Skip to main content Accessibility help
×
Hostname: page-component-586b7cd67f-l7hp2 Total loading time: 0 Render date: 2024-11-23T01:25:24.536Z Has data issue: false hasContentIssue false

13 - Indicators of Coherence and the Interpretation of CIL

from Part III - CIL and Its Interpretation in the Normative Universe

Published online by Cambridge University Press:  aN Invalid Date NaN

Marina Fortuna
Affiliation:
Rijksuniversiteit Groningen, The Netherlands
Kostia Gorobets
Affiliation:
Rijksuniversiteit Groningen, The Netherlands
Panos Merkouris
Affiliation:
Rijksuniversiteit Groningen, The Netherlands
Andreas Føllesdal
Affiliation:
Universitetet i Oslo
Geir Ulfstein
Affiliation:
Universitetet i Oslo
Pauline Westerman
Affiliation:
Rijksuniversiteit Groningen, The Netherlands

Summary

Interpretation is ubiquitous in everyday life. We constantly interpret a variety of objects. Interpretation is central to the practice of international law, too. Arguing about international law’s content is the everyday business of international lawyers, and this often includes arguing about the existence and content of norms of customary international law (CIL). Although a number of scholars recognise that CIL can be interpreted, disagreements remain as to the precise methods and extent of CIL interpretation. Such disagreements are born of a common concern to secure competently made, coherent and accurate interpretations of CIL, given the latter’s non-textual nature. This chapter aims to explore in a preliminary manner two related questions regarding CIL interpretation: (1) Is it necessary, or even possible, to strive towards coherence in the interpretation of CIL? (2) Are there any possible indicators of (in-)coherence in that respect? Providing answers to these questions depends on how one understands coherence in the first place, including its relation to legal reasoning. A substantial part of the chapter will therefore deal with that as well.

Type
Chapter
Information
Customary International Law and Its Interpretation by International Courts
Theories, Methods and Interactions
, pp. 299 - 325
Publisher: Cambridge University Press
Print publication year: 2024
Creative Commons
Creative Common License - CCCreative Common License - BYCreative Common License - NCCreative Common License - ND
This content is Open Access and distributed under the terms of the Creative Commons Attribution licence CC-BY-NC-ND 4.0 https://creativecommons.org/cclicenses/

1 Introduction

Interpretation is ubiquitous in everyday life. We constantly interpret a variety of objects, including texts, signs, gestures, works of art, intentions, and human practices and interactions. Interpretation is central to the practice of international law, too. Arguing about international law’s content is the everyday business of international lawyers, and this often includes arguing about the existence and content of norms of customary international law (CIL). Although a number of scholars recognise that CIL can be interpreted,Footnote 1 disagreements remain as to the precise methods, modalities, and extent of CIL interpretation. Such disagreements are born of a common concern to secure competently made, coherent, and accurate interpretations of CIL, given the latter’s non-textual nature.

My aim in this chapter is to explore in a preliminary manner two related questions regarding CIL interpretation: (1) is it necessary, or even possible, to strive towards coherence in the interpretation of CIL? And (2) are there any possible indicators of (in-)coherence in that respect? Providing answers to these questions depends on how one understands coherence in the first place, including its relation to legal reasoning. A substantial part of the chapter will therefore deal with that as well.

Coherence is here envisaged as a state of axiological compatibility existing among a set of independent beliefs, statements, reasons for action, and so on, which are expressed at any level of abstraction. The degree of compatibility of the set is a function of the existence of internal structures of mutual, reciprocal support among the elements of the set, so that some of the elements are supported by other elements and vice versa.Footnote 2

Based on this understanding, I make three broad claims. First, coherence is related to, yet still independent of, concepts like consistency, correctness, and comprehensiveness (and its corollary, predictability); and, moreover, it has two dimensions (one substantive, the other methodological) which are both relevant in law (Section 2). Second, this makes coherence particularly relevant to legal reasoning, where its dual dimension translates into a similar dual role that is also at once substantive and methodological (Section 3). Given this, coherence in law is more than a simple goal or a results-oriented ideal. It is also a method of constructing one’s reasoning and of deliberating about one’s interpretative choices. It is therefore both possible and, indeed, necessary to strive for coherence in the interpretation of CIL if one wishes to claim authority for, and persuade others of, one’s interpretation. Third, striving for coherence in the interpretation of CIL means being cognisant of three kinds of processes – namely, framing, contextualising, and iteration/reflexivity (Section 4). These can serve as indicators of (in-)coherent interpretations; however, they themselves are interpretative in nature and therefore can be subject to debate and reasonable disagreement.

2 The Independent Concept of Coherence

We tend to place value on coherence because it implies that something, or someone, makes sense and is intelligible. Being incoherent, by contrast, causes frustration and confusion. Coherence is thought to be a desirable attribute to have in virtually every aspect of one’s life. The legal field is no exception, where there exists a vast literature on coherence in law in general and a large consensus that coherence suits law and legal reasoning particularly well.Footnote 3 However, coherence remains a largely under-theorised concept in the field of international law. One can, for example, find passing references to coherence in the ILC’s work on the fragmentation of international law. Therein, the ILC indicates that there exists a link between coherence and the principle of systemic integration under Article 31(3)(c) of the Vienna Convention on the Law of Treaties (VCLT).Footnote 4 Beyond that, however, coherence seems to be generally regarded by the ILC as co-terminous with mere legal security and predictability, which is to say it is seen as a formal principle devoid of its own independent content.Footnote 5 One sees a similar picture in other areas of international law, notably international investment law, where concerns of incoherence in arbitral awards have been raised for years alongside concerns about a perceived lack of judicial consistency, correctness, and predictability.Footnote 6 Here, too, coherence is regarded as valuable, yet the content given to it is often interchangeable with that of legal certainty, predictability, and legal authority.Footnote 7

Therefore, before we can ponder the role of coherence in interpreting CIL, one key question must be answered: what is coherence and what is its role in legal reasoning more generally? This section seeks to answer the first half of that question (what is coherence?) by demarcating coherence’s scope from other, related concepts like legal consistency, correctness, and comprehensiveness (and its corollary predictability).Footnote 8 The goal is not simply to reduce confusion by disentangling the meaning of concepts but also to set proper expectations of the concept of coherence in the process. The second half of the question (the role of coherence in legal reasoning) is dealt with in Section 3.

2.1 Coherence and Consistency in Law

Consistency means absence of contradiction.Footnote 9 Propositions are consistent if each can without contradiction be asserted in conjunction with every other proposition in the same set, and with their conjunction.Footnote 10 Consistency is thus an absolute and logical property: any two propositions either are or are not consistent. By contrast, the conditions of satisfaction of coherence are less precise. The quality of being coherent is often conveyed through mental images and metaphors.Footnote 11 It is often said, for instance, that something is coherent if it hangs together well, if its parts fit and are mutually supportive, if it is intelligible, if it flows from or expresses a unified viewpoint; whereas the same thing would be incoherent if it is unintelligible, fragmented, or disjointed.Footnote 12 More generally, an attribute of coherence is that the object of enquiry ‘makes sense’, whereas an attribute of incoherence is that it does not.Footnote 13

Neil MacCormick once used the following humorous example to make the distinction between consistency and coherence stand out more clearly. MacCormick asks us to imagine a house where the cleaning rules are as follows: leave everything as untidy as possible on Mondays, Wednesdays, and Fridays, but tidy everything up to the highest perfection on Tuesdays, Thursdays, and Saturdays (Sundays are reserved for resting). These rules can be easily observed, in the sense that none infringes upon the other, but they surely give rise to an unreasonable, indeed absurd, way of living, since they do not seem to serve any immediately discernible end – in short, they make no sense. The image of coherence painted by this example, and the general descriptions above, is that of a network – an interdependent web – between propositions and their justification standing or falling together.Footnote 14

Transposed to law, it is therefore possible to restate the quality of being coherent – that is, a function of the ability to ‘make sense’ – as the degree to which a set of propositions are rationally related to each other and to the ordering values or principles that are thought to justify them – which MacCormick usefully summarised in the pithy phrase ‘axiological compatibility among two or more rules, all being justifiable by reference to some common principle or value’.Footnote 15

Aiming for axiological compatibility involves a conscious effort to bring two or more propositions into accord with each other. As such, coherence is not an absolute property but rather a matter of degree and interpretation. Whereas consistency ‘is or is not’, coherence can be ‘more or less’. In the same vein, unlike coherence, consistency as a logical property remains agnostic towards the merits of the positions that are made consistent.Footnote 16 Thus, ironically, greater consistency may actually perpetuate injustice. It follows that between consistency and coherence, consistency is the subordinate and dependent concept. For example, a perceived inconsistency between two judicial decisions may be interpreted away if one goes higher in the level of abstraction and considers the more general norms, principles, or values that have justified them. By the same token, two legal positions that appear consistent because they yield the same outcome may still not cohere if one looks at the norms, principles, or values that have justified them.

