Sydney Law Review
The Authority of International Law — Lifting the State Veil
The legitimate authority of international law, ie its ability to generate moral duties of obedience for its subjects whether states, international organisations or individuals, has become a subject of growing interest among international legal scholars and legal theorists over the past fifteen years or so. The initial difficulty most accounts face has to do with the reality of international law itself qua decentralised and non-hierarchical legal order. Stemming from those complexities, a second difficulty pertains to the concept of legal authority itself. This article presents a single concept of legal authority and a set of justifications for the legitimacy of international law that can not only accommodate the complexity of contemporary international law, but also make sense of it in the context of competing claims to legitimacy made over the same people by national, regional and international legal orders. The key to the authority of international law in a pluralist legal order lies, the author argues, in lifting the state veil. This implies focusing on the individual as the ultimate subject of authority in international law. The article’s argument unfolds in three steps. It starts by presenting the conception of legal authority the article is based on, and in particular a revised co-ordination-based version of Joseph Raz’s service conception of authority. It then argues that this autonomy-based account of authority best explains the legitimacy of international law by focusing on four key features of legal authority in the international context: the multiplicity of international subjects and law-makers and their relationship; the role of co-ordination in the justification of international law’s authority; the piecemeal nature of authority and the role of state consent in that context; and, finally, the protection of state sovereignty and its compatibility with the authority of international law. In the third and final section, the article addresses borderline cases, and in particular relativism-based exclusions of international legal authority and exceptions to that authority based on justified international disobedience.
‘The further we depart from the picture of international laws as being created solely by states and as dealing solely with the relations of states to one another — and the more seriously we take the idea that human beings, not states, are the ultimate objects of moral concern — the clearer it becomes that a satisfactory account of the legitimacy of international law must include more than an explanation of why States ought to regard the international institutions through which law is made as having the right to rule. More precisely, appreciating the new face of international law shows just how inadequate the traditional framing of the question of the legitimacy of international law is. The question is much broader than “Why should states consider international law binding?”’ (emphasis in original)
The authority of international law, and more precisely the legitimate authority or legitimacy of international law as I will understand it here, is international law’s ‘right to rule’; the exercise of which binds its subjects by imposing duties of obedience.
Qua practical authority, international law provides its subjects with reasons for action. Those reasons for action are moral reasons. Moral duties to obey the law qua law ought to be carefully distinguished from moral duties to obey the law because its content is moral, on the one hand, and from legal duties to obey the law, on the other. While the latter are important duties that can co-exist with the former, they do not capture the core of legitimate authority, ie a moral duty to obey the law not because it is morally correct but qua law. Importantly, the existence of international law’s moral right to rule is the result of an objective evaluation: international law may have legitimate authority whether or not its subjects think it does and whether or not they have consented to its authority. It is not therefore, the perceived or sociological legitimacy of international law this article is concerned about, but its normative legitimacy.
Of course, like domestic law, international law usually benefits from de facto authority by the mere fact of exercising power over its subjects and/or being effectively complied with in practice. This pertains independently from its actual legitimacy and right to authority. By virtue of its legal validity, international law also lays a claim to legitimate authority and that claim largely explains, and to a certain extent reinforces its de facto authority. The fact that international law inherently makes a claim to legitimacy does not, however, entail that it actually or entirely possesses it, or even that it is capable of possessing it under realistic conditions. Nor, conversely, would international law’s lack of legitimacy deprive it of its legal status. In most cases where international law lays a general claim to obedience, it claims more legitimacy than it can effectively have. In fact, there may even be a deeper hiatus between the extent of international law’s claim to legitimate authority and the effective scope of its legitimacy than there is in national law.
