5 - Uses of the Work of International Law Commission on State Responsibility in International Investment Arbitration

Summary

The Articles on State Responsibility for Internationally Wrongful Acts (‘ARSIWA’) constitute an experiment in international law-making. Unlike other successful projects of the International Law Commission (‘ILC’), such as its work on the law of treaties and diplomatic and consular relations, the ARSIWA have not yet led to the adoption of a multilateral treaty. Yet, their text is cited commonly as the authoritative statement of the law on state responsibility with investment tribunals being by far their most frequent users. This well-recorded paradox calls for a reflection on the ways in which investment tribunals make use of the ARSIWA. This chapter examines the methods which investment tribunals explicitly or implicitly employ when using the ARSIWA in order to determine the content of rules of general international law on state responsibility. The chapter then proceeds to critically assess these findings from two perspectives: the overarching aims of the law of state responsibility and the doctrine of sources of international law. The chapter synthesises these empirical and doctrinal insights into a proposal for a common framework for the use of ARSIWA.

1 Introduction

The Articles on State Responsibility for Internationally Wrongful Acts (‘ARSIWA’) constitute an experiment in international law-making. Unlike other successful projects of the International Law Commission (‘ILC’), such as its work on the law of treaties and diplomatic and consular relations, the ARSIWA have not yet led to the adoption of a multilateral treaty.¹ Yet, their text is cited commonly as the authoritative statement of the law on State responsibility with investment tribunals being by far their most frequent users. To put this into perspective, in a 2017 report to the UN General Assembly, the UN Secretariat identified 392 publicly available decisions of various bodies which make reference to the ARSIWA including those of the ICJ, the ICC, the WTO, international and hybrid criminal tribunals, and human rights courts and treaty bodies.² Although the report does not provide a specific number for investment arbitrations, it records 264 arbitral decisions referencing ARSIWA (72.5%) with investment tribunals accounting for the majority of these references.³ At the same time, the interpretation and/or application of ARSIWA is one of the most common issues arising in investment arbitration. In numerical terms, 444 cases have led to a decision since 2000 including cases in which no issues of responsibility arose or in which the reasoning of the decision is not public.⁴ The present study has traced at least 200 decisions issued in the same period citing ARSIWA or its previous versions. The extent of the practice attests to the central importance of the formally unwritten law of State responsibility and its interpretation in investment proceedings. Indeed, investment tribunals refer to this body of law to determine a variety of key issues including whether the acts forming the basis of the claim belong to the respondent state, what are the consequences of a finding against the respondent state, or whether there are circumstances precluding a finding against the respondent state or calling for mitigation. The sheer volume of the practice also raises broader questions about the ways in which investment tribunals engage with the identification and development of unwritten international law, that is, customary international law and general principles of law.

This chapter examines and also critically assesses the methods which investment tribunals explicitly or implicitly employ when using the ARSIWA in order to identify rules of general international law on State responsibility and determine their content. It does so by building upon, and adding to, the work of the UN Secretariat.⁵ Section 2 introduces the problem of ARSIWA’s lack of formal status and its implications for the use of ARSIWA in international adjudication. Section 3 surveys whether and how investment tribunals justify their reliance on ARSIWA. Section 4 highlights the variety of ways in which ARSIWA are used in the process of the determination of the content of applicable rules of law. Section 5 discusses the outcomes of this survey against the analytical backdrop of the unity of the law on State responsibility and the law relating to sources of international law. The chapter argues in favour of a principled use of ARSIWA in investment proceedings based on the distinction between the ascertainment of the legal status of a normative proposition contained therein and the determination of the content of a normative proposition whose status is undisputed.

2 Formal Status of ARSIWA and Its Methodological Corollaries

From a ‘formal’ perspective, the ARSIWA, as a document originating from the International Law Commission, does not possess any binding force.⁶ In terms of the Statute of the ICJ, the ARSIWA constitutes ‘teachings of the most qualified publicists’ that can be used as ‘subsidiary means for the determination of rules of law’.⁷ Yet, in fact, international courts and tribunals tend to attach to ARSIWA much more weight than the label of ‘teachings’ would normally suggest and often treat them as uncontroversial statements of applicable rules of law.⁸ This section briefly maps out the available justifications and methodologies for the use of ARSIWA in investment arbitration and points out why the practice of investment tribunals matters.

To start, investment tribunals are constituted for the settlement of a dispute between a State and a national of another State arising out of an investment which the parties have consented to submit to arbitration.⁹ Whether investment tribunals can apply in this context also other rules of international law including customary international law and general principles of law is a procedural issue.¹⁰ Relevant procedural rules commonly uphold in the first place the autonomy of the parties to determine the applicable law,¹¹ but provide residually for the application of ‘applicable rules of international law’ in the absence of an agreement or when the tribunal determines such law to be appropriate.¹² Besides, even when the rules of general international law are not deemed directly applicable to a specific issue, a tribunal might decide to take them into account as relevant rules for the interpretation of the applicable IIA.¹³ Questions of identification of the international law on State responsibility arise against this procedural background. By implication, investment tribunals turn to ARSIWA for the purposes of identifying rules of international law external to the IIA in question or otherwise applicable to the case. Accordingly, investment tribunals tend to justify their reliance on ARSIWA by using the terminology of ‘formal’ sources of international law, albeit with varying degrees of sophistication.

