The chivalric pursuit of coherence in international law

1. Introduction

9 June 1663. Gottfried Wilhelm Leibniz defends his Bachelor thesis entitled the Metaphysical Disputation on the Principle of Individuation at the University of Leipzig. This was Leibniz’s first major work. It was completed under the supervision of Jacob Thomasius who is said to have instilled in Leibniz his famous drive to reconcile modern philosophy with the philosophy of Aristotle, Plato, the Scholastics and the Renaissance humanist tradition. In his thesis, Leibniz argued that any real thing is simply singular by virtue of its metaphysical subcomponents, which are singular by themselves (this is the claim one can always identify the simple unities that compose the whole). ¹ At the end of his Metaphysical Disputation on the Principle of Individuation, Leibniz came to state that the ‘essences of things are like numbers’. The modern concept of coherence was born. ²

9 June 2023. Somewhere on a server used by Cambridge University Press, one of the leading refereed international law journals rolls out an article – the main genre of the field – by James Devaney (hereafter the author) on the possibility of a coherent decision-making in the determination of facts by the International Court of Justice (hereafter the Court). The article, written by one of the new authoritative voices on international dispute settlement of the early twenty-first century, articulates a descriptive and normative claim about the possibility for the Court to make coherent factual determinations through a rational process, which, the author argues, is achieved by specifying the parties’ competing factual claims, before evaluating the coherence of the competing claims against relevant coherence principles and selecting as justified the most coherent of these claims. ³

Between 9 June 1663 and 9 June 2023 exactly 360 years have passed. There have been so many generations, revolutions, inventions, destructions, reconstructions, transformations, discriminations, repressions, abominations in these 360 years. And yet, these are 360 years which the reading of this 2023 article makes one experience as two consecutive days. It is submitted here that the author’s feat is not only to flatten time. It is also to make his readers’ feel that the concept of coherence, as designed by Leibniz and his modern followers, has thrived uncontested for almost four centuries. It is as if humankind had lived all these years without ever encountering the compelling challenges of modernity’s main patterns of thought. In that sense, the author makes one travel though time while simultaneously razing the heritage of Butler, Spivak, Adorno, Foucault & co to the ground.

Whilst I cannot help remaining dumbfounded each time I come across crude modernism in international legal thought in the twenty-first century, it is not my aim in the following pages to recall how coherence – and the idea of unity on which it is built – is always bound to collapse under the weight of the normative presuppositions it is derived from. ⁴ Instead, I want to give this article a very different twist and argue that it takes a great deal of chivalry to vindicate coherence in twenty-first century international legal thought. I mean it seriously. This new piece published by the Leiden Journal of International Law denotes bravery and unwavering ambition. Although I will show in this article that any pursuit of coherence is bound to remain a fantasy, I recognize that the author does not demur in front of the (uncountable) objections that could possibly be raised against his coherence framework. Throughout the whole article, he sticks to his sword and proudly walks past the postmodern night.

It must be acknowledged that such chivalric behaviour constitutes an attitude that is rife in international legal thought and practice. ⁵ In fact, the author finds himself in good company when he vindicates the embrace of one of the most modern patterns of thought, namely coherence. ⁶ For that reason, his article, albeit of a much greater sophistication than most of the modernist literature published nowadays, will be applauded by the vast majority of its readership for both its ambition and its erudition. But this is precisely what chivalry is about. It is courage and unwavering ambition but also loyalty to the establishment and the values of the day. ⁷

In the present article, rather than recalling the compelling arguments against the very possibility of coherence, I want to unpack three singular components of the author’s chivalric vindication of a coherence framework. I will show that the article’s erudite defence of modern coherence, and its correlative obliteration of modernity critique, are articulated around three moves: A move towards meaning (Section 2), a move towards context (Section 3), and a move towards discipline (Section 4). These three moves are discussed in turn in the following sections. In spelling out the main moves made by the author in his vindication of a coherence framework for factual determinations by the Court, I have the ambition to turn Leibniz modernism over its head and show the dead end to which all the singular moves of the author’s coherence framework lead. In other words, I aim at showing that modern thought, even when crafted by its most talented masters, fails in each and every of the singular moves that composes it. This article ends with a few concluding remarks (Section 5).