2.2 Coherence and Correctness in Law

Unlike coherence and consistency, coherence and correctness in law are generally not conflated with each other at a conceptual level or treated as interchangeable. However, because coherence is not agnostic towards the merits of the propositions that are made coherent, it is sometimes proposed that a requirement associated with coherence is the ability to arrive at right answers. That is to say, it is sometimes proposed that a coherent whole should be able to determine the single correct outcome in particular cases.Footnote 17 However, this is not precisely the case, and the statement thus requires more nuance so as to avoid introducing distorting expectations of coherence. While there exists correlation between coherence and correctness, it is doubtful that this extends to causation. To see why, we must further distinguish between determinate and demonstrable correctness.

Something is determinately correct when the reasons offered in its support compel it, which is to say when they carry the greatest weight out of all other competing, plausible reasons supporting it.Footnote 18 Determinate correctness is an instance of so-called ontological objectivity, which seeks to describe the way things are – their nature and existence – in this case by making particular reference to their ability to produce accurate answers.Footnote 19 By contrast, something is demonstrably correct when its determinate, or at least plausible, correctness can be demonstrated to the satisfaction of everyone, or virtually everyone. Demonstrable correctness is thus an instance of so-called epistemic objectivity, which refers to the state of rational agents having formed a justified belief regarding the correctness of the object in question.Footnote 20 Determinate and demonstrable correctness are not absolute properties, but matters of degree. The stronger the reasons offered in support of an outcome, the more they would determine that outcome and, by extension, the more likely it would be that others are persuaded of the outcome’s correctness.

The above applies to the legal field as well. Law has an arguable character, meaning that statements about its content are in reality arguments supported by reasons whose relative strength and manner of deployment can vary.Footnote 21 Evaluating the force of competing legal arguments is bound to be a matter of degree and calls for the exercise of judgement.Footnote 22 Correctness in law, therefore, cannot be simply reduced to a mere pronouncement (e.g. ‘the law demands x’, or ‘party A must bear liability’) but is rather a combination of the content of the pronouncement and the content and structure of the reasons offered in supporting it.Footnote 23 The quality of correctness in law is always determined and demonstrated relatively to a set of strung-together premises and assumptions acting as reasons favouring some outcomes or courses of action over others.

Because coherence is axiological compatibility with guiding norms, values, or principles, coherence is thus linked to legal correctness in the following manner. One person’s ‘correct’ interpretation or decision may still be thought to cohere better than another’s, given the justification offered for each interpretation or decision, and given the method used to arrive at them. In other words, the existence of a coherent process of reasoning would mean that its outcome has a claim to correctness, but also that such a claim would not be absolute since another person’s process of reasoning may have been performed better, thus making the latter person’s proposed outcome arguably more coherent and with a stronger claim to correctness.

It follows, then, that coherence and correctness in law are not necessarily equivalent properties. An otherwise coherent process of justification may still result in an outcome which we may regard as legally incorrect for being in tension with collective attitudes and beliefs. One may use an appropriate method and link together the various aspects of the legal system in a competent manner (e.g. by properly identifying the relevant materials, legal norms, or past practices) but may nonetheless make a substantive, value-laden interpretation of them, with which at least some participants in the legal system may not agree. By the same token, determining that an outcome is coherent cannot be exclusively a function of the extent to which the decision is considered to be correct in its substance. One may reach a correct decision in substance, which everyone would agree that the law ought to provide, but do so through contradictory reasoning or through a not well-justified process, where the various elements of the legal system have not been linked together adequately or accurately. In such a case, we would be inclined to conclude that coherence is lacking, despite the moral appeal that the outcome itself may have.

2.3 Coherence and Comprehensiveness in Law

Yet a third requirement often associated with coherence is that of comprehensiveness, meaning the absence of gaps or uncertainties and the corresponding ability to readily supply an answer to each and every question that may be raised.Footnote 24 In the legal field, comprehensiveness can be linked to the value of predictability – that is, the ability of the law’s addressees to determine in advance the legal consequences of their actions. Comprehensiveness and predictability are properties that obtain in degrees. Enhancing comprehensiveness, in the sense of expanding the range of readily available answers to distinct questions of law, can result in greater predictability for the legal system’s addressees.

It may be a reasonable assumption to make that if a legal system is sufficiently comprehensive and thus exhibits a high degree of predictability, then it would also be coherent. However, it can be doubted whether such a connection would necessarily exist. Comprehensiveness does seem to improve the chances of overall coherence but probably cannot guarantee it, and, conversely, coherence does seem to improve the chances of eventually experiencing comprehensiveness over time (and hence predictability) but probably cannot guarantee that either.Footnote 25 As was the case with correctness previously, correlation does not equal causation.

The above conclusions follow logically from the previous examination of the links between coherence, consistency, and correctness. Because coherence is characterised by axiological compatibility, as opposed to mere non-contradiction, the predictability that would normally result from a nominally consistent line of interpretations or applications of the law does not mean that these would necessarily exhibit coherence also. Furthermore, one need not be fully comprehensive in scope, or perfectly predictable at the outset, to be coherent. This again applies especially in law, considering the well-recognised need for change and progressive development following the occurrence of novel situations not covered in existing legal instruments or following a change in collective attitudes and beliefs. By the same token, the inverse also holds true. Coherence does not preclude the need at times to strike a different path from what past consistent practice may indicate. Nor does coherence preclude the existence of gaps in the law.

2.4 Summing-up

How do consistency, correctness, and comprehensiveness fare as determinants of coherence? In the abstract, one may be inclined to say that the occurrence of all three is necessary to describe an object of study as coherent in ideal theory. Thus, the perfect co-existence of all three elements will necessarily render the object of study perfectly coherent in an ideal situation. Conversely, the absence of all three elements would render the object incoherent. However, based on the preceding analysis, the same conclusion does not seem to follow in practical, non-ideal situations. Indeed, three general points can be made in this respect.

First, although certainly desirable for coherence to exist, consistency is not a necessary or sufficient requirement of it. The presence or absence of consistency does not necessarily mean that coherence obtains or that coherence is forfeit. In fact, consistency seems to be the dependent concept when seen against coherence. Often, it is a sense of coherence that makes us either disregard contradictions as being only apparent (and thus capable of being interpreted away) or, indeed, call them out as being true.Footnote 26

Second, correctness is likewise not a necessary determinant of coherence. Although there exists a correlation between legal correctness (both determinate and demonstrable) and coherence, this does not extend to causation, given the law’s arguable character and the scope within it for debate and reasonable disagreement. Legal pronouncements may not always be determinately correct, yet they could still exhibit congruence with some of the system’s principles or values and be supported by a competently drafted statement of reasons. Incoherence resulting from a complete disregard of a system’s norms, practices, and principles, or from a complete disregard of the ways of structuring and presenting legal arguments, are likely to be rare.

Third, the same considerations apply also with respect to comprehensiveness as a determinant of coherence. Some unpredictability resulting from an absence of comprehensiveness is not necessarily a sign of incoherence. The law is always a work-in-progress and unpredictable outcomes may be caused by legal actors using novel doctrines, arguments, or theories that are not yet settled or widely accepted in the current state of the discipline.