In the first place, the well-recorded discrepancy between the ‘formal’ status of ARSIWA as a source of law and their effective ‘authority’ in the context of investment arbitration calls for a more detailed survey into how investment tribunals justify their use of ARSIWA.¹⁴ With respect to customary international law, the ILC has opined – in a somewhat self-aggrandising manner – that ‘a determination by the Commission affirming the existence and content of a rule of customary international law may have particular value, as may a conclusion by it that no such rule exists’.¹⁵ This is so notwithstanding the fact that the ILC outputs cannot constitute evidence of State practice or opinio juris in and of themselves, as they do not originate directly, or even indirectly, from States.¹⁶ Moreover, as the late Judge Crawford recounts, the ARSIWA ‘have been derived from cases, from practice, and from often unarticulated instantiations of general legal ideas’.¹⁷ This implies that the ARSIWA are not necessarily a monolith from the perspective of sources of international law. Even assuming that some normative propositions included in the ARSIWA do not strictly adhere to the standards of identification of customary international law, tribunals might still have recourse to them as articulations of underlying general principles of law.¹⁸ In a nutshell, there are multiple justifications available to investment tribunals for their use of ARSIWA.

Besides, apart from why investment tribunals rely so much on ARSIWA, a closely related question is also how they make use of ARSIWA. In their final form, the ARSIWA consist of provisions articulated in prescriptive terms much like the draft of a treaty.¹⁹ It could be argued that an international court or tribunal having recourse to them should approach them in a way akin to any document having binding effect.²⁰ In so doing, the court or tribunal should follow a methodology that builds upon the rules of treaty interpretation but also accounts for the fact that the ARSIWA do not originate directly from States but from a technical body of the UN.²¹ However, the traditional perception of ILC outputs as ‘teachings’ is inimical to according any particular value to the views of the ILC as such, the ILC being a body of legal experts not of representatives of States.²² Rather, a decision-maker should focus on the evidence that the ILC adduces for the existence of a rule and reconstruct the content of the rule in question on the basis of that evidence.²³ In more practical terms, the ILC outputs on the topic not only encompass the ARSIWA and their accompanying commentaries, but also a multitude of documents including the previous reports of the Commission, comments by governments, the summary records of discussions within the plenary including the reports of drafting committees, and the reports of the special rapporteurs.²⁴ A combined reading of these documents reveals ‘titanic disagreements’ on virtually all issues, which are imprinted in the carefully balanced language of the final text.²⁵ In this respect, the traditional label of ‘teachings’ provides little guidance as to how to navigate through all these materials in determining applicable rules of law, as it treats all these materials indistinctly.²⁶

In this context of contestation, the premise of this study is that the practice of investment tribunals, as the most frequent users of ARSIWA, can further elucidate the connections between the ARSIWA and the framework of sources of international law, and shed light on the methodology for their proper use. As a preliminary point, the practice of investment tribunals regarding ARSIWA has been paralleled to ‘a drowning man …grab[bing] a stick in the sea in the hope of having certainty’.²⁷ More fundamentally, it has been argued that the power of the international judge to resolve an international dispute necessarily entails a certain degree of discretion as to the identification of applicable rules and the determination of their content.²⁸ Hence, the tendency of investment to rely on ARSIWA could be dismissed as an instantiation or corollary of such discretion or ‘expediency’ in international judicial decision-making.²⁹ Yet, whilst it is arguable that international judges enjoy more leeway than domestic ones in identifying and determining the content of applicable rules of law, they are not entirely uninhibited by any rules or principles.³⁰ In resolving competing claims as to the identification and determination of the content of rules of State responsibility, investment tribunals have the incentive –and often do– justify thoroughly their legal findings.³¹ Apart from dispelling any impression of bias or arbitrariness, investment tribunals have the strong incentive to pre-empt the annulment of the award or to prevent any domestic obstacles in the implementation of their decision.³² Therefore, the practice of investment tribunals engaging with ARSIWA cannot be reduced easily to mere expediency, but could constitute evidence of existing or emerging rules or principles, or at least, good practices, for the identification and interpretation of applicable rules of unwritten international law.

3 Patterns of Justification for the Use of ARSIWA in International Investment Arbitration

At first glance, a majority of investment tribunals give specific reasons for relying on ARSIWA, albeit a significant number of tribunals are silent on the matter. In purely quantitative terms, out of a sample of 205 decisions surveyed for the purposes of this study, 144 provide separately a justification for their reliance on ARSIWA or its previous versions. Conversely, 61 decisions cite the ILC Articles without explicitly providing a reason for doing so. On the one hand, the significant number of decisions that lack any justification for reliance on ARSIWA give support to the argument that the distinction between identification and determination of content is not water-tight not only in theory but also in practice.³³ However, a closer look at the reasoning of the tribunals depicts a much more diverse and nuanced picture than a numerical presentation suggests. In some cases, the tribunals’ stance towards the nexus between ARSIWA and the sources of international law becomes apparent from the context of the reference or the decision notwithstanding the lack of an expressis verbis or clear-cut justification. On the other hand, the tribunals provide a wide variety of justifications which do not always consist of a clear link between the ARSIWA and the ‘formal’ source of the applicable rule. Besides, the same decision might follow different lines of reasoning with respect to different provisions of ARSIWA or rely on ARSIWA only partly for the identification of applicable rules on State responsibility.³⁴ Mindful of these difficulties, the present subsection attempts to flag up common patterns in the ways in which investment tribunals justify their reliance on ARSIWA, whereas the following subsection focuses on how they use ARSIWA for the determination of the content of the applicable rule.

From the outset, tribunals recognise that the ARSIWA have no formally binding status as such either implicitly or, less often, explicitly.³⁵ However, remarkably, this survey has found only one award that explicitly characterises the ARSIWA – and, particularly, the ILC’s Commentary – as ‘works of highly qualified writers’,³⁶ despite this characterisation being relatively uncontroversial in theory with respect to works of the ILC.³⁷ The underlying reason for this discrepancy seems to be the ensuing discrepancy between the relative value which such sources are to be accorded generally in the determination of applicable rules according to the ICJ Statute and the effective authority of ARSIWA in the context of investment arbitration.

