3 - Misinterpreting Customary International Law

Summary

With the inescapable constructive dimension of CIL, courts serve a key function in the development of CIL through their interpretations and their interpretative methodologies. Sometimes courts are mistaken in their interpretation of CIL. This contribution explores three main questions: (1) what is misinterpretation? (2) how and why do misinterpretations take place? and (3) what are the potential consequences of misinterpretation of CIL? These all converge in the underlying question of whether there are detectable objective standards for the determination of misinterpretation or whether such observation is always a subjective one – anchored on a disagreement on the values which lie at the core of international law. To preserve legal certainty and predictability in the interpretation of CIL, it is crucial to recognise that courts are not infallible, may lack the requisite methodological tools and occasionally just get CIL wrong. The indeterminacy of CIL renders futile the attempts to pin down precise conditions for its validity, and simultaneously, leaves open the definition of misinterpretation of CIL.

1 Introduction

This chapter explores the misinterpretation of customary international law (CIL) and its practical and normative consequences. I focus on three main questions: (1) what is misinterpretation? (2) how and why do different misinterpretations take place? and (3) what are the potential consequences of misinterpretation of CIL? These all converge in the underlying question of whether there are detectable objective standards for the determination of misinterpretation or whether such observation is a subjective one – anchored on a disagreement on the values which lie at the core of international law.

In exploring these questions, I combine doctrinal study with empirical examples. Additionally, the different potential consequences of misinterpretation call for a normative evaluation: whether misinterpretation renders a norm invalid or illustrates lex ferenda? Could misinterpretation create an authoritative verdict of the status of law, even against its flawed premise – ‘a corrupt pedigree’, a term coined by Fernando Teson?¹ And why does pedigree matter for CIL – is there something beyond institutional formality conferring authority and legitimacy?

By its nature, CIL is constantly evolving – the customary process is continuous. While in its purest form interpretation of CIL may consist of an analysis of an already ascertained rule (and its elements), interpretation of CIL in most cases inevitably includes an element of construction at that particular point in time – it is difficult to distinguish between the formation, identification and interpretation of CIL.² As Anthony D’Amato has noted in relation to CIL: ‘there is an interrelation between law-formation and law interpretation’.³

According to traditional methodology, CIL emerges spontaneously ‘like a path in a forest’.⁴ It has been suggested, somewhat convincingly, that the identification and interpretation of CIL have taken a strategic turn, potentially arising from the proliferation of international interactions and norm-interpreters and -entrepreneurs.⁵ The theories of ‘modern CIL’⁶ have attempted to explain and justify the broadened methodology, which utilises CIL to advance political, ethical, economic and other aims. Some of such attempts may in fact encourage and expand potential misinterpretations of CIL, with reliance and application of elements far removed from the common understanding of state practice and opinio juris. The effect is not relevant only in the methodology but also in the outcomes: with the utilisation of different interpretative methodologies by different courts and other norm-interpreters,⁷ the resulting identification of a rule of CIL and/or its subsequent interpretation could be highly inaccurate, due to either a genuine mistake in the interpretive methodology or an aspiration to apply a progressive norm disguised as customary rule for moral, ethical, policy, or other extra-legal reasons.

Following the ‘CIL as a path’ metaphor, the interpreter of a norm may misidentify, say, a dried-up stream as a path, designate a minor trail as a fully-fledged path, find a path where there is none, or call a man-made walkway a path. These present examples of misinterpretations without delving into their underlying motives. It is, however, useful to analyse reasons behind a misinterpretation as they may have a direct bearing on the consequences flowing from it, how it is received and responded to by the international community, and for the determination of whether it is representative of ‘a corrupt pedigree’ or a matter of sluggish methodology.

While Articles 31 to 33 of the Vienna Convention on the Law of Treaties (VCLT) serve as the starting point for addressing interpretation in international law – and these rules have crystallised as part of CIL in their own right⁸ – they provide little guidance on how to interpret CIL rules. Just as well, courts have adopted and adapted their own approaches and methodologies on how to go on about interpreting CIL, and not always consistently even within the same institution. In this chapter I intentionally avoid delving into the discussion of what constitutes interpretation: this is accomplished by other authors in this volume.⁹ One may criticise my approach for cutting corners or reversing the analysis while pursuing exactly the same result: exploring what interpretation is not. I accept that the critique may be warranted. The purpose here is, however, not to provide an ample understanding of the misinterpretation of CIL but to initiate conversation on how interpretation may go awry and what it may do to the validity and legitimacy of CIL.