International law can therefore remain valid and even retain its de facto authority without its claim to authority being justified and its authority legitimated. In fact, the threat of the use of power is an even more widespread means of securing collective action around certain directives among international actors than it is in the domestic context. More importantly, there may be other (instrumental and non-instrumental) reasons for the attitude of respect or recognition developed by some of its subjects which are distinct from authoritative reasons. One may mention state consent or strategical reasons for state compliance with international law. Furthermore, there may be moral reasons to create and support just international institutions because they are just or because they are fair. There may even be moral reasons to create and sustain the rule of international law independently from that law’s legitimacy. However, all those reasons ought not be confused with moral reasons to obey the legal rules generated by those institutions.
All the same, the existence of an entirely illegitimate, albeit valuable for different reasons, legal order would not be sustainable in the long run. The law’s distinctive contribution to the advancement of other valuable goals lies precisely in successfully laying down authoritative directives to reach those goals. Furthermore, the fact that valid law necessarily claims to be legitimate implies that it should be capable of being authoritative and hence be produced so that it can be. In those conditions, ensuring the legitimacy of international law has a key influence on the organisation of international law-making processes. Finally, due to the increasingly direct impact of international law norms on individuals in areas previously covered by legitimate national law, a legitimacy gap is gradually widening and the legitimation of international law in those areas has become more pressing.
Unsurprisingly therefore, the claim to and justification of the authority of international law has become a central concern for international legal scholars ever since the late 1980s. It was said to replace antecedent and more ontological discussions about the nature of international law; after those discussions had been dominating the theorising of international law since the 1950s. For years, indeed, international legal scholarship had debated over the absence, among other things of: reliable enforcement mechanisms in international law, an unfettered and habitually-obeyed sovereign, and a Kelsenian Grundnorm or a Hartian rule of recognition. This methodological change certainly had a liberating effect — international legal scholars could turn to questions of justice, and identify or in some cases, negate the values sustaining international law without having to fight back sceptical views about its legal nature.
It has become clear today, however, that both discussions are intrinsically connected and cannot be seen as alternatives. A good understanding of the legality and of the sources of international law implies explaining how international law is able to claim legitimacy and could be capable of possessing authority. But the reverse is also true — explaining the latter claim, and how it can be justified in some cases at least, also implies understanding how international law works in the first place, at the risk of otherwise undermining the possibility of recognising any legitimate international law at all in current circumstances. Thus, if sceptical views about the legal nature of international law have become less pressing, it is not so much by virtue of a ‘post-ontological’ turn in international law scholarship, but mainly because contemporary jurisprudence and theoretical accounts of the nature of law no longer attempt to separate legal validity too strictly from legitimacy. As a result, whereas the authority of international law may at first sight sound like an eminently normative topic, it may also turn out to give rise to a conceptual discussion given the inseparability of both questions in matters relating to law.
A primary difficulty facing any explanation of the legitimacy of international law lies in the identification of a concept of legitimacy that can account for the legitimacy of (at least some part of) international law. The differences between international and national law in that respect are well-known — international law is mostly the product of horizontal interstate law-making practice and, to be more precise, of different interstate practices. Some are more akin to making intersubjective contractual promises and others to general legal rule-making for all subjects of international law. As a result, international law is said to lack a centralised and hierarchical ensemble of law-making institutions and processes that may be equated with domestic law-making authorities and legislating procedures, on the one hand, and is mostly exempt from sanctions backing up its norms, on the other.
To make things more complex, new forms of international law have arisen in recent times that question the exclusivity of the horizontal interstate law paradigm: subjects no longer include states making law for other states, but also international organisations (IOs) and individuals as subjects of rights and obligations. With respect to its objects, international law no longer pertains only to interstate relations, but also to intrastate relations and therefore, directly regulates the life of individuals alongside domestic law. Finally, and maybe as a result, the sources of international law and its law-making processes have become more diverse and have developed to include, besides the ‘famous three’ (treaties, customary law and general principles), general multilateral interstate law-making processes that often associate individual actors, and unilateral legislation by IOs. Arguably, those very sources which include subjects of international law other than states, regulate matters previously covered by domestic law only and often through majority rule. In terms of normativity as well, international law no longer offers a unified face. International legal norms can bind subjects universally or not ( eg erga omnes and omnium duties) and to varying degrees (eg jus cogens norms and, more controversially, soft law).