Before sketching out the moves through which the author moulds his coherence framework, a definitional observation is warranted. Modern coherence, that is as Leibniz construes it, is understood here as referring to the quality of a practice or an object that forms a unified whole which can always be decomposed in identical singular subcomponents, such decomposable unity enabling the finding of logic, systematicity, consistency or even paradoxes. ⁸ In that sense, thinking in terms of coherence, as the modern mind does, is a type of thinking that postulates, projects and constructs the unity of what it is applied to. ⁹ Whether such postulation, projection, and construction of unity ever succeeds is precisely what the following paragraphs will discuss.

2. Defining coherence (and the fall into post-structuralist meaninglessness)

Unsurprisingly, the article makes it one of its main tasks to define the very idea of coherence that guides the coherent factual determination process that it vindicates. Yet, and this is one of the article’s greatest master strokes, the author, in seeking to substantiate the meaning of his coherence framework, manages to always elude any definition of coherence, let alone of the type of unity that such coherence rests on. Indeed, coherence, he argues, is achieved through ‘a move from the factual claims made by the parties to the selection by the Court of the hypothesis which explains the facts better than any other explanation’, that is ‘the selection of the best (most coherent) explanation’. ¹⁰ Coherence is thus, according to the article, the outcome of a choice for the most coherent explanation. The author adds that such realization of coherence through the selection of the most coherent explanation is itself the result of the use of ‘coherence principles that are relevant in relation to explanatory coherence (the pertinent form for the purposes of reasoning about legal fact’. ¹¹ After mentioning and illustrating these seven principles of coherence conducive to the determination of the most coherent explanation – symmetry, explanation, analogy, data priority, contradiction, competition, and acceptance, ¹² the author indicates that following the principle of competition corresponds to choosing the claim ‘which best explains the disputed fact, in accordance with recognized coherence principles’. ¹³ He similarly points out that following the principle of acceptance entails that ‘the acceptability of a claim … depends on its coherence with other claims in that system’. ¹⁴ In sum, coherence is achieved through coherent explanations which are themselves achieved through the adoption of coherence principles, themselves defined by coherence. In other words, for the author, coherence means … coherence.

Interestingly, evading the definition of coherence is not only achieved through the abovementioned self-referential move whereby coherence refers to coherence. It is also enabled by deferring the meaning of coherence to other – similarly undefined and self-referential – idioms. In fact, the author indicates that coherence, for the sake of his framework, is achieved through ‘inference to the best explanation rather than any objective truth or probabilistic assessment’, ¹⁵ which is a process that builds on ‘responsibilist’ and ‘contextualist’ frameworks. ¹⁶ This corresponds, according to the article, to the Court making ‘factual determinations that can be rationally justified’ ¹⁷ and stressing on that occasion that truth-value can be attached to the concept of coherence through ‘interdependence of knowledge, whether that be knowledge of the content of our own minds or the minds of others, or knowledge of the world’. ¹⁸ In sum, coherence, for the sake of the author’s framework, is an aggregate of responsibilism, contextualism, rationality, truth-value, interdependence of knowledge, etc., each of these concepts being themselves left in a semantic limbo in the article.

The result of the abovementioned self-referential or meaning-deferring moves is a very sophisticated definition-avoidance, which is very reminiscent of the post-structuralist fate of language: ¹⁹ coherence is nowhere defined and its meaning is nowhere to be found because it is always deferred to itself or to other forms whose meaning is similarly left hanging in the air. The meaninglessness in which coherence is left throughout the article, and the sophisticated self-referral and deferring moves through which such meaninglessness is achieved calls for two observations.

First, in perpetually eluding the meaning of coherence, the author mimics the judicial body that he places at the centre of his universe, namely the Court. ²⁰ Indeed, what is a judgment of the Court if not an elegant textual assemblage that constantly pushes back the meaning of a word or a text to another word or text. In that sense, the author, by always pushing back the meaning of coherence, simulates the very role of those to whom his coherence framework is addressed. In doing so, he ensures that judges of the Court will find themselves very much at home in his coherence framework. Second, the meaninglessness in which the author leaves his coherence framework bears upon the author’s – surprisingly scant and minimal – discussion of the possible objection of indeterminacy. ²¹ In fact, rejecting any possible objection based on the indeterminacy of the concept of coherence, the article claims that ‘concepts are not so malleable so as to be meaningless, but rather that through intersubjectivity we can to a greater or lesser extent construct a core meaning’. ²² It is argued here that such discussion of the objection of indeterminacy completely misses the point. It is not that coherence could possibly have an indeterminate content. It is rather that coherence, for the sake of the coherent factual determination process advocated by the author, has no content at all, whether determinate or not determinate. There is thus no question of indeterminacy arising at all, for coherence has been given no content by the article in the first place. ²³