In the first place, tribunals use the ARSIWA in the process of identifying rules stemming from customary international law. In most cases, investment tribunals affirm the applicability of ARSIWA, because they ‘reflect’,³⁸ ‘codify’,³⁹ ‘state’,⁴⁰ ‘restate’,⁴¹ ‘express’,⁴² ‘formulate’,⁴³ ‘articulate’,⁴⁴ ‘represent’,⁴⁵ or ‘are declaratory of’⁴⁶ rules or principles of customary international law on State responsibility. Very often, these findings are couched in axiomatic terms without any further explanation or are reasoned in such vague terms so as to amount to little more than assertions.⁴⁷ When they do reason such findings, tribunals tend to uphold the authority of ARSIWA because of the evidence they rely upon,⁴⁸ their particular drafting process,⁴⁹ or their subsequent reception in practice.⁵⁰ In this latter respect, several tribunals have taken note of the fact that the General Assembly has annexed ARSIWA to a resolution and commended them to the consideration of States.⁵¹ Other tribunals refer to the pronouncements of other international courts or investment tribunals finding that certain provision of ARSIWA reflect customary international law.⁵² Whatever the specific line of reasoning, the common thread between these decisions is the finding that ARSIWA or a specific provision has decisive value for the identification of customary international law on this matter. In practical terms, the normative propositions contained in ARSIWA are treated as having the status of – or, more precisely, as materially identical with – rules of customary international law.

Second, a significant number of decisions use ARSIWA as means to identify rules of general application without explicitly utilising the terminology of customary international law. On the one hand, there are decisions mentioning ARSIWA in the context of identification of general principles of law. In this respect, tribunals usually declare that certain provision of ARSIWA or a statement in the Commentary is generally recognised in domestic legal systems without engaging in any detailed comparative examination or independently assessing its transposability in international law.⁵³ On the other hand, quite often, tribunals turn to ARSIWA as means for the identification of international law without deciding or clarifying what is the particular status of the underlying rules. In most cases, it is impossible to discern whether there are any legal reasons for such ambiguity or whether it is just a product of idiosyncratic drafting.⁵⁴ However, in some cases, the context of the decision reveals deeper concerns about the applicability of the rules on State responsibility reflected in ARSIWA in the investor-State context.⁵⁵ So, for instance, in the Jan de Nul Award, the Tribunal declared that the General Assembly resolution, to which ARSIWA are annexed, ‘is considered as a statement of customary international law on the question of attribution for purposes of asserting the responsibility of a State towards another State’.⁵⁶ Despite this finding, it held that ARSIWA was ‘applicable by analogy to the responsibility of States towards private parties’.⁵⁷ A similar issue arose in Vestey where Venezuela argued that the principle of full reparation reflected in Art 31 ARSIWA was inapplicable in the case of unlawful expropriation.⁵⁸ In response, the Tribunal held that ‘while the ILC Articles govern a State responsibility vis-à-vis another State and not a private person, it is generally accepted that the key provisions of the ILC, such as Article 31(1) can be transposed in the context of the investor-State disputes.’⁵⁹ Relying on judicial decisions applying the principle of full reparation in the context of expropriation,⁶⁰ the Tribunal concluded that ‘Venezuela must provide full reparation under customary international law’.⁶¹ In all these cases, tribunals treat ARSIWA as definitive statements of applicable rules of law, albeit with some ambiguity as to the precise source of such rules. Apart from general principles of law common to domestic laws, it is not apparent whether tribunals apply customary international law or refer to an altogether different category of general principles of law emanating from within international law.⁶²

Another strand of decisions seems to take into account the parties’ mutual reliance on ARSIWA in their submissions so as to affirm their applicability in the proceedings. Notably, some tribunals appear to treat the parties’ agreement as the sole basis for the application of a provision of ARSIWA in the proceedings without any finding on whether such provision actually reflects a rule stemming from a ‘formal’ source of international law.⁶³ So, for instance, in EDF, the Tribunal found it unnecessary to take a position on ‘the theoretical question of how far the various aspects of ILC Article 25 codifi[ed] customary defenses related to necessity’.⁶⁴ The Tribunal took note that ‘both sides in this arbitration stipulate[d] that the Tribunal‘s analysis should take as applicable legal norms the State of Necessity defen[c]e presented by the contours articulated in ILC Article 25’ and found that ‘the standards urged by both sides, as providing applicable norms’.⁶⁵ In these cases, the emphasis of tribunals seems to be on the relevance of party autonomy for the determination of applicable law rather than its identification on the basis of the relevant rules of international law. Yet, it could be argued that the parties’ agreement has only evidentiary value as to the status of the rule reflected in certain provision of ARSIWA so as to obviate the need for further independent analysis.⁶⁶ So, for instance, the annulment committee in Continental Casualty upheld the legal findings of the Tribunal on applicable law on the basis that ‘it was not disputed by either party that Article 25 of the ILC Articles codified the customary international law principles, and the Tribunal proceeded on this basis’.⁶⁷

This overview shows that investment tribunals tend to justify their reliance on ARSIWA by using the terminology of ‘formal’ sources of international law. However, although the vocabulary of ‘formal’ sources is used or alluded to, it can be concluded that investment tribunals do not seem to treat ARSIWA as a monolith in the process of identification of applicable rules. In most cases, tribunals accord ARSIWA decisive value in the process of identification of rules of customary international law on State responsibility. Other tribunals evoke the terminology of general principles of law or are ambiguous in this respect. What is more, tribunals usually avoid wide-ranging conclusions and focus on whether a specific provision of ARSIWA reflects an applicable rule. That said, a noteworthy insight gained from the empirical analysis is that these variations are not limited to certain parts or provisions of ARSIWA. Rather, they exist with respect to a variety of issues such as the rules on attribution of conduct, the breach of an international obligation, the circumstances precluding wrongfulness, or the provisions relating to reparation and implementation of responsibility.