Capturing the definition and examples of misinterpretation is like chasing a moving target – as with interpretation, the elements may be in flux, the circumstances and narratives changing, and the line between genuine and fake CIL – and correct interpretation and misinterpretation – often fluid. With CIL identification (and possibly subsequent interpretation), one can usually find evidence to support what one is looking for – but so can the opposite party. Transplanting a correct interpretation reached at a given point in time into a later case may in fact provide the very premise for misinterpretation even when the methodology of the initial interpretation has been accurate per se: the act of interpreting CIL requires the interpreter to analyse the practice and opinio juris at a specific point in time. By definition, CIL can develop continually and therefore the interpreter needs to look beyond the matter or dispute at hand to get a broader vision of the applicable evidence of the elements of CIL. This argument runs somewhat parallel to Article 30 of the VCLT on the application of successive treaties relating to the same subject matter and to its Article 31 (3) (a) and (b) on subsequent agreement and practice in the interpretation of a treaty: the commentaries call for consideration of what is appropriate in particular circumstances and for caution in resorting to effective interpretation, noting that ‘even when a possible occasion for [principles and maxims’] application may appear to exist, their application is not automatic but depends on the conviction of the interpreter that it is appropriate in the particular circumstances of the case’,¹⁰ and that to resort to extensive or liberal interpretation ‘might encourage attempts to extend the meaning of treaties illegitimately on the basis of the so-called principle of “effective interpretation”’.¹¹ The commentaries also touch upon consequences of such extensive interpretation, with a reference to the 1950 Interpretation of Peace Treaties Advisory Opinion of the International Court of Justice (ICJ), where the court emphasised that to adopt an interpretation which ran counter to the clear meaning of the terms would not be to interpret but to revise the treaty.¹² In a similar manner, to interpret CIL ‘effectively’ or in a way that runs counter to established practice and opinio juris, could either constitute a revision of the rule of CIL (if it is shown that new practice and opinio have emerged) or, as is the focus here, result in misinterpretation (if practice and opinio do not sufficiently support the new interpretation). As noted by the arbitrator in the Island of Palmas case: ‘The same principle which subjects the act creative of a right to the law in force at the time the right arises, demands that the existence of the right, in other words its continued manifestation, shall follow the conditions required by the evolution of law.’¹³

Moreover, methodological questions on whether a solid definition of ‘interpretation’ ought to be stipulated in order to address misinterpretation cannot be dismissed. While the purpose of this chapter is not to extensively delve into these questions, they run in the background of this inquiry and occasionally surface; must we pre-determine the conditions of validity of CIL before we can analyse its misreadings? How, by whom and why should the interpretation of CIL rule be deemed invalid? What can the examples of misinterpretation of CIL tell us about the rules of interpretation?

In discussing the substance of CIL, I use ‘norms’ when the legal validity is uncertain or they appear in a space of conceptual ambiguity – is it CIL or not, is it a legal rule or simply social practice or aspiration? ‘Rules’ refer to those norms which have, to the best of our knowledge and assessment, materialised or crystallised as a part of CIL.

2 From Methods of Interpretation to Misinterpretation

Interpretative exercises in customary international law have been described as ‘methodological mayhem’ resting on the flexibility of methodological uncertainty,¹⁴ and creating an environment of ontological doubt.¹⁵ The same goes for misinterpretation. The orthodox purpose of interpretation is to clarify the intentions of parties. While in treaty law this may be a feasible – if not an easy – task, in CIL identifying and clarifying the intentions of parties is practically impossible, not least for the absence of records of travaux preparatoires. This may depend, however, on the theory of formation of CIL: whether one accepts that CIL forms ‘like a path in the forest’ or whether CIL may arise through a focused, intended and continued practice of actors in international law. In the latter occasion, tracing cognisant practices may be possible, even if based on speculation.