The above leads to the conclusion that coherence is an independent concept having its own content. Were it not, we could only perceive it to the extent, if any, that it entails or is entailed by consistency, correctness, and comprehensiveness. That is to say, the conclusion ‘coherent or incoherent’ would necessarily follow from any statement about the law, so long as we considered that statement to be consistent, correct, or comprehensive (predictable). However, this does not appear to be the case in practice. To the contrary, we are ultimately not precluded from perceiving coherence in practice or from debating about coherence meaningfully, even when some of the above-identified elements are satisfied to a lesser extent or even entirely missing.

In the same vein, striving for coherence in the determination and interpretation of CIL should not mean that consistency, correctness, or predictability will be necessary outcomes, although they may be probable and, indeed, highly hoped-for outcomes. Does this then make coherence practically irrelevant when engaging in CIL interpretation or, for that matter, in any form of legal reasoning? Section 3 argues that it does not.

3 Practical Legal Reasoning and the Dual Role of Coherence

Legal reasoning is the principal means through which lawyers argue about the existence and content of law. This makes it natural to enquire about the role that coherence may have to play therein. Yet, answering this question presupposes answering a more fundamental question first: what kind of reasoning is legal reasoning? There are two candidate-types to consider in this respect: the theoretical and the practical. A fundamental difference between the two relates to their respective conditions of validity and the outcomes that each type of reasoning seeks to achieve.

Starting from the proposition that, much like other kinds of normative reasoning, legal reasoning is practical rather than merely theoretical, this section argues that seeing legal reasoning as practical leads to the conclusion that coherence has a dual role to play therein, a role that is at once substantive and methodological. The section concludes by hinting at three interpretative processes acting as possible indicators of (in-)coherence in legal reasoning, termed ‘framing’, ‘contextualisation’, and ‘iteration’/‘reflexivity’. Their importance for the interpretation of CIL is subsequently examined in Section 4.

3.1 Theoretical versus Practical Reasoning

In theoretical argumentation, one adduces reasons for or against forming a belief about what is or is not the case in reality.Footnote 27 Theoretical reasoning is thus reasoning principally about questions of fact, explanation, and prediction.Footnote 28 As such, it has a backward-looking and a forward-looking dimension. It is backward-looking when it looks at events that have occurred and asks why or how they have occurred. It is forward-looking when it uses past observations and the rules of logic to attempt a true determination about what is going to happen in the future. Given its format and reliance on the rules of logic, theoretical reasoning can be challenged in two ways: first, substantively, when the premises used are not accurate, thus leading to a necessarily inaccurate outcome (inaccurate assessment or collection of data); or, second, formally, when the syllogism itself violates the rules of logical consequence or entailment (impermissible moves between otherwise accurate premises).

Practical reasoning is different. Practical reasoning is reasoning about what to do when faced with a problematic situation. As such, it does not stop at formulating an opinion or belief about what the case may be, as theoretical reasoning does, but continues until one commits to a course of action. Put differently, practical reasoning attaches to a project an actor has (be it a problem requiring solution or a goal to be achieved) and seeks to identify the course of action that would lead to the satisfactory completion of that project.Footnote 29 From this fundamental point, a number of key differences between theoretical and practical reasoning emerge.

First, whereas the ends in a theoretical inference are clearly defined and certain (i.e. ascertaining the truth of a proposition or accepting it as a matter of belief), this is not always the case in a practical inference. Indeed, when we contemplate possible courses of action in practice, we often come across phenomena like complexity, uncertainty, instability, uniqueness, or value-conflict.Footnote 30 In these situations the ends cannot always be taken as given – which is to say, without some prior attempt at ascertaining their desirability or appropriateness as ends to be achieved in the first place.Footnote 31 Therefore, in resolving a practical problem, an actor must also deliberate in order to prioritise or reconcile competing ends, including also the requirements underpinning the available courses of action to meet these ends.Footnote 32 Reconciling and prioritising may lead the actor to choose a different course of action and to even amend their original goal. Practical deliberation thus requires the actor to simultaneously determine how to do what seems like a good idea of doing as well as whether it really is a good idea to do it in the first place.Footnote 33

Second, because ends in practical contexts are not beyond challenge or amendment, rationality in practical reasoning is not based only on logical entailment (as in theoretical reasoning) but also on plausibility. This means that although the outcome of a properly conducted theoretical inference will be the same for everyone, this is not a guarantee in a practical inference.Footnote 34 Two actors may both reason correctly from a logical point of view but still arrive at different courses of action, since there may be multiple suitable ways to achieve the same intended goal. Accordingly, rationality in a practical inference is a function of the degree to which a desirable end x is achieved by a course of action y, subject to any competing or countervailing reasons for not doing y given the possible existence of an additional, also desirable, end z. By enlarging the scope of the debate in this manner, the rationality of a practical inference may be affirmed or challenged.Footnote 35

Third, because of the above characteristics, practical inferences can be challenged in three ways: first, substantively, when the premises used are not accurate; second, formally, when the inference itself violates the rules of logical consequence or entailment; or third, defeasibly, through the introduction of additional contexts to the enquiry that do not necessarily make existing premises inaccurate yet, on balance, render them inappropriate and replaceable by better premises.Footnote 36

3.2 Legal Reasoning as Practical Reasoning

Traditional depictions of legal reasoning in international legal academia have closely tracked the theoretical model of reasoning just described.Footnote 37 Legal reasoning is thus often depicted as a chain of logical deduction consisting of a syllogism of the form ‘if p then q’, where p stands for a proposition of law and q stands for the consequence that follows from it.Footnote 38 The reasoner’s role is to input the set of facts that match the applicable proposition of law in order to get the legally prescribed solution. The more difficult the case, the more complex the chain of reasoning. Although this picture of legal reasoning can be challenged for being outdated and overly formalistic, it remains the prevailing narrative when it comes to CIL. That is, in relation to CIL, the above picture is often reversed by saying that the reasoner’s task is to identify the rule of CIL that is hidden in the raw facts of state practice in a process of logical induction.Footnote 39 Thus, the fundamentally syllogistic ‘if p then q’ format of reasoning is retained.

However, although often presented outwardly as a chain of demonstrable theoretical reasoning, legal reasoning better fits the format of practical reasoning.Footnote 40 Indeed, legal reasoning exhibits the three main characteristics of practical reasoning identified previously.

In the first place, legal reasoning intends to lead to action and not simply to the formation of a belief about the way things are in the current state of the law. States typically invoke law and make arguments about the law to justify taking a course of action in the future, to justify having taken a course of action in the past, and to block other states’ justifications for following one course of action rather than another. Similarly, when seised of a case, international adjudicators do not simply declare the law in the abstract; they commit themselves to a course of action by handing down a decision in favour of a disputing party, which is tailored to the particular circumstances of the case presented before them.

In the second place, legal reasoning is a purposive activity. And yet, as the ILC has also affirmed, the law’s ends are not always clearly defined, indisputably certain, or entirely uniform.Footnote 41 Take international investment law. The preambles of international investment agreements indicate that states enter into them for a variety of reasons, including, for instance, to protect their investors abroad, to stimulate investment flows in their territory, or to contribute to their economic development. The choice to give foreign investors access to investor-state dispute settlement has also been justified on multiple grounds, including as an additional incentive to invest in another countryFootnote 42 or as a way to depoliticise investment disputes by removing them from the domain of diplomatic protection.Footnote 43

The same considerations apply, arguably with greater force, in the case of CIL. As Gorobets points out, customary norms are community norms.Footnote 44 Therefore, even when it is a small group of states or, indeed, a single state that originally sparked the creation of a CIL norm (one can think of the 1945 Truman proclamation on the continental shelf here), these states may not be regarded as the norm’s authors in exactly the same way as the contracting parties to a treaty are that treaty’s authors.Footnote 45 The juridical goals of the original ‘creators’ of a CIL norm are not immune from re-appraisal and re-interpretation by all subsequent actors who may seek to invoke and apply that norm in practice.

The conclusion would not be different even if one were to focus on purely doctrinal goals associated with the law – such as ensuring legal certainty, justice, or the rule of law – and sought to determine how these should be promoted through legal reasoning. We would still need to perform an act of interpretation of these abstract goals case by case, in light of the institutional role of the international actor who advances the legal argument in each instance. In short, interpretability of ends appears to be a ubiquitous feature of legal reasoning.