4 Patterns of Interpretation in the Use of ARSIWA in International Investment Arbitration

4.1 Content-Determination Pursuant to the ‘Text’ of ARSIWA

Whether they provide a justification or not, tribunals rely on ARSIWA primarily as means to determine the content of applicable rules on State responsibility. In the case of treaties, the determination whether a text or statement has the formal hallmark of a treaty entailing binding obligations, on the one hand, and the determination of the meaning of a binding treaty provision, on the other, clearly involves different considerations so much so that it is possible to speak of two distinct juridical operations governed by different rules.⁶⁸ However, the ARSIWA are not a treaty. As we have seen, tribunals justify their applicability in investment arbitration as expressions of unwritten international law, that is, customary international law or, less often, general principles of law. In this respect, it has been maintained in theory that it is impossible to identify a rule of unwritten international law without, at the same time, determining its content.⁶⁹ Conversely, rules of unwritten law are not amenable to interpretation, this operation presupposing the existence of a text.⁷⁰ As a corollary, the determination of the content of a rule depends on the very same means as the identification of a rule and requires the establishment of State practice and opinio juris or of recognition and transposability, as the case may be.⁷¹ This section (and the subsections that comprise it) shows that these theoretical considerations can explain the practice of investment tribunals relating to ARSIWA only partially. Although decisions differ to a large extent as to how they use ARSIWA in the determination of the content of applicable rules, it is possible to discern certain trends. One discernible trend that is not easily reconcilable with these considerations is investment tribunal’s emphasis on a ‘textualist’ approach when using ARSIWA to determine the content of applicable rules on State responsibility.

To start with, the role of interpretation in determining the content of applicable rules of State responsibility is not usually pronounced and intertwines with the ways in which tribunals use ARSIWA in this process. According to the classification laid down in Art 38(1) ICJ Statute, ILC works constitute ‘subsidiary means for the determination of rules of law’.⁷² Albeit investment tribunals never refer to this categorisation explicitly (but for one singular exception),⁷³ some of them effectively use ARSIWA as such by merely citing them to support a determination that certain normative proposition found in judicial pronouncements or other sources reflects a rule of international law.⁷⁴ These tribunals seem to treat ARSIWA indistinctly as secondary evidence in the determination of unwritten international law without any methodological explanation as to the steps followed in this process.

However, in many cases, the ways in which tribunals engage with ARSIWA in the process of determination of applicable rules goes beyond indirect reliance. Most notably, there is an abundance of decisions in which tribunals proceed to apply provisions of ARSIWA as self-explanatory to the facts of the case. Thus, for instance, many tribunals quote Article 4 ARSIWA as a representation of applicable law and continue to determine whether a person or an entity is a State organ or not relying on the characterisation provided in the domestic law of the relevant State.⁷⁵ Similarly, tribunals have applied without any discussion a variety of provisions of ARSIWA including those on attribution of conduct,⁷⁶ on the time of the breach,⁷⁷ on circumstances precluding wrongfulness,⁷⁸ and on reparation.⁷⁹ In the same vein, tribunals often identify in a provision of ARSIWA an applicable rule of law and then refer to judicial pronouncements as means to determine the meaning of that provision. To illustrate this point, in the Jan de Nul award on the merits, the Tribunal considered ILC’s provision that a given conduct is considered an act of State ‘if the person or group of persons is in fact acting on the instructions of, or under the direction or control of, that State in carrying out the conduct’.⁸⁰ After characterising ARSIWA ‘a statement of customary international law’, it proceeded to further clarify the meaning of the provision.⁸¹ The Tribunal held that ‘[i]nternational jurisprudence is very demanding in order to attribute the act of a person or entity to a State, as it requires both a general control of the State over the person or entity and a specific control of the State over the act the attribution of which is at stake; this is known as the “effective control” test’, citing in support the findings of the ICJ in Nicaragua.⁸² Subsequent awards reproduce the Jan de Nul formula more or less verbatim,⁸³ even though the ‘general control’ prong of the test does not feature explicitly in the text of Article 8 ARSIWA, its Commentary or the pronouncements of the ICJ after Nicaragua.⁸⁴

In these two groups of cases, it is difficult to discern what precise juridical operation is at play, but two alternatives are conceivable from an analytical perspective. On the one hand, the lack of any separate analysis on the content of the applicable rule is suggestive of the absence of an intermediate step between identification of a rule of customary international law or general principle of law and its application.⁸⁵ Similarly, the reliance on judicial pronouncements can be construed as an extension of the determination of State practice/opinio juris or recognition/transposability, as the case may be, albeit implicitly and on the basis of secondary evidence.⁸⁶ After all, judicial decisions, much like ARSIWA, constitute ‘subsidiary means’ for the determination of applicable rules.⁸⁷ On the other hand, these tribunals not even purport to determine the content of the applicable rule on State responsibility through an independent analysis of State practice and opinio juris or a comparative survey. Rather, they proceed to apply the formulations of the ILC to the facts of the case as if they were a binding text.

