3 - The Identification of Customary International Law, and International Investment Law and Arbitration

Summary

The chapter is based on a survey of state practice of CIL identification in connection with execution and annulment proceedings. The chapter proceeds in three main parts. Part I examines the nature of state practice in the form of judicial decisions, addressing, among others, the question of whether and to what extent a decision by a domestic court may be seen separately or concurrently as constitutive of practice, in the terms of ICJ Statute Article 38(1)(b), and/or as a subsidiary means, within the meaning of ICJ Statute Article 38(1)(d). Part II provides an overview of selected features of the surveyed body of state practice. This overview shows how domestic courts and states parties to execution and annulment proceedings before those courts approach CIL identification and rely on ICJ Statute Article 38(1)(a)-(c) to ascertain whether a certain source does constitute one of the sources of law recognised in those subparagraphs of ICJ Statute Article 38. Part III investigates the significance of the surveyed state practice, with a particular focus on some of the wider implications this state practice might have with respect to broader debates on CIL identification and the sources of international law, and shows a bi-directional interaction between general international law and international investment law.

1 Introduction

State practice in support of general rules governing the identification of customary international law (CIL), a subset of secondary rules of general international law epitomised by the so-called ‘two-element approach’ to CIL identification, is often neglected. Nevertheless, its nature and significance raise important issues. As with other secondary rules, decisions of international courts and tribunals are often uncritically assumed to be a sufficient basis for rules governing CIL identification. Yet, despite their varying degrees of authoritativeness, such decisions are not a sufficient, let alone a necessary, condition for establishing general rules on CIL identification. By contrast, State practice remains a necessary condition to establish the existence and content of that subset of rules, as with any other international law rules.¹

These issues not only arise as a matter of general international law, but also where sub-systems of particular international law are applied. Paramount among those sub-systems is international investment law. Insofar as international arbitration remains the preferred method for the settlement of foreign investment disputes, post-award proceedings commenced before domestic courts afford an important, though heretofore insufficiently explored, opportunity to enquire into actual general practice on CIL identification, attributable to State organs from different branches of government, participating in such proceedings in various capacities.

This chapter, based on a survey of selected State practice in connection with post-award proceedings, examines how the interplay between general international law and international investment law may have a bearing on the understanding of major general rules governing CIL identification. The body of practice on which the chapter focuses is not only confined to practice of judicial organs in the form of decisions by domestic courts hearing post-award proceedings, but also of executive organs, in the form of pleadings by States appearing in post-award proceedings. The aforementioned surveyed State practice is analysed through the prism of selected literature, decisions of international courts and tribunals, including those of the International Court of Justice (ICJ), and, in particular, to an extensive extent, the work of the International Law Commission (ILC) on CIL Identification,² including the reports of the ILC Special Rapporteur on this subject, notably as discussed by States.

The remainder of this chapter is divided into four parts. Part 2 examines the nature and significance of State practice in connection with proceedings before domestic courts, with a particular reference to post-award proceedings in the field of international investment law and arbitration. It discusses, in greater detail, various general issues concerning the primacy of State practice over decisions of international courts and tribunals, despite the latter’s often prevailing role in analyses of CIL, and the various roles State practice may play, in the form of domestic court decisions or conduct in connection with domestic court proceedings. Part 3 proceeds in two sections. The first section provides an overview of uses of ICJ Statute Article 38(1) by investor-State arbitral tribunals and, more importantly, by States in connection with those proceedings. The second section analyses actual instances of State practice in connection with post-award proceedings. It shows how that practice may have an impact on the overarching question of whether secondary rules on CIL identification have a basis in actual State practice. This is also raised in the practice discussed in the first section. Part 4 concludes with some suggestions for further research.

2 State Practice in Connection with Proceedings Before Domestic Courts: Nature and Significance

This part examines the nature of State practice in the form of judicial decisions, addressing, among others, the questions of whether and to what extent a decision by a domestic court may be seen separately or concurrently regarded as a constitutive element of custom, be it practice and/or acceptance as law, under ICJ Statute Article 38(1)(b), and/or as a subsidiary means, under ICJ Statute Article 38(1)(d),³ respectively. This part further examines the nature of other State practice in connection with proceedings before domestic courts.

While, as mentioned in Part 1, decisions of international courts and tribunals tend to be the exclusive or, if not so, the preferable basis for CIL identification,⁴ they may only constitute a subsidiary means for the determination of CIL rules, among other rules of international law.⁵ By contrast, decisions of domestic courts may play a twofold role⁶ in two spheres: internally, with respect to a custom and any resulting CIL rule(s), they may constitute general practice and/or acceptance as law in support thereof (as constitutive elements of that custom, both at its formative stage and, once in force, as requirements to identify any resulting CIL rule(s), under the ‘two-element approach’); and externally, with respect to any existing CIL rule(s) to whose creation they did not contribute, they may constitute a subsidiary means for the determination of the existence, scope and/or content of any such other CIL rule(s).⁷