In the third place, legal reasoning is defeasible. Some legal questions can be genuinely difficult to answer. People who are competent and well informed about the law can still reasonably disagree about its content, despite starting from similar premises and advancing formally valid syllogisms. When lawyers disagree, they do so not only by challenging the accuracy of each other’s chosen premises or the logical consequence of their syllogism, but also by pointing out missing premises or missing contexts in a competing argument, which may render that argument less plausible or convincing.Footnote 46

3.3 A Dual Role for Coherence

To recap, legal reasoning is practical in the sense that it seeks to motivate a situated actor, who has certain goals, into committing to a particular course of action. In practical reasoning, an interactive and iterative relationship exists between the actor’s goals, the course of action chosen, and the mental steps taken in between – all these are, moreover, open to deliberation and modification in the course of reasoning. Thus understood, legal reasoning exhibits important coherence-related features. Because it is practical, legal reasoning spans out in a web of clustering reasons for action, rather than extending linearly as a chain. Moreover, by having interpretable goals, legal reasoning relies on axiological compatibility between these goals, the reasons that justify pursuing them, and the appropriateness of the steps taken to achieve them in practice. In other words, like practical reasoning, legal reasoning is meant to lead one to a conclusion about what ought to be done (or not done) given one’s evaluation of a problematic situation to be resolved or a goal to be achieved. The strength of a legal argument thus comes principally from the force of the connection – the degree of mutual supportiveness – between these various clusters of reasons.Footnote 47 The latter, in turn, is a function not only of the desirability of the reasons used but also of the way in which these reasons have been structured and made to interact in pursuit of the chosen goal.

What role, then, does coherence play in legal reasoning? I believe it plays a dual role, which is at once substantive and methodological. On the one hand, when faced with a legal problem, one must frame the legal question at issue; identify the normative context, including the governing values, principles, and other norms of conduct, that will guide the decision-making process; and, in so doing, also reflect on one’s institutional role and aims (including on their limits) in the legal system. Coherence here implies an overall congruence with the body of knowledge and value prevalent in the legal system in question,Footnote 48 and thus operates in a clearly normative way as a substantive principle. It assists in formulating, even if in the abstract, the end, purpose, or state of affairs to be achieved and the reasons supporting it in principle.

On the other hand, having come up with the end, purpose, or state of affairs to be achieved, one must then proceed by following an iterative process whereby one identifies concrete courses of action to fit the chosen end, purpose, or state of affairs; considers the impact on the identified actions of any countervailing ends, purposes, or states of affairs; in the process potentially reassesses the desirability of the end, purpose, or state of affairs originally chosen; and, if so, reformulates amended courses of action, as appropriate, to fit the reassessed end, purpose, or state of affairs to be achieved. The idea of coherence here takes on a methodological hue. It describes a reflexive process of internal deliberation and a manner of structuring all relevant elements into a justifying narrative that seems at least plausible, if not compelling, to the community at large.

Singled out from the above are three mental processes associated with coherence in law as potential indicators: (1) acts of framing, (2) acts of contextualising, and (3) acts of iteration/reflexivity. Section 4 offers some preliminary thoughts on how these three processes may fit within the interpretation of CIL.

4 Indicators of (In-)Coherence and the Interpretation of CIL

There exists a link between coherence’s dual role in legal reasoning and the interpretation of CIL. In particular, the three interpretative processes of framing, contextualising, and reflecting can act as indicators of the coherence or incoherence of one’s proposed interpretation of a CIL norm. The word ‘indicators’ is key, however, as it points to something less than a determinative test. Indeed, it is argued here that acts of framing, contextualising, and reflecting are themselves interpretative in nature and their successful deployment can therefore be the subject of debate and reasonable disagreement. Moreover, my listing of the three processes in the above manner does not mean to indicate a strict separation between them or their strictly sequential application in practice. Rather, I see these processes as three interdependent facets of the same interpretative operation.

An additional clarification must also be offered before proceeding. Perhaps unsurprisingly, the very term ‘interpretation’ can itself be subject to interpretation and to different understandings in different fields. In a broad sense, any act of apprehension of meaning and any ‘reading’ of a situation, no matter how commonplace, intuitive, or unmediated, can be said to involve interpretation.Footnote 49 This is not the common understanding of interpretation in international law. Interpretation in the latter is often understood in a narrower sense predicated on the existence of doubt about the meaning or the proper application of a pre-existing rule.Footnote 50 In this narrower sense, CIL interpretation can be seen as taking place only after a CIL rule has been identified;Footnote 51 whereas the broader sense would see interpretation as permeating the entire lifecycle of CIL, from its formation and identification to the apprehension of its meaning now and in the future. The latter approach thus leaves open the interesting, but still somewhat controversial, possibility of state practice being a subject of interpretation also.Footnote 52

In what follows, I have in mind the broader sense of interpretation just identified. I understand interpretation as elucidating the content of legal norms, but also as a reasoned ‘reading’ of a situation populating it with meaning. I recognise that my use of the term has implications about the way one understands the concept of law itself. However, I will not offer any arguments on that here, nor, more generally, on the plausibility of shifting to an interpretative conception of international law.Footnote 53

4.1 Framing

Practice is the foundation of CIL’s existence and content: it is practice that becomes law if it is accepted as such by states.Footnote 54 Yet, identifying the elements of practice to figure into one’s ascertainment or interpretation of a CIL norm does not proceed on a blank slate. Rather, it presupposes a prior framing of a legal question in need of an answer, or of a problematic situation in need of resolution, forming the contours of a ‘project’ that the interpreter intends to complete as their legal goal.Footnote 55 The ‘project’, appropriately framed, helps to kick-start and guide the enquiry methodologically as well as substantively.Footnote 56 Thus understood, acts of framing are interpretative in at least three ways.

In the first place, framing helps to formulate legally acceptable goals to be achieved through the ascertainment of the content of a CIL norm. As Hakimi points out, international actors do not approach CIL as detached observers, but rather as ‘advocates advancing their own preferences’.Footnote 57 An understanding of the interpreter’s institutional role is therefore important. Different international actors have different institutional roles, affecting the kind of questions that can be asked, the legal problems that need resolving, or the arguments that can be made. For instance, states enjoy greater freedom in devising policy goals to be pursued through legal means. States can thus generate arguments about possible new CIL norms, or about new meanings to be assigned to existing CIL norms, simply by acting and by justifying such action against existing practice or against their reasoned views about the action’s appropriateness or desirability in the legal system.Footnote 58 By contrast, the institutional role of international adjudicators is more constrained. When determining the content of a CIL norm, adjudicators can proceed acceptably only by attempting to reconstruct the intentions of those states that have engaged in relevant practice and substantive argumentation in relation to a putative CIL norm.Footnote 59

In the second place, framing helps to impose order and priority in the virtual infiniteness of state practice. One cannot hope to isolate those elements of state practice that would be relevant for examination, or to assess their legal relevance, unless one has already framed with some degree of specificity a legal problem to resolve or a legal goal to achieve. If, for instance, one wishes to advocate for a new, less strict content of the CIL minimum standard of treatment of aliens in the context of foreign investment protection, one should look for evidence of such content, among others, in the development of investment treaty provisions over time, interventions made by states before international forums such as UNCITRAL or UNCTAD, positions taken by states in arbitration or in non-disputing state party submissions made in the context of investment disputes, relevant diplomatic exchanges, or even judgments by domestic courts addressing grievances of foreign investors.Footnote 60

At the same time, the weight, authority, or priority of different sources of evidence of practice are all relative, meaning that some materials of practice may only be relevant prima facie and that an assessment must be made case by case.Footnote 61 Thus, in the third place, framing the legal issue appropriately also helps the interpreter to determine the kind of practice they should establish or look for.Footnote 62 For example, in Nicaragua v. Honduras, the ICJ considered in obiter that in certain circumstances the application of the CIL principle of uti possidetis could be extended to the delimitation of maritime boundaries (as opposed to being confined to the determination of territorial boundaries), in particular when this is in connection with historic bays and territorial seas.Footnote 63 No new evidence of state practice or opinio juris was offered for this statement, yet arguably none was needed given the framing of the issue. That is, extending the application of uti possidetis from the determination of territorial boundaries to the delimitation of some maritime boundaries may be said to have been a continuation, by analogy, of the ICJ’s prior conclusion, in Frontier Dispute, that uti possidetis is a ‘principle of a general kind which is logically connected with … decolonization wherever it occurs’.Footnote 64 Put differently, from the point of view of the juridical aim of securing respect for existing boundaries at the moment of independence,Footnote 65 determining a maritime boundary between two newly independent states following decolonisation is not unlike determining a territorial boundary between them.Footnote 66

In short, framing kick-starts the interpretative enterprise by singling out what is, prima facie, legally relevant in determining the content of a CIL norm. In doing so, one necessarily ‘bakes in’ certain assumptions, notably about the existing state of the law, about acceptable ends to be pursued, and about one’s institutional role and capacities in the international legal system.Footnote 67 Framing thus goes hand in hand with an often-implied process of normative contextualisation.