Thus, either the concept of legitimacy that is chosen accounts for the legitimate authority of both national and international law, but in a way that can capture not only their differences but also the sheer diversity of international law-making itself, or two (or more) separate concepts of legitimacy are used in each case. Most authors currently writing about the legitimacy of international law rightly choose and use one of the concepts of legitimate authority developed for the domestic context. First, this matches a conceptual requirement: the transitive application of the same concept of law in the national and international legal orders implies using the same concept of legal legitimacy across the board. However, the reference to a single concept of legitimate authority also corresponds, second, to the nature of the autonomy of the international legal order. In the past few years, there have been fundamental changes in international legal practice and using a single concept of legitimacy corresponds to the new circumstances of international law. While international law still covers numerous areas of interstate relationships, it has also permeated the material and personal spheres of national law. International and national law now largely share the same objects and subjects, as international legal norms apply increasingly to legal areas previously regarded as exclusively domestic, and to states and IOs, but also to individuals. In those circumstances, international legal norms apply (directly or indirectly) to individuals within domestic legal orders without always being transformed into domestic legal norms. In those circumstances, the legitimacy of international law cannot be assessed on its own, but in a context of many overlapping legal orders. This cannot be done according to a different concept of legitimacy from the one that applies to the norms stemming from domestic legal orders.
A second problem looms large, however. Even if this conceptual unity in different or in the same legal order(s) is granted, most accounts of the legitimacy of international law do not look at it from the perspective of individual legal subjects. They use the same concept of legitimacy as for domestic law, but transpose it to the interstate level and apply it to states only. As a result, they focus on authority for states the way domestic accounts focus on authority for individuals, without referring to states’ relationship to individuals. Even when authors have realised the importance of looking at individual subjects in this context, it is per se, the way they would address domestic legal authority for individuals and not by reference to individuals’ normative relationship to states.
Again, the difficulty stems from the reference to international law qua law of interstate relations exclusively, thus ignoring new areas of international legal regulation that affect national law directly and hence individuals, whether directly or, more often, indirectly. In any case, when states are bound by obligations of international law, their institutions and citizens are bound indirectly and have to comply with them through the actions of their state. As a result, assessing the legitimacy of international law on an interstate basis only and that of national law on a state-individual basis is unhelpful. The legitimacy of international law can only be understood if the reasons for action it provides to all subjects of that authority are assessed at the same time. Of course, this is not to deny that international law may provide different subjects with different reasons. However, because the relationship between those subjects is one of constituency, separating the justifications for those reasons or, worse, eluding to some of them blinds an essential connection between them. When a state is bound by an international legal norm, its institutions and citizens are bound at the same time, whether directly or indirectly, and this must necessarily affect in return the way in which a state can be bound.
In this article, I would like to present a single concept of legal authority and a set of justifications for the legitimacy of international law that can not only accommodate the complexity of contemporary international law, but also make sense of it in the context of competing claims to legitimacy made over the same people by national, regional and international legal orders. To do so, the article purports to clarify who the subjects of authority are in international law by distinguishing carefully between law-makers and legal subjects and by replacing individuals at the core of the inquiry into the legitimacy of international law.
The article’s argument unfolds in three steps. It starts by presenting the conception of legal authority the article is based on, and in particular a revised co-ordination-based version of Joseph Raz’s service conception of authority. It then argues that this account best explains the legitimacy of international law by focusing on four key features of legal authority in the international context: the multiplicity of international subjects and law-makers and their relationship; the role of co-ordination in the justification of international law’s authority; the piecemeal nature of authority and the role of state consent in that context; and, finally, the protection of state sovereignty and its compatibility with the authority of international law. In the third and final section, the article addresses borderline cases, and in particular relativism-based exclusions of international legal authority and exceptions to that authority based on justified international disobedience.