Hence, in contrast to decisions of domestic courts, which may play up to three roles (ranging from the formation of either constitutive element of custom, to the identification of ensuing CIL rules and to other determinations), decisions of international courts and tribunals may only play the role of serving as a subsidiary means for the determination of CIL,⁸ let alone their role, if any, as ‘material source’ for certain CIL rules, with whose creation they may be deemed to be so associated. And, even in their capacity as subsidiary means, there are limitations to the weight decisions of international courts and tribunals can have with respect to CIL determinations. The United States of America, for instance, points out in a commentary on the ILC’s work on CIL Identification, that ‘[e]ven the International Court of Justice does not offer interpretations of customary international law that are binding on all States’.⁹ Furthermore, the United States observes, ‘a tribunal might accept without analysis that a rule is customary based on nothing more than the absence of a dispute between the parties’.¹⁰ And, relatedly, the United States points to the fact that State practice in connection with proceedings before international courts and tribunals – as might also happen before domestic courts, may have to be weighed in view of ‘the context of litigation, [in which] States may choose to assert or decline to contest that rules are customary in nature for reasons of litigation strategy rather than out of a thorough assessment that such rules are customary in nature’.¹¹ In sum, this comparatively limited role of decisions of international courts and tribunals renders more incomprehensible the tendency to overlook decisions of domestic courts.

Decisions of courts of States, often interchangeably referred to as ‘domestic’, ‘internal’ or ‘national’, and the questions of whether and in what forms they constitute State practice for the purposes of custom formation, its evidence, and, latterly, the identification of resulting CIL rules, have given rise to various questions, addressed by States themselves,¹² international courts and tribunals, and the ILC.

Furthermore, decisions of domestic courts have been widely recognised as a form of State practice, as evidenced in decisions of both national and international courts and tribunals alike.¹³ To the extent that decisions, like other similar forms of conduct, are verbal in nature, the question of their character as a form of State practice overlaps with the debate over whether practice can only consist in ‘physical’ acts or may also comprise ‘verbal acts’. While a detailed discussion of this problem exceeds the scope of this part, it suffices to observe that the ILC has concluded that there is sufficient support in State practice and decisions of international courts and tribunals to hold that verbal acts may constitute State practice. This proposition has found support among States, as evidenced in their comments, and elicited the interest of some of them. Israel, for instance, agrees with the inclusion of verbal acts as State practice, with some caveats,¹⁴ and suggests that it be defined as ‘verbal conduct (whether written or oral) […] when such conduct itself is regulated by the alleged customary rule’.¹⁵

Having observed this, the key factor for a domestic judicial decision to constitute an instance of state practice of a given State is that that decision emanate from a (judicial) organ of that State. Hence, attributability, as opposed to other properties of decisions often discussed, such as quality of reasoning or finality,¹⁶ is essential. Some States have argued in favour of a more stringent criterion, requiring not only attributability to any judicial organ, but a certain (high) position in a given State’s judicial hierarchy. Israel, for instance, considers that ‘only high courts’ final and definitive decisions (ie, that cannot be further appealed) should be taken into account’.¹⁷ Nevertheless, this proposed criterion may be an overstatement, if it concerns the necessary conditions for a decision to constitute State practice, and may be best portrayed as a criterion for attributing weight to the respective decision, without denying its character as a form of State practice, provided that it be attributable and remain in force (ie, if not final, at least not reversed on appeal or cassation).¹⁸ In sum, as the ILC concludes, ‘[d]ecisions of national courts at all levels may count as State practice’, without prejudice, as discussed above, to recalling that ‘it is likely that greater weight will be given to the higher courts’.¹⁹

Other factors which may be of relevance include the nature²⁰ and subject-matter of the alleged CIL rule at issue.²¹ For example, Israel has suggested that ‘decisions of higher national courts […] would only constitute practice or opinio juris in and of themselves when the issue in question concerns the conduct or view of judicial bodies (such as the dismissal of a lawsuit by reason of immunity)’.²² Israel’s suggested criterion of a concordance between a State organ’s scope of competence, in this case a judicial one, and the purported CIL rule’s subject-matter, appears reasonable. In particular, this criterion lends further support to the suitability of decisions of national courts as practice in support of secondary rules of CIL. In fact, secondary rules, such as those on immunity, tend to fall within the purview of judicial organs, thus paving the way for relying on their decisions in order to establish State practice in support of secondary rules on CIL identification.

Decisions of domestic courts may constitute a form of acceptance as law, as well, as seen in decisions of international and national courts and tribunals alike.²³ In its aforementioned work, the ILC had relied on domestic court decisions to establish the existence of acceptance as law.²⁴ Furthermore, not only may decisions of domestic courts constitute a form of evidence of acceptance as law in themselves, but they may also contain other separate forms of such evidence, such as ‘public statements made on behalf of States’.²⁵

The assessment of whether and to what extent domestic court’s decisions express (or evidence, as the case may be) the acceptance as law on the part of the respective State raises important and, to a certain extent, unresolved, issues. Latterly, among other criteria, ILC Special Rapporteur Wood calls for a cautious analysis as to whether, in the words of Moremen, whom he cites approvingly, acceptance as law presumably expressed in a given domestic court’s decision ‘derives from international law, from domestic law, or from domestic auto-interpretation of international law’.²⁶