As we know, in the North Sea Continental Shelf cases¹⁶ the ICJ evaluated the basic parameters for CIL based on Article 38 (1) (b) of the ICJ Statute, which have become the reference point for the traditional account of CIL. The ICJ articulated the elements of CIL as follows: ‘Not only must the acts concerned amount to a settled practice, but they must also be such, or be carried out in such a way, as to be evidence of a belief that this practice is rendered obligatory by the existence of a rule of law requiring it.’¹⁷ This was adapted in the Nicaragua case,¹⁸ which saw a more flexible approach, in particular regarding the relationship and the chronological order of emergence of practice and opinio juris. The ICJ reaffirmed the two-element approach in the Jurisdictional Immunities of the State case, stating, with a reference to its previous case law,¹⁹ that ‘the existence of a rule of customary international law requires that there be “a settled practice” together with opinio juris’.²⁰ Now, beyond the basic identification of CIL and its elements, the ICJ has not contributed a great deal to the science of interpretation of CIL, nor – luckily – does it offer obvious examples of misinterpretation either,²¹ although some have claimed that ‘the identification practice of the International Court of Justice for customary norms deviates from the traditional definition of customary law in Art. 38 (1) lit. b of the ICJ Statute’.²² It has been suggested that ‘progressive determinations of CIL [by courts] are generally unproblematic when States are in the dock [as opposed to international criminal law proceedings]’.²³ This statement is simplistic and misleading. First, determinations and interpretations of CIL, especially when delivered by a court with a high authority such as the ICJ, unavoidably influence not only the development of international law but also methods and techniques on how CIL is interpreted subsequently by other international and national courts. Second, while the principle of legality has a heightened relevance in (international) criminal proceedings,²⁴ it is not redundant in inter-state adjudication. In addition, the ICJ – and any other court deciding inter-state cases – has a duty to uphold core tenets of the rule of law, such as consistency, predictability and non-arbitrariness. Adopting progressive determinations of CIL has the potentiality to obstruct these fundamental principles in international judicial decision-making, regardless of whether it is states or individuals in the dock.

Some have argued that the Draft Conclusions on Identification of Customary International Law by the International Law Commission (ILC) constitute a ‘statement of the principles guiding the interpretation of CIL’.²⁵ This characterisation is inaccurate. The difference between ‘formation’ and ‘identification’ has been discussed at length in the ILC’s Special Rapporteur’s First and Second Reports on CIL,²⁶ but ‘interpretation’ has simply not received similar attention; while the work of Sir Michael Wood is invaluable in setting out the issues, controversies and principles for the identification of CIL, it does little to inform the interpretation of existing rules of CIL. This caveat was highlighted also, inter alia, in the Comments and Observations by the Government of the Netherlands to the Draft Conclusions in 2018, which notes that ‘it does not become clear whether the process for identifying the existence of a rule is the same as the process for determining the content of that rule’.²⁷ As has been discussed in this chapter and extensively elsewhere over the course of the lifespan of international legal scholarship, the process of CIL interpretation, nevertheless, overlaps with the process of identification in a complex manner.

In my previous work,²⁸ I have addressed different categories of identification and interpretation of CIL: First, courts may find customary international law by employing the traditional method of assessing state practice supported by opinio juris. Second, they may place more weight on opinio juris over practice – often in this context understood to include a broad spectrum of different considerations. Third, they may deduce customary rules from treaties, national legislation and other (binding or nonbinding) documents. And fourth, courts may refer to previous case law as a confirmation of the customary status of a norm, without in fact assessing the actual findings of practice and opinio juris at that point in time.²⁹ The main criticism regarding the CIL methodology concerns the lack of proper analysis of the elements of custom: courts rendering assertions without justifications, either intentionally, negligently, or, as often seems to be the case, rather casually. This serves as a background for the ensuing analysis of misinterpretation of CIL, which bridges the methodological considerations with the underlying rationales. Identifying examples of misinterpretation is to a large extent related to the way one perceives the functions and limits of international law. For those adopting traditional reading of CIL,³⁰ many more cases of misinterpretation may be detectable than to those with leanings towards the ‘modern approach’³¹ or ‘the sliding scale approach’.³²

There are two dimensions to misinterpretation. Misinterpretation can refer to, on the one hand, to the process and the outcome of the process, and on the other, to the law ascertainment and content determination:³³

1. (1) The act of misinterpretation and its motives – affects institutional legitimacy.

2. (2) The substance of the misinterpretation and its consequences – affects the validity of the norm, depending on its reception by relevant actors, and hence, depends on the conditions of validity imposed by the normative framework.