In this respect, the conciseness of analysis can also be construed as an emanation of a textual approach towards ARSIWA in a way that parallels known approaches of treaty interpretation. In other words, the tribunals’ line of reasoning consists conceivably of the application of the terms of a provision whose source of legal validity (CIL or general principle of law) has already been determined, because they deem its ordinary meaning sufficiently clear.⁸⁸ To illustrate this point, in Tulip, the Tribunal examined whether the termination of a contract by a company owned predominantly by a State agency was attributable to Turkey. The Tribunal accepted that ‘the ILC Articles constitute a codification of customary international law with respect to the issue of attribution of conduct to the State’.⁸⁹ Turning to Article 8 ARSIWA, the Tribunal focused on its text and decided that ‘[p]lainly, the words “instructions”, “direction” and “control” are to be read disjunctively’.⁹⁰ The fact that a State agency owned the majority share of the company in question entailed that the company was under the control of the Turkish State in the sense that Turkey was capable of exercising a degree of control to implement governmental policies.⁹¹ Nonetheless, the Tribunal held that such control was insufficient for attribution, because, according to the ILC’s Commentary, the State must ‘us[e] its ownership interest in or control of a corporation in order to achieve a particular result’.⁹² In the subsequent annulment decision in Tulip, the committee clarified that ‘the tribunal, in interpreting Article 8, took into account the ILC Commentary’ and upheld the analysis of the Tribunal finding that ‘[it] correctly interpreted Article 8’.⁹³ Such textual analysis of ARSIWA with reference to the ILC Commentary as an authoritative interpretative aid is most common in the practice of investment tribunals, even if they are not always as explicit as the Tulip committee when justifying their methodological choices.⁹⁴ Therefore, there is evidence to suggest that such ‘textualist’ approach constitutes essentially an interpretative operation.

Along similar lines, it is possible to argue that tribunals frame their reliance on previous case law in terms of interpretation because such previous pronouncements only purported to interpret rules whose existence was undisputed. So, for instance in El Paso, Argentina argued that the Tribunal exceeded its powers by relying on case law to determine the applicable standard of reparation under the law on State responsibility, despite judicial decisions’ lack of binding status beyond the confines of a specific case.⁹⁵ The annulment committee dismissed this claim on the basis that ‘[a]rbitral tribunals must resort to different methods of interpretation to decide the dispute’ before them and, in the event, the tribunal relied on previous case law only ‘to be helped in its interpretation’.⁹⁶ More generally, tribunals seem to single out a provision of ARSIWA as the embodiment of an unwritten rule, whilst the reference to international jurisprudence only comes after this determination without justification, as if it were only an interpretative aid.

4.2 Content-Determination through the Application of Means Akin to Treaty Interpretation

Apart from such textual approach, it is not uncommon for tribunals to invoke or evoke other interpretative principles commonly used in the process of treaty interpretation. First, a few tribunals have recourse to materials produced in the long preparatory process of ARSIWA as another means to determine the content of the applicable rules on State responsibility. Second, investment tribunals often refer to means of interpretation applied to treaties focusing on the context or object and purpose of ARSIWA.

References to the discussions within the ILC leading up to the adoption of ARSIWA are not particularly frequent in investment decisions. In some cases, tribunals rely on the record of discussions in order to determine whether the silence of ARSIWA also implies a determination by the ILC that certain concept or proposition does not form part of international law.⁹⁷ Thus, in Alghanim, the investor invoked the distinction between ‘obligations of conduct’ (ie those that prescribe or proscribe a specific conduct) and ‘obligations of result’ (ie those that require the achievement of a specific result irrespective of the conduct adopted) that appeared in previous drafts of ARSIWA.⁹⁸ The Tribunal took note of the critical stance of the last Special Rapporteur and the deletion of the distinction from the final draft of ARSIWA and concluded that the distinction did not form part of customary international law.⁹⁹ Furthermore, in other cases, tribunals purport to rely on the preparatory works of ARSIWA as a means to interpret a provision of ARSIWA. For instance, in the LG&E decision, the Tribunal started from the determination that Article 25 ARSIWA reflects the standard of necessity in international law.¹⁰⁰ It, then, proceeded to discuss each of the elements of Article 25 ARSIWA referring exclusively to the opinions of the ILC Special Rapporteurs and other individual members of the ILC.¹⁰¹ From a methodological perspective, it is possible to maintain that these tribunals merely examine all available secondary evidence without endorsing any firm distinction between identification and determination of the content of the applicable rule.¹⁰² However, in cases like the LG&E, the analysis clearly emphasises the intention of the ILC in adopting certain provision rather than the evidence upon which the ILC relied. For this reason, these findings are not easily explicable under the mainstream view on the identification of customary international law or general principles of law.¹⁰³ Rather, these tribunals seem to resort to the record of discussions of the ILC more as an interpretative aid in a way that parallels the use of travaux préparatoires in the context of treaty interpretation.

More conspicuously, investment tribunals often refer to interpretative principles which pertain to the ‘the spirit, purpose and context of the clause or instrument in which the words are contained’.¹⁰⁴ The Tribunal’s approach in Devas relating to the attribution of conduct of a State-owned company to India provides a very illustrative example. In this case, the Tribunal noted that the text of Article 8 ARSIWA only mentioned ‘persons or group of persons’, but made no reference to ‘entities’ like, for instance, Article 5 ARSIWA establishing also a rule of attribution of conduct.¹⁰⁵ Nonetheless, the Tribunal considered that ‘it is generally recognized in modern legal systems that “person” includes not only a natural person but also a legal person’ and that several Institutes of Internal Auditors (IIAs) included corporations in their definition of ‘persons’.¹⁰⁶ In methodological terms, the Tribunal referred to other rules of international law, which it deemed relevant for the interpretation of the rule reflected in ARSIWA, in a way akin to the context of a treaty.¹⁰⁷ Furthermore, the Tribunal remarked that ‘it would make no sense to impose a restrictive interpretation that would allow a State to circumvent the rules of attribution by sending its direction or instruction to a corporate entity rather than a physical person or group of physical persons’.¹⁰⁸ Instead, it opted for a different interpretation noting that even in the case of corporations the instructions or direction would be received and acted upon by natural persons (ie the directors and agents of the corporation).¹⁰⁹ From a doctrinal viewpoint, the Tribunal chose out of two available interpretations the one that gave full effect to Article 8 ARSIWA in what appears to be a straightforward application of the interpretative principle of effectiveness (ut res magis valeat quam pereat or effet utile).¹¹⁰

Moreover, several tribunals often proceed to construct provisions of ARSIWA on the basis of broader considerations, which they deem as cross-cutting in the law of State responsibility. For instance, several tribunals invoke the stability of international obligations as a stepping stone for a restrictive interpretation of the customary defence of necessity.¹¹¹ Another set of illustrative decisions declare that the purpose of an award of interest under Article 38 ARSIWA is to ‘ensure full reparation’ and proceed to award compound interest.¹¹² This is so notwithstanding the fact that Article 38 ARSIWA is silent on the matter and the ILC Commentary clearly favours the award of simple interest.¹¹³ These findings seem to evoke the object and purpose or the ratio of ARSIWA or of specific provisions in order to determine the meaning of the applicable rule in a way that parallels known approaches to treaty interpretation.¹¹⁴ The common thread that binds all these pronouncements together is the blending of a literal reading of ARSIWA with contextual or teleological considerations that mirrors the process dictated by the rules of treaty interpretation.