2. The Concept of Legal Authority
Joseph Raz’s seminal and refined account of authority, and of legitimate legal authority in particular, constitutes a useful starting point for any discussion of international law’s authority.
In a nutshell, the Razian concept of legal authority is comprised of two elements: A has legitimate authority over C when A’s directives are (i) content-independent and (ii) exclusionary reasons for action for C. In other words, the directives are authoritative reasons for action, first, by virtue of the fact that A issued them and not because of the content of any particular directive, and, second, because these reasons are not simply to be weighed along with other reasons that apply to C but, instead, have the normative effect of excluding some countervailing reasons for action.
The authority of those reasons for action is justified, according to Raz, if two conditions are fulfilled: (i) the dependence condition (DC); and (ii) the normal justification condition (NJC). Both conditions are intrinsically related. First, A’s directives have to match reasons that apply to C independently of A’s directives. Second, A has legitimate authority over C if the latter would better conform with those reasons that apply to him or her if he or she intends to be guided by A’s directives than if he or she does not. So, an authority is legitimate when its subjects would likely better conform with the reasons that apply to them by treating the authority’s directives as content-independent and exclusionary reasons for action than if they did not. This is what is meant by the so-called ‘service conception’ of legitimate authority: it facilitates its subjects’ conformity with the (objective) reasons that already apply to them and hence respects their autonomy. By autonomy, I mean having and exercising the capacity to choose from a range of options.
Among the content-independent reasons that may trigger the application of the NJC, one usually mentions the authority’s epistemic expertise, its cognitive, decisional or volitional ability or its co-ordinative ability. The specific justification will vary depending on the circumstances and the concrete ability of each legal norm or set of legal norms. This explains the piecemeal nature of law’s legitimate authority in Raz’s account, ie the fact that the law cannot have general legitimate authority over all subjects at one given time, and this realisation is quite illuminating when contrasted with the law’s general claim to authority.
Of course, other reasons to comply with the law may co-exist and complement the cases where legal norms are truly legitimate, but those reasons such as consent-based or promise-based reasons are not legitimate reasons in a strict sense. Scope precludes rehearsing the many shortcomings of the consent-based justification of political authority, whether consent is thought of as express or tacit, and as actual or hypothetical. It suffices to mention an important one. Consent to be governed by a malevolent authority is similar to promising to do something morally wrong; neither generates the reasons it might under better conditions. As a result, Raz argues persuasively that consent to a legal authority is effective as a source of obligations only if the authority respects autonomy and hence satisfies an independent test of legitimacy. Of course, consent can still have some impact on legitimacy; it can strengthen obligations to obey and can express a citizen’s trust in their government.
One of the major content-independent sets of reasons for action that can be provided by a public authority in the legal context is a salient set of co-ordinative reasons. When there is an independent reason to co-ordinate in circumstances of disagreement, legal authority can help identify some of the conflicting reasons for action or orderings of reasons as salient and hence can help legal subjects co-ordinate over them.
In fact, as I have argued elsewhere, co-ordination on issues of common concern is a much more common requirement in the pluralist circumstances of contemporary politics than legal theorists are usually ready to concede. In conditions of pervasive and persistent reasonable disagreement about justice, the creation of a legal order as a means of general co-ordination over matters of justice is actually in itself a requirement of justice. Law constitutes the best co-ordination mechanism one may think of; this has to do with its decisive and expressive ability, but also, although not only, with the sanctions it can provide in case of non-conformity. As a result, the reason to co-ordinate over certain issues is not restricted to certain contexts and legal areas where co-ordination problems arise, but it is a more general reason to constitute a legal system and to abide by the rules of that legal system as a whole, whether or not those rules effectively solve co-ordination problems in practice.