While misinterpretation has an inherently negative sound to it, it can be a necessary stage in the development and normative change of CIL rules. For an existing CIL rule to change, the practice and/or opinio juris ought to differ from, contradict or go beyond lex lata, providing initially a consideration of lex ferenda. Now, even when practice and/or opinio juris may not have changed or the evidence thereof is mixed, the interpreters (usually courts) may pitch in to spearhead the change in a catalytic manner: such activity may constitute an example of misinterpretation of CIL, closely related to the misidentification and misevaluation³⁴ of the elements of CIL and leading to a potential misrepresentation of a norm. The interpreters may find evidence of practice and/or opinio juris where there is none, exaggerate their prevalence and impact, or declare a norm as CIL without further ado. The breadth, depth, scope and applicability of CIL may be incorrectly set out: for instance, a regional custom may be (mis)interpreted as universally applicable, a general principle of law may be mistakenly awarded customary status, or a jus cogens norm may be characterised as CIL even in the absence of widespread and consistent practice.³⁵ The same could occur in reverse: downplaying practice and/or opinio juris, to hinder the emergence of an undesired rule of CIL even when the elements would point to its crystallisation.

Does the finding of misinterpretation presuppose a cognisant misinterpreter? No: misinterpretation by definition is not concerned with motivations, but it simply refers to ‘the act of forming a wrong understanding of something that is said or done, or an example of a wrong understanding’.³⁶ In any case, evidence of deliberate misinterpretation of CIL is rare and mostly misinterpreters have adopted a lazy methodology or ignored rules of interpretation in evaluating practice and/or opinio juris.

When analysing the notion of misinterpretation in CIL, we can also break it down to the elements: is it the CIL rule as a whole, or practice or opinio juris that is being misinterpreted? What is the relative relevance of misinterpreting practice or opinio juris? Misinterpreting the element of practice – being usually more quantifiable – may be more obvious than that of the more fluid, subjective, element of opinio juris. It may also be more consequential, as it is viewed – at least by many of us – as the very bedrock of custom.

Now, it is not terribly difficult to find cases of national courts taking liberties in their interpretation of the concept and rules of CIL,³⁷ although most often these courts ‘simply assert, without citing persuasive practice authority, the existence of a customary norm’.³⁸ At the international level one may expect to see more cross-referencing and recognise potential consequences of misinterpretation for international law. The next section sets out categories, with selected examples, where international courts and tribunals have overstepped methodological limits to the effect that may constitute misinterpretation.

3 Categories of Misinterpretation

Based on my initial research findings, three types of misinterpretation are identifiable. First, the extension (or reduction) of CIL through exitus acta probat – the end-justifies-the-means approach – where analysis of the elements of CIL is modified to fit the desired outcome and the elements are substituted or complemented with resort to extra-legal tools and concepts. This approach often finds support among the more modern-liberal theories of CIL. Second, I have identified the negligent interpretation, which may amount to misinterpretation when the norm-interpreter labels a norm as CIL without further analysis of the elements and where in fact opposite practice and opinio juris might be observable. Finally – and luckily evidence of this remains scarce – there is the fallacious method of misinterpretation, where the interpreter finds false CIL or considers flawed or incomplete evidence of its elements. All the three categories contain overlapping dimensions – it may be hard to distinguish whether the norm-interpreter was merely negligent or plain wrong, or where the end-justifies-the-means approach crosses over to the delivery of fallacious interpretation. In particular, regarding the latter example, ideological leanings may cloud the legal astuteness of determining whether the interpreter has acted in good faith or not, and wherein lies the line between an actual legal error and consequentialist bending of the rules of interpretation to achieve a morally desirable outcome. Also, a fallacious interpretation of CIL has a much higher chance of success to flourish through subsequent interpretations and practice when it leads to ‘good’ outcomes – for instance, an ‘effective’ interpretation expanding the scope of a human right can be expected to be met with more praise than an argument to the opposite effect.

A misinterpretation may be discoverable in subsequent proceedings by the same or another court. In May 2010, the Extraordinary Chambers in the Courts of Cambodia (ECCC) held that the mode of responsibility of Joint Criminal Enterprise (JCE III)³⁹ did not exist under CIL in 1975–79, and consequently was not applicable in the proceedings of that court.⁴⁰ The ECCC limited JCE III by declaring that there was not enough evidence of its customary nature, at least not in 1975–79, thus dismissing the ICTY’s argumentation in Tadić by illustrating that the Tadić court had in fact invented that category of criminal responsibility.⁴¹ In analysing the concept of JCE, the ECCC first noted that it must consider ‘not only whether JCE existed under customary international law at the relevant time, thus being punishable under international criminal law, but also whether it was sufficiently foreseeable and accessible to the Charged Persons’.⁴² It then examined the findings in Tadić, other ICTY cases, and case law dealing with the crimes committed in World War II, stating in relation to JCE III that ‘[h]aving reviewed the authorities relied upon by Tadić in relation to the extended form of JCE (JCE III), the Pre-Trial Chamber is of the view that they do not provide sufficient evidence of consistent State practice or opinio juris at the time relevant to Case 002’.⁴³ This approach was further confirmed in November 2016 by the Supreme Court Chamber of the ECCC.⁴⁴