The conduct of States in connection with proceedings commenced before domestic courts is also a form State practice, along other forms of ‘executive’ State practice, so-called since it emanates from organs belonging to the executive branch of a government, as opposed to its ‘legislative’ or ‘judicial’ branches (following Montesquieu’s tripartite model of governmental functions). The attributability of a conduct to a State organ, as opposed to that conduct’s connection with the proceedings, remains the key to the characterisation of that conduct as a form of State practice. This implies, among others, that conduct not attributable to a State, even if it is performed in connection with proceedings before domestic courts, and has an actual bearing on the questions of international law raised in those proceedings, does not constitute a form of State practice. As Special Rapporteur Wood aptly observes,

while individuals and non-governmental organizations can indeed ‘play important roles in the promotion of international law and in its observance’ (for example, by encouraging State practice by bringing international law claims in national courts or by being relevant when assessing such practice), their actions are not ‘practice’ for purposes of the formation or evidencing of customary international law.²⁷

This statement by the Special Rapporteur finds support among States commenting upon his work, as exemplified by Singapore’s comments to similar effects.²⁸ Singapore expressly confines ‘practice that contributes to the formation, or expression of rules of customary international law’ to that of States, to the explicit exclusion of that of ‘non-State actors’.²⁹

3 State Practice in Connection with Investor-State Dispute Settlement Proceedings and CIL Identification: The Interaction between Award and Post-Award Practice

This part provides a survey of practice of identification of CIL in connection with investor-State dispute settlement proceedings (ISDS) under a number of international investment agreements (IIAs). The practice surveyed not only studies that of ISDS arbitral tribunals, in the form of their decisions at various stages of the proceedings and, where applicable, of post-award proceedings before international law organs, such as International Centre for Settlement of Investment Disputes (ICSID) annulment committees, but also examines the practice of States in connection with those proceedings. The latter body of practice comprises not only submissions which are widely regarded as forms of State practice in connection with ISDS proceedings, epitomised by submissions pursuant Article 1128 of the North American Free Trade Agreement (NAFTA), but also practice of State pleadings before domestic courts, in post-award proceedings, where applicable. This part proceeds in two sections.

3.1 State Practice in Connection with CIL Identification in Investor–State Dispute Settlement Proceedings

This section examines the practice of ISDS arbitral tribunals and ICSID annulment committees, on one hand, and that of States, most prominently in the form of NAFTA Article 1128 submissions and submissions of a similar nature under other IIAs, on the other hand. A key criterion for the identification of this practice has been the reliance, whether explicit or implicit, on ICJ Statute Article 38(1), including its subparagraph (b), concerning CIL identification.

There is a set of instances of State practice questioning the widespread tendency towards CIL identification merely based on the findings of international courts and tribunals. Notably, these various instances of State practice place emphasis on the role of the two-element approach as a criterion for determining whether and to what extent CIL identification on the basis of decisions of international courts and tribunals is permissible. In her application for annulment of the award in CMS Gas Transmission Company v The Argentine Republic, Argentina invoked ICJ Statute Article 38(1)(d) to argue that ‘even if the cited references were correct and sufficiently supported, that would not cure the Tribunal’s failure to express its reasoning, since the authorities and the case law are secondary sources of international law’.³⁰ In its NAFTA Article 1128 submission in Eli Lilly and Company v Government of Canada, Canada specifically stated that ‘the NAFTA Parties have repeatedly asserted their agreement that the decisions of international investment tribunals are not a source of State practice or opinio juris for the purpose of establishing a new customary norm’.³¹ As Canada noted more specifically in Bear Creek Mining Corporation v Republic of Peru,

[t]he decisions and awards of international courts and tribunals do not constitute instances of State practice for the purpose of proving the existence of a customary norm and are only relevant to the extent that they include an examination of State practice and opinio juris.³²

El Salvador, in its non-disputing party submission in Spence International Investments et al v The Republic of Costa Rica, having recalled the Dominican Republic-Central America Free Trade Agreement (CAFTA-DR) parties’ reliance on the two-element approach to CIL identification, further observed that, ‘while decisions of arbitral tribunals that discuss State practice might be useful as evidence of the State practice they discuss, arbitral decisions can never substitute for State practice as the source of customary international law’, adding that CIL identification claims so substantiated are even more tenuous where those decisions ‘themselves contain no analysis of State practice or opinio juris’.³³ Indeed, as Canada observed in a response to NAFTA Article 1128 submissions in Mesa Power Group LLC v Government of Canada, ‘the awards of investment tribunals do not qualify as state practice for the purposes of proving the existence of a rule of customary international law’.³⁴ In its observations regarding the Award on Jurisdiction and Merits issued by the Bilcon of Delaware v Canada Tribunal, Canada challenged that Tribunal’s assessment of CIL in connection with its interpretation of NAFTA Article 1105 on grounds that, as ‘all three NAFTA parties have consistently agreed, decisions of arbitral tribunals can describe and examine customary international law, but they are not themselves a source of customary international law’. More specifically, and like El Salvador, Canada argued that ‘[t]he decisions upon which the Bilcon majority relied, and in particular, the decision of the Tribunal in Merrill and Ring v Canada, do not conduct the required analysis of customary international law’.³⁵ Conducting a similar analysis of the soundness of an arbitral tribunal’s identification of CIL, the United States argued in ADF Group Inc v United States of America that