3. Foregrounding context-dependency (and the return to pre-Kantian empiricism)

Following a very modern pattern of thought, the article makes constant references to context, arguing for instance that in making coherent factual determinations the Court must take context into account ²⁴ or that coherence is in constant need of contextualization. ²⁵ Likewise, it is said in the article that the epistemic responsibility at the heart of the coherence framework promoted in the article ²⁶ has a contextualized nature ²⁷ and that the duties which the author ascribes to the Court in its pursuit of coherence are said to be context-dependent duties. ²⁸ The author similarly stresses that what counts as epistemically responsible behaviour is shaped by the context in which the legal fact-finder is operating. ²⁹ He also argues that context is what provides the outer limits of the coherence framework. ³⁰ In the same vein, he claims that the idea that the Court must achieve contextual justice. ³¹

The centrality of context-dependency in the author’s coherence framework is obviously nothing extraordinary, for it also is a common feature of international legal thought and practice. For most international lawyers, if Article 31 of the Vienna Convention on the Law of Treaties provides that a treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context, there must be such a thing as a context of any legal text or legal practice. For international lawyers, context-independency – and thus deferring meaning to what is outside the text or the practice being interpreted – is a very common hermeneutic attitude.

Albeit the expression of a very ordinary pattern of thought, the centrality of context-dependency in the author’s coherence framework remains very unsettling in my view, for at least three reasons. First, it is yet another expression of the elementary belief that context exists out there and can be ascertained independently from the text or practice whose interpretation it supports or guides. In that sense, it is a plain dismissal of the – now common – argument that context is always produced through the text or practice whose interpretation it supports and guides, always being secondary to it. ³²

Second, and more alarmingly, such context-dependency, whether in the author’s coherence framework or in international legal thought and practice in general, reveals the embrace of some very crude empiricism. ³³ By crude empiricism, I mean a belief in the idea that there is a reality-in-itself and that this reality-in-itself is accessible and cognizable provided that the right methods are used. For the author, the context is an outside reality that is available to the legal interpreter and which can guide her hermeneutic work. The author’s belief in a reality-in-itself that is accessible and cognizable is further revealed by his claim that ‘(a)ny factual determinations which do not bear any relation to reality are of limited value (if any value at all) to the Court’. ³⁴ So for him, there is a reality out there which the Court must attune its factual determinations to. Here it is not Leibniz that the article would please but our good old Descartes! It is noteworthy that, in foregrounding some premodern cartesian realism, the author repudiates all those works that have shown that reality is a creation of methods, ³⁵ the verification of a postulated theory, ³⁶ a product of ideology, ³⁷ an effect of the discourse, ³⁸ the result of what has been made visible, ³⁹ etc. In doing so, he even refutes Kant’s rejection of the possibility of an absolute knowing by the self of the world-in-itself. ⁴⁰ Kant seems to have lost all its lustre in the author’s eyes. This is not modernity but pre-modernity at its peak.

Third, the author making of context-dependency a cornerstone of his coherence framework is equally bewildering as it further defers the very meaning of coherence. Indeed, on top of being left in meaninglessness, ⁴¹ his concept coherence is simultaneously supposed to earn its (non-)content from an ever changing and never defined context. The fall of the author’s framework into meaninglessness continues in a tailspin…

4. Disciplining judges (and the invention of a pre-Galilean epistemology)

From the very start, the article makes clear that his coherence framework is both normative and descriptive, insisting that he provides both ‘a theoretical account of the Court’s fact-finding process, and a normative argument for what it should do in the future’. ⁴² Some of the article’s modern – and especially Kelsenian ⁴³ – readers will probably bemoan such epistemological instability. It must be acknowledged that, in conflating the normative and the descriptive, the author indulges in what I have called elsewhere some ‘self-confirming thinking’ ⁴⁴ , that is a move of convenience whereby the word and the world come to confirm one another. ⁴⁵ As far as I am concerned, I see no reason to be concerned with the author’s conscious and fully assumed conflation of the normative and the descriptive, for I think that this is also the inevitable fate of any descriptive argument ⁴⁶ while reflecting a common operating feature of the language. ⁴⁷