4.3 Interpretation of the Law on State Responsibility and Its ‘Rule-ness’ in International Investment Arbitration

It has been shown so far that it is not uncommon for investment tribunals to hold explicitly or implicitly, through the use of relevant methods, that the determination of the content of the applicable rules of State responsibility also requires a process of interpretation notwithstanding its unwritten character.¹¹⁵ A further question that arises is whether the practice of investment tribunals evidences the existence of rules or principles governing this process.

Remarkably, some decisions not only accept that the determination of the content of rules of State responsibility also involves interpretation, but they also clearly suggest that such process is governed by relevant rules or principles of international law. The award on jurisdiction in ST-AD is telling as to this general point. In this case, the Tribunal enunciated that ‘every rule … of international law must be interpreted in good faith’.¹¹⁶ Applying this rule of interpretation to the requirement of exhaustion of local remedies under customary international law,¹¹⁷ the Tribunal found that ‘[t]his rule is interpreted to mean that applicants are only required to exhaust domestic remedies that are available and effective’.¹¹⁸ Similarly, several decisions relating to the customary defence of necessity, which is articulated in Article 25 ARSIWA, raise explicitly the issue of interpretation.¹¹⁹ Most notably, in Enron, the annulment committee censured parts of the award discussing whether measures taken by Argentina were the only way to address the economic crisis constituting the situation of necessity and whether Argentina contributed to that crisis.¹²⁰ According to the committee, the ‘only way’ and ‘non-contribution’ requirements spelled out in Articles 25(1)(a) and 25(2)(b) ARSIWA, respectively, were ‘capable of more than one interpretation’.¹²¹ The committee held that the Tribunal ‘was necessarily required, either expressly or sub silentio, to decide or assume the correct interpretation in order to apply the provision to the facts of the case’.¹²² It, thus, concluded that the Tribunal committed an annullable error by not laying down its own interpretation of these requirements.¹²³ Inversely, in EDF, the annulment committee admitted as a matter of principle that the ‘the concept of “only means” is open to more than one interpretation’.¹²⁴ It held that ‘[i]n the light of the principle that necessity is an exceptional plea which must be strictly applied (a principle expressly stated in paragraph 1171 of the Award), … “only” means “only”; it is not enough if another lawful means is more expensive or less convenient’.¹²⁵ Although the committee held that failure to elaborate on the issue of interpretation did not constitute an annullable error, it nonetheless recognised the application of a principle or rule of interpretation to the customary rule of necessity according to which ‘exceptions to general principles are to be interpreted restrictively’.¹²⁶

By contrast, some decisions eschew a precise determination as to the content of the applicable rule of State responsibility by limiting themselves to taking note of the parties’ stance in the proceedings. In some cases, the lack of an independent analysis on the content of the applicable rule seems to be elicited by factual or evidentiary considerations. So, in Inmaris, the Tribunal identified the applicable rule in Article 31(1) ARSIWA that provides for ‘the obligation to make full reparation for the injury caused by the internationally wrongful act’.¹²⁷ As to the issue of causation, Ukraine proposed two tests, namely, one based on proximity and another based on foreseeability.¹²⁸ The Tribunal declined to determine the applicable test finding that the relevant acts caused harm to the investor ‘under either standard discussed by the respondent’.¹²⁹ However, in other decisions, the agnostic stance towards interpretation appears to stem from legal considerations. To give an illustrative example, in Suez annulment decision, the crucial issue was whether the Tribunal failed to apply the proper law, in the event, Article 25 ARSIWA on the state of necessity. The committee conceded as a matter of principle that the ‘only way’ and ‘non-contribution’ requirements appearing in Article 25 ARSIWA ‘are indeed susceptible to a certain degree of interpretation’.¹³⁰ However, it emphasised that no party raised any issue of interpretation in the proceedings before the Tribunal. On this basis, the committee concluded that ‘an interpretation issue that was not raised by the Parties cannot be considered “outcome-determinative” with the consequence that a failure to address such issue would amount to a manifest excess of powers under Article 52(1)(b)’.¹³¹

The committee’s reasoning in Suez echoes other annulment decisions that distinguish between ‘disregarding the proper law’, which constitutes an annullable error, from ‘misapplication of the proper law’, which does not.¹³² In the case of applicable treaty provisions, annulment committees also occasionally examine whether tribunals disregarded any applicable rules of interpretation, despite allowing them ample deference as to the application of such rules in the specific case.¹³³ The Suez case raises the question whether and how this distinction can be applied with respect to applicable rules of unwritten international law, such as those under the law of State responsibility. In this respect, as has been shown, the Enron annulment decision clearly suggests that a tribunal must pay some regard, either explicitly or sub silentio, to the principles upon which it bases its determination of the content of the applicable rules of unwritten international law.¹³⁴ What is more, the EDF committee traced back the interpretative principle applied by the Tribunal despite finding that this was beyond the scope of annulment review.¹³⁵ By contrast, whilst acknowledging a distinction between identification of applicable rules of unwritten law and their interpretation, the Suez committee relied exclusively on the parties’ stance in the underlying proceedings to decide whether an issue of interpretation was ‘outcome-determinative’. In this way, it remained entirely agnostic as to the existence of rules or principles of interpretation of unwritten international law.