As a result, the kind of co-ordination at stake here is (partial-conflict) co-ordination over moral concerns when people disagree reasonably over them and therefore have an independent reason to co-ordinate over a common take on those issues if they know others will do so as well and can identify what all of them will co-ordinate over — even if this means not doing things the way they separately think is correct. As a result, common objections pertaining to the shortcomings of the Lewis model of co-ordination in cases of classic co-ordination problems and co-ordination over arbitrary matters are easily met. First, co-ordination in the cases concerned implies co-ordinating over pre-existing objective moral reasons, and not over mere subjective interests and preferences. Second, the co-ordinative scheme can provide a separate set of dependent albeit exclusionary reasons and need not simply choose either of the existing sets of reasons; without the law, participants would not be able to act together on an abstract set of reasons.
Scope precludes expanding here on the question of law and co-ordination, but any form of co-ordination over matters of common concern cannot be judged in the same way. Democratic co-ordination provides the most legitimate mode of co-ordination in circumstances of reasonable disagreement over matters of justice. Democratic legitimacy is another dimension of legal legitimacy that escapes most recent accounts of law’s authority that are still, and questionably so, focused on a hierarchical divide between rulers and governed. In any case, since democracy is incremental and rarely fully realised, this account of co-ordination-based authority does not exclude less or non-democratic forms of legitimate co-ordination.
In a nutshell, democratic decision-making is intrinsically valuable because it respects basic political equality. More precisely, majority rule provides all participants with an equal chance of giving salience to their own views over what ought to be done over matters of common concern and thus by taking turns in the decision-making process. It may even in certain deliberative conditions be vested with epistemic qualities. In this respect, it is important to stress that, in a democracy, the authority of law does not only apply to the relationship between citizens and law-making institutions, but also between the latter qua officials. Following the co-ordinative model just presented, the reasons that apply between them reflect those that apply between law-making institutions (A) and citizens (C1, C2, etc). To allow citizens to co-ordinate over the directives given by A, officials in A ought to co-ordinate over it as well and defer to the first directives given by any of their colleague officials within A.
If all this pertains, (democratic) co-ordination provides one the main justifications for the law’s authority. It is a justification that encompasses some of the others mentioned above, including epistemic expertise and executive or volitive ability in certain cases, but it also applies much more broadly than most. This has consequences for the piecemeal approach to legal authority presented before — although law’s legitimate authority is not necessarily as general as the law claims it is, its scope is much broader than conceded by proponents of the Razian account. All this does not, however, prevent the co-ordination-based account from co-existing with other reasons for respect and recognition that usually fill the gap between the law’s general claim to legitimate authority and the piecemeal scope of its objective legitimacy. Nor does it mean that other justifications of authority cannot apply on an individual basis and complement the co-ordination-based legitimacy of the law.
Importantly, the co-ordination-based approach to authority just presented ought not be understood as an alternative to the Razian account described before, but, on the contrary, as a re-interpretation of that account in circumstances of ordinary law-making and public authority. This re-interpretation requires specifying the concept of justified authority and its conditions, in order to accommodate the way the law provides a whole class of subjects, and not each of them separately, with reasons for co-ordinated action over matters of justice and common concern. Interestingly, most of these points have actually been taken on board by Raz’s most recent re-statement of his account of legal authority.
First, the pre-emptive nature of authoritative reasons should be read so as to accommodate the need to identify, in ordinary circumstances of political law-making, both the existence of an issue of common concern over which co-ordination is needed and the existence of an authority able to provide a salient point over which others will co-ordinate, before its authority can actually be confirmed. Prima facie, this would seem to contradict Raz’s contention that if an authority is to make an authoritative determination in a case of co-ordination, it is precisely because it pre-empts the subjects’ own reasoning on the need to do so. Accordingly, it would undermine the whole point of authority to have to identify it as a co-ordinative authority before it can effectively be such. However, according to Waldron, Raz’s pre-emption point can be satisfied if we distinguish serially the recognition of an authority as co-ordinative from that authority’s determination of what is to be done about it. This requirement of knowability of public authority has since been conceded by Raz in his general re-statement of his service conception of authority.