[c]ontrary to the Pope tribunal’s suggestion that the sheer number of BITs could evidence the existence of a rule of customary international law, all three NAFTA Parties agree that State practice alone – without a showing of opinio juris – cannot give rise to a rule of customary international law’.³⁶

In particular, the United States rejected the above mono-elemental approach, which satisfies itself with the proposition that the growing set of BITs amounts to CIL on foreign investment, as it specifically argued that, ‘[b]ecause the Pope tribunal made no effort to determine the existence of opinio juris, its reasoning as to the BITs and customary international law is faulty’.³⁷ This echoes Canada’s proposition to a similar effect.³⁸ Furthermore, this is consistent with the United States’ emphasis on the need for establishing ‘the twin requirements of State practice and opinio juris’, as discussed in the ILC’s Second Report on CIL Identification.³⁹ The aforementioned denials of, or qualifications of the limited relevance of, CIL identification solely based on decisions of international courts and tribunals is not without prejudice to their role in aid of treaty interpretation. In this vein, the Tribunals in Sempra Energy International v The Argentine Republic and Camuzzi International SA v The Argentine Republic noted that arbitral tribunals partake in treaty interpretation, which ‘is not the exclusive task of States’, contrary to what Argentina had argued, since interpretation ‘is precisely the role of judicial decisions as a source of international law in Article 38(1) of the Statute of the International Court of Justice, to which the Respondent refers’.⁴⁰

There are other instances of State practice which focus on the very question of the legal basis and content and scope of the two element approach to CIL identification as such. The following instances are notable for their implicit and explicit reliance on ICJ Statute Article 38(1), particularly its subparagraph (b).

Some instances of State practice rely on ICJ Statute Article 38(1)(b) implicitly. In a submission pursuant to Article 10.20.2 of the CAFTA-DR in Michael Ballantine and Lisa Ballantine v The Dominican Republic, the United States noted that ‘Annex 10-B to the CAFTA-DR addresses the methodology for interpreting customary international law rules covered by the agreement’, and added that ‘[t]his two-element approach – State practice and opinio juris – is “widely endorsed in the literature” and “generally adopted in the practice of States and the decisions of international courts and tribunals, including the International Court of Justice”.⁴¹

Other instances of State practice explicitly invoke ICJ Statute Article 38(1)(b). In Lone Pine Resources Inc v Government of Canada, Canada, referring to the NAFTA parties’ respective NAFTA Article 1128 submissions, specifically indicated that the NAFTA parties’ understanding as to the applicability of the two-element approach, including as to the burden of proving each constitutive element, ‘finds its source in Article 38 of the Statute of the International Court of Justice’.⁴² Similarly, and more specifically, Canada argued in Eli Lilly and Company v Government of Canada that, ‘[p]ursuant to Article 38(1)(b) of the Statute of the International Court of Justice, customary international law has two constitutive elements: (1) extensive, uniform and consistent general practice by States; and (2) belief that such practice is required by law (opinio juris)’.⁴³ In its NAFTA Article 1128 submission in Loewen Group, Inc and Raymond L. Loewen v The United States of America, Mexico, quoting ICJ Statute Article 38(1)(b), started its analysis of CIL identification by stating that ‘Article 38 of the Statute of the International Court of Justice describes customary international law’.⁴⁴ More specifically, Canada submitted, ‘Article 38(1)(b) of the ICJ Statute identifies the two essential elements of custom: practice and opinio juris’.⁴⁵ Indeed, Canada argued, ‘the provisions at issue in this case contained in the more than 1800 BITs and in the ICSID Convention in existence have not been transformed into rules of customary international law consistent with Article 38(1)(b) of the ICJ Statute’.⁴⁶

The question of the significance of a proper determination of the relevant source of law, particularly where CIL rules are arguably involved, has also been addressed in the surveyed practice. The Annulment Committee in Venezuela Holdings, BV, and others v The Bolivarian Republic of Venezuela found that the BIT at issue, which contained an ‘explicit reference … to ‘the general principles of international law’ … is presumably to be understood as pointing in turn to one of the sources of law enumerated in Article 38(1) of the Statute of the International Court of Justice’. In that Annulment Committee’s view, ‘[i]t is the Tribunal which makes its own addition to the Treaty list by adding in a mention of customary international law’.⁴⁷ For this Annulment Committee,

the Tribunal gives no indication of where it derives the authority to make what looks like a modification – or indeed an expansion – of the source rules laid down in the Article, nor does the Tribunal state what criterion it has in mind to use in order to decide (when the case arises) whether or not to ‘include customary international law’.⁴⁸

Such an ‘expansion’, this Annulment Committee observed, can be evidenced by the fact that ‘[i]n Article 38(1) of the ICJ Statute, the sub-paragraph referring to “international custom” stands separate and distinct from the sub-paragraph referring to “general principles”.⁴⁹ Based on the above considerations, this Annulment Committee found that