To summarise, investment tribunals employ an array of methods in the course of determining applicable rules on unwritten international law on State responsibility. At the crux of these diverse approaches seems to lay a latent, or sometimes explicit, distinction at an analytical level between the identification of applicable rules on State responsibility and their interpretation with ARSIWA operating as the focal point. This distinction is not merely a theoretical one, but it has both substantive and procedural implications. First, the use of ARSIWA as an object of interpretation has allowed investment tribunals to develop the law on State responsibility in ways that contradict, or cannot be inferred from, the evidence adduced by the ILC. The award of compound interest in investment arbitration is a case in point. Second, a comparative survey between investment awards reveals that new divisions have spun out as to the choice of the proper means of interpretation. The cases on the content of the necessity defence are paradigmatic. Third, the casting of disputes on the content of the law on State responsibility in terms of interpretation by reference to ARSIWA has also played a role in the context of annulment proceedings. In this respect, the division between annulment committees is not so much whether customary international law or general principles of law on State responsibility can be interpreted; annulment committees readily accept this point. Rather, the main point of contention seems to be the existence and interplay of applicable interpretative rules or principles so as to enable annulment review, delineate its scope, and determine its operation.

5 Interpretation as a Balancing Exercise between Centrifugal Forces in International Investment Arbitration and the Unity of Law on State Responsibility

The previous section has identified a variety of ways in which investment tribunals use ARSIWA in determining applicable rules on State responsibility. In the main, investment tribunals favour a formalist approach: they tend to justify their reliance on ARSIWA as means to determine the applicable rules on State responsibility on the basis of the rules on the identification of international law. Moreover, many investment tribunals go even further and proceed to apply, explicitly or implicitly, (meta-)rules or (meta-)principles of interpretation on ARSIWA in order to determine the content of the applicable rule. At the same time, some decisions remain agnostic with regards to the issue of interpretation. This section argues that the fundamental consideration of the unity of the law on State responsibility militates in favour of its uniform application and against a case-by-case approach. The theory of sources of international law provides several footholds for the interpretation of rules of unwritten international law.

To start, the fundamental premise of the law on State responsibility is its unity: ‘every internationally wrongful act of a state entails the international responsibility of that state’.¹³⁶ Yet, an agnostic approach as to the methodology for the determination of the content of the applicable rule can severely distort its content. To illustrate this point, in Bayindir, the Tribunal referred to ‘the international rules of attribution reflected in Articles 4, 5 and 8 [ARSIWA]’, which it found applicable ‘as expressing current customary international law’.¹³⁷ In its analysis on Article 8 ARSIWA, the Tribunal held:

the Tribunal is aware that the levels of control required for a finding of attribution under Article 8 in other factual contexts, such as foreign armed intervention or international criminal responsibility, may be different. It believes, however, that the approach developed in such areas of international law is not always adapted to the realities of international economic law and that they should not prevent a finding of attribution if the specific facts of an investment dispute so warrant.¹³⁸

In the specific case of Article 8 ARSIWA, the ICJ has ruled that the ‘effective control’ test cannot be displaced in the absence of an applicable lex specialis.¹³⁹ Yet, at first glance, the Bayindir panel was not obligated to follow the ICJ’s ruling and an inductive examination of the available primary and secondary evidence at the time would have led to conflicting results.¹⁴⁰ What is striking in this award is that the Tribunal put aside the much debated ‘effective control’ test purporting that it is a factual issue pertaining to the application of the customary rule. This enabled the Tribunal to infuse its analysis of the content of the applicable rule of responsibility with asides about the speciality of international economic law in a way that runs counter to the fundamental consideration of the unity of the law on State responsibility.¹⁴¹ The Bayindir award stands out as an outlier, investment tribunals accepting this premise virtually unanimously. Accepting the premise of the unity of the law on State responsibility, which the ICJ has upheld as a rule of customary international law, militates against a case-by-case approach and reveals the need for a consistent methodology for the determination of the content of rules on State responsibility.¹⁴²

Consistently with the mainstream approach concerning the sources of international law, the determination of the applicable rules of the law on State responsibility must take place in principle on the basis of an inductive examination of all available evidence.¹⁴³ However, at the same time, a purely inductive approach towards the identification of unwritten international law has certain limitations. So, with respect to the identification of customary international law, the ILC has concluded that ‘the two-elements approach does not preclude an element of deduction as an aid’ particularly ‘when considering possible rules of customary international law that operate against the backdrop of rules framed in more general terms that themselves derive from and reflect a general practice accepted as law or when concluding that possible rules of international law form part of an “indivisible regime”’.¹⁴⁴ More emphatically, the current special rapporteur on general principles of law has opined that ‘deduction is …the main criterion to establish the existence of a legal principle that has a general scope’.¹⁴⁵ The fact that the ILC perceives deduction as part and parcel of the process of identification of a rule of unwritten international law is somewhat less relevant. The key point is the acknowledgment of a juridical operation analytically distinct from induction that consists of the determination of the content of the applicable rule on the basis of an inference from a normative proposition whose status is undisputed.

