Australian Year Book of International Law
By invoking customary international law, partisans of global justice have hoped to establish the legal status of certain global justice principles (or sufficiently close approximations to them) without encountering the problems that beset reliance on treaties (which require the consent of the states bound by them) or pure ethical reasoning (which, in lacking a grounding in institutional facts, is problematic as a source of law). This is because a customary norm does not require the consent of all states in order to come into existence; but, when it has come into existence, it is opposable against all states without exception (provided the norm is not a regional custom). However, the existence of a customary norm is not determined purely by ethical reasoning; instead, it must have a requisite grounding in general state practice and opinio juris. Nevertheless, ethical reasoning can still play a vital role in the process of determining whether a customary norm has come into being: this is because that process can be understood as an interpretative one that aims to present the raw data of state practice and opinio juris in its best moral light.
Of course, this entire procedure crucially presupposes a broadly traditional understanding of custom, one that involves the two elements of usus and opinio. But the very availability of that conception has been subject to long-standing scepticism, and this renders problematic any attempt to harness customary international law to a global justice agenda. Thus, in a recent book offering a radical reinterpretation of custom within the framework of rational choice theory, Jack Goldsmith and Eric Posner dismiss the traditional conception in the space of little more than a page. They identify four familiar ‘conceptual problems’, the first three of which concern indeterminacy regarding the nature, and requisite levels, of the state practice and opinio juris needed for norm-formation. But these problems are not obviously insuperable, especially in light of possible institutional remedies (eg an enhanced role for the International Court of Justice (ICJ) in making authoritative determinations of customary law). But the fourth objection – that ‘there is no convincing explanation of the process by which a voluntary behavioral regularity transforms itself into a binding legal obligation’ – is of a different order:
Opinio juris is described as the psychological component of customary international law because it refers to an attitude that states have toward a behavioral regularity. The idea of opinio juris is mysterious because the legal obligation is created by a state’s belief in the existence of the legal obligation. Opinio juris is really a conclusion about a practice’s status as international law; it does not explain how a widespread and uniform practice becomes law.
Unless this objection can be met, the traditional conception of custom will stand revealed as gravely flawed, thus leaving it vulnerable to a reconceptualisation along ‘rational choice’ lines that, as Goldsmith and Posner are eager to insist, would render it singularly inapt as a means for pursuing a global justice agenda. The aim of this paper is to offer a solution to this problem.
First, we need to clarify the nature of the problem. Contrary to what Goldsmith and Posner suggest, the problem is not that there is no explanation of how states’ belief that a norm has legal status can lead to that norm’s having legal status. After all, just as there can be a rule of recognition in a domestic legal system that identifies as law certain enactments of the Queen-in-Parliament, so too there can be a rule of recognition in the international legal system that identifies as law certain norms that are, inter alia, believed to be law by a sufficient number of states. Instead, the problem arises with the nature of the traditional explanation of how opinio juris contributes to the emergence or revision of customary international law.
The problem is this. On the traditional understanding of opinio juris, the emergence (or revision) of customary international law depends upon states mistakenly believing, or successfully pretending mistakenly to believe, that the pattern of behaviour to which they are referring is already lawful. Now, of course, this is not a literal contradiction in the traditional conception of custom itself. There is nothing contradictory in affirming that for a norm to become legally valid, states must (successfully pretend to) mistakenly believe that it is already legally valid. So the traditional conception is not, as is often said, ‘paradoxical’ in a literal sense; it involves no self-contradiction in the explanation of custom-formation. But the problem identified is certainly a serious flaw in that conception. This is because it is a plausible demand on any adequate source of law that it fulfils a requirement of transparency. In other words, its successful operation in generating new law or revising existing law must not necessarily depend on mistaken beliefs (or pretended such beliefs) on the part of the agents that create or revise the law as to what it is that they are doing.
