged that: [in order to determine] whether a customary rule of international law has crystallised one must also delve into other elements. In particular, one must look to the behaviour of States, as it takes shape through agreement upon international treaties that have an import going beyond their conventional scope or the adoption of resolutions by important intergovernmental bodies such as the United Nations, as well as the enactment by States of specific domestic laws and decisions by national courts.200 The use of “one must also” suggests that customary international law cannot only be evidenced by one type of act— domestic court judgments—alone, but must be supported by more evidence or other forms of evidence. What is unclear is whether the STL understood that

195. See Corte Suprema de Justicia de la Nacioń [CSJN] [National Supreme Court of Justice], 28/8/2004, “Enrique Lautaro Arancibia Clavel case, s/ homicidio calificado y asociacioń ilıćita y otros—,” La Ley [L.L] (2004-F-292) (Arg.). 196. See Zrig v. Canada, [2002] 1 F.C. 559 (Can.); Suresh v. Canada, [2002] 1 S.C.R. 3 (Can.). 197. See Bouyahia Maher Ben Abdelaziz et al., Judgment No. 1072, Cass. crim., sez. I (Ct. Cass, Ital., Jan. 17, 2007). 198. See Cavallo, No. 140/2002 (Sup. Ct. Mex., June 10, 2003). 199. See Almog v. Arab Bank, 471 F. Supp. 2d 257, 284 (E.D.N.Y. 2007); United States v. Yunis, 924 F.2d 1086 (D.C. Cir. 1991). 200. See Prosecutor v. Ayyash, Case No. STL-11-01/I, Interlocutory Decision on the Applicable Law: Terrorism, Conspiracy, Homicide, Perpetration, Cumulative Charging, ¶ 87 (Spec. Trib. for Lebanon Feb. 16, 2011). One drawback to the approach of the STL is the easy mixing of methodologies for customary international law and general principles of law, leading us to wonder which source is being examined at each turn. This practice is reminiscent of the ICTY, which the STL cites for support of this approach. See id. ¶¶ 91-99 (“Consistent national legislation can be another important source of law indicative of the emergence of a customary rule.”) (citing Prosecutor v. Furundzˇija, Case No. IT-97-17/1-A, Judgment, ¶¶ 177-78 (Int’l Crim. Trib. for the Former Yugoslavia Dec. 10, 1998)); Prosecutor v. Kunarac, Case No. IT-96-23, Appeal Judgment, ¶ 439 (Int’l Crim. Trib. for the Former Yugoslavia, Feb. 22, 2001). See also Prosecutor v. Erdemovic´, Case No. IT-96-22-A, ¶ 57 (Int’l Crim. Trib. for the Former Yugoslavia, Oct. 7, 1997) (joint separate opinion of Judge McDonald and Judge Vohrah) (“[I]t is generally accepted that [a] comprehensive survey of all legal systems of the world [is not required,] as this would involve a practical impossibility and has never been the practice of the International Court of Justice or other international tribunals which have had recourse to Article 38(I)(c) of the ICJ Statute.”). Perhaps it would be better if the ICTY and STL accepted the invitation of Simma and Alston and acknowledged that they were in fact searching for general principles of international law. See Simma & Alston, supra note 8, at 88, 96, 102-06.

2014]

495


selecting various types of acts was necessary before it could find customary international law acts or whether for the question asked, the tribunal simply needed more evidence of practice and looked elsewhere to find it, or that the tribunal was already satisfied with the proof of customary international law and was only citing additional evidence to buttress the judgment against the expected criticism of the judgment that would follow. E.

Sampling

After determining what acts or facts constitute suitable evidence for sampling, and what the hypothesis is, a court examining a question of customary international law proceeds to actual sampling, i.e., selecting evidence submitted to it that is or is not probative. In forming the sample pool, courts conducting such an exercise are unclear what size of a sample pool is required for any given rule of customary international law.201 Thus, if they are truly engaging in an inductive exercise, courts could be criticized for potentially creating a biased, non-random pool of data that is self-fulfilling, i.e., they only select instances of state practice that support the rule they wish to find.202 Riccardo Pavoni argues that, while the Court did refer to practice (in this situation case law) from Canada, France, Slovenia, New Zealand, Poland and the United Kingdom, it missed case law from the U.S. He concludes that this omission was deliberate, since U.S. practice pointed in the opposite direction of the Court’s finding on the law.203 Pavoni argues that had the Court actually selected a data pool of state practice that represented the practice in the real world, it would have been forced to conclude other than it wanted. Courts do not seem to make use of any randomization techniques at all, let alone a computer-generated randomization. Certainly traditional and modern customary international law agree on this point. Without a prior randomized decision on the sampling method, the sampling might also fall victim to the problem of only surveying data

201. See Worster, supra note 74. 202. See, e.g., Stanford v. Kentucky, 492 U.S. 361, 370 n.1 (1989) (While “[t]he practices of other nations, particularly other democracies, can be relevant to determining whether a practice uniform among our people is not merely a historical accident, but rather so ‘implicit in the concept of ordered liberty’ that it occupies a place not merely in our mores, but, text permitting, in our Constitution as well, . . . they cannot serve to establish the first Eighth Amendment prerequisite, that the practice is accepted among our people.” (internal citations omitted)). 203. See Pavoni, supra note 119 (criticizing the Court for failing to cite or distinguish U.S. practice).

496

[Vol. 45


that is readily available or that merely gives legal force to existing global power relations.204 This problem would most likely favor the practice, not of states generally in a “utopia-apology” sense, but of states with sufficient means to document and diffuse their practice.205 This author has elsewhere argued that this potential weakness might be explained by re-understanding the “specially interested” state as the states most active in diffusing norms important to them.206 Some authors have also suggested that the only states that should be sampled are those that generally respect certain fundamental ethical principles,207 or that truly understand customary international law (e.g., in the decisions of their courts).208 It has also been argued that the selection of states for sampling purposes should not be random, but representative, i.e., that the states should be purposefully selected in order to create the best generalizable abstraction of practice and opinio juris on a global scale.209 This can be interpreted to mean diversity.210 All of these views argue for

204. See Worster, supra note 74; see also Treves, supra note 120, at 78. 205. See Worster, supra note 74; see also SCHACHTER, supra note 8, at 35-36; KOSKENNIEMI, FROM APOLOGY TO UTOPIA, supra note 1, at 355; MOHAMMED BEDJAOUI, TOWARD A NEW INTERNATIONAL ECONOMIC ORDER 51-52 (1979); Int’l L. Ass’n., Statement of Principles, supra note 14, ¶ 2; Kelly, Twilight, supra note 17, at 469, 520-21; Antonio Cassese, Ex iniuria ius oritur: Are We Moving Towards International Legitimation of Forcible Humanitarian Countermeasures in the World Community?, 10 EUR. J. INT’L L. 23 (1999); Jonathan I. Charney, Anticipatory Humanitarian Intervention in Kosovo, 93 AM. J. INT’L L. 834, 835-37 (1999); Tasioulas, Relative Normativity, supra note 14, at 121-23; Oscar Schachter, New Custom: Power, Opinio Juris and Contrary Practice, in THEORY OF INTERNATIONAL LAW AT THE THRESHOLD OF THE 21ST CENTURY: ESSAYS IN HONOUR OF KRZYSZTOF SKUBISZEWSKI 531, 536 (Jerzy Makarczyk ed., 1996); Charney supra note 13, at 537, 544-45; Chodosh, supra note 19, at 102. 206. See Worster, supra note 74. 207. See LEPARD, supra note 122, at 155. 208. See, e.g., Pinochet III, 2 W.L.R. 827, 866 (H.L. 1999) (opinion of Lord Hope). 209. Representativity has been understood to involve a cultural, regional diversity of states whose practice is examined. See Charney, supra note 13, at 543-45 (arguing that larger number and diversity of states, addressing interconnected global problems necessitates a modified view on customary international law). 210. See North Sea Continental Shelf (Ger. v. Den.; Ger. v. Neth.), 1969 I.C.J. 3, 227 (Feb. 20) (dissenting opinion of Judge Lachs); 2 ICRC CUSTOMARY IHL, supra note 120, at xliv-xlv, li; SCHACHTER, supra note 8, at 36; KOSKENNIEMI, FROM APOLOGY TO UTOPIA, supra note 1, at 355 (arguing that the nature of the formation of customary international law acts to apologize for the behavior of more powerful Western states); BEDJAOUI, supra note 205, at 51-52; Int’l L. Assoc., supra note 14, princ. 14 cmt. (d)-(e); Treves, supra note 120, at 35 (“It has been argued that this generality must include States representing the main legal, economic, and political orientations and geographical areas”). However, it may be that the representative quality is not so crucial, but the authority of the actor. See Duncan Hollis, Why State Consent Still Matters: Non-State Actors, Treaties, and the Changing Sources of International Law, 23 BERKELEY J. INT’L L. 137 (2005) (identifying authority as playing a far

2014]