4.2 Contextualising

Implied in framing is contextualising. The latter involves seeing one’s object of enquiry against an accepted system of background knowledge so as to infer normative content out of it. In the case of CIL, contextualisation takes place when seeking to determine the existence of a CIL norm by carving out an area of relevant practice, and also when seeking to determine the meaning or scope of an already established CIL norm by re-assessing practice that has already been identified in the past. In that sense, like framing, contextualisation in CIL interpretation is indispensable, since mere regularity of conduct cannot lead to a normative conclusion on its own and in the absence of some background context.Footnote 68 That is, a collection or a pattern of state conduct cannot by itself be transformed into a reason to act in any particular way absent a rule, goal, or principle acting as the normative context that creates the necessary connections between the observable instances of conduct.Footnote 69

Contextualising practice is an interpretative act, in that it can in principle take place at various levels of abstraction. A pattern of conduct may be identified as relevant state practice when seen against broad areas of international law that already include a network of rules, goals, or principles (e.g. use of force, treatment of aliens abroad) as well as against specific international legal rules or doctrines (e.g. uti possidetis, minimum standard of treatment of aliens). Fixing the level of abstraction thus becomes a choice of critical normative importance. The chosen level of abstraction determines not only whether one has identified enough evidence of practice to formulate a CIL norm, but also whether the identified evidence is of the right sort or whether altogether different evidence must be looked for instead.

Take the ICC appeals chamber decision in Al Bashir, which held that there was no CIL rule of head-of-state immunity before international courts (as opposed to before domestic courts) that is applicable in the horizontal relationship between states in situations where the ICC has issued an arrest warrant against a head of state.Footnote 70 Put differently, for the appeals chamber, head-of-state immunity would not apply even if the ICC had requested the surrender of a head of state who hailed from a non-state party to the ICC. The ICC appeals chamber concluded so after pointing to the fact that no evidence of practice or opinio juris could be adduced for the existence of a CIL rule of head-of-state immunity in these situations.

For present purposes, one can point out the following vis-à-vis the content-determinative nature of contextualisation. The ICC seems to have initially framed the pertinent legal question adequately (i.e. does head-of-state immunity apply between state parties and non-state parties to an international court when that court has issued an arrest warrant against a head of state?), yet seems to have contextualised its assessment of observable practice narrowly compared to the question (i.e. is there evidence to suggest a rule that states recognise immunity for each other’s heads of state when an international court has issued an arrest warrant?).Footnote 71 Having done so and having observed no such evidence, the ICC appeals chamber came to the inescapable conclusion that no rule of CIL exists mandating immunity in these situations.

The outcome could have been markedly different had the contextualisation taken place at a higher level of abstraction, for instance by considering the underpinnings of the doctrine of head-of-state immunity. If one takes the view that the doctrine derives from the principle of sovereign equality, then the fact that the immunity may be raised before an international rather than a domestic court becomes immaterial. Instead, what would be critical is whether the head of state whose immunity is waived is a national of a state party to the international court in question. If they are not, the CIL rule on immunity would still apply, since waiving it would violate the principle of sovereign equality.Footnote 72 Moreover, making the latter argument does not require one to search for additional evidence of state practice or opinio juris. In fact, it makes the kind of evidence that the appeals chamber was searching for in this instance beside the point.

Crucially, despite the critical interpretative importance of contextualisation, there appear to exist no determining criteria, and no defined metarules, for what the appropriate level of abstraction is in each case a CIL interpreter contextualises practice.Footnote 73 The case rather seems to be that the CIL interpreter, much like other practitioners grappling with doubt or ambiguity in law and elsewhere, must make such determinations on the spot, in the act of interpreting, and without the benefit of hindsight.Footnote 74 In the absence of guiding metarules, a conscious and active process of reflection may thus be the only kind of safeguard an interpreter of CIL has.

4.3 Reflexivity

Thinking reflectively consists in ‘turning a subject over in the mind and giving it serious and consecutive consideration’, says John Dewey.Footnote 75 According to Dewey, reflective thinking can be a useful method to safeguard one’s inferences from falling into obvious errors.Footnote 76 Perhaps the best-known contemporary exposition of reflective thinking comes from John Rawls and his concept of reflective equilibrium as a technique of justification.

Roughly put, the idea behind reflective equilibrium is that we must work back and forth in our reasoning between various levels of abstraction, including between our judgements about particular instances or cases, the principles that we believe underpin those judgements, and the theoretical considerations that we believe bear on accepting our judgements and principles, while revising any of these elements whenever necessary in order to achieve an acceptable level of convergence among them.Footnote 77 An acceptable level of convergence, according to Rawls, is ‘wide’ rather than ‘narrow’. That is to say, we should not seek only mere consistency among our judgements, but we should also strive to ensure that some of these judgements provide the best justification for the others. Therefore, it is acceptable for prior judgements to be modified or new judgements to be added during this iterative process.Footnote 78 The goal is to achieve an ‘optimal’ equilibrium before stopping the enquiry. We arrive at an optimal equilibrium when the component judgements, principles, and theories of our reasoning are such that we are not inclined to revise any further, because, taken together, they have the highest degree of acceptability or credibility.Footnote 79

The reflective equilibrium technique has strong similarities with the understanding of coherence put forward in this chapter as interdependence, mutual supportiveness, and a web of propositions exhibiting axiological compatibility. Not only is interdependence of judgements a core feature of reflective equilibrium, but also the latter’s iterative process precisely seeks to ensure that one’s judgements converge by being mutually supportive and able to stand together as a unit, as the concept of coherence indicates. In short, reflexivity and coherence are conceptually linked. Exhibiting reflexivity in one’s practice is conducive to coherent reasoning, which, in turn, is conducive to a well-justified outcome, substantively as well as methodologically.

A reflexive attitude must thus permeate the entire process of CIL interpretation, permitting and indeed instructing interpreters to critically evaluate and, if need be, amend their interpretative choices and moves before fixing the exact scope of a concrete CIL rule to be applied in the case at hand.Footnote 80 No aspect of the process is beyond the possibility of re-assessment and amendment, including, critically, the framing of the legal issue/goal, the chosen level of contextualisation, and the concomitant normative weight assigned to every observable piece of evidence of practice. This makes CIL judgements essentially axiological overall.Footnote 81 Every interpretative step taken in ascertaining the content of CIL is to be assessed for its conformity to the identified normative background, its consistency with the implications created by earlier argumentative steps, and its practical consequences (whether desirable or problematic) – all combining to yield an appropriate resting place for the interpretative enquiry.Footnote 82

I must make a caveat here. It is not at this point possible to offer a practical example of reflective thinking in CIL interpretation. Much like argumentation in general, putting forward legal arguments or handing down judicial decisions on the content of CIL are performative acts to some extent. That is to say, when expressed publicly, legal argumentation and judicial decision-making are constructed so as to persuade an audience that the argument or decision in question contains well-chosen premises and that the outcome follows rationally from these premises.Footnote 83 There is an inherent difficulty in examining reflective thinking taking place live in practice from that point of view, since any prior act of active reflection during the argument-forming or decision-making process would presently be obscured to the audience.Footnote 84 The audience may then only make reasonable retrospective inferences about the reflective thinking likely at play behind the finalised version of the argument or decision presented to them.