[t]he Tribunal manifestly exceeded its powers to the extent that it held that general international law, and specifically customary international law, regulated the determination and assessment of the compensation due to the Mobil Parties for the expropriation of their investment in the Cerro Negro Project, in place of the application of the provisions of the BIT.⁵⁰

Indeed, this Annulment Committee emphasised that the aforementioned shortcomings were ‘so seriously deficient both in their reasoning and in the choice and application of the appropriate sources of law under the governing Bilateral Investment Treaty as to give rise to grounds for annulment under Article 52(1) of the ICSID Convention’.⁵¹ In particular, this Annulment Committee concluded, ‘the “manifest” nature of this failure is shown by the inadequacies in the Tribunal’s reasoning for the choice of applicable law, in both its positive (the law chosen) and negative (the law rejected) aspects’.⁵² The aforementioned conclusions led to this Annulment Committee’s decision to partly uphold ‘the request for the annulment of the portion of the Award dealing with compensation for the expropriation of the Cerro Negro Project’.⁵³ This Annulment Committee’s reasoning is notable not only for its materiality to the decision, but also for its reliance on the categories set out in ICJ Statute Article 38(1), and, in particular, the significance of specifically basing findings on custom as a source of law separate from general principles of law, even though CIL typically contains general principles.

The significance of not only finding State practice of reliance on ICJ Statute Article 38(1)(b), but also of establishing this practice is not engaged in by virtue of a conventional legal obligation under the ICJ Statute, is exemplified by the United States challenge of reliance on the ICJ Statute qua treaty. The United States, in ADF Group Inc v United States of America, stated that ‘there is no basis in international law for the Pope tribunal’s analysis of the phrase “international law” in Article 1105(1) based solely on the reference to that term in the Statute of the International Court of Justice, a treaty not related to the NAFTA’. In this vein, the United States submitted, ‘context includes the text of the treaty and certain related instruments, but does not include unrelated treaties’.⁵⁴ Indeed, the United States argued, ‘[c]ontrary to the Pope tribunal’s approach, Article 38 does not purport to define the term ‘international law’ in any event’.⁵⁵ While the United States focused on its understanding of the purpose of ICJ Statute Article 38(1), and did not place emphasis on the absence of an obligation to apply it qua treaty to CIL identification in connection with proceedings under NAFTA, the observation that the ICJ Statute is ‘unrelated’ to the NAFTA does raise the question of the legal basis for applying ICJ Statute Article 38(1) outside ICJ proceedings, an issue to which the following section turns.

3.2 State Practice in Connection with Post-Award Proceedings

This section provides an overview of selected features of the surveyed State practice. It shows how domestic courts and States that are parties to post-award proceedings before those courts approach CIL identification, relying on ICJ Statute Article 38(1)(b). As mentioned in the conclusion to the previous section, there is a genuine need for identifying the legal basis for applying ICJ Statute Article 38(1)(b) outside ICJ proceedings, including, if any, qua a statement reflecting any CIL rules on CIL identification.

The instances discussed in greater detail below happen to particularly relate to Argentina’s challenge of ISDS arbitral decisions before Belgian and German courts, and form the focus of this section. They add to domestic decisions adopted in various jurisdictions in connection with ISDS proceedings and, broadly, other international arbitrations involving States as respondents. Without entering into a fuller survey and discussion of such decisions, two cases are worthy of mention.

In Swissbourgh Diamond Mines (Pty) Limited & 8 ors v Kingdom of Lesotho, the Supreme Court of Singapore’s Court of Appeal made a number of observations concerning the nature of ISDS proceedings and the interplay of treaty and custom within international investment law’s hybrid framework. This judgment decided an appeal against a decision adjudicating on a setting aside application challenging an award made by an ad hoc international arbitration tribunal constituted under the auspices of the Permanent Court of Arbitration and seated in Singapore pursuant to Art 28 of Annex 1 to the Protocol on Finance and Investment of the Southern African Development Community.⁵⁶ The Court of Appeal made a number of relevant, general, propositions, namely, that ‘[i]nternational investment law is a hybrid legal construct uniquely placed at the crossroads of domestic and international law and of private and public law’, and that ‘[t]he dispute resolution mechanisms and substantive rules of investment protection provided for in the growing body of investment treaties enable such investors to bring proceedings against host States for alleged breaches of investment treaty obligations’.⁵⁷ Furthermore, and with particular reference to CIL’s place in ISDS proceedings, the Court of Appeal observed that

[w]hile these treaties are unusual in the sense that States party to them undertake obligations that may be enforced by private individuals, this is generally subject to the qualification that an investor would not be permitted to bring a claim against the State unless certain jurisdictional requirements provided for either under the treaty or as a matter of customary international law are first satisfied.⁵⁸