497


a study that is not a purely random selection of state practice. However, this explicitly rejects the approach of scientific studies that deliberately choose randomization as the best way to achieve representativity. This conclusion means that the supposedly inductive study that results was the result of a study structure built on deduction. It could be that studies on customary international law that apply this method are simply employing a stratified sampling method to ensure that the sample pool draws upon certain factors.211 The need for distribution of samples over various geographic areas and legal systems suggests that stratified sampling would be the method most appropriate for an analysis of customary international law.212 However, even in a stratified sampling method, only the broad classifications are established (e.g., individuals within certain racial groups, professions, zip codes, etc.), whereas the precise instances (e.g., Tom, Dick, and Harry) are still subjected to randomized sampling, which no studies of customary international law appear to employ. In any event, taking a stratified sampling approach clearly privileges certain values over others at the outset,213 and limits the sampling pool. Where there is a distribution of selection across geography, the value selected is geography. It is not clear that the location of a state at a certain place in the globe is a helpful modification to the sampling, though it is aiming at producing a universal rule. Limiting the pool to other criteria, such as general compliance with human rights, operates quite differently. At its logical extreme, the restricting of the potential sampling pool assures that only certain states can be selected and fatally undermines the random nature of the sampling process, if we agree that universality is an objective of the sampling. If this limitation on sampling is indeed the case, and it appears that it is, then the process is not truly inductive. Instead it is a balancing of inductive and deductive processes. Inductive in the sense that a random sampling is being used; deductive in that the states eligible for selection are being limited based on value judgments deduced from fundamental principles. Furthermore, the step of moving from the samples that are in hand to a general rule can still remain an inductive

more prominent role in adoption of norms than sources theory has yet acknowledged and bringing non-state actors into the authority discussion). 211. See Yates et al., supra note 70. 212. See North Sea Continental Shelf (Ger. v. Den.; Ger.v. Neth.), Judgment, 1969 I.C.J. 3, 227 (Feb. 20) (dissenting opinion of Judge Lachs); 2 ICRC CUSTOMARY IHL, supra note 120, at li. 213. See Martti Koskenniemi, “Intolerant Democracies”: A Reaction, 37 HARV. INT’L L. J. 231, 232-33 (1996).

498

[Vol. 45


one, but the assembling of the data pool is not. The act of assembling the sampled data also involves deductive steps. The actor must assess two things. First, the actor must determine whether the identified act is an act that qualifies as potentially evidencing state practice. This step can be straight-forward, e.g., if domestic judgments can evidence state practice, then the particular evidentiary material presented or located is, in fact, a judgment. While this step is usually obvious in the case of Supreme Court judgments or other bodies that are clearly judicial organs, it is less so for other items, e.g., disciplinary proceedings under a domestic professional licensing scheme run by a private, membership-based organization that enjoys a public monopoly on the practice of the profession.214 We also do not usually examine the judicial quality of the decisions to determine whether they truly qualify as decisions of courts. For example, we could examine the impartiality and independence of judges;215 observation of fair trial rights (e.g., right to submit an argument); the judicial appointing authority, mechanisms and oversight, etc., to determine if the body was

214. See, e.g., ILC, Draft articles on Responsibility of States, supra note 155, art. 5 (“The conduct of a person or entity which is not an organ of the State under article 4 but which is empowered by the law of that State to exercise elements of the governmental authority shall be considered an act of the State under international law, provided the person or entity is acting in that capacity in the particular instance.”) (citing, inter alia, Hyatt Int’l Corp. v. Iran, 9 Iran-U.S. C.T.R. 72, 88-94 (Iran-U.S. Cl. Trib., 1985); CODIFICATION CONFERENCE, BASES OF DISCUSSION, Responsibility of States, supra note 187, at 90. In the context of the WTO Agreements, it needed to be clarified that measures undertaken by these types of bodies are measures subject to the WTO Agreements and thus for which the Member State could be found responsible. See General Agreement on Trade in Services art. I(3)(a), Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 1B, 1867 U.N.T.S. 154. Of course, clarifying that question suggests that there was a question on the matter to begin with. 215. See, e.g., Kyprianou v. Cyprus, 2005-XIII Eur. Ct. H.R. 47; Assanidze v. Georgia, 2004-II Eur. Ct. H.R. 155; Sacilor Lormines v. France, 2006-XII Eur. Ct. H.R. 121; Sramek v. Austria, Appl. No. 8790/79, ¶ 42, (Eur. Ct. Hum. Rts., Oct. 22, 1984); Piersack v. Belgium, Appl. No. 8692/79, ¶¶ 43-4 (Eur. Ct. Hum. Rts., 1982); see also Gonzales del Rio v. Peru, Inter-Am. Ct. H.R. (ser. C) No. 263/1987, ¶ 52 (Oct. 28, 1992); Prosecutor v. Furundzˇija, Case No. IT-97-17/1-A, Judgment, ¶¶ 177, 189, 191 (Int’l Crim. Trib. for the Former Yugoslavia July 21, 2000). Even where the ECHR found the cases to be violations of the right to a fair trial (impartiality and independence of judges in this case), the court never contemplated that these were not judicial decisions (though the ECHR never reached the further question of whether these impugned judicial decisions could contribute to customary international law). It might risk creating a circular analysis where court decisions that do not provide a fair trial are held to not be judicial decisions, which in turn means that they are not tested for compliance with fair trial protections in the first place.

2014]

499


truly judicial, and thus qualifying as evidencing state practice.216 This inquiry could be completed through induction, but it is usually not.217 Generally, courts examining this question apply an apparently implied deductive approach where such matters are excluded or at least that the Court remains willfully blind to these concerns. Second, the actor must determine whether an act that does qualify is probative of state practice and/or opinio juris.218 We do not assess multiple instances where certain evidentiary items were or were not probative to establish a rule on what is or is not probative, and then apply that rule to the evidentiary material before the court. Instead the court first looks at the material itself and deduces whether or not the material, based on its inherent qualities, can be placed on one or the other side of the scale. While it has been argued that the International Criminal Tribunal for the former Yugoslavia (ICTY) in the Tadic´ decision on jurisdiction was relying more heavily on opinio juris than practice, and thus employing the “modern” approach to customary international law, the tribunal actually appears to be saying that it was searching for better quality evidence of the true practice of states.219 In the Terrorism decision, the STL considered the apparent motivations of the state in manifesting its practice220 and articulations

216. Note that in the U.S. many judges are elected into office by universal suffrage, see generally CHRIS W. BONNEAU & MELINDA GANN HALL, IN DEFENSE OF JUDICIAL ELECTIONS (2009), and that the Lord Chancellor in the U.K. was, formerly, simultaneously the presiding officer of the upper chamber of Parliament, and the nation’s highest judicial appointing authority, see Constitutional Reform Act, 2005, c. 4, pt. 2 (U.K.). Of course this aspect of the Lord Chancellor is historically based in the ancient nature of the office, arising at a time when judicial, legislative and executive roles and functions where more blurred. 217. See Jurisdictional Immunities of the State (Ger. v. It.: Greece Intervening), Judgment, 2012 I.C.J. 99, 143, ¶¶ 55, 72-77 (Feb. 3). 218. See Pinochet III, 2 W.L.R. 827 (making an assessment of whether the particular practice is probative of the practice at issue) (“Finally, none of the international criminal court decisions (e.g., of Nuremberg and Tokyo) ‘deal[s] with the question of the immunities of incumbent Ministers for Foreign Affairs before national courts where they are accused of having committed war crimes or crimes against humanity . . . .’”). 219. See Prosecutor v. Tadic´, Case No. IT-94-1-l, Decision on Defence Motion for Interlocutory Appeal on Jurisdiction, ¶ 99 (Int’l Crim. Trib. for the Former Yugoslavia, Oct. 2, 1995) (arguing that the unreliability of certain sources of evidence of state practice during wartime demands that the tribunal rely more heavily on alternate sources, including official statements, military manuals and judicial decisions, among others). 220. See Prosecutor v. Ayyash et al., Case No. STL-11-01/I/AC/R17bis, Interlocutory Decision on Applicable Law: Terrorism, Conspiracy, Homicide, Perpetration, Cumulative Charging, ¶ 91 (Spec. Trib. for Lebanon Feb. 16, 2011) (“As was rightly noted by a great authority in international law, Dionisio Anzilotti, ‘laws that ensure a certain conduct of a State towards other States and which are not motivated by special interests of that State (for as a rule no State does for

500

[Vol. 45


of the practice as being of universal concern.221 Thus, the overall process was still an inductive search for evidence, but the decision on what sources constituted better quality evidence was deductive.222 Again, this is not to say that induction could not be used in deciding on the rules regarding better quality evidence, but simply that it was not used. F.