To illustrate this point, according to the ILC, the provisions of ARSIWA relating to the content of State responsibility are ‘without prejudice to any right, arising from the international responsibility of a State, which may accrue directly to any person or entity other than a State.’¹⁴⁶ Nonetheless, investment tribunals commonly apply the rules on reparation reflected in ARSIWA, even if this involves an element of interpretation as an intermediate step.¹⁴⁷ Notably, in Quiborax, the issue arose whether investment tribunals can issue a declaratory award as a form of reparation. The Tribunal referred to Articles 34 and 37 ARSIWA and enunciated that ARSIWA ‘restate customary international law and its rules on reparation have served as guidance to many tribunals in investor-State disputes’.¹⁴⁸ It specified that ‘the remedies outlined by the ILC Articles may apply in investor-State arbitration depending on the nature of the remedy and of the injury which it is meant to repair.’¹⁴⁹ In this respect, it cautioned that ‘some types of satisfaction as a remedy are not transposable to investor-State disputes’.¹⁵⁰ In particular, it held that ‘the type of satisfaction which is meant to redress harm caused to the dignity, hono[u]r and prestige of a State, is not applicable in investor-State disputes.’¹⁵¹ The Tribunal concluded, thus, that ‘[t]he fact that some types of satisfaction are not available does not mean that the Tribunal cannot make a declaratory judgment as a means of satisfaction under Article 37 [ARSIWA], if appropriate’.¹⁵² In practical terms, the tribunal essentially engaged in the interpretation of Article 37 ARSIWA as a rule of customary international law referring expressly to its wording and its object and purpose. Whether it did so in order to determine its transposability in the context of investor-State arbitration or whether it interpreted it as a directly applicable rule of customary international law is less relevant for practical purposes.

The important point is that the determination of the applicable rules of State responsibility is composed of two analytically distinct operations with ARSIWA being at the crux of the analysis. On the one hand, the determination of the status of a normative proposition contained in ARSIWA involves an inductive analysis of evidence.¹⁵³ In this respect, the practice of investment tribunals is an unambiguous attestation that ARSIWA enjoy a ‘présomption de positivité’.¹⁵⁴ In practical terms, challenging the status of most normative propositions contained in ARSIWA would require evidence indicating lack of generality of practice/opinio juris or recognition, as the case might be. On the other hand, the determination of the content of a normative proposition contained in ARSIWA also involves an element of interpretation. Specifically, it involves criteria that parallel the process of treaty interpretation. The key difference is that the intention of the drafters of ARSIWA – that is, the ILC – is less relevant in this process because the formal foundation of the normative propositions contained therein is only the consuetudo of States.¹⁵⁵ In practical terms, this is reflected in the findings of investment tribunals emphasising the text, context, and object and purpose of ARSIWA or a specific provision at the expense of the multitude of materials leading up to their adoption or even, in case of inconsistency, the ILC Commentary.¹⁵⁶ Therefore, the process of interpretation is not an unprincipled process, even if the relevant principles and their interaction are still in a process of elaboration and refinement (not unlike the rules of treaty interpretation before the adoption of the Vienna Convention on the Law of Treaties (VCLT)).

To recap, the premise of the law on State responsibility is inimical to an agnostic approach with respect to the identification and interpretation of the rules on State responsibility. The crucial normative concern is the unmaking of the law on State responsibility through its blending with considerations special to specific sub-systems, in this case, international investment law. In this respect, adherence to ARSIWA is only a starting point. Rather, the key point is that the determination of the applicable rules on State responsibility on the basis of ARSIWA is also a principled process to which interpretation is an inextricable part.

6 Conclusion

Undeniably, investment tribunals spearhead the consolidation of the general law of State responsibility through their widespread endorsement of ARSIWA. This comparative analysis of the use of ARSIWA in investment arbitration has shown that the discrepancy between the current form of ARSIWA and their effective authority should not be overstated. In the main, investment tribunals do justify their reliance on ARSIWA using the vocabulary of the sources of international law. In fact, the formal status of most normative propositions contained in ARSIWA has been tested and analysed from every possible angle by hundreds of investment panels. Whilst admittedly this has led to divergent views as to theoretical points, the preponderance of decisions converges into treating ARSIWA as authoritative statements on the law on State responsibility.

What emerges as a new challenge is ensuring the uniformity of the law on State responsibility within this context of pervasive use. The focus of States, investors, and tribunals seems to have moved away from ‘grand questions of principle’ towards ‘the boring small print’ of responsibility rules like the meaning of control, contribution, injury, causation, or damage.¹⁵⁷ In this respect, the traditional approaches on the identification of customary international law and general principles of law have certain limitations. This survey has shown that investment tribunals are increasingly aware that, through their use and elaboration of ARSIWA, they are engaging in the interpretation of the formally unwritten law on State responsibility. The realisation that interpretation is also a principled process guided by international law will ensure the unity and consistent development of the law on State responsibility even within the specialised context of investment arbitration.

In fact, this survey has shown that elements of such a principled methodology are already present in the practice of investment tribunals, even if in incipient form. In the first place, the determination that a normative proposition contained in ARSIWA reflects a rule of customary international law or a general principle of law is only a starting point. This determination entails that the ordinary meaning of the terms of ARSIWA is important in specifying the content of the formally unwritten rule. In this respect, tribunals seem mindful that this is essentially an interpretative endeavour guided by considerations of good faith. Second, another overarching consideration is that rules on State responsibility do not exist in a vacuum but form part of an internally coherent law of State responsibility that should be deemed compatible with the rules and principles of international law which this law is meant to make operational. As a corollary, whilst tribunals tend to confirm the status of a normative proposition contained in ARSIWA in a piecemeal fashion (ie for each provision separately), this does not constitute the end of the analysis. Tribunals also take into account the immediate and broader context of each provision alongside the object and purpose of the entirety of ARSIWA and of each provision within the broader system of ARSIWA when clarifying its content. Third, tribunals seem also to appreciate that the anticipated outcome of this innately interpretative exercise is not the determination of the intention of the drafters of ARSIWA, but the concretisation of rules and principles emanating from the legally relevant conduct of States. Whilst they seem to accord particular value to an ILC’s determination denying binding status to a certain normative proposition, they use materials produced in the run-up to the adoption of ARSIWA only exceptionally and in a supplementary fashion in the context of the interpretation of a normative proposition contained in the ARSIWA.