In Democratic Republic of the Congo and others v FG Hemisphere Associates LLC, the Court of Final Appeal of the Hong Kong Special Administrative Region considered the customary status of rules on absolute or restrictive immunity.⁵⁹ While the Court of Final Appeal found that ‘[w]hether the state immunity available in the courts of Hong Kong is absolute or restrictive is a question of common law’, and that ‘[t]he correct answer does not depend on it being a rule of customary international law’,⁶⁰ it made a number of findings concerning the nature of CIL. Paramount among those findings are the Court of Final Appeal’s propositions that ‘there may well be areas in which … international custom proves more important than treaties’,⁶¹ and, crucially for this chapter’s purposes, that ‘a rule of domestic law in any given jurisdiction may happen to result from a rule of customary international law or it may happen to precede and contribute to the crystallisation of a custom into a rule of customary international law’.⁶²

Turning to the cases in post-award ISDS proceedings initiated by Argentina before Belgian and German courts, a more in depth analysis is warranted, particularly of Argentina’s arguments before the Belgian Court of Cassation.

In K v The Argentine Republic, the Third Chamber of Germany’s Federal Constitutional Court’s Second Senate adjudicated on two constitutional complaints initiated by the Republic of Argentina.⁶³ The Chamber made a number of observations concerning the nature of CIL, which are worthy of analysis. Among others, the Chamber observed, ‘[g]eneral rules of international law are rules of universally applicable customary international law, supplemented by the traditional general legal principles of national legal orders’, and, crucially, that ‘[w]hether a rule is one of customary international law, or whether it is a general legal principle, emerges from international law itself, which provides the criteria for the sources of international law’.⁶⁴ The latter proposition is notable for aptly emphasising the role of international law as legal regulation of the conditions for existence of a source of law, including as to CIL-identification.

Furthermore, the Chamber applied the two-element approach to its analysis of the customary status of the rules on state of necessity. Indeed, having stated that the ‘[i]nvocation of state necessity is recognised in customary international law in those legal relationships which are exclusively subject to international law’, the Chamber, however, found ‘there is no evidence for a state practice based on the necessary legal conviction (opinio juris sive necessitatis) to extend the legal justification for the invocation of state necessity to relationships under private law involving private creditors’.⁶⁵

The Chamber, more specifically, went on to address each of the elements of custom, making a number of relevant general propositions in its process of CIL-ascertainment.

As for State practice, the Chamber observed, ‘[a] general legal principle cannot be verified absent a corresponding embodiment in actual legal practice’.⁶⁶ This general observation was preceded by the Chamber’s discussion of the value of international decisions.

The Chamber noted that ‘[t]he practice of international courts does not constitute an adequate basis for the recognition of an objection of state necessity towards private individuals’.⁶⁷ The Chamber made this observation having acknowledged that ‘the rulings of international tribunals have always been used as indicators of the existence of customary international law’,⁶⁸ and, more specifically,

[t]he rulings of international courts are, as a rule, major indications that certain rules of international law are anchored in customary law because – frequently in contrast to rulings of national courts – they deal with the qualification and application of specific norms under international law.⁶⁹

The Chamber’s use of the words ‘indications’ and ‘indicators’ correctly characterises the role of international decisions in CIL-identification, importantly avoiding a conflation between law-making and law-ascertaining roles, insofar as international courts and tribunals are concerned. The Chamber added that

[w]hilst courts such as the International Court of Justice or the International Tribunal for the Law of the Sea are, as a rule, restricted by their charters to settling those international-law matters which relate to relations between two or more states or other subjects of international law, international tribunals may also deal with cases which relate to economic disputes between states and private individuals.⁷⁰

The Chamber further specified the conditions under which international decisions may be considered appropriate ‘indicia’. Indeed, the Chamber observed, ‘disputes, [in which] the ruling was consequently based on the international-law relationship between two states’ lead to international decisions which are unsuitable as indicia of State practice, since such ‘purely international proceedings cannot be used as indicia in the assessment of state practice concerning the direct defence of state necessity vis-à-vis private persons for the direct disputes in front of national courts that are customary today’.⁷¹ Crucially for the Chamber’s final finding, it observed that ICSID decisions, despite involving ‘claimants … [which] were legal entities subject to private law … [n]onetheless, … do not provide any indications of the transferability of a plea of state necessity to private-law relations’.⁷² The Chamber emphasised that this distinction followed, among others, from the legal position of investors under international investment agreements, which the Chamber characterised as comprising ‘an obligation … which is owed not directly to the private applicant, but to his or her home state, although the protective purpose of the agreement targets the interests of private investors’.⁷³

Having delimited the proper place of international decisions in CIL-identification, the Chamber did conduct ‘[a]n inspection of national case-law on the question of state necessity [which] also fails for lack of agreement to suggest that the recognition of state necessity impacting on private-law relationships is established in customary law’.⁷⁴ The Chamber also considered scholarship on the question of relevant State practice, concluding that, although

scholarly literature takes the view, in agreement with international and national case-law, that necessity is recognised by customary law … [t]he relevant literature also distinguishes, however, between recognition in relations between states on the one hand and recognition as a legal justification in relations with private individuals on the other.⁷⁵

In sum, ‘as the evaluation of state practice undertaken to verify customary law has revealed’,⁷⁶ the Chamber concluded, ‘there is no rule under international customary law which recognises the transferability of the defence of necessity from relationships under international law to relationships under private law’.⁷⁷

As for acceptance as law, the Chamber noted that, while

[t]he ILC Articles on State Responsibility [(ASR)] … [which] also [cover] state necessity under international law … [were] accepted by the United Nations General Assembly on 12 December 2001 [t]his, however, leads neither eo ipso to customary-law application, nor to legally binding application for another reason, but may serve as an indication of a legal conviction as is necessary to form customary law.⁷⁸

This observation, although not preventing the Chamber from otherwise recognising the character of the ASR as codificatory of customary international law,⁷⁹ is notable for confining the role of UN General Assembly resolutions to the role of evidence, and not in themselves constitutive, of opinio juris.