Assessing the Sufficiency of the Evidence

Having established the sample pool of data, the court will then proceed to examine what has been determined to be probative evidence of state practice and opinio juris. This is where the court can, or may, truly move to induction. The court will look at multiple instances of a certain rule being proposed, followed, or protested to determine whether the sample pool is widespread and consistent enough to constitute a rule predicting future outcomes. Interestingly some courts refer to the examination of state practice as a deductive step,223 though it is not entirely clear why they would, other than the identification of the deductive aspects of the analysis that this Article highlights. As noted above, it appears that courts and tribunals consider the density, uniformity, consistency, “extensivity” and representativity in this step of assessing the sufficiency of the evidence rather than

the other States, without gaining any advantage, more than it believes it must do)’ constitute ‘very important indicia about the existence of a customary rule . . . .’” (citing 1 ANZILOTTI, COURS, supra note 155, at 108; 1 P. GUGGENHEIM, TRAITE DE DROIT INTERNATIONAL PUBLIC 51 (1953))). 221. Id. (“However, the mere existence of concordant laws does not prove the existence of a customary rule, ‘for it may simply result from an identical view that States freely take and can change at any moment.’ Thus, for instance, the fact that all States of the world punish murder through their legislation does not entail that murder has become an international crime. To turn into an international crime, a domestic offence needs to be regarded by the world community as an attack on universal values (such as peace or human rights) or on values held to be of paramount importance in that community; in addition, it is necessary that States and intergovernmental organisations, through their acts and pronouncements, sanction this attitude by clearly expressing the view that the world community considers the offence at issue as amounting to an international crime.” (citation omitted)). 222. See Military and Paramilitary Activities in and Against Nicaragua (Nicar. v. U.S.), 1986 I.C.J. 14 (June 27); see also Treves, supra note 120, ¶ 30 (“Particularly significant are manifestations of practice that go against the interest of the State from which they come, or that entail for them significant costs in political, military, economic or other terms, as it is less likely that they reflect reasons of political opportunity, courtesy etc.”). 223. See Pinochet case, ¶ 58 (“[The Lords] had been unable to deduce from this practice that there exists under customary international law any form of exception to the rule according immunity from criminal jurisdiction and inviolability to incumbent Ministers for Foreign Affairs where they are suspected of having committed war crimes or crimes against humanity . . . .”).

2014]

501


assessing the admissibility of the evidence. The ICJ has stated that Although the passage of only a short period of time is not necessarily, or of itself, a bar to the formation of a new rule of customary international law on the basis of what was originally a purely conventional rule, an indispensable requirement would be that within the period in question, short though it might be, State practice, including that of States whose interests are specially affected, should have been both extensive and virtually uniform in the sense of the provision invoked . . . .224 Although we could read the analysis above in a very restrictive sense, requiring both “extensive and virtually uniform” practice only in cases of customary rules that are built on a basis of a conventional rule or only customary rules of limited pedigree, the language above has not been interpreted as such. It does appear that the ICJ understands that the extensive and uniform requirement goes to the weight of the evidence rather than the admissibility thereof because the Court links those requirements to the “formation of a new rule of customary international law” rather than to evidencing that such a rule already exists.225 In particular, the PCIJ held in the Wimbledon case that a customary practice had been shown by reference to extremely limited practice.226 Had the above factors gone to the admissibility of the evidence, the Court should have refused to consider the evidence on the grounds that it was not extensive. Instead, the Court did not refuse to consider the evidence, implicitly ruling that the evidence was admissible, and that act suggests that the Court later weighed the factors in the assessment of the sufficiency of the evidence.

224. See North Sea Continental Shelf (Ger. v. Den.; Ger. v. Neth.), 1969 I.C.J. 3, 43 (Feb. 20). 225. Similar conclusions can be reached by examining other statements of the Court. See, e.g., Military and Paramilitary Activities in and Against Nicaragua (Nicar. v. U.S.), 1986 I.C.J. 14, ¶ 185 (June 27); Fisheries Case (U.K. v. Nor.), 1950 I.C.J. Reps. 116, 131, 138; Reservations to Convention on Prevention and Punishment of Crime of Genocide, Advisory Opinion, 1951 I.C.J. 15, 25 (May 28). 226. See S.S. “Wimbledon” (U.K., Fr., It. & Japan v. Ger.), 1923 P.C.I.J. (ser. A) No. 1 (Aug. 17). See also Nottebohm (Liech. v. Guat.), 1955 I.C.J. 4 (Apr. 6) (leaving unclear what evidence was relied upon); Asylum (Colom. v. Peru), 1950 I.C.J. 266, 278 (Nov. 20) (internal inconsistencies discounted); Lotus (Fr. v. Turk.), 1927 P.C.I.J. (ser. A) No. 10 (Sept. 7) (viewing limited practice as reliably representative of the rule).

502

[Vol. 45


1.

Inductive Analysis

It is only here that the arguments about traditional and modern customary international law analysis take on significance insofar as that discussion focuses on the relationship between induction and deduction, practice and opinio juris.227 For example, the STL argued that there could be a presumption in favor of opinio juris supporting custom where there was concordant practice.228 Furthermore, the distinction appears to focus even more narrowly on cases where there may be practice but no opinio juris, or the reverse. Where practice and opinio juris run together, the traditional and modern debate has less importance. Furthermore, the distance between modern and traditional customary international law analysis, and between state practice and opinio juris, presumes that the two are elements of a rule, not factors. If they are factors, then heavy reliance on one when the other is weaker is perfectly acceptable. It does appear that the Court usually requires proof of both which is not the approach we would expect for factors,229 however, that approach has not necessarily been consistent.230 Based on the above, it would appear that practice and

227. See North Sea Continental Shelf (Ger. v. Den.; Ger.v. Neth.), Judgment, 1969 I.C.J. 3, 44 (Feb. 20) (dissenting opinion of Judge Sørensen); D’Amato, supra note 8 (arguing that the modern approach upsets the traditional priority of practice over opinio juris); Jennings, Identification, supra note 19, at 3; Simma & Alston, supra note 8, at 83, 88, 96, 102-06; Bodansky, supra note 21, at 116-19; Charney, supra note 13, at 543-47; Chodosh, supra note 19; Charlesworth, supra note 19; Kirgis, supra note 14, at 149; Tasioulas, Relative Normativity, supra note 14; Kelly, Twilight, supra note 17, at 451. 228. See Prosecutor v. Ayyash, Case No. STL-11-01/I, Interlocutory Decision on the Applicable Law: Terrorism, Conspiracy, Homicide, Perpetration, Cumulative Charging, ¶ 101 (Spec. Trib. for Lebanon Feb. 16, 2011) (citing M. Sørensen, Principes de droit international public, 1976-III REC. DES COURS HAGUE ACAD. INT’L L. 51; Int’l L. Ass’n, Statement of Principles, supra note 14, at 40 (holding that “if there is a good deal of State practice, the need (if such there be) also to demonstrate the presence of the subjective element is likely to be dispensed with”)) (“According to him [Max Sørensen] one should assume as a starting point the presumption of the existence of opinio juris whenever a finding is made of a consistent practice; it would follow that if one sought to deny in such instances the existence of a customary rule, one must point to the reasons of expediency or those based on comity or political convenience that support the denial of the customary rule.”). However, the ILA then backed away from the proposed rule in a footnote. “Unless there are grounds for considering that the practice does not count towards the formation of a rule— e.g. because it is a usage of mere comity.” Id. at n.98. 229. See Jurisdictional Immunities of the State (Ger. v. It.; Gr. Interv.), Judgment, 2012 I.C.J. 99, ¶ 55 (Feb. 3). 230. See North Sea Continental Shelf (Ger. v. Den.; Ger. v. Neth.), 1969 I.C.J. 172 (dissenting opinion of Judge Tanaka) (deriving opinio juris from practice); Military and Paramilitary Activities in and Against Nicaragua, 1986 I.C.J. ¶ 183 (June 27) (deriving opinio juris from state acts). See also

2014]

503


opinio juris are hybrid factor-elements. If indeed, the Court is applying an inductive method, we might look for a Bayesian confirmation theory, i.e., evaluating competing hypotheses for their ability to better explain the sampled data (and predict outcomes).231 The use of the term “contradiction”232 suggests that the Court has undertaken an “enumeration of cases” inductive approach. This is so because the Court could only find contradiction if it was counting instances evidencing one position within the data pool and contrasting them to instances evidencing the opposite. The Court does not say as much, but we are left to guess whether it also takes the logical step, or not, of then projecting that ratio of instances in favor or not in the data pool onto the population as a whole. This would appear to be the case, though courts do not seem to ever articulate a probability of the accuracy of the projection that supports extending the sample’s ratio to the population. In addition, this conclusion means that the Court necessarily believes that any remaining, unexamined instance will fall into either one of the opposing conclusions—a lingering problem of the hypothesis discussed above. It could be that the court is applying predictive inference. As has been well discussed by others, customary international law straddles a strange divide between being descriptive and prescriptive.233 The descriptive side appears to be measured by the enumeration of cases approach discussed above. However the prescriptive side appears to be a predictive inference. Not only does the actor need to conclude that, based on the qualities of the rule in the sample pool, the population as a whole has the same qualities in the same proportions, but that any future case will also be governed by the same rule that the, backwardlooking, data pool evidenced. Perhaps it is more accurate to say that this prescriptive side of the analysis is inference by analogy. Rather than identify the existence of the rule in the current population, it might identify the existence of underlying qualities that establish the rule

Prosecutor v. Tadic´, Case No. IT-94-1-l, Decision on Defence Motion for Interlocutory Appeal on Jurisdiction, ¶ 99 (Int’l Crim. Trib. for the Former Yugoslavia, Oct. 2, 1995). 231. See generally PAUL HORWICH, PROBABILITY AND EVIDENCE (1982); Rosenkrantz, supra note 51; Teller, supra note 51; Fain, supra note 51. 232. See Jurisdictional Immunities of the State (Ger. v. It.), Judgement, 2012 I.C.J. 99, ¶ 72 (Feb. 3), available at http://www.icj-cij.org/docket/files/143/16883.pdf (“While not directly concerned with the specific issue which arises in the present case, these judicial decisions, which do not appear to have been contradicted in any other national court judgments, suggest that a State is entitled to immunity in respect of acta jure imperii committed by its armed forces on the territory of another State.”). 233. See KOSKENNIEMI, FROM APOLOGY TO UTOPIA, supra note 1, at 41-42; Roberts, supra note 8.