Nevertheless, this does not take away from the importance of reflexivity as good practice of an interpreter’s moral responsibility, particularly in light of international law’s arguable character and legal reasoning’s practical nature.

5 Conclusion

This chapter has argued that it is both necessary and possible to strive towards coherence in the interpretation of CIL, once coherence is properly understood as an independent concept having both a substantive and a methodological dimension. This dual dimension of coherence is critical to law and legal reasoning. It means that coherence in law is more than a mere goal or a vague, results-oriented ideal. It is also a method of constructing one’s reasoning and of deliberating about one’s interpretative choices. The dual dimension is especially important when reasoning in practical settings (and legal reasoning is one such example), where one commits to action and where interpretability of ends is a core aspect of the process. The chapter has concluded by arguing that striving for coherence in the interpretation of CIL means being cognisant of three kinds of processes – namely, framing, contextualising, and iteration/reflexivity. These can serve as indicators of the coherence or incoherence of proposed CIL interpretations. However, the three processes are themselves interpretative in nature and therefore may be the subject of debate and reasonable disagreement.

To be sure, thinking about coherence, legal reasoning, and interpretation in the manner argued here raises fundamental questions about the nature of international law and the prevailing doctrine of sources. These are important and heavily implicated issues in the present analysis, which, however, I have not begun to address in this chapter. What can be said for the moment is that, if persuasive, the present analysis gives us at least an impetus to re-examine key tenets of legal positivism in international law and to re-assess the degree to which they comport with how international actors seem to reason and argue about international law in practice.

Footnotes

1 eg P Merkouris, J Kammerhofer, and N Arajärvi (eds), The Theory, Practice, and Interpretation of Customary International Law (Cambridge University Press 2022); O Chasapis Tassinis, ‘Customary International Law: Interpretation from Beginning to End’ (2020) 31 EJIL 235; P Merkouris, ‘Interpreting the Customary Rules on Interpretation’ (2017) 19 ICLR 126; D Alland, ‘L’interprétation du droit international public’ (2013) 362 RdC 41, 82–88; A Orakhelashvili, The Interpretation of Acts and Rules in Public International Law (Oxford University Press 2008) ch 15; R Kolb, Interprétation et création du droit international: Esquisse d’une herméneutique juridique moderne pour le droit international public (Bruylant 2006) 219ff; C de Visscher, Théories et réalités en droit international public (4th edn, Pedone 1970) 171–72.

2 To note, the requisite strength of the mutually supporting relations cannot be determined in the abstract but rather depends on the standards of rationality prevalent in the domain of human endeavour in which one finds oneself. The standards of rationality in ethics or morality, for example, are not necessarily the same as those found in scientific rationality and empirical proof.

3 For a non-exhaustive list, see A Amaya, The Tapestry of Reason: An Inquiry into the Nature of Coherence and Its Role in Legal Argument (Hart 2015); N MacCormick, Rhetoric and the Rule of Law (Oxford University Press 2005); J Hage, ‘Law and Coherence’ (2004) 17 Ratio Juris 87; J Raz, ‘The Relevance of Coherence’ (1992) 72 BULR 273; R Alexy and A Peczenik, ‘The Concept of Coherence and Its Significance for Discursive Rationality’ (1990) 3 Ratio Juris 130; R Dworkin, Law’s Empire (Harvard University Press 1986); BB Levenbook, ‘The Role of Coherence in Legal Reasoning’ (1984) 3 Law Philos 355. See generally Y Radi, ‘Coherence’ in J d’Aspremont and S Singh (eds), Concepts for International Law: Contributions to Disciplinary Thought (Edward Elgar 2019) 105 (and additional references therein).

4 ILC, ‘Fragmentation of International Law: Difficulties Arising from the Diversification and Expansion of International Law, Report of the Study Group of the International Law Commission’ (13 April 2006) UN Doc A/CN.4/L.682, 211 para 419.

5 eg Footnote ibid 248 para 491 (‘Coherence is, however, a formal and abstract virtue. For a legal system that is regarded in some respects as unjust or unworkable, no added value is brought by the fact of its being coherently so.’).

6 UNCITRAL, ‘Report of Working Group III (Investor-State Dispute Settlement Reform) on the Work of Its 36th Session (Vienna, 29 October–2 November 2018)’ (6 November 2018) UN Doc A/CN.9/964, 6–11 paras 25–63.

7 eg C Schreuer, ‘Coherence and Consistency in International Investment Law’ in R Echandi and P Sauvé (eds), Prospects in International Investment Law and Policy (Cambridge University Press 2013) 391 (‘Coherence and consistency are desirable qualities in any legal system. A legal system is coherent if its elements are logically related to each other and if it shows no contradictions. A legal system is consistent if it treats identical or similar situations in the same way and if it gives equal treatment to the participants in the system.’). Similarly, C Schreuer, ‘Diversity and Harmonization of Treaty Interpretation in Investment Arbitration’ in M Fitzmaurice, O Elias, and P Merkouris (eds), Treaty Interpretation and the Vienna Convention on the Law of Treaties: 30 Years On (Martinus Nijhoff 2010) 139 (‘The need for a coherent case law is evident. It strengthens the predictability of decisions and enhances their authority.’).

8 To note, this chapter does not intend to offer a full account of the concept of coherence or an exhaustive analysis of its different facets. For an effort in that direction, see Amaya (Footnote n 3).

9 LA Kornhauser and LG Sager, ‘Unpacking the Court’ (1986) 96 Yale LJ 82, 103.

10 MacCormick (Footnote n 3) 190.

11 eg JM Pérez Bermejo, ‘Coherence: An Outline in Six Metaphors and Four Rules’ in M Araszkiewicz and J Šavelka (eds), Coherence: Insights from Philosophy, Jurisprudence and Artificial Intelligence (Springer 2013) 93; L Moral Soriano, ‘A Modest Notion of Coherence in Legal Reasoning: A Model for the European Court of Justice’ (2003) 16 Ratio Juris 296.

12 K Kress, ‘Coherence’ in D Patterson (ed), A Companion to Philosophy of Law and Legal Theory (2nd edn, Wiley-Blackwell 2010) 521; S Bertea, ‘The Arguments from Coherence: Analysis and Evaluation’ (2005) 25 OJLS 369, 371–72 (and additional references therein); Raz (Footnote n 3) 276.

13 MacCormick (Footnote n 3) 189–93.

14 Pérez Bermejo (Footnote n 11) 97 (‘This image [of the net] describes the structure and organization of the coherentist systems … [as] highly interconnected cells or neurons, sustained through the mutual support of all [their] elements and permanently open to learning by reacting to any external input.’); Raz (Footnote n 3) 287 (‘[W]e attribute beliefs, goals, and actions to people, not singly but in interdependent clumps. This interdependence means nothing other than a presumption of coherence.’). Similarly, but in a different disciplinary context (text linguistics), see RA de Beaugrande and WU Dressler, Introduction to Text Linguistics (Longman 1981) ch V para 23 (‘Coherence will be envisioned as the outcome of combining concepts and relations into a network composed of knowledge spaces centred around main topics.’).

15 See the analysis in MacCormick (Footnote n 3) 189–93, 230–31.

16 T Schultz, ‘Against Consistency in Investment Arbitration’ in Z Douglas, J Pauwelyn, and JE Viñuales (eds), The Foundations of International Investment Law: Bringing Theory into Practice (Oxford University Press 2014) 297.

17 Kress (Footnote n 12) 528–29. Kress calls this the ‘completeness’ requirement of coherence.

18 On compelling reasons, see W Lucy, ‘Adjudication’ in JL Coleman, KE Himma, and SJ Shapiro (eds), The Oxford Handbook of Jurisprudence and Philosophy of Law (Oxford University Press 2004) 230–31.