The transparency requirement itself expresses a general constraint on the legitimate deployment of power. Political legitimacy demands that exercises of political power, including paradigmatically acts of law-making, must be publicly assessable in terms of standards that appropriately constrain political decision-making. A corollary of their being assessable in this way is a transparency requirement, according to which exercises of political power must be sincerely and accurately presented and defended by their agents as the acts that they are. In particular, legislators must have a clear-sighted appreciation, and make an honest avowal, of what it is that they are doing when they make laws. This is a prelude to their providing a justification for their legislative acts and to the public evaluation of those justifications. But the traditional conception of the genesis of custom involves an in-built transgression of this transparency requirement, since it is a condition of the creation of custom that those engaged in creating it mistakenly believe (or successfully present themselves as mistakenly believing) that the norm in question is already law.
Nor will many critics of the traditional understanding of custom find it surprising that it should fail a test of transparency. The latter test, they will say, presupposes a modernist understanding of law-creation as a deliberate act that is subject to public standards of legitimacy. Customary international law, they might continue, is a remnant of a primitive legal world-view, one in which legal norms can emerge through an essentially spontaneous or ‘organic’ process of accumulated communal practice. This is not a process that either invites or bears much critical, reflective scrutiny; hence, it is hardly likely to be consistent with modern notions of legitimacy. So, it is not only the concentration of power in the hands of unaccountable governmental and bureaucratic elites controlling organisations such as the World Trade Organization (WTO) or the World Bank that poses a significant threat to international legitimacy in the contemporary world. The process of customary norm creation does so as well. And, in the latter case, it seems to be a threat that is integral to that very process.
Needless to say, given their agenda of consistently sidelining ethical concerns in explaining international law, this is not at all how Goldsmith and Posner spell out the significance of the fourth problem. They are content to view it as an unsolved intellectual puzzle. Still, it is this failure of transparency that makes it a serious one. Moreover, if the problem can receive no cogent response, one may well have to concede that we should turn our backs on custom as a source of international law, especially if we are looking to it as a means of legalising norms of global justice. For it would be no small irony to pursue a global justice agenda though a mechanism that is inherently flawed in terms of legitimacy. In any case, I hope enough has been said to justify resisting the studied, but unconvincing, insouciance that some international lawyers adopt in response to the question of custom’s genesis. A good recent example is a survey article by Hugh Thirlway in which he dismisses the problem of communis error facit jus with the bland reassurance that ‘[t]he process by which customary rules change and develop thus presents theoretical difficulties; but it is a process which does occur’. Presumably, Thirlway means to consign the problem, along with that of identifying the ultimate norms on which the international law rests, to the category of problems that are of ‘purely academic interest’. Such an attitude inevitably provokes a sceptical backlash that seeks to jettison customary international law as traditionally conceived; this is especially so when customary norms are being increasingly invoked against states that have not consented to them. A better approach is to tackle the question head-on. And doing so will also require, strangely enough, that we broach the second question dismissed as ‘purely academic’ by Thirlway.
Fortunately, I believe, there is a relatively straightforward solution to the problem. This involves adopting a disjunctive conception of the content of opinio juris, one that distinguishes two broad types of cases: 1) cases where opinio juris concerns the creation or revision of customary international law, and 2) cases where opinio juris concerns the persistence across time of a norm that has already come into existence at some earlier stage. For the problem that has been identified does not implicate category two cases, that is, there seems to be nothing obviously perplexing about the claim that, once a customary norm of international law has come into existence, its persistence as a legal norm depends upon continued state practice and opinio juris, and that the content of the opinio juris should include as a component the belief that the relevant pattern of behaviour is already lawful. Thus, the content of opinio juris has to be understood in disjunctive terms, as relating either to situations of creation/revision or to those of continued existence, or (in cases where the legal status of the norm is uncertain) to both. The flaw, therefore, in the traditional rendering of opinio juris was to assume that, variation in the content of the norm being accepted aside, it is a unitary attitude, that is, one whose content does not vary as between situations of creating a legal rule and those of following or endorsing a legal rule that is taken already to exist.