Second, pertaining to the DC, the correspondence between the subject’s reasons and those given by the authority cannot be direct, if independent reasons to co-ordinate have to be recognised. As Raz has since conceded, however, all the dependence thesis requires is that we have the abstract reasons which the public authority gives us new opportunities to pursue, even though we did not have the opportunity to do so on our own beforehand.
Finally, and more importantly, before a public authority can satisfy the NJC, it has to be recognised as a public authority that others will regard as such and around which they will be co-ordinated. This has to take place before the authority can provide a salient point of co-ordination, but this may contradict Raz’s condition according to which an authority should not have to be identified as an authority before it gives reasons which are effectively legitimate and actually becomes an authority. Such an account of authority would simply undermine the point of having an authority in the first place.
According to Waldron, however, this could be fixed by reference to Raz’s considerations about de facto authority and the role of power as a necessary albeit insufficient condition for legitimate authority. For the NJC to apply in cases of co-ordination over matters of common concern, it is useful to distinguish between the individual application of the NJC and the additional requirement that a large number of people regard the NJC as being satisfied as well. As Waldron argues,
[t]he normal justification of public authority, in particular, has two levels to it: the first level might be given by something like Raz’s [NJC]; but the second level requires in addition some sense that a large number of the people who would be governed by the putative authority if it were an authority do actually accept that it satisfies [the NJC] ... I believe this second level test is not something we can just fold back into the first level, laying out as one of the reasons on which the justification of authority is dependent. It operates as a different sort of reason. (emphasis in original)
This last point also explains how the NJC can be reconciled with the proposed account of democratic co-ordination. Prima facie, indeed, what looks like a procedural account of legitimacy would seem to run against Raz’s more substantive theory of legitimacy. This objection fails on one major count: the proposed justification of authority builds upon Raz’s substantive concept of legitimate authority and is not purely procedural. What a legal authority does, when understood along those lines, is provide legal subjects with reasons to co-ordinate over an abstract set of reasons they share objectively even if they disagree about it or its internal ordering concretely. In circumstances of reasonable disagreement about issues of justice and other matters of common concern, they will be able to abide by their own reasons better overall if they co-ordinate, getting their turn in identifying a salient point of co-ordination, than if all of them decide (even correctly) for themselves in each case. This explains how the NJC can be respected individually in each case, besides also having to be shared by all participants.
This reasoning even applies to a non-epistemic account of democratic deliberation, even if the procedure itself does not necessarily improve the chances of reaching the correct result in all cases. What matters in circumstances where no epistemic guarantees can be provided, is to make sure all perspectives are given equal respect and get a chance to become the group’s co-ordinating position. In those circumstances and given the need to choose a single set of rules for all on most issues of common concern, the best way to comply with one’s own reasons is to co-ordinate with others. As a result, what satisfies the NJC is not so much the epistemic quality of the process, but its co-ordinative ability in circumstances where there is a matter of common concern over which there is an independent reason to co-ordinate. Due to our epistemic limitations we are better off co-ordinating than trying to work things out on our own.
Thus, contrary to what Tasioulas argues, if the NJC wants to be a ‘faithful reflection of our epistemic situation,’ it needs to factor in our epistemic disagreements and the need for co-ordination by a public authority as its primary feature. This interpretation of the NJC does not, however, turn it into an empty requirement by condoning any co-ordination procedure that turns out to be effective. The democratic procedure is based on and protects individual autonomy in conditions of political equality and disagreement over matters of common concern.
3. Elements of International Legal Authority
Thanks to its minimal and flexible features, the revised Razian account of democratic co-ordination-based authority provides an instructive account of the legitimacy of international law. In what follows, I would like to argue that it does not only accommodate the differences between international and national law, on the one hand, but also accounts for the diversity of norms of international law, on the other.