Earlier the same year, however, the Appeals Chamber of the ICTY reconfirmed its interpretation of JCE III as a mode of liability under CIL by noting that ‘the third category of joint criminal enterprise has existed as a mode of liability in customary international law since at least 1992 and that it applies to all crimes consistently confirmed in the Tribunal’s subsequent jurisprudence’.⁴⁵ In responding to the defendant’s challenge of the customary nature of JCE III, the Appeals Chamber stated that ‘this contention is essentially premised on his suggestion to depart from the existing jurisprudence on the basis of his misconstruction of the law’.⁴⁶ This could be viewed as the Appeals Chamber’s reaction to the debate surrounding JCE III and presents a clear expression of its position vis-à-vis interpretation of CIL by the ECCC and many scholars.

Even when acknowledging that nearly twenty years passed between the commission of crimes in Cambodia and in the former Yugoslavia – and the applicable rule of CIL needs to be determined with reference to those points in time – these cases nonetheless show CIL’s ambiguity and the challenges it poses to interpretation. The drastic departure of the ECCC from the ICTY jurisprudence brings uncertainty on the actual status of the rule and raises the question of the implications of such diverse interpretations for future cases dealing with modes of criminal liability.⁴⁷ This goes on to illustrate that there may not be objectively one right answer to how and what to interpret. The method and consequently the outcome may depend on the interpreter’s approach to CIL: which element carries the most weight and which evidence is included in the assessment of practice and/or opinio juris. Both interpretations of the status of JCE III can be objectionable on these grounds. As noted by Verdier and Voeten, ‘an attempt to justify a breach by reference to the rule’s ambiguity is likely to be interpreted as a violation by the counterparty and (some) third parties’.⁴⁸ Regarding the treatment of the JCE III neither of the tribunals can be accused of sluggish methodology even if one could be persuaded to view ICTY’s early judicial activism as falling into the exitus acta probat category. The interpretation of JCE III as CIL has, however, been repeated in the jurisprudence of the ICTY in numerous subsequent cases. Below, under ‘Consequences of Misinterpretation’, I will discuss the relevance of repetitive judicial practice in the context of potentially ‘corrupt pedigree’ of CIL.

Similar discourse involving possible misinterpretations of CIL took place internally between different chambers of the ICTY and the Special Court for Sierra Leone (SCSL), in relation to the requirement of specific direction as a part of the actus reus of aiding and abetting.⁴⁹ The ICTY and the SCSL, within a space of less than a year, interpreted this requirement under CIL reaching opposite outcomes: Whilst the ICTY Appeals Chamber held in Perisić that specific direction is a part of actus reus of aiding and abetting, it did not make any explicit reference to its status under CIL. The SCSL Appeals Chamber in Taylor, on the other hand, stated that:

[i]n the absence of any discussion of customary international law, it is presumed that the ICTY Appeals Chamber in Perisić was only identifying and applying internally binding precedent. … [T]he ICTY Appeals Chamber’s jurisprudence does not contain a clear, detailed analysis of the authorities supporting the conclusion that specific direction is an element of the actus reus of aiding and abetting liability under customary international law.⁵⁰

In effect, the SCSL called out the ICTY Appeals Chamber’s misinterpretation of CIL. The ICTY Appeals Chamber, in a decision shortly after, adopted the SCSL interpretation, mending its own prior misinterpretation.⁵¹ Hence, this is an example of the second category of misinterpretation where the ICTY had discussed a rule, presuming it part of CIL without a further analysis of its elements and when in fact a further analysis would have uncovered opposite practice and opinio juris.