504

[Vol. 45


(state practice and opinio juris), then project the likelihood of the same underlying qualities into the future and from that projection draw an analogy that the rule will also be present there. In either understanding, the prescriptive function also appears to be inductively derived. Again, this problem of the form of the induction harkens back to the initial difficulty in determining whether the question is one of law or fact, a question of the existence of the rule or the proof of the existence of the rule, and the role of the elements in that determination. Here we can draw the reader’s attention to the distinction made above: that the inductive process is applied only during the step of assessing the sample pool of data on practice and opinio juris, in an effort to reach a descriptive and predictive conclusion about what the practice and opinio juris is likely to be over the population and in future instances. The next step is the transformation of the likely universal practice and/or opinio juris into a prescription. This author disagrees that the inductive, descriptive step here does not include opinio juris. In fact, there is a survey of the opinio juris and an inductive inference to the opinio juris held by states generally. However, Kammerhofer is correct that the descriptive conclusion alone is not sufficient to reach the prescriptive conclusion.234 2.

Deductive Analysis

Despite this process being essentially inductive, deduction is still involved. The court will use deduction to assess each item of evidence to determine whether it satisfies the above criteria of actor, class and probative value, but will also deductively assess what the item of evidence is proving.235 That is to say that the court will classify certain manifestations of behavior as state practice and/or opinio juris based on

234. See Kammerhofer, supra note 6, at 537, 544-45. 235. Often courts simply look to subsidiary sources of the law to determine the existence of rule of customary international law. This practice is not necessarily a repudiation of the inductive method, provided that the subsidiary source relied on (or the subsidiary source that the subsidiary source relied on) also relied on inductive reasoning. See, e.g., Van v. Publ. Pros., S.G.C.A. 47, ¶ 91 (Sing. Ct. App., Oct. 20, 2004), reprinted in I.L.D.C. 88 (SG 2004); RESTATEMENT (THIRD) OF THE FOREIGN RELATIONS LAW OF THE UNITED STATES § 702 (1987). It would also appear that a failure to contest the existence of a rule of customary international law could be conclusive of the existence of the rule, at least insofar as the relevant proceedings are concerned. See, e.g., Van v. Publ. Pros. (2004), S.G.C.A. 47, ¶ 91 (“It is quite widely accepted that the prohibition against cruel and inhuman treatment or punishment does amount to a rule in customary international law. The Prosecution also has not made any assertions to the contrary.”). However in Van, the Court did proceed to examine one subsidiary source that confirmed the existence of the rule. See id.

2014]

505


its understanding of what items qualify as either. For example, in the Jurisdiction Immunities Case, the ICJ examined the relevance of the European Convention on State Immunity and the United Nations Convention on the Jurisdictional Immunities of States and their Property.236 The Court considered article 11 of the European Convention237 in light of article 31.238 The Court concludes that this context “makes clear that it is not confined to that matter and excludes from the scope of the Convention all proceedings relating to acts of foreign armed forces.”239 This is deductive. The court did not look to instances of state practice where in fact article 11 in light of article 31 was read in this way, but rather deduced logically that this resulting rule followed. The Court looked to the Explanatory Report on the Convention to support its logical conclusion.240 Indeed, the Court also looked at instances of court judgments interpreting this same provision, but it is unclear whether the court was using those judgments as a sample of state practice or proof that its deductive logic held.241 A similar

236. See Jurisdictional Immunities of the State (Ger. v. It.), Judgment, 2012 I.C.J. 99, ¶ 72 (Feb. 3), available at http://www.icj-cij.org/docket/files/143/16883.pdf (citing European Convention on State Immunity, May 16, 1972, 1495 U.N.T.S. 182; E.T.S. No. 74; United Nations Convention on the Jurisdictional Immunities of States and their Property, G.A. Res. 59/38 Annex, U.N.G.A.O.R. 59th sess., Supp. No. 49, U.N. Doc. A/59/49 (not yet in force)). 237. Id. ¶ 67 (citing id. art. 11): A Contracting State cannot claim immunity from the jurisdiction of a court of another Contracting State in proceedings which relate to redress for injury to the person or damage to tangible property, if the facts which occasioned the injury or damage occurred in the territory of the State of the forum, and if the author of the injury or damage was present in that territory at the time when those facts occurred. 238. Id. (citing id. art. 31) (“Nothing in this Convention shall affect any immunities or privileges enjoyed by a Contracting State in respect of anything done or omitted to be done by, or in relation to, its armed forces when on the territory of another Contracting State.”). 239. Id. 240. Id. Once again, the Court could look to state practice affirming that it may look at explanatory reports, and it may have done so in chambers, but in its judgment, the Court appears to rest confidently on deduction. 241. Id. ¶ 68: Courts in Belgium (judgment of the Court of First Instance of Ghent in Botelberghe v. German State, 18 February 2000), Ireland (judgment of the Supreme Court in McElhinney v. Williams, 15 December 1995, [1995] 3 Irish Reports 382; ILR, Vol. 104, p. 691), Slovenia (case No. Up-13/99, Constitutional Court, para. 13), Greece (Margellos v. Federal Republic of Germany, case No. 6/2002; ILR, Vol. 129, p. 529) and Poland (Judgment of the Supreme Court of Poland, Natoniewski v. Federal Republic of Germany, Polish Yearbook of International Law, Vol. XXX, 2010, p. 299) have concluded that Article 31 means that the immunity of a State for torts committed by its armed forces is unaffected by Article 11 of the Convention.

506

[Vol. 45


blended inductive/deductive analysis was undertaken for article 12 of the United Nations Convention.242 In his critique of the Terrorism Decision by the Special Tribunal for Lebanon, Ben Saul argued, inter alia, that the STL Appeals Chamber had relied on certain domestic laws as evidence of customary international law that violated human rights law. Saul argued that “[i]t is problematic to say the least for the Appeals Chamber to seek to evidence custom by relying on national laws that violate existing international human-rights law. Such national laws are simply unlawful under international law (to the extent of their inconsistency with the ICCPR).”243 This author would disagree with the implication of the second sentence. Simply because a domestic statute might violate an international norm does not mean that the statute does not exist at all for other purposes. In the best dualist tradition, the international norm might be violated, but that violation does not necessarily deprive the statute of force as law, albeit only domestically. While it is correct to say that ex turpi causa non oritur actio,244 we are not debating the existence of a right or cause of action at law, but rather the creation of law.

242. Id. ¶ 69 (including the observation, “[n]o State questioned this interpretation” without clearly explaining why this fact is important for either deductive or inductive reasoning). 243. See Ben Saul, Legislating from a Radical Hague: The United Nations Special Tribunal for Lebanon Invents an International Crime of Transnational Terrorism, 24 LEIDEN J. INT’L L. 1 (2011) (referring to, inter alia, the laws of Egypt, Peru, Uzbekistan, and Iraq and citing, e.g., Concluding Observations of the UN Human Rights Committee: United States of America, ¶ 11, UN Doc. CCPR/C/USA/ CO/3 (2006); Algeria, ¶ 11, UN Doc. CCPR/C/79/Add.95 (1998); Egypt, ¶ 8, UN Doc. CCPR/C/ 79/Add.23 (1993); DemocraticPeoples’ Republic of Korea, ¶ 14, UN Doc. CCPR/CO/72/PRK (2001); Portugal (Macao), ¶ 12, UN Doc. CCPR/C/79/Add.115 (1999); Peru, ¶ 12, UN Doc. CCPR/ C/79/Add.67 (1996); Special Rapporteur, On the Promotion and Protection of Human Rights and Fundamental Freedoms While Countering Terrorism, ¶¶ 27-28, 45-47, 56, 62, UN Doc. E/CN.4/ 2006/98 (Dec. 28, 2005) (by Martin Scheinin); High Comm’r for Human Rights, Rep. on the Protection of Human Rights and Fundamental Freedoms While Countering Terrorism, ¶¶ 20-23, UN Doc. A/HRC/8/13 (June 18, 2008)). 244. See Legal Status of E. Greenland (Den. v. Nor.), 1933 P.C.I.J. (ser. A/B) No. 53, at 95 (Apr. 5) (Anzilotti, J., dissenting) (“An unlawful act cannot serve as the basis of an action at law.”); see also Diversion of Water from the Meuse (Neth. v. Belg.), Judgment, 1937 P.C.I.J. (ser. A/B) No. 70, (ser. C) No. 81, ¶ 240 (June 28); Military and Paramilitary Activities in and Against Nicaragua (Nicar. v. U.S.), Judgment, 1986 I.C.J. 14, ¶ 270 (June 27) (Schwebel, J., dissenting); United States Diplomatic & Consular Staff in Tehran (U.S. v. Iran), 1980 I.C.J. 53, 53-55, 62-63 (May 24) (dissenting opinions of Judges Morozov and Tarazi). In addition to ex turpicausa non orituractio, international law also recognizes ex injuria jus non oritur. See Accordance with Int’l Law of the Unilateral Declaration of Independence in Respect of Kosovo, Advisory Opinion, 2010 I.C.J. 523, ¶¶ 132-37 (July 22) (separate opinion of Judge CançadoTrindade) (citing Paul Guggenheim, La validite´ et la nullite´ des actes juridiques internationaux, 74 REC. DES COURS HAGUE ACAD. INT’L L. 191,

2014]