In Argentine Republic v NMC Capital, Argentina appeared before the Court of Cassation of Belgium.⁸⁰ Argentina claimed that the decision it impugned had violated the customary rule of ne impediatur legatio. In particular, Argentina argued, the impugned decision had breached the ‘rule of customary international law binding at the very least on the Argentine Republic and the Kingdom of Belgium by virtue of which the immunity from execution of which diplomatic missions of a foreign State benefit must be the object of a specific waiver’.⁸¹ By failing to acknowledge the ‘autonomous character of the immunity from execution of bank accounts of foreign diplomatic missions’, Argentina concluded, the impugned decision had breached various treaty provisions including, specifically, ICJ Statute Article 38(1)(b).⁸²

Argentina elaborated on her view that ICJ Statute Article 38(1)(b) had been specifically breached. In order to make better sense of this part of Argentina’s argument, it is worth bearing in mind the distinction between primary and secondary rules.

First, Argentina had maintained that

the immunity from execution of which the bank accounts of a diplomatic mission benefit results from the international customary rule ne impediatur legatio which seeks to guarantee the efficient accomplishment of the functions of diplomatic missions, independently of the general immunity from execution of which foreign States benefit.⁸³

Secondly, Argentina argued, ‘the binding force of this international custom as source of international law is consecrated by article 38, § 1st, b), of the Statute of the International Court of Justice, annexed to the Charter of the United Nations of 26 June 1945’.⁸⁴ For Argentina, the violation of ICJ Statute Article 38(1)(b), resulted, more specifically, from the fact that ‘the judgement decides that there does not exist international custom by virtue of which the immunity of execution of which the bank accounts of diplomatic missions of a foreign State benefit should be the object of specific waiver’.⁸⁵ The judgment, Argentina maintained, had failed to acknowledge the specificity of the waiver since

it deducts that the general waiver of the claimant to her immunity from execution with regard to the defendant necessarily implies a waiver of her immunity from execution as it concerns to the bank accounts of her diplomatic mission in Belgium, notwithstanding that this latter immunity from execution had not been the object of a specific waiver.⁸⁶

In sum, Argentina concluded, by reason of its failure to require a specific waiver, ‘the judgement breaches the aforementioned international custom […] as well as of article 38, § 1st, b) of the Statute of the International Court of Justice’.⁸⁷

The Court of Cassation considered that Argentina’s ground for cassation, as formulated above, was ‘well-founded’.⁸⁸ The Court of Cassation considered that the judgment had not verified ‘that the sums seized were destined to aims other than the functioning of the diplomatic mission of the claimant’.⁸⁹ Furthermore, the Court of Cassation observed that the judgment in deciding

that the general waiver […] extends to properties of this diplomatic mission, including its bank accounts, without requiring an express and special waiver concerning these properties, violates Articles 22, 3, and 25 of the Vienna Convention of 18 April 1961 and the international customary rule of ne impediatur legatio.⁹⁰

While the Court of Cassation refrained from explicitly discussing Argentina’s claim of violation of ICJ Statute Article 38(1)(b), the Belgian Attorney General, in an opinion regarding Argentina’s request for cassation, agreed with Argentina’s claim as to the existence and content of the CIL rule of ne impediatur legatio.⁹¹ In the opinion, the Belgian Attorney General, like Argentina, invoked ICJ Statute Article 38(1)(b), and referred generally to ‘the constitutive elements of custom: (1) repetition during a sufficient period of time and within the framework of certain acts or behaviours called precedents, and (2) the opinio juris sive necessitatis’.⁹²

That the Court of Cassation was silent on Argentina’s argument that ICJ Statute Article 38(1)(b) had itself been breached does not indicate a refusal to discuss that provision. In another decision also involving Argentina, the Court of Cassation analysed ICJ Statute Article 38(1)(b). In NML Capital v Argentine Republic,⁹³ the Court of Cassation stated that

[b]y virtue of article 38, § 1st, b), of the Statute of the International Court of Justice, annexed to the Charter of the United Nations of 26 June 1945, the International Court of Justice, the mission of which is to settle in accordance with international law the disputes which are submitted thereto, applies international custom as proof of a practice generally accepted as being law.⁹⁴

In particular, it rejected the fourth strand of the second ground for cassation in support of which NMC Capital alleged that the impugned decision had violated Article 38(1)(b), ICJ Statute,⁹⁵ for

it does not result from this provision that the state judge who identifies and interprets an international customary rule is obliged to verify, in his decision, the existence of a general practice, admitted by a majority of states, which would be the origin of this customary rule.⁹⁶