Likewise, an example of misinterpretation, subsequently identified by another authority, can be found in the EU case law. The rules which constitute an expression of customary international law are binding upon the EU institutions and form part of the EU legal order and as such: ‘[CIL] is regularly interpreted and applied by the Court as an “integral part” of EU law’.⁵² De Burca has observed that ‘CIL was cited by the CJEU in twenty-one cases’ (as of October 2015).⁵³ A recent search on EUR-LEX reveals that the number stands now at thirty.⁵⁴ Interestingly, the Court of Justice of the European Union and the Advocate General have made some remarks about misinterpretation of CIL. For example, without delving further into EU case law, in Front Polisario, the Advocate General considered and accepted⁵⁵ the argument put forth by the council and commission that ‘the General Court misinterpreted customary international law, as it did not cite any legal basis requiring the EU institutions to verify that the other party to the agreement has complied with the principle of permanent sovereignty over natural resources and the primacy of the interests of the inhabitants of non-self-governing territories’.⁵⁶

A striking example of a misinterpretation of CIL – or misidentification as the limits may be fluid – through flawed methodology undercutting the element of practice took place in the ICTY Trial Chamber judgment Prosecutor v. Kupreškić, discussing the prohibition of reprisal attacks against civilians:

Admittedly, there does not seem to have emerged recently a body of State practice consistently supporting the proposition that one of the elements of custom, namely usus or diuturnitas has taken shape. This is however an area where opinio iuris sive necessitatis may play a much greater role than usus, as a result of the aforementioned Martens Clause. In the light of the way States and courts have implemented it, this Clause shows that principles of international humanitarian law may emerge through a customary process under the pressure of the demands of humanity or the dictates of public conscience, even where State practice is scant or inconsistent.⁵⁷

The court, going against the traditional understanding of CIL and its elements, allowed for inconsistent practice to suffice in its finding of CIL and suggested that opinio juris is of a higher value than state practice. It relied almost exclusively on opinio juris in its interpretation of the CIL rule on reprisals, while simultaneously broadening the internal nature of opinio juris and its limits to include extra-legal considerations, such as ‘elementary considerations of humanity’, which, the ICTY considered, should be fully used when interpreting and applying loose (customary) international rules.⁵⁸ It further noted a customary rule of international law had emerged ‘due to the pressure exerted by the requirements of humanity and the dictates of public conscience’.⁵⁹

Such misinterpretation originates, most likely, from two objectives: on the one hand, the court saw evidence of a horrific act which was, however, not explicitly covered by the rules of international law at the time and felt a moral duty to rectify this – to bring the perpetrators to account for their actions, to deliver justice for the victims, to contribute to deterring future atrocities and to enhance international criminal law as a social pedagogical imperative, and, possibly, to contribute to the development of the law. At the same time, the mandate of the ICTY limits its jurisdiction to the application of international humanitarian law that is ‘beyond any doubt part of customary law so that the problem of adherence of some but not all States to specific conventions does not arise’.⁶⁰ Hence, the ICTY could only apply law that it considered to have already crystallised as CIL, which then led it down the path of broadening the methodology of identifying and interpreting norms so as to fit them under the umbrella of CIL – at times like in Kupreškić amounting to misinterpretation of CIL at the intersection of an exitus acta probat (end-justifies-the-means) approach and the intentionally deceiving method of misinterpretation.

Another example of a potential misinterpretation of CIL stems from the 6 May 2019 judgment of the Appeals Chamber on the International Criminal Court (ICC) in the Jordan Referral in the Al Bashir case. The court discussed the customary status of Article 27 (2) of the Rome Statute and concluded that Head of State immunity under customary international law does not apply in international courts and tribunals.⁶¹ It stated that: ‘[t]here is neither State practice nor opinio juris that would support the existence of Head of State immunity under customary international law vis-à-vis an international court. To the contrary, such immunity has never been recognised in international law as a bar to the jurisdiction of an international court.’⁶² The decision was not unanimous and in their joint dissenting opinion Judges Luz Del Carmen Ibáñez Carranza and Solomy Balungi Bossa stated that ‘[i]t is thus clear that the international community as a whole has consistently rejected the invocation of Head of State immunity for the commission of international crimes’, continuing somewhat contentiously, that ‘[u]nder customary international law, immunity can never result in impunity for grave violations of the core values consolidated in international human rights law’.⁶³

Despite the ICC’s approach that there is no CIL rule supporting the applicability of immunities for international crimes in an international court, there is no general understanding on the status of this rule, as illustrated by the several amicus curiae briefs submitted on the issue at the ICC, strong political resistance from the African Union,⁶⁴ and some previous jurisprudence of the court itself. For instance, in the South Africa Decision in the Al Bashir case, the ICC Pre-Trial Chamber II held that ‘[t]he Chamber is unable to identify a rule in customary international law that would exclude immunity for Heads of State when their arrest is sought for international crimes by another state, even when the arrest is sought on behalf of an international court’.⁶⁵ Further to this, the practice of states on the matter remains far from consistent, constant and uniform, and expressions of opinio juris are few and far between.⁶⁶