507


International law does recognize that a provision in a treaty could be rendered inapplicable by customary international law lex specialis, where, necessarily, the states are acting out of conformity with the treaty.245 In such a situation, the state would be violating international law, yet that act would still have norm creating force. In a similar way, state laws that violate international law should not be so casually dismissed as irrelevant since they might be contributing to a new rule. That is, of course, unless the enactment of those laws have been specifically identified or, better yet, conceded to be, violations of international law.246 Regardless of our conclusion on Saul’s argument, the debate is essentially over the tribunal’s reliance on certain samples of state practice and whether a law that violates an international norm can be used to evidence state practice. This author concludes that it can because the existence of the law is being used as evidence of another law (international law), not, in the strict sense, for its own normative value. However, this conclusion is largely a deductive one. At least insofar as they are assessing facts, international tribunals do not appear to apply a “fruit of the poisonous tree” exclusion principle.247 We could deduce an analogous practice for the types of evidence admissible to prove customary international law. Saul continues: “There is no opinio juris suggesting that excessive national terrorism definitions may lawfully derogate from international human-rights standards and thereby be relevant to custom formation.”248 He believes that we need

226-27, 230-31, 256 (1949); HERSCHLAUTERPACHT, RECOGNITION IN INTERNATIONAL LAW 420-21 (1947); Gabcˇikovo-Nagymaros Project (Hung. v. Slovk.), Judgment, 1997 I.C.J. 7, 54 (Sept. 25)). 245. See, e.g., Delimitation of the Continental Shelf (U.K. v. Fr.), 18 R.I.A.A. 3, ¶ 47 (Perm. Ct. Arb. 1977) (“Consequently, only the most conclusive indications of the intention of the parties to the 1958 Convention to regard it as terminated could warrant this Court in treating it as obsolete and inapplicable as between the French Republic and the United Kingdom in the present matter.”). Saul does not argue specifically that human rights as jus cogens would preclude any lex specialis of customary international law, though it can be understood from his emphasis on the violated norms being norms of human rights law. This argument is also not convincing. Again, we are discussing the creation of law, not the conformity with it. Regardless of the foregoing, human rights norms do permit the lex specialis of international humanitarian law. 246. See Military and Paramilitary Activities in and Against Nicaragua (Nicar. v. U.S.), 1986 I.C.J. 14 (June 27). 247. See Worster, supra note 176; Corfu Channel (U.K. v Alb.), Judgment, 1949 I.C.J. 4, 13-15, 32-35 (Apr. 9); Prosecutor v. Jelisic´, Case No. IT-95-10-A, Judgement, ¶¶ 34-37 (Int’l Crim. Trib. for the Former Yugoslavia July 5, 2001); Ga¨fgen v. Germany, 2010 Eur. Ct. H.R. 759; but see Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, G.A. Res. 39/46, U.N. GAOR, 39th Sess., Supp. No. 51, U.N. Doc. A/39/51, at 15 (providing a ban on evidence produced by torture). See Saul, supra note 243, at 8. 248. See id.

508

[Vol. 45


customary international law to determine whether the method of assessing an element of customary international law is correct. More specifically, he seems to push for an inductive assessment of the scientific method used in another inductive assessment. It is not clear whether there needs to be opinio juris on this point at all, if we accept the conclusion that the types of evidence can be, and often are, a deductive matter. 3.

Burden of Proof

In the assessment of customary international law and evidence in favor of the law at issue, the Court must also apply some notion of a burden of proof. Though it may seek to derive the content of the rule via induction, it must implicitly have in mind a tipping point where the evidence of the rule arising from the sample pool is sufficient. As observed above, this tipping point is inherently a value judgment.249 Reference to domestic assessments of domestic customary law is not particularly helpful for this question.250 Both traditional and modern customary international law agree that there needs to be evidence of the customary norm sufficient to some degree of confidence to find the existence of the norm at issue. Only in very rare circumstances have courts indicated that there is some vague, poorly articulated burden of proof.251 Some courts have

249. See Rudner, supra note 74; Worster, supra note 74. 250. In some cases, the burden is clearly placed on the party asserting the existence of the law. E.g., Nigerian Evidence Act, 1945, § 14(1). 251. See, e.g., Jurisdictional Immunities of the State, 2012 I.C.J. 99, ¶ 72 (Feb. 3) (“[T]hese judicial decisions, which do not appear to have been contradicted in any other national court judgments, suggest that . . .”, a statement of analysis which in itself implicitly suggests that the court is weighing the quantity and perhaps quality of national court judgments against each other.). Apparently courts of differing national jurisdictions can contradict each other, but this implies that they are opposed and/or enter into dialogue with each other. As a conclusion, this dialogue is perfectly acceptable, but it is reached through deduction, not induction. Though we might stretch our minds to read a burden of proof in these words, it is difficult to identify with any confidence. The burden of proof of customary international law appears to be partly the basis for the concerns of Weisburd and Kelly. See Arthur A. Weisburd, Customary International Law: The Problem of Treaties, 21 VAND. J. TRANSNAT’L L. 1 (1988) (and the ensuing conversation between Weisburd and D’Amato: Anthony A. D’Amato, Custom and Treaty: A Response to Professor Arthur A. Weisburd, 21 VAND. J. TRANSNAT’L L. 459, 473 (1988); Anthony A. D’Amato, A Brief Rejoinder; id. at 489); Kelly, Twilight, supra note 17, at 451. For a recent attempt to assess the degree of consensus around certain practice, see LEPARD, supra note 122, at 103, 158-59 (arguing from game theory that harmony or coordination games— with incentives to cooperate—require a lower level of consensus whereas prisoner’s dilemma

2014]

509


held that the rule must be “conclusively” proved,252 or “clearly and firmly established,”253 or simply “sufficient.”254 In assessing whether the burden of proof has been met, courts and scholars sometimes apply inferences and presumptions either in favor of one of the two elements of customary international law (or sometimes inferences of presumptions of customary international law, skipping over the need to prove the individual elements).255 The ILA has concluded that where there is

games—with incentives to defect—a higher degree of consensus is necessary); Francesco Parisi & Nita Ghei, The Role of Reciprocity in International Law, 36 CORNELL INT’L L.J. 93 (2003) (applying game theory to questions of state compliance and focusing on reciprocity). Also see the work of Jack L. Goldsmith and Eric A. Posner using game theory to argue that states comply with international law where and because it suits their self-interest in that situation, not because of opinio al See Goldsmith & Posner, supra note 8. Lastly, this author has also attempted to articulate a theory for assessing the degree of consensus drawing on sociological studies on the formation of critical mass. See Worster, supra note 74. Where the Court is assessing an international norm in the context of a case of individual criminal responsibility, the burden of proving the case must necessarily be very high. See, e.g., Prosecutor v. Ayyash, Case No. STL-11-01/I, Interlocutory Decision on the Applicable Law: Terrorism, Conspiracy, Homicide, Perpetration, Cumulative Charging, ¶ 91 n.152 (Spec. Trib. for Lebanon Feb. 16, 2011) (“However it is important to note that in the case of individual criminal responsibility, the tribunal is obliged to apply the principle of favor rei, although it is unclear whether this principle applies in the assessment of state practice and opinio juris related to the existence of a rule of customary international law.”). It is unclear to what degree this burden is different from the burden usually applied for questions of customary international law. As a result of the greater likelihood that a tribunal will find a rule not to exist when contrary to the interests of the accused, perhaps conclusions of international criminal tribunals on questions of customary international criminal law are less helpful as subsidiary sources of international law generally. 252. See Lotus (Fr. v. Turk.), 1927 P.C.I.J. (ser. A) No. 10, at 26 (Sept. 7) (“In the Court’s opinion, the existence of such a rule has not been conclusively proved.”). 253. See Van v. Publ. Pros. (2004) S.G.C.A. 47, ¶ 88 (Ct. Appl., Sing., Oct. 20, 2004) (“Any customary international law rule must be clearly and firmly established before its adoption by the courts.”); id. ¶ 92 (“Indeed, the passage quoted above shows that there is not enough evidence at this time to show a customary international law prohibition against the death penalty generally.”). In determining the existence of the rule, the court assessed the number of retentionist countries at seventy one, the number of completely abolitionist countries at seventy seven, the number of countries abolitionist for ordinary crimes only at fifteen and the number of countries that can be considered de facto abolitionist at thirty three. Id. It concluded “[t]he number of States retaining the death penalty was almost equal to the number of States that had abolished it.” Id. This information on state practice was all drawn from a single UN report, see U.N. Secretary General, Question of the Death Penalty: Report of the Secretary-General, UN Doc. A/HRC/21/29 (2003). 254. See Case C-366/10, Air Transp. Ass’n of Am. v. Sec’y of State for Energy & Climate Change, ¶ 106 (Dec. 21, 2011), available at http://eur-lex.europa.eu/LexUriServ/LexUriServ. do?uri⫽CELEX:62010CJ0366:EN:HTML (“[I]nsufficient evidence exists to establish . . . the principle of customary international law . . . .”). 255. See S.S. “Wimbledon” (U.K., Fr., It. & Japan v. Ger.), 1923 P.C.I.J. (ser. A) No. 1, at 25 (Aug. 17) (inferring opinio juris from widespread and consistent practice); Nottebohm (Liech. v.