Among the various key questions usually addressed in any account of legal authority, there are four main features that need to draw attention in the international context. In an autonomy-based account of legal authority, key questions and objections pertain to: whose autonomy it is we are concerned about, how that autonomy can be said to be enhanced by authority, how consent relates to the best exercise of one’s autonomy and, finally, whether autonomy can sometimes be said to be best protected on its own. In international terms, and mutatis mutandis, this means that one ought to be concerned about: the identity of the subjects of authority in international law and in particular the relationship between states and individuals, the justification of the authority vested in international horizontal law-making processes and in particular the democratic co-ordination justification, the role of additional reasons for respect for the law one encounters in international law and in particular the role of state consent, and, finally, the compatibility between the service conception of authority and state sovereignty.
One of the most important challenges facing any account of the authority of international law lies in the definition of its subjects, ie the definition of those subjects in authority and of those submitted to their authority. In an account of authority based on autonomy, it is essential to identify whose autonomy we are concerned with.
In the domestic context, the subjects of authority are usually taken to be the law and, more exactly a centralised set of state institutions, on one side, and individuals or groups of individuals, on the other. Given the prevalence of the vertical model of political authority, the (state) institutions v individuals model of authority is usually transposed without further thought in the legal context. While it matches the reality of legislative politics, the model has been criticised, however, for personifying law-making authority. This implies obfuscating the identity between law-makers and legal subjects in a democratic legal order, on the one hand, and eluding the co-ordinative relationship between law-making institutions, on the other.
In the international context, the vertical model has become an important source of confusion. To start with, there is no centralised and hierarchical law-making process, but many and without a ranking, ranging from treaty-making and customary law-making to unilateral law-making by IOs. Second, the law-makers are manifold and are implicated to different extents in those different processes; they range from states and IOs to individuals. The same applies to the subjects of those laws which are diverse; they range from natural persons to collective entities like states and IOs. Third, contrary to what applies at the domestic level, not all subjects of international law are subjects of authority, ie subjects to (legitimate) duties to obey the law; some are merely subjects of rights, while others may be subjects to duties to obey, but without the means to claim their rights and duties at the international level. This is the case, according to traditional approaches, of individuals in international law.
Finally, at least in traditional international law-making processes such as treaties or customary law, given the overlap between states qua law-makers and states qua legal subjects, there is prima facie no clear separation between law-makers and legal subjects. As a result, the relationship seems horizontal as opposed to the vertical domestic relation of authoritative law-making. Of course, this could also be said of law-making in a democracy where individuals author their own laws. However, in the latter case, individuals do not act in their private capacity but as citizens or as representatives, whereas in the case of states in international law, the traditional view is that they make international law as one would enter into a private contract.
Regrettably, this complexity is not well reflected in recent accounts of the authority of international law. Most accounts focus on the subjects to whom authoritative laws apply and elude the question of whose authority it is. Those few accounts that discuss law-making institutions include among them states and IOs (and other non-governmental actors), without distinguishing between them and without dissociating their roles between different law-making processes. As we will see, states play a very different role in international law-making depending on whether they are acting on their own or are part of an institutionalised law-making process in an IO. Further, even with respect to subjects of authority, most recent accounts explain the legitimate authority of international law for the main subjects of international law, ie, states. Only a few authors concentrate on individuals and even fewer address both states (and/or IOs) and individuals.
It is essential, however, to lift the state veil if one is to understand the scope of the authority of international law. A state may or may not be conceived as an entity ‘over and above’ the people who constitute it, but if states can act, and be held in duty, this is only because there are people involved. This is true even if legal doctrine, in treating corporate entities such as states as legal persons, neglects the relationship those entities have to people. The potential moral effect of the law on people is in need of justification. This does not mean that all legitimate authority of international law is, in the end, authority over individuals. It means that practices of ascribing duties to collectives of people like states must make moral sense and this means that the moral position of individuals cannot be ignored.