The Al-Bashir case and its discussion on immunities touches on a core tenet of interpretation: it all hinges on the expectations. What tasks and results should international courts, other norm-interpreters and international law in general, deliver? For instance, while many scholars debate the ICC’s recent decision on Heads of State immunity under CIL,⁶⁷ advocates for ending impunity, human rights organisations and several scholars have cheered at the decision, viewing it as very much the correct interpretation of CIL of immunities in international tribunals.⁶⁸ This is to show how different levels and categories of misinterpretation will be most definitely welcomed by one audience or another. Interpreters, of course, are aware of this and can strategically adjust the method of interpretation and, consequently, the ensuing norm, to address the target audience.⁶⁹ In this context, Andrea Bianchi has noted that ‘interpretive strategies [are] adaptable and flexible enough to serve different purposes’.⁷⁰

Finally, misinterpretations may be politically motivated with little legal foundation or objective. They may be just plain wrong, with flawed methodology and conclusion. As noted above, examples of such outrightly erroneous misinterpretation of CIL at the international level remain scarce. One example could be devised from the torture debate in the early 2000s in the USA.⁷¹ Similarly, in vain, a few scholars tried to argue in 2003 that the Bush Doctrine – as some still recall meaning preventative war with pre-emptive strikes and so on – had developed into CIL, or even into instant CIL.⁷² These misinterpretations did not come from internationally authoritative sources nor did they do much more than stir some debate and yield material for scholarly articles. They remain cautionary tales of getting CIL wrong but also attest to the resilience of traditional sources theory and the approach of international legal scholarship in preserving the core of what constitutes CIL, even when recognising that due to their morally distasteful nature they were bound to invoke strong opposition.

4 Consequences of Misinterpretation

The implications and consequences of misinterpretation vary depending on the original form of misinterpretation. Corrupt pedigree and self-fulfilling prophecy are by no means mutually exclusive but may complement one another in a fertile environment. They may be reflections of different stages of the customary process: originating from a corrupt pedigree, but resulting in a self-fulfilling prophecy when viewed with subsequent hindsight. What, then, are the consequences of misinterpretation?

Practice – as some examples set out in Section 3 and many in this volume illustrate – suggests that the interpretative rules of CIL may be fluid and open to interpretation themselves (just like Articles 31–33 VCLT have been given various interpretations). Hence, the methods do differ and even when employing the same set of methods or rules of interpretation, different interpreters may reach different outcomes. Interpretation is inevitably connected to cognitive frames and social needs,⁷³ and extra-legal considerations are omnipresent and impact the process and result. It depends, not least, on the interpreter’s position towards CIL how explicitly these considerations inform his or her interpretative methodology.

Fernando Teson offers Nicaragua’s determination of the rule of non-intervention as an example of CIL based on corrupt pedigree, in which the court failed to cite practice, precedent or consensus.⁷⁴ The ensuing practice, precedents and consensus consolidate and perpetuate the legal error, and themselves become sort of precedents creating a corrupted chain of legal justification. As opposed to Teson, who claims that the only kind of fake custom that has the power to generate genuine custom is the false legal statements made by states if they are then clearly followed widely by the international community,⁷⁵ I argue that the same must go for decisions of international courts, but only if subsequent practice confirms the rule. This idea can be implicitly found also in the ILC Draft Conclusions, which note that while the practice of international courts and tribunals is not state practice, pleadings by states in those forums can be.⁷⁶ Hence, when a state would refer to the case law containing the ‘corrupt pedigree’ in front of a court, this could be viewed as constituting such a confirmatory subsequent practice. As Harlan Cohen notes, ‘precedent must be understood within practice [– or community of practice –] of international law [and its] force derives solely from the desirability of the rule reflected in it’.⁷⁷ If the precedent is not cited, followed or endorsed in any way, it has very little authority on its own. At best, it can produce a strong presumption that the interpretation is in fact the rule, creating a compliance-pull. This depends on multiple factors: the quality of legal reasoning, the clarity of interpretation, adherence to prior interpretations, and how well the interpretation fits within the broader legal framework, the aspirations of the parties and the potential burden it imposes on those parties.⁷⁸ Thus, even with corrupt pedigree, a norm may under favourable conditions eventually spawn into real custom.