510

[Vol. 45


a U.N. General Assembly Resolution on a point of customary international law, a rebuttable presumption of customary international law arises.256 The ILA also concluded that the weight of a particular item of evidence might be related to which branch of government, in a separation of powers theory, manifested the act.257 The ECJ takes the curious approach of considering whether the parties to the dispute have contested the existence of the rule,258 which

Guat.), Judgment, 2d Phase, 1955 I.C.J. 4, 22 (Apr. 6). See also North Sea Cont’l Shelf (Ger. v. Den.; Ger. v. Neth.), 1969 I.C.J. 218, 231 (Feb. 20) (dissenting opinion of Judge Lachs); North Sea Cont’l Shelf (Ger. v. Den.; Ger. v. Neth.), 1969 I.C.J. 241, 246-47 (Feb. 20) (dissenting opinion of Judge Sørensen); HERSCH LAUTERPACHT, THE DEVELOPMENT OF INTERNATIONAL LAW BY THE INTERNATIONAL COURT 380 (1958); CLIVE PARRY, THE SOURCES AND EVIDENCES OF INTERNATIONAL LAW 62 (1965); Richard Baxter, Treaties and Custom, 129 I REC. DES COURS HAGUE ACAD. INT’L L. 25, 69 (1970). Cf. C. WILFRED JENKS, THE PROSPECTS OF INTERNATIONAL ADJUDICATION 253-54 (1964); CHARLES DE VISSCHER, THEORY AND REALITY IN PUBLIC INTERNATIONAL LAW 441 n.19 (1968); Akehurst, supra note 13, at 32-34. 256. See Legality of the Threat or Use of Nuclear Weapons, 1996 I.C.J. 226, ¶¶ 70-71 (July 8) (“[U.N.G.A. resolutions] can, in certain circumstances, provide evidence important for establishing the existence of a rule or the emergence of an opinio juris.”); JORGE CASTAN˜EDA, THE LEGAL EFFECTS OF UNITED NATIONS RESOLUTIONS 192-93 (1969) (“[T]he resolutions are binding, not in the sense that they created new obligations, but in the sense that they are the expression and the legally irrefutable proof of general principles of law that are obligatory.”); Int’l L. Ass’n, Statement of Principles, supra note 14, princ. 29 (“Resolutions of the General Assembly expressly or impliedly asserting that a customary rule exists constitute rebuttable evidence that such is the case.”). But see Voting Procedure on Questions Relating to Reports & Petitions Concerning the Territory of South West Africa, 1955 I.C.J. 84 (June 7) (separate opinion of Judge Klaestad); Voting Procedure on Questions Relating to Reports & Petitions Concerning the Territory of South West Africa, 1955 I.C.J. 90 (June 7) (separate opinion of Judge Lauterpacht). 257. See Int’l L. Ass’n, Statement of Principles, supra note 14, princ. 9 cmt (e) It can happen, particularly in countries where there is a separation of powers, that the position of the judiciary (or of the legislature) conflicts with that of the executive. This is a matter of what weight is to be attached to the various instances of the State’s practice . . . . In the ultimate analysis, since it is the executive which has primary responsibility for the conduct of foreign relations, that organ’s formal position ought usually to be accorded more weight than conflicting positions of the legislature or the national courts. But it should also be noted that, in such a case, the internal uniformity or consistency which is needed for a State’s practice to count towards the formation of a customary rule may anyway be prejudiced . . . Thus, the ILA understands acts by different government offices to not qualify as state acts, but to assert relative legal value. 258. See Case C-366/10, Air Transp. Ass’n of Am., ¶ 105 (“Nor has the existence of those principles of international law been contested by the Member States, the institutions of the European Union, the Republic of Iceland or the Kingdom of Norway in their written observations or at the hearing.”); Id. ¶ 106 (“[T]he United Kingdom Government and, to a certain extent, the German Government dispute the existence of such a principle.”).

2014]

511


might also be said of the ICJ.259 However, the ECJ has elsewhere suggested that proof of customary international law is not a mere evidentiary matter, the factual existence of which could be conceded. In Racke the ECJ said that “the rules of customary international law . . . are binding upon the Community institutions and form part of the Community legal order.”260 However, proof of the law—as the ECJ has required—is generally applied in situations of proof of foreign law, where the foreign law can be a question of fact or law.261 If international law formed a monist part of the EU legal order, then it should not need to be proved as a foreign law sufficient to overcome a factual burden of proof. In the STL Terrorism Decision, the tribunal quite explicitly stated that “[h]owever significant these judicial pronouncements may be as an expression of the legal view of the courts of different States, to establish beyond any shadow of doubt whether a customary rule of international law has crystallised one must also delve into other elements . . . .”262 which suggests that the STL might see the burden of proof as “beyond a shadow of a doubt.” However, it seems that the tribunal did not view that as a mandatory burden of proof of customary international law, but rather that in order to establish the rule for the standard of proof, if it was asked for it, the tribunal would need greater proof than it had offered so far. It then went on to provide this further evidence sufficient to meet the burden. From the exact language, it appears that the tribunal did not see “beyond a shadow of a doubt” as a legal standard for proof of customary international law, nor that it saw the

259. See, e.g., Certain Questions of Mutual Assistance in Criminal Matters (Djib. v. Fr.), Judgment, 2008 I.C.J. 177 (June 4) (where neither France nor Djibouti were parties to the Vienna Convention, neither challenged the applicability of the content of the same rules under customary international law, and the Court found the rules to apply). 260. See Case C-162/96, Racke v. Hauptzollamt Mainz, 1998 E.C.R. I-3655, ¶ 46. 261. See, e.g., Fed. R. Civ. P. 44.1 (providing that the court must treat the question of foreign law as a question of law, not fact); In re Magnetic Audiotape Antitrust Litig., 334 F.3d 204, 209 (2d Cir. 2003) (per curiam) (refusing to accept Korean law when insufficiently pleaded); Curley v. AMR Corp., 153 F.3d 5, 14 (2d Cir. 1998) (reversing the District Court for relying on the law of the forum state, not the applicable foreign law). But even where foreign law is treated purely as a question of law, that conclusion does not preclude a court from applying a de facto burden of proof of the law. See, e.g., Milton Pollack, Proof of Foreign Law, 26 AM. J. COMP. LAW 470, 471 (1978) (“Accordingly, it is reasonable to assign most of the burden of demonstrating foreign law to the parties, and not the Court. I leave for another occasion the nice issue of which party should be obliged to carry the load.”) (NB: Pollack was a Judge of the Southern District of New York). 262. See Prosecutor v. Ayyash, Case No. STL-11-01/I, Interlocutory Decision on the Applicable Law: Terrorism, Conspiracy, Homicide, Perpetration, Cumulative Charging, ¶ 86 (Spec. Trib. for Lebanon Feb. 16, 2011).

512

[Vol. 45


proof of customary international law as a question of fact, but rather employed the term for mere rhetorical effect. Unfortunately, the use of such a specific term of art for rhetorical effect is probably not a best practice in judicial opinions because of just this type of confusion. In all of the foregoing analysis of the sufficiency of state practice, courts often look to other principles as an aid in satisfying some unclear burden in interpreting the existence and content of a rule of customary international law.263 The court may consider the value or the importance of the rule264 and whether such a proposed rule is logical or not.265 In fact, in the Gulf of Maine case and subsequent cases, the ICJ appears to go so far as to suggest that certain customary international law norms might be excused from the evidentiary process altogether

263. Lotus (Fr. v. Turk.), 1927 P.C.I.J. (ser. A) No. 10, at 25 (Sept. 7) (“A corollary of the principle of the freedom of the seas is that a ship on the high seas is assimilated to the territory of the State the flag of which it flies, for, just as in its own territory, that State exercises its authority, upon it, and no other State may do so . . . ”).This consideration could, however, be the complementary assessment of principles of law, rather than the deduction of customary principles. In either event, the deductive reasoning process is being applied in the context of analyzing the rules of customary international law. 264. See Jurisdictional Immunities of the State (Ger. v. It.), Judgment, 2012 I.C.J. 99, ¶ 57 (Feb. 3) (“The Court considers that the rule of State immunity occupies an important place in international law and international relations.”); Racke, 1998 E.C.R. ¶ 50 (citing The GabcˇikovoNagymaros Project (Hung. v. Slovk.), Judgment, 1997 I.C.J. 7, ¶ 104 (Sept. 25)) (“The importance of that principle has been further underlined by the International Court of Justice, which has held that ‘the stability of treaty relations requires that the plea of fundamental change of circumstances be applied only in exceptional cases.’”); Kirgis, supra note 14, at 148 (citing, inter alia, Fisheries Jurisdiction Case (U.K. v. Ice.; FRG v. Ice.), 1974 I.C.J. 3, 175 (July 25)) (“When the stakes are not as high, international decision makers have not been as quick to find restrictive customary rules.”). 265. See Juris. Immunities of the State, 2012 I.C.J. Reps. __, ¶ 57 [The rule of State immunity] derives from the principle of sovereign equality of States, which, as Article 2, paragraph 1, of the Charter of the United Nations makes clear, is one of the fundamental principles of the international legal order. This principle has to be viewed together with the principle that each State possesses sovereignty over its own territory and that there flows from that sovereignty the jurisdiction of the State over events and persons within that territory. Exceptions to the immunity of the State represent a departure from the principle of sovereign equality. Immunity may represent a departure from the principle of territorial sovereignty and the jurisdiction which flows from it. See also Pinochet case (No. 3) ¶¶ 56-58; Ayyash, Case No. STL-11-01/I, ¶ 86 (citing, inter alia, Report to the Counter-Terrorism Committee (Iran), UN Doc. S/2001/1332, at I (Dec. 27, 2011) (“As a preliminary matter, there is no doubt that there is a commonly shared agreement on the need to ‘fight international terrorism in all its forms and irrespective of its motivation, perpetrators and victims, on the basis of international law’”). For a recent attempt to assess the degree of consensus around certain practice, see LEPARD, supra note 122, at 103, 158-59.