So, to take type one cases, opinio juris can be characterised as broadly involving the following content:
[OJ1] the creation of an international legal rule according to which the specified pattern of behaviour would be lawful is ethically justified, and such a legal rule should be created by means of a process that involves general state practice consistent with it and an ethical endorsement by states of its establishment as a legal rule.
Alternatively, in type two cases, where a state purports to act in accordance with (or purports to invoke) a rule that is already part of the corpus of customary international law, the content of its opinio juris would be as follows:
[OJ2] the specified pattern of behaviour is in accordance with an international legal rule that has been appropriately created through general state practice and opinio juris [i.e. OJ1], and that rule is ethically justified.
Both OJ1 and OJ2 presuppose a broader, systemic ethical commitment on the part of a state whose opinio juris is in question, that is:
[SC] the process of creating international legal rules through general state practice and opinio juris is an ethically legitimate one, i.e. it provides a pro tanto reason for the exercise of power within the international system.
This disjunctive interpretation of opinio juris introduces clarity by differentiating the state’s attitude to the creation of a new norm, or the modification of an existing norm, from its endorsement of a norm that has already become law, where the latter bears on the question whether that norm remains legally valid. It thereby solves the problem of transparency without implausibly attributing overly complex attitudes to states. And it does so in a way that makes good sense of familiar statements, such as the ICJ’s statement in the North Sea Continental Shelf Cases, that for opinio juris to exist ‘[t]he States concerned must … feel that they are conforming to what amounts to a legal obligation’. On the account I have offered, this statement (and countless others like it that have been uttered by the Court, other judicial bodies, governments, international legal practitioners and scholars, and so on) does not have to be straightforwardly rejected as false or incoherent. Instead, it can be accepted as true (insofar as it refers to OJ2, at least in those cases where the relevant legal norm is obligation-imposing), but incomplete (insofar as it neglects OJ1). To the extent that interpretative charity counts in favour of an explanation, this recommends the disjunctive account of opinio juris.
A further advantage of that account is that it can explicitly accommodate situations where a state is uncertain of the legal standing of the relevant rule on the basis of which it acts or which it endorses. Here an opinio juris that is itself disjunctive in content (ie the proposition ‘OJ2 or, if the norm is not already law, OJ1’ is affirmed) may be ascribed to the state. Finally, this account leaves open the possibility that a state’s mistaken belief that X is a norm of customary law can contribute to the process whereby X becomes a legally valid customary norm (or, at any rate, this possibility is not excluded by anything that I have said so far). But this sort of error is not a necessary feature of the account of the genesis of custom that is made possible by the disjunctive interpretation, which is precisely the problem that confronted the ‘traditional’ understanding of opinio juris.
Consider now an objection to the foregoing analysis, one that takes issue with the liberal use of the adjective ‘ethical’ in characterising the beliefs and commitments schematically represented by OJ1, OJ2 and SC. After all, why should it be thought that an ‘ethical’ matter is at stake, requiring an ‘ethical’ justification, in the case of each individual customary law or even in endorsing the general process whereby customary laws are generated? The answer lies in the capaciousness of my understanding of the ethical. It is not, as in a popular tabloid construal, confined to life-and-death matters or sexual relations. Nor it is confined, in line with an increasingly popular lawyer’s fallacy, to matters implicating human rights. In using the word ‘ethical’, I refer quite broadly to the domain of reasons that bear on an agent that derive from a proper regard for human interests, both his own and, especially, those of others. Particularly significant among these reasons are those that are obligation-imposing. The latter reasons are categorical and pre-emptive, that is, reasons that apply to us independently of how we are subjectively orientated and which, in their normative force, are not simply weighed in with other reasons, but rather exclude some of them. The other assumption I make is about the implicit standing that OJ1, OJ2 and SC accord to ethical justification, namely, that it is broadly objectivist in character. The objectivist aspiration of ethical thought I take to be an essential component of that mode of thought’s self-understanding. It basically involves the claim that it is possible to assess ethical beliefs and commitments as true or false, justified or unjustified, by reference to standards that are not arbitrary or simply given, but which can instead be rationally vindicated against competing standards. Whether that aspiration can be fulfilled or not is another matter, but I take the aspiration to be inherent in our ordinary ethical self-understanding, differentiating it in one important way from expressions of personal taste.