19 MH Kramer, Objectivity and the Rule of Law (Cambridge University Press 2007) ch 1. Conversely, something is indeterminate when the reasons that can be offered for or against it are all of equal strength and weight, so that no decision or choice is ultimately better than any other. To note, being ontological states, determinacy and indeterminacy are positive claims that must be argued for rather than merely assumed as default positions (on this, see R Dworkin, ‘Objectivity and Truth: You’d Better Believe It’ (1996) 25 Philos Public Aff 87, 129–31).

20 Kramer (Footnote n 19) 17.

21 On law as a ‘culture of argument and interpretation’, see JB White, ‘Law as Language: Reading Law and Reading Literature’ (1982) 60 Tex L Rev 415, 436.

22 MacCormick (Footnote n 3) 14–15.

23 eg compare the majority decision in Oil Platforms (Iran v USA) (Judgment) [2003] ICJ Rep 161 [31]–[78]; with Oil Platforms, Separate Opinion of Judge Higgins [40] – [54] and Separate Opinion of Judge Kooijmans [41]–[63]. In Oil Platforms, the court’s majority concluded that the destruction by the United States of two Iranian oil platforms, under suspicion of involvement in the attack on two US vessels, was not a legitimate exercise of self-defence under the CIL on the use of force. In their separate opinions, Judges Higgins and Kooijmans concurred with the final outcome (that is, the United States could not rely on the law on the use of force to justify its actions), yet took issue with what they saw as several systemic flaws in the decision – namely (i) that the majority had implicitly changed the framing of the dispute, from that of a commercial dispute brought under a commercial treaty between Iran and the United States, to one about the unlawful use of force; (ii) that the majority had used Article 31(3)(c) of the VCLT inappropriately to incorporate by reference the totality of the CIL on the use of force as applicable law; and (iii) that, in so doing, the majority had violated its jurisdictional mandate. Put differently, and despite their ultimate agreement with the decision’s dispositif, for Judges Higgins and Kooijmans, the Oil Platforms decision was on the whole incorrect given the process of reasoning and justification followed by the majority.

24 Kress (Footnote n 12) 528.

26 MacCormick (Footnote n 3) 190 (‘A story can be coherent on the whole and as a whole, though it contains some internal inconsistencies – and in this case, the sense of the overall coherence of the story may be decisive for us in deciding which among pairs of inconsistent propositions to disregard as anomalies in an overall coherent account or opinion.’ (footnote omitted)).

27 N MacCormick, ‘Argumentation and Interpretation in Law’ (1993) 6 Ratio Juris 16, 16. MacCormick uses the term ‘speculative’ to refer to theoretical reasoning.

28 RJ Wallace, ‘Practical Reason’, The Stanford Encyclopedia of Philosophy (Spring edn, 2020) <https://plato.stanford.edu/entries/practical-reason/> accessed 12 June 2024.

29 V Descombes, Le Raisonnement de l’ours: Et autres essais de philosophie pratique (Seuil 2007) 21.

30 DA Schön, The Reflective Practitioner: How Professionals Think in Action (Basic Books 1983) 39.

32 Descombes (Footnote n 29) 23–24.

33 Footnote ibid 26; Schön (Footnote n 30) 39–40 expresses the same idea by pointing out that the principal preoccupation in theoretical contexts (which Schön calls the ‘model of technical rationality’) is problem-solving, whereas the principal preoccupation in practical contexts is both problem-solving and problem-setting.

34 Descombes (Footnote n 29) 23.

35 Footnote ibid 27. Consider a simple example. My desire to keep myself cool during a warm day makes my decision to turn on the air-conditioning logical and rational. However, upon reflection I may decide that my desire to keep cool will have to be reconciled with my equally strong desire to maintain a low carbon footprint in my daily activities. I may even decide that maintaining a low carbon footprint must take priority because of climate change. In this case, turning on the air-conditioning is incompatible with my amended goal, thus making it a less desirable course of action than simply opening the window and letting a cool breeze into the room. I may therefore conclude that what I ought to do in this case – what I am compelled to do, given my amended objectives – is to open the window.

36 For an argument in favour of reason-based logic as a framework for practical reasoning in international law, see Chapter 2.

37 eg G Abi-Saab, ‘Cours général de droit international public’ (1987) 207 RdC 9, 214.

38 cf N MacCormick, Legal Reasoning and Legal Theory (Oxford University Press 1978) 23.

39 eg Merkouris (Footnote n 1) 134–37; Y Dinstein, ‘The Interaction between Customary International Law and Treaties’ (2006) 322 RdC 243, 265; AE Roberts, ‘Traditional and Modern Approaches to Customary International Law: A Reconciliation’ (2001) 95 AJIL 757, 758; MH Mendelson, ‘The Formation of Customary International Law’ (1998) 272 RdC 155, 181; Abi-Saab (Footnote n 37) 176–77 (distinguishing the inductive process of ‘traditional custom’ from the deductive process of ‘new custom’); C de Visscher, Problèmes d’interprétation judiciaire en droit international public (Pedone 1963) 16; also ILC, ‘Draft Conclusions on Identification of Customary International Law, with Commentaries’ (30 April–1 June and 2 July–10 August 2018) UN Doc A/73/10, 126 para 5; Delimitation of the Maritime Boundary in the Gulf of Maine Area (Canada/USA) (Judgment) [1984] ICJ Rep 246 [111].

40 In fact, the outward depiction of legal reasoning as a syllogism of a logical deduction or induction is likely the performative aspect of the whole process, insofar as one has already worked out the way to a conclusion and is now putting things in form to demonstrate the validity and accuracy of the conclusion to others. By the same token, the syllogistic format may also be an exercise in authority – for instance, by a court that wishes to project the image that the law actually demands, and indeed, has always demanded, the outcome this court has pronounced.

41 ILC, ‘Fragmentation of International Law’ (Footnote n 4) 23 para 34 (‘[Legal reasoning] cannot be understood as reaffirming something that already “exists” before the systemic effort itself. There is no single legislative will behind international law. Treaties and custom come about as a result of conflicting motives and objectives – they are “bargains” and “package-deals” and often result from spontaneous reactions to events in the environment. But if legal reasoning is understood as a purposive activity, then it follows that it should be seen not merely as a mechanic application of apparently random rules, decisions or behavioural patterns but as the operation of a whole that is directed toward some human objective. Again, lawyers may disagree about what the objective of a rule or a behaviour is. But it does not follow that no such objective at all can be envisaged. Much legal interpretation is geared to linking an unclear rule to a purpose and thus, by showing its position within some system, to providing a justification for applying it in one way rather than in another.’ (emphasis in original)).

42 A Broches, ‘The Convention on the Settlement of Investment Disputes between States and Nationals of Other States’ (1972) 136 RdC 331, 348.

43 Corn Products International, Inc v United Mexican States (Award) ICSID Case No ARB (AF)/04/1 (18 August 2009) Separate Opinion of Andreas F Lowenfeld [1].

44 K Gorobets, ‘Practical Reasoning and Interpretation of Customary International Law’ in P Merkouris, J Kammerhofer, and N Arajärvi (eds), The Theory, Practice and Interpretation of Customary International Law (Cambridge University Press 2022) 374–75.

46 eg Z Douglas, ‘Other Specific Regimes of Responsibility: Investment Treaty Arbitration and ICSID’ in J Crawford, A Pellet and S Olleson (eds), The Law of International Responsibility (Oxford University Press 2010) 824. Therein Douglas criticises the decision in Hussein Nuaman Soufraki v The United Arab Emirates (Award) ICSID Case No ARB/02/7 (7 July 2004), not by challenging the accuracy of any of the tribunal’s chosen premises or the logical consequence of its syllogism, but by pointing out two things: first, that the tribunal’s decision in fact implies that the nationality requirements of the ICSID Convention track the doctrine of nationality of claims under diplomatic protection (implied, missing premise in the decision); and second, that this disregards the intentions of the ICSID Convention drafters, who did not see an equivalency between the Convention and diplomatic protection (introducing as countervailing reason the relevance of the parties’ intentions in the course of treaty interpretation).