2014]

513


based on the values they represent.266 Perhaps even issues such as game theory—i.e., whether the situation is a prisoner’s dilemma or a coordination problem, etc.— can characterize certain state-to-state interactions and thus establish differing burdens of proof for the existence of certain customary norms.267 Certain additional understandings, such as the Lotus principle, may come into play in the assessment of the sufficiency of the evidence.268 Admittedly, it is unclear whether these considerations go to the sufficiency of practice and opinio juris for the formation of the rule, to the sufficiency of evidence for the proof of the rule, or whether those two measures are the same. These deductive considerations influence the inductive process by coloring the quality of the inductive leap. Rules that are important and logical are more likely to be found than the reverse.269 It would seem that where a norm is logical, because it can be deduced from another norm or social condition, the burden of proving the custom is proportionately diminished.270 In essence, a logical norm requires a smaller sample pool. Thus the inductive method is not completely abandoned, but rather its application is modified by deductive conclusions.271 Use of a burden of proof is necessary if the inductive method is to reach a result. However, the actual burden applied does not appear to be derived inductively. It could be derived by analyzing the practice of states in establishing a burden of proof for proving customary international law from evidence of practice, but it does not. Judge Lachs was explicit in his dissent in the North Sea Continental Shelf cases that where the customary international legal norm cannot be proved the Court should still consider whether the norm can be reached through deduction.272 However, Lachs may have meant that where a norm can

266. See supra notes 104-05. 267. See LEPARD, CUSTOMARY INTERNATIONAL LAW, supra note 122, at 152-53, 158-61. 268. See Arrest Warrant of 11 April 2000 (Dem. Rep. Congo v. Belg.), Judgment, 2002 I.C.J. Reps. 137, ¶ 10 (dissenting opinion of Judge Van den Wyngaert) (assessing the evidence of the rule) (“The Court has, in effect, decided that there is evidence of a rule protecting the immunity of Foreign Ministers and no evidence of departing from this in the case of war crimes or crimes against humanity.”); id. ¶ 13 (also arguing that the Lotus principle plays a role in assessing the sufficiency of the evidence). 269. See Jurisdictional Immunities of the State, 2012 I.C.J. Reps. __; Delimitation of the Maritime Boundary in the Gulf of Maine Area (Can. v. U.S.), Judgment, 1984 I.C.J. 246 (Jan. 20). 270. Id. 271. See Treves, supra note 120, ¶ 20. 272. See North Sea Continental Shelf (Ger. v. Den.; Ger. v. Neth.), 1969 I.C.J. 172, 179 (Feb. 20) (dissenting opinion of Judge Tanaka) (“In the event that the customary law character of the principle of equidistance cannot be proved, there exists another reason which seems more

514

[Vol. 45


be deduced the degree of proof of the norm through induction is less demanding. In assessing the sample pool, the court must determine whether an act that is not in conformity with the apparent rule collapses the rule or buttresses it insofar as the act is acknowledged to be unlawful. Even where the formation of the apparent rule is damaged by inconsistent practice, the Court has held that the practice need not be entirely consistent.273 Following Karl Popper’s premise that any induction is vulnerable to negation by proof of inconsistency, this conclusion suggests that the analysis is not truly inductive.274 Scientifically speaking, a single non-conforming event should refute the rule entirely.275 The fact that the Court finds rules where there are minor inconsistent practices suggests one of two possibilities: either the Court is not truly applying an inductive approach or it is applying the inductive approach but finds the practice to be an outlier, i.e., a practice so far outside the usual patterns that it is irrational, chaotic, random, and perhaps not properly considered as the kind of practice the Court was assessing in the first place.276 As such, even a scientist applying a careful inductive analysis would be confident to ignore the item of data. We can understand this apparent contradiction with the inductive method as actually not a contradiction at all. The rule—that a small measure of inconsistency in practice is tolerable— can be confirmed inductively, but that the satisfaction of the rule in a particular case need not be inductively verified. If the court does have in mind some notion of a tipping point, even perhaps remaining unarticulated, the court must find that the point

cogent for recognizing this character. That is the deduction of the necessity of this principle from the fundamental concept of the continental shelf.”). 273. See Military and Paramilitary Activities in and Against Nicaragua (Nicar. v U.S.), Judgment, 1986 I.C.J. 14, ¶ 186 (June 27) (“The Court does not consider that, for a rule to be established as customary, the corresponding practice must be in absolutely rigorous conformity with the rule.”). 274. See, e.g., Karl R. Popper, Science: Conjectures and Refutations, in PHILOSOPHY OF SCIENCE: THE CENTRAL ISSUES 3-10 (J.A. Curd & Martin Cover eds., 1998); Popper & Miller, Proof, supra note 38. This concern may also be motivating Weisburd; see Weisburd, Customary International Law, supra note 251; Weisburd, Reply to D’Amato, supra note 251. 275. Of course, if the determination of customary international law were indeed a science, Karl Popper might ask why then it does not seek to disprove its hypotheses regarding the existence of law rather than confirm it. For Popper, observed behavior is simply interpretation of experiences in the frame of one’s theory, leading to the conclusion that the observed sees confirmation in everything. As long as there can be no evidence that disproves the hypotheses, then it cannot be a science. See Popper, supra note 274; Popper & Miller, Proof, supra note 38. 276. See Military and Paramilitary Activities in and Against Nicaragua, 1986 I.C.J. ¶ 186.

2014]

515


has been reached by applying Ockham’s razor.277 Some criticism of the STL decision is essentially that it was not Bayesian, i.e., that it did not sufficiently consider that the concordance of national legislation and judicial decisions could be more simply explained as not being founded in opinio juris.278 Of course, this argument presupposes that this explanation is somehow simpler. Clearly the STL saw the other explanation as simpler. This conclusion about a presumption in favor or against opinio juris as the simplest explanation for concordance is deductively based on a certain view of international law, in short, a view that international obligation is considered by states in choosing which action to take.279 This view on international law is most likely a gross oversimplification of a matter that likely varies widely from state to state and situation to situation. 4.

Essentially Inductive Step in the Analysis

The Legality of the Threat or Use of Nuclear Weapons advisory opinion is sometimes cited as an example of the application of a deductive approach (with the dissents by Schwebel and Higgins opposing the deductive approach the Court took).280 Schwebel argued in dissent that the Court interpreted the practice incorrectly. Though Schwebel’s dissent has been cited as a complaint against the encroachment of the deductive method, he makes no such charge. In fact, he argues that the relevant practice is the practice of five of the world’s major Powers, of the permanent members of the Security Council, significantly supported for almost 50 years by their allies and other States sheltering under their nuclear umbrellas . . . that together rep-

277. See supra notes 90-92 (discussing the requirement to seek the inference producing the simplest explanation for the data). 278. See generally Saul, Legislating, supra note 243. 279. See Prosecutor v. Ayyash, Case No. STL-11-01/I, Interlocutory Decision on the Applicable Law: Terrorism, Conspiracy, Homicide, Perpetration, Cumulative Charging, ¶ 91 (Spec. Trib. for Lebanon Feb. 16, 2011) (“However, the mere existence of concordant laws does not prove the existence of a customary rule . . . . To turn into an international crime, a domestic offence needs to be regarded by the world community as an attack on universal values (such as peace or human rights) or on values held to be of paramount importance in that community; in addition, it is necessary that States and intergovernmental organisations, through their acts and pronouncements, sanction this attitude by clearly expressing the view that the world community considers the offence at issue as amounting to an international crime.” (citation omitted)). 280. See Roberts, supra note 8.

516

[Vol. 45


resent the bulk of the world’s military and economic and financial and technological power and a very large proportion of its population.281 He adds that “[t]his practice has been recognized, accommodated and in some measure accepted by the international community.”282 Thus, he is identifying the value being placed on the practice as problematic, rather than the methodology for assessing the law. In fact, by highlighting the aspects of the practice as he does, he is applying a deductive reasoning technique to appreciate the quality of the practice. It has also been observed that Higgins, in her dissent, was arguing against the Court’s appreciation of the reality of state practice. This is not correct. Higgins did not oppose the methodology of sampling state practice to reach applicable rules, but rather she opposed the Court’s failure to apply the rules to the facts.283 When we turn to the majority’s assessment, we see reliance on General Assembly Resolutions for determining the controlling rules of customary international law, but this does not mean that the court is applying an overall deductive approach. The Court surveys the resolutions—an inductive technique—and concludes that “[e]xamined in their totality, the General Assembly resolutions put before the Court declare that the use of nuclear weapons would be ‘a direct violation of the Charter of the United Nations’; and in certain formulations that such use ‘should be prohibited.’”284 The Court then continued to appreciate that “[t]he focus of these resolutions has sometimes shifted to diverse related matters; however, several of the resolutions under consideration in the present case have been adopted with substantial numbers of negative votes and abstentions,”285 so that a

281. Legality of the Threat or Use of Nuclear Weapons, 1996 I.C.J. 311 (July 8) (dissenting opinion of Vice-President Schwebel). 282. Id. 283. Legality of the Threat or Use of Nuclear Weapons, 1996 I.C.J. 583, ¶¶ 9-10 (July 8) (dissenting opinion of Judge Higgins) (“At no point in its Opinion does the Court engage in the task that is surely at the heart of the question asked: the systematic application of the relevant law to the use or threat of nuclear weapons. It reaches its conclusions without the benefit of detailed analysis. An essential step in the judicial process—that of legal reasoning— has been omitted.”). See also id. ¶ 19 (“To show how it reached its findings in paragraph 2E of the dispositif, the Court should, after analysing the provisions of humanitarian law concerning the protection of combatants, then systematically have applied the humanitarian rules and principles as they apply to the protection of civilians.”); id. ¶ 25. 284. Legality of the Threat or Use of Nuclear Weapons, 1996 I.C.J. 226, ¶ 71 (July 8). 285. Id.