This statement, at first, appears to deny the applicability of the two-element approach, since the rule whose identification is at issue is expressly characterised as one of CIL. Yet, it might be construed as partially accurate, to the extent that ICJ Statute Article 38(1)(b) qua treaty provision is, indeed, only binding on the ICJ as such. Furthermore, as Belgium observed with respect to this particular decision, in the decision whose cassation was sought, ‘the Court of Appeals, notably invoking a jurisprudence of the Court of Cassation confirming the existence of an international custom, sufficiently responded to the question’.⁹⁷

In Partenreederei MS “Neptun” GmbH & Co KG v Arquimedes Lazaro R,⁹⁸ the Court of Cassation rejected the recourse of cassation, including in particular the third ground of cassation, whereby the applicant adduced that Article 38(1)(b), ICJ Statute, had been violated by the decision impugned,⁹⁹ for the claimant ‘wrongly assumed that the formula “international custom” employed by the appeals judges refers to an international custom as source of international law in the sense of Article 38 of the Statute of the International Court of Justice’.¹⁰⁰ This response not only did not address the content of ICJ Statute Article 38(1)(b), like the Court of Cassation’s observation in NML Capital v Argentine Republic, but also went on to deny that the custom at issue was an international custom at all, unlike the custom involved in NML Capital v Argentine Republic.

The above instances of practice add to cases, also before the Belgian Court of Cassation, in which ICJ Statute Article 38 is invoked by the parties, but not dealt with in the decision, as illustrated by JPA and consorts v Kingdom of the Netherlands and De Nederlandsche Bank,¹⁰¹ where Article 38(1)(b) was relied upon by the claimant.¹⁰²

It is worth noting that the aforementioned instances of State practice are also in addition to a growing body of provisions in bilateral investment agreements in which general rules on CIL identification are expressly stated. While a discussion of the value of this practice is beyond the scope of this chapter, which has focused on practice in connection with post-award proceedings, it is worth bearing in mind that the following provisions in bilateral investment treaties specifically refer to the ‘two-element approach’: Article 5(2), China/Mexico BIT (2008); footnote 6 to Article 4(1), Singapore/Colombia BIT (2013); footnote 4 to Article 3(1), Burkina Faso/Singapore BIT (2014); footnote 1 to Article 4(2), Mexico/Singapore BIT (2009); footnote 6 to Article 4(1), Singapore/Colombia BIT (2013); and footnote 4 to Article 3(1), Burkina Faso/Singapore BIT (2014), among others.

To sum up, the aforementioned proceedings before the Belgian Court of Cassation are noteworthy. They involve practice in connection with national judicial proceedings by executive organs of both the Argentine and Belgian States. In particular, these forms of executive State practice are notable for their direct relevance to the content of the two-element approach, as a secondary rule treated distinctly from the CIL rule at issue (in this instance also secondary rule, on immunity from execution). Indeed, both the Argentine State in its pleadings, and the Belgian State through the Attorney General’s opinion, were in agreement as to the applicability and violation of the two-element approach, and both invoked ICJ Statute Article 38(1)(b), thus showing that, at the very least, this provision has a significance not qua treaty provision (the only issue actually explicitly touched upon in the Court of Cassation’s respective decision), but as a statement of the two-element approach. Therefore, these two instances of actual State practice lend support to the ILC’s statement of the two-element approach. And, together with the aforementioned, growing, instances of investment treaty practice, they show the potential of international investment law and arbitration as a sub-system of particular international law which contributes to the strengthening of key secondary rules of general international law, such as those governing CIL identification, including the two-element approach, which lies at the core of CIL identification.

4 Some Concluding Reflections

This chapter has investigated the significance of the surveyed State practice, with a particular focus on some of the wider implications it might have with respect to broader debates on CIL identification. Notably, it has shown that the very question of the applicability and content of ICJ Statute Article 38(1)(b), and the two-element approach to CIL identification, which is associated to this provision, have been raised and addressed with increasing sophistication by States in connection with ISDS proceedings.

This practice also shows that, to a certain extent, arguments about the applicability and scope of the two-element approach, the main basis for CIL identification, as opposed to the interpretation of previously identified CIL rules, has some hermeneutic dimensions. Such hermeneutic dimension raises questions calling for further research including the extent to which that dimension is a form of interpretation on an equal footing with CIL, let alone treaty, interpretation, in particular a form of ‘existential’ interpretation – blurring the distinction between identification and interpretation, or rather an exercise in ‘characterisation’ –in the same sense as private international law proceeds when categorising certain rules.

Furthermore, a bidirectional interaction between general international law and international investment law has been observed, insofar as State practice in connection with the latter sheds light on the former.¹⁰³ This departs from the common view that only general international law has an impact on sub-systems of particular international law. This interplay is highly significant, since it adds to the basis in actual State practice of general secondary rules. The potential for wider contributions of State practice in post-award proceedings with respect to secondary rules of general international law is thus worthy of further research.