47 Pérez Bermejo (Footnote n 11) 97–98; Soriano (Footnote n 11) 311ff.

48 A Aarnio, The Rational as Reasonable: A Treatise on Legal Justification (D Reidel Publishing 1987) 189–90.

49 MacCormick (Footnote n 27) 19–20. More generally, see A Bianchi, ‘Textual Interpretation and (International) Law Reading: The Myth of (In)determinacy and the Genealogy of Meaning’ in PHF Bekker, R Dolzer, and M Waibel (eds), Making Transnational Law Work in the Global Economy: Essays in Honour of Detlev Vagts (Cambridge University Press 2010) 34.

50 eg RK Gardiner, International Law (Pearson Longman 2003) 79; MacCormick (Footnote n 27) 19–20.

51 eg Merkouris (Footnote n 1) 135–36; Orakhelashvili (Footnote n 1) 496–97.

52 eg Gorobets (Footnote n 44) 375; Chasapis Tassinis (Footnote n 1) 241–47.

53 However, for an argument moving in that direction, see C Giannakopoulos, Manifestations of Coherence and Investor-State Arbitration (Cambridge University Press 2022) chs 2–3.

54 ILC, ‘Draft Conclusions’ (Footnote n 39) Draft Conclusion 2.

55 For a more detailed exploration of this thesis, see also Chapter 1.

56 Also R Kolb, ‘Selected Problems in the Theory of Customary International Law’ (2003) 50 NILR 119, 130 (‘[T]here is no induction without a prior element of axiomatic or deductive reasoning. It is impossible to induce anything if the framework within which the induction shall take place is not defined.’).

57 M Hakimi, ‘Making Sense of Customary International Law’ (2020) 118 Mich L Rev 1487, 1507. Stated more provocatively, Kolb (Footnote n 56) 133 (‘[C]ustom is not an objective reality emerging from a bundled set of facts, but a subjective projection of beliefs grounded in values to the extent these are not contradicted by practice.’).

58 eg ‘Proclamation 2667: Policy of the United States with Respect to the National Resources of the Subsoil and Sea Bed of the Continental Shelf’ (28 September 1945) 10 Fed Reg 12305, recitals 1 (policy rationale), 3 (policy rationale), and 4 (normative basis).

59 Staying with the continental shelf example, see North Sea Continental Shelf Cases (Germany/Netherlands; Germany/Denmark) (Judgment) [1969] ICJ Rep 3 [63] (recognising the CIL status of the continental shelf, among other reasons, because Articles 1–3 of the 1958 Convention on the Continental Shelf were at that time regarded by states as reflecting at least emerging rules of CIL).

60 On the various forms of admissible practice, see ILC, ‘Draft Conclusions’ (Footnote n 39) Draft Conclusion 6.

61 Footnote ibid Draft Conclusions 11–14.

62 On the importance of framing in this respect, see also KA Johnston, ‘The Nature and Context of Rules and the Identification of Customary International Law’ (2021) 32 EJIL 1167, 1172–74.

63 Territorial and Maritime Dispute between Nicaragua and Honduras in the Caribbean Sea (Nicaragua v Honduras) (Judgment) [2007] ICJ Rep 659 [232]–[234].

64 Frontier Dispute (Burkina Faso/Republic of Mali) (Judgment) [1986] ICJ Rep 554 [23].

66 More generally on the prevalence of analogical reasoning in the determination of CIL, see Kolb (Footnote n 56) 131; but for a view against, see Orakhelashvili (Footnote n 1) 496. On the various interpretative techniques used in determining CIL, see generally Orakhelashvili (Footnote n 1) 497ff; P Merkouris, ‘Interpreting Customary International Law: You’ll Never Walk Alone’ in P Merkouris, J Kammerhofer, and N Arajärvi (eds), The Theory, Practice, and Interpretation of Customary International Law (Cambridge University Press 2022); as well as Chapter 5.

67 Echoing Westerman (Chapter 1), one might then indeed say that opinio juris counts double in the determination of CIL. For a similar view, see Johnston (Footnote n 62) 1181–83.

68 GJ Postema, ‘Custom in International Law: A Normative Practice Account’ in A Perreau-Saussine and JB Murphy (eds), The Nature of Customary Law: Legal, Historical and Philosophical Perspectives (Cambridge University Press 2007) 285 (‘Thus, mere regularities of behaviour taken alone – the usus or “state practice” of international law discourse – not only fail to constitute customs of international law, they fail to constitute customs of any sort, including those of “comity”, because they fail to constitute norms.’ (emphasis in original)). Generally, on the context-dependent nature of meaning-making, see JR Searle, ‘The Background of Meaning’ in JR Searle, F Kiefer, and M Bierwisch (eds), Speech Act Theory and Pragmatics (D Reidel Publishing 1980).

69 JC Hage, Reasoning with Rules: An Essay on Legal Reasoning and Its Underlying Logic (Springer 1997) 75–77; also Kolb (Footnote n 56) 130.

70 The Prosecutor v Omar Hassan Ahmad Al Bashir (Judgment in the Jordan Referral re Al-Bashir Appeal) ICC-02/05-01/09-397-Corr (6 May 2019) [1]–[2], [113]–[117].

71 Narrowly contextualising practice may be regarded as a symptom of what Hakimi calls the ‘rulebook conception’ of CIL, which, according to her, dominates CIL orthodoxy. For Hakimi (Footnote n 57) 1497, the ‘rulebook conception’ presupposes that CIL manifests entirely as a body of rules, meaning that ‘a proposition can be CIL only if it applies more or less in the same way in all cases of a given type, rather than vacillates without discernible criteria from one situation to the next’.

72 D Akande, ‘International Law Immunities and the International Criminal Court’ (2004) 98 AJIL 407, 417; also Chasapis Tassinis (Footnote n 1), 265–66.

73 Kolb (Footnote n 56), 131–33; similarly, Hakimi (Footnote n 57) 1506–10 (arguing that CIL has no secondary rules – that is to say, rules that determine when the two-element test of state practice and opinio juris has been satisfied); and for a critical assessment of the ICJ’s CIL methodology in essence along similar lines, S Talmon, ‘Determining Customary International Law: The ICJ’s Methodology between Induction, Deduction and Assertion’ (2015) 26 EJIL 417.

74 Schön (Footnote n 30) 49ff.

75 J Dewey, How We Think: A Restatement of the Relation of Reflective Thinking to the Educative Process (2nd edn, DC Heath 1933) 3.

76 Commenting on the importance of method in particular, Footnote ibid 166.

77 J Rawls, A Theory of Justice (revised edn, Harvard University Press 1999) 18.

78 Footnote ibid 42–43; also J Rawls, Justice as Fairness: A Restatement (E Kelly ed, Harvard University Press 2001) 29–32.

79 N Daniels, ‘Reflective Equilibrium’, The Stanford Encyclopedia of Philosophy (Summer edn, 2020) <https://plato.stanford.edu/archives/sum2020/entries/reflective-equilibrium/> accessed 12 June 2024.

80 cf Schön (Footnote n 30) 163–64.

81 Hakimi (Footnote n 57) 1507–08 (citing to the ICRC’s study on rules of customary international humanitarian law); and, more generally, Kolb (Footnote n 56) 130–31 (noting the level of ‘axiology and subjectivity’ entering into the CIL process).

82 cf Schön (Footnote n 30) 93–102.

83 E Jouannet, ‘La Motivation ou le mystère de la boite noire’ in H Ruiz Fabri and JM Sorel (eds), La Motivation des décisions des juridictions internationales (Pedone 2008) 267; and from a broader epistemological perspective, Dewey (Footnote n 75) 128–29.

84 The research methodology required to identify active reflection in the course of one’s practice goes beyond the scope and aims of the present chapter.

Save book to Kindle

To save this book to your Kindle, first ensure [email protected] is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about saving to your Kindle.

Note you can select to save to either the @free.kindle.com or @kindle.com variations. ‘@free.kindle.com’ emails are free but can only be saved to your device when it is connected to wi-fi. ‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

Find out more about the Kindle Personal Document Service.

Available formats
×

Save book to Dropbox

To save content items to your account, please confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about saving content to Dropbox.

Available formats
×

Save book to Google Drive

To save content items to your account, please confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about saving content to Google Drive.

Available formats
×