One can now see the appropriateness of the ethical aspect of OJ1, OJ2 and SC. It is not simply a convenient way of avoiding the problem of transparency by moving illegitimately from judgments of legality to judgments of morality. On the contrary, as Joseph Raz has argued, the only way of explaining the normativity of ordinary, ‘committed’ legal statements quite generally is as expressing ethical beliefs (whether sincere or otherwise). And the reason for this is that legal statements make claims that potentially involve serious implications for the interests of others, so that they have to be interpreted as purporting to express some justified ethical claim that licences that impact – for example, a moral obligation that obtains in virtue of the law. And this presupposes a wider ethical endorsement (akin to SC) of the rule of recognition by reference to which legal norms are identified. If this is true in the case of ordinary legal statements, it is a fortiori true in the case of statements about the law that should be established.
I have argued that the supposed ‘paradox’ of customary international law poses a genuine threat to the legitimacy of that source of law. It is therefore not a problem to be brushed aside as merely of ‘academic’ interest. Fortunately, a solution to the paradox is ready to hand, in the shape of a disjunctive interpretation of opinio juris. With this challenge to its legitimacy met, a recognisably traditional conception of custom – conceptualised along the interpretative lines I have advocated elsewhere –offers the greatest hope for conferring universal legal force on principles of global justice.
[∗] An abbreviated version of the argument presented in this comment appeared in: J Tasioulas, ‘Customary International Law and the Quest for Global Justice’ in A Perreau-Saussine and J B Murphy (eds), The Nature of Customary Law: Philosophical, Historical and Legal Perspectives (2007) 320.
[∗∗] Fellow and Tutor in Philosophy, Corpus Christi College, Oxford.
 I use the expression ‘global justice’ in a broad sense, to encompass moral values and principles, including but not limited to the minimalist values of peaceful co-existence and co-operation, that should be respected by all states whether in their internal affairs (transnational justice) or in their relations with other states (international justice). These requirements include norms establishing human rights, authorising humanitarian intervention and assistance, enjoining environmental protection, and so on.
 An exception may need to be made for states that are ‘persistent objectors’ to the putative customary norm during the process of its formation; but this exception does not extend to customary norms that are principles of jus cogens, which reflect some of the weightiest requirements of global justice.
 I develop such an interpretative account of customary international law in: J Tasioulas, ‘Customary International Law and the Quest for Global Justice’ in A Perreau-Saussine and J B Murphy (eds), The Nature of Customary Law: Philosophical, Historical and Legal Perspectives (2007). For an earlier presentation of the view, see J Tasioulas, ‘In Defence of Relative Normativity: Communitarian Values and the Nicaragua Case’, (1996) 16 Oxford Journal of Legal Studies 85.
 J L Goldsmith and E A Posner, The Limits of International Law (2005) 24.
 H Thirlway, ‘The Sources of International Law as a Consensual Bond’ in M Evans (ed), International Law (2003) 128.
 Ibid 129.
 North Sea Continental Shelf (Federal Republic Germany v Denmark: Federal; Republic Germany v Netherlands)  ICJ Rep 3  (20 February). It is, of course, a mistake to suppose that opinio juris always involves the claim that the relevant pattern of behaviour is obligatory, as opposed to some other possible deontic modality, eg rights, permissions, etc.
 For another interpretation of opinio juris that emphasises its ‘ethical’, or at least ‘practical’ rather than strictly legal, character but ignores the need for a disjunctive account, see the sophisticated discussion in J Finnis, Natural Law and Natural Rights (1980) 238.
 See J Raz, Practical Reason and Norms (1975).
 See J Tasioulas, ‘The Legal Relevance of Ethical Objectivity’ (2002) 47 American Journal of Jurisprudence 211.
 See J Raz, ‘Hart on Moral Rights and Legal Duties’ (1984) 4 Oxford Journal of Legal Studies 123.