More distinctive is probably the moral programme that the author attaches to his coherence framework. Indeed, under the banner of what he calls the ‘responsibilist’ dimension of framework, ⁴⁸ the author spells out a series of ‘epistemic duties’. ⁴⁹ Although the author claims that ‘these are not legal duties … (n)or are they even (or necessarily) moral duties’, ⁵⁰ I can hardly figure out what distinguishes such ‘epistemic duties’ from a set of moral prescriptions for the very epistemic community concerned. What is more, and most remarkably, these ‘epistemic duties’ are worded as a creed. In fact, they include ‘a duty to be rational believers’, ⁵¹ ‘a duty to believe in evidence that has been arrived at through a rational process’, ⁵² ‘a duty to believe all (and only) claims that are supported by evidence’, ⁵³ ‘a duty to behave in ways that will maximize the number of justified beliefs and minimize their unjustified beliefs’, ⁵⁴ etc. This means that these duties are not only prescriptions to behave in a certain way but are also prescriptions of … what judges ought to believe! In my view, this is a belief system in all but name. This is why, simply claiming these duties are epistemic duties does not suffice to conceal the moralism which comes to permeate the author’s coherence framework.

Irrespective of whether such epistemic duties are of a moral character or not, what matters here is that the author, by virtue of the creed associated to his coherence framework, ends up performing the role of a priest who dictates to judges what they should believe. Surely, for a modern mind like the author, such pastoral role constitutes an oddity. Yet, in international legal thought and practice, this is not unheard of. After all, I have always felt that international law has all the trappings of a belief system. ⁵⁵ Likewise, I am not at all baffled by the article’s endeavour to discipline judges. In fact, that the author seeks to order international legal practice through a set of duties is nothing idiosyncratic, for a discipline always constitutes itself by the very discipline and order it enables in the production of truth-claims. ⁵⁶ From this perspective, the author simply wants to be among those who compose the very prescriptions constitutive of the international legal discipline. And, in that regard, he certainly is not the first international lawyer to flirt with such disciplinary ambition. For these reasons, the metamorphose of the author’s coherence framework into a belief system that morally prescribes what judges ought to believe is not what fascinates me the most here.

What I find most intriguing is that the article, in inventing a new belief system meant to discipline judges and sustain his coherence framework, moves from a very modern posture – a quest for coherence for the sake of truth-making in the name of reality – to a medieval posture whereby beliefs come to govern factual determinations. ⁵⁷ Indeed, in setting up a new belief system for judges engaged in factual determinations, even if only to sustain a coherence framework, the article provides the groundwork for a new trial of Galileo Galilei as well as a condemnation of all factual determinations that would not follow the belief system in place. ⁵⁸ After projecting a model of international dispute settlement processes governed by Leibzinian coherence, the author accordingly proposes that international dispute resolution be reduced to a pre-Galilean practice whereby beliefs govern factual determinations.

5. Concluding remarks

As the previous sections have shown, the author has not always proved a uniformly loyal knight in the performance of his chivalric defence of coherence. Indeed, he started his journey through the postmodern night as a modern defender of the Leibnizian coherence. However, on his way, he indulged in some kind of post-structuralist meaninglessness, as was discussed in Section 2. Later on, he did not balk at showing some affinity for Descartes and crude empiricism, as was demonstrated in Section 3. Eventually, while seeking to back up his coherence framework with a belief system, he displayed sympathy for the pre-Galilean primacy of belief systems over factual determinations, as was emphasized in Section 4. The author, in the course of his knightly campaign for a new coherence framework, has thus been found making medieval, pre-modern, ultra-modern, and post-modern moves.

In the last part of his article, the author – mimicking a practice of the Court in case of non-appearance ⁵⁹ – engages with what he deems the most likely objections against his coherence framework. It is submitted at the very last stage of this article that the objections which the author tries to anticipate are maybe not the most fundamental counterarguments that he should have been concerned with. In my view, his attention should have rather been turned to the three impossible moves that he makes to sustain his Leibnizian idea of coherence and which have been discussed in the previous sections. And yet, the limitations of the last part of his article should, however, not strip it from its greatest merit, i.e., its chivalric ambition to uphold coherence in the factual determinations by the Court. As was said at the start of this article, upholding Leibnizian coherence after 360 years of debate and critique is no small task. May this story of Devaney’s tilting at coherence resonate among all those other knights who could possibly be tempted by similar chivalric pursuits.