Antonio Cassese has suggested that acquiescence to a misinterpretation would have the same effect as affirmation through practice and consensus, as is the case with the formation of CIL⁷⁹ – where ‘silence equals consent’. I am sceptical of this position for it sidesteps any requirement of actual practice, basing the existence of CIL merely on ‘combination of a string of decisions … coupled with the implicit acceptance or acquiescence of all the international subjects concerned’.⁸⁰ Teson argues against such flex CIL methodology, stating that ‘citing a multitude of non-binding documents does not turn a proposed norm into a binding customary norm because is neither anchored in state practice nor is the object of a universal and specific consensus’.⁸¹ Repetition alone does not confer normativity or legitimacy! In order to sustain a level of legitimacy, the interpreter cannot exclusively refer to their own practice and create a cyclic self-asserting method of interpretation – an external confirmation or affirmation is required, even when the end result may be what I refer to as a self-fulfilling prophecy. While it may be likely that with multiple decisions discussing the same rule at least some would enter a detailed analysis of practice and opinio juris, we cannot base a coherent conceptual analysis of CIL merely on judicial practice, neglecting evidence of practice or its absence. This would place a completely new burden on states to react to decisions of courts (and potentially other norm-interpreters) in order to ensure they will not be bound by the so-called CIL arising from their decisions.

Subsequent state practice and consensus can legitimise the misinterpretation, which has then served as a catalyst for change. When this is the case, the discovery of past misinterpretation – classifying a norm under CIL when it has not yet so crystallised – does not denounce its subsequent normative validity, if it has been followed as if it were already part of CIL. So, in determining the consequences of misinterpretation, we must go back to the roots of CIL before and after the act of misinterpretation, to look at the practice and opinio juris, and to assess whether sufficient affirmation exists which renders the legal basis of the CIL rule. This affirmative consequence of misinterpretation was noted already at the International Military Tribunal for the Far East in 1953, with Judge Pal stating that ‘law also can be created illegally otherwise than by the recognized procedures … any law created in this manner and applied will perhaps be the law henceforth’.⁸² Misinterpretation can, naturally, lead to positive as well as negative outcomes. It may be that a progressive approach and dynamic interpretation, even when considered incorrect either methodologically or substantively, directs the development of practices and beliefs towards more just processes and outcomes.⁸³ Even if the rule was not customary at the initial point, the subsequent practice may override the initially faulty interpretation as the norm gains wider usage, which is supported by illustrations of opinio juris.⁸⁴

5 Concluding Remarks

Is it possible to avoid misinterpretation in international law and if not, how can the negative consequences be mitigated? For CIL to develop, some instances of misinterpretation may be inevitable, serving as test cases on what states and other actors perceive to be the acceptable limits of the law at a given time. Determining the relevance of misinterpretation is a retrospective exercise, and in that sense, if courts deliberately enter unknown, uncertain or even outright incorrect legal terrain, they risk facing accusations of judicial activism or ‘effective interpretation’, which may affect their institutional legitimacy; at the same time, if all plays out well, their (mis)interpretation may instigate the crystallisation of a new (and maybe better!) CIL rule.

Identifying misinterpretation is a challenging task, which depends on the underlying approach to CIL. If one concedes that CIL forms through various ways and that its elements rest on a broad range of evidence beyond settled understanding of what amounts to state practice and opinio juris, one is likely to also accept the wide-ranging methods of interpretation of CIL – and consequently, find less occurrences of misinterpretation. To remain faithful to the traditional notion of CIL – which is embodied in practice – and to preserve legal certainty and predictability, it is crucial to recognise that courts are not infallible, sometimes lacking the requisite methodological tools, and occasionally just getting CIL wrong. The indeterminacy of CIL renders futile the attempts to pin down precise conditions for its validity, and simultaneously, leaves open the definition of misinterpretation of CIL. Misinterpretation, in general, diminishes foreseeability and consistency of the law. It may also, however, push the actors to develop the law. With the inescapable constructive dimension of CIL, courts implicitly serve a key function in the development of CIL through their interpretations and their interpretative methodology.

The misinterpretation of a customary norm, which is subsequently followed by states and other entities as if it were part of CIL, creates a self-fulfilling prophecy – a self-generating crystallisation of a rule. Even if the rule was not customary law embedded in practice and opinio juris at its ‘inception’, the subsequent practice and acceptance eradicates the mishap of the initial faulty interpretation and legitimises the rule as part of CIL. On the other hand, as examples from international criminal tribunals illustrate, a later decision may denounce the misinterpretation and correct the course of the customary process and norm development, or the misinterpretation will remain an unfortunate but soon forgotten misstep, neither to be restored nor repeated.