2014]

517


clear opinio juris on the part of states cannot be discovered there. This conclusion results from a classic empirical study of what states have stated through General Assembly Resolutions, and what the consequences are.286 In considering customary international humanitarian law, the Court undertook the same essentially inductive approach. The Court discovered the “cardinal principles” of international humanitarian law by surveying the texts of international humanitarian law conventions287 and observed that, although these principles are in accord with “elementary considerations of humanity,”288 the treaties have “enjoyed a broad accession” so that the principles “are to be observed by all States whether or not they have ratified the conventions that contain them, because they constitute intransgressible principles of international customary law.”289 Though we might take issue with the discovery of principles of law rather than rules of law, this analysis is still essentially inductive. The difficulty is that labeling the “modern” technique applied in the Nicaragua case as “deductive” is not really correct.290 We might argue over the quantity and quality of the state practice surveyed,291 but that is not a question of induction or deduction, but rather a problem of the application of the inductive method: a problem with forming the sample pool or identifying the degree of evidence necessary, or how to treat outliers. Those issues might be deductive problems inherent in the overall induction, but that does not mean that the entire process has now changed to one of deduction. Also, we might argue over whether the Court should look at statements of rules in the General

286. See MERON, HUMAN RIGHTS, supra note 151; Bruun, supra note 151; Lillich, supra note 15 (arguing that statements of international organizations or of states in international fora may be evidence of custom, though unclear whether those statements specifically are evidence of practice and opinio juris.). 287. See Legality of the Threat or Use of Nuclear Weapons, 1996 I.C.J. 226, ¶ 78 (July 8). 288. Id. ¶ 79. 289. Id. (citing Corfu Channel (U.K. v Alb.), Judgment, 1949 I.C.J. 4, 22 (Apr. 9)). 290. Initially, the labels are not accurate from a chronological standpoint. Compare S.S. “Wimbledon” (U.K., Fr., It. & Japan v. Ger.), 1923 P.C.I.J. (ser. A) No. 1, at 25 (Aug. 17) (applying a “modern” approach with little to no inductive analysis); Corfu Channel Case (U.K. v Alb.) (Merits), 1949 I.C.J. Reps. 4, 22 (Apr. 9) (applying “elementary considerations of humanity”, an apparently deductive approach, in 1949), with Jurisdictional Immunities of the State (Ger. v. It.), Judgment, 2012 I.C.J. 99, ¶¶ 55-72 (Feb. 3), available at http://www.icj-cij.org/docket/files/ 143/16883.pdf (applying a far more “traditional” inductive approach than the court did in the older cases from 1923 and 1949). 291. See Weisburd, Customary International Law, supra note 251.

518

[Vol. 45


Assembly,292 but again that does not transform the inquiry from one of induction to one of deduction. Taking a sample of statements and looking for the consistency that underlies them, as evidence of a general rule, is inherently inductive. The choice of which forms of evidence can aid in establishing state practice and/or opinio juris, and the determination of whether each item of evidence goes to one or other element of customary international is a deduction, but a necessary one within the overall inductive approach. G.

Conclusion on Whether a Prescriptive Rule has been Established or Not

The final step is that the assessor of law needs to conclude whether the descriptive and, now, predictive, conclusion is also prescriptive.293 This step returns to the initial deduction that a descriptive customary practice can be a prescriptive norm. Here as well, the actor might consider applying an inductive approach. There are now a great many examples of cases where courts or other actors have assessed past practice to reach the conclusion that the prescriptive step is generally undertaken. Of course, this induction suffers from the same weaknesses as all of the foregoing: it assumes the validity of the induction regarding induction. If we dig deeper, we find that this step essentially rests on a further deduction. This deduction begins with the premise that a strong description of practice with opinio juris produces a prescription. The difficulty is that the assessment of customary international law, and any law for that matter, already bridges the gap between being descriptive and prescriptive. This tension is not inherently a problem of the inductive and deductive methods alone. It is primarily a problem of law as a customary practice or in relying on empirical description of the functioning of the law to prescribe how the law will function in the future. Because we have this tension between description and prescription, we must necessarily embrace a balance between induction and deduction.

292. See Jimeńez de Arećhaga, supra note 10, at 48; Eduardo Jimeńez de Arećhaga, Custom, in CHANGE AND STABILITY IN INTERNATIONAL LAW-MAKING 1, 2-3 (Antonio Cassese & Joseph H.H. Weiler eds., 1988); Akehurst, supra note 13, at 6-7; Charney, supra note 13, at 544-45. 293. Kammerhofer, supra note 6 (“Therefore it can be argued that the inductive approach is redundant, since we cannot discover by induction the higher echelons of law by looking at behaviour and claims alone . . . .”) (citing KELSEN, ALLGEMEINE THEORIE, supra note 7, at 4, 221).

2014]

519


V.

CONCLUSION

If we take an inductive analysis of what courts are doing when they assess customary international law to determine whether they are applying inductive or deductive methods, we find a curious tension and interplay between the two reasoning methods. In practice it does not appear that any state, tribunal, or scholar applies a purely deductive technique for the assessment of international law.294 This Article has not attempted to argue that international courts and tribunals are incorrectly applying customary international law. Rather, this article has observed that the actual assessment of custom shows a mixed deductive and inductive process, and that observations on the “traditional” and “modern” approaches to the assessment of customary international law overlook the deep way the processes are intermingled. In much of the literature, traditional and modern customary international law have been viewed as polar opposites, forcing the jurist to choose one side or the other, or employ an inconsistent eclectic approach that suits their politics. The traditional approach is said to be an inductive process favoring state apology; the modern approach is said to be a deductive approach favoring utopia. In fact, the various cases where customary international law was assessed are remarkably similar in that the inductive and deductive approaches are intermingled. Even an avowedly traditional/inductive approach is partly based on deductive conclusions; even an avowedly modern/deductive approach needs some evidence of state practice to inductively verify the hypothesis. The existence of this mixed methodology diffuses the supposed harsh tension in customary international law between traditional and modern approaches. Indeed, there are slightly differing approaches, but those approaches are realized through a number of layered and integrated issues. If we accepted the hypothesis that a party wishing to apologize for state practice would apply an inductive analysis to instances of state practice, we would expect an inductive approach on all of those issues, but this approach is not taken. The reverse is also true. Thus, although there is some tension between the approaches, they are heavily tempered by the eclectic and mixed approach commonly found. Following Schwarzenberger, the approach is to a degree dictated by the question and needs of the

294. See Kammerhofer, supra note 6, at 542 (“[F]ar more scholars today hold relatively clear inductive views than pure deductive views. While natural law and the like may still be popular, a derivation of a legal system from pure reason alone is not to be found; an element of human interaction is present in every theory.”).

520

[Vol. 45


international legal system. Schwarzenberger did not reject the deductive method entirely.295 In fact, he even left the door open for the evolution to other legal analytical methods.296 What we see is that fundamental, foundational questions about the methodology, process of analysis, and forms of evidence are largely deduced, whereas questions about the empirical verification of practice and opinio juris are largely induced. This conclusion means that no analysis of customary international law can be completely classified as deductive or inductive. Instead, each act of customary international legal analysis employs a blend of approaches across a handful of questions. In doing so, it balances the tension between apology and utopia, and realizes the benefit of the differing approaches. We can only critique the balance struck in each case and argue whether it is a just balance.

295. See Schwarzenberger, supra note 22, at 566-70 (“In advocating the use of the inductive method in international law, it is as well to make clear what this approach to international law does not mean. It does not mean a complete renunciation of the deductive method. It does not mean a cult of precedents. It is not identical with the case-law method. Even the most experimental of sciences use the deductive method. Scientists in these fields, however, are usually aware of the fact that, unless and until verified, such deductions are but provisional.”); see also E. A. Whittuck, Professor Oppenheim, 1 BRIT. Y.B. INT’L L. 1, 2 (1920) (“The right method is to abstract the principles from the decisions, and then to quote the decisions themselves as examples of the application of the principles.”); Oppenheim, supra note 2, at 341. 296. See Schwarzenberger, supra note 22, at 570 (“The time will equally come when the inductive method will have served its purpose and made way for other methods which may then more closely correspond to the needs of the science of international law. The test by which, at any given moment, any of these methods stand or fall is simple. They must be judged by their results. Given the same qualifications on the part of those who apply any of these methods, ‘the right method secures the best results, and it is these we are aiming at.”).

2014]

521