In what follows, the relationship between international law-makers and international legal subjects is addressed in more detail. Of course, the co-ordination-based justification of authority defended here relies on a democratic model of law-making where legal subjects participate in a collective law-making exercise or, at least, where institutions are proxies for legal subjects in that exercise. As a result, the distinction between law-making institutions and legal subjects is merely expository. A second caveat is in order: The co-existence of different subjects of authority implies different authoritative reasons. Although all subjects of authority are discussed here, scope precludes discussing the specific justifications of authority for each of them later on in the article and I shall focus mostly on individuals hereafter.
International law’s authority is the authority of international legal norms, but also accordingly of their law-making processes and hence of the institutions involved in those processes. According to Buchanan, indeed,
[i]nstitutional legitimacy is primary insofar as the legitimacy of particular laws or of a corpus of law depends on the legitimacy of the institutions that make, interpret, and apply the laws (although legitimate institutions may sometimes produce illegitimate laws).
In contrast to the domestic context, there is more than one law-making process in international law, but also more than one law-making institution. Both are related, but the same international law-making process, such as treaty law-making, may itself involve more than one law-making institution and within each institution more than one legal subject. In a nutshell, international law-making institutions can consist of or include IOs, states or even individuals, although in the latter case states or IOs are usually involved at the same time. Even in the case of unilateral law-making in IOs, states are often directly implicated as well, and not only indirectly through the constitutive act of the IO.
First, states are the original law-makers in international law. This is at least the case for the main sources of international law such as treaty-law, customary law and general principles of international law. Traditionally, states are understood as being both the primary law-makers and the primary legal subjects in international law, and hence as making laws for themselves on a horizontal basis. This horizontal conception of international law-making fits the voluntarist or consensualist view of international law whose authority is allegedly based on state consent: states make laws like individuals enter into contracts and they are only bound by what they have acquiesced to. However, unlike individuals directly or indirectly involved in the domestic law-making context who act qua officials either as representatives or as citizens and not in their private capacity, this approach sees the involvement of states in international law-making qua individuals rather than qua officials.
Such an approach is deeply misleading on more than one count. First, and as Waldron rightly argues, states are not just makers of the international legal order as private individuals would be the makers of a web of contractual promises; they are also its officials. Except in the cases where IOs’ institutions are involved in international law-making, international law has few institutional resources of its own. It depends on states for the making, but also for the enforcement of its provisions. Governments are the officials or officers of the international legal system.
Second, although states are free, rational agents, they are not themselves human individuals and cannot therefore act as those would. In the last resort, states are not the bearers of ultimate value. They exist for the sake of human individuals. In the international context, states are recognised by international law as trustees for the people committed to their care. This, of course, becomes clear from international human rights law, but it is also the point of most norms of international law. Ultimately, international law is oriented to the well-being of human individuals, rather than to the freedom of states.
True, important portions of international law still arise out of treaties. And some view treaties as contracts between states acting in a private capacity. It is a banal observation nowadays that all treaties are not, however, analogous to contracts between businesses or individuals. This analogy can make sense with respect to bilateral treaties that regulate particular aspects of trade, for example. This is what is meant by contract-like treaties. In other areas, however, treaty-making is much more like participating in legislation than like striking a commercial bargain. This is certainly true of multilateral treaties that are sometimes referred to as legislation-like treaties; when institutionalised, these treaties have treaty-based bodies and use majority rule.
Finally, states cannot be viewed as independent law-makers separately from the individuals they ought to care for as officials and whom they bind indirectly when binding themselves. The fact that states no longer make laws only for themselves but also directly for other international subjects, such as IOs or individuals, is further evidence of their role as officials. One may think of customary law, for instance, or of other international legal sources that generate duties for subjects other than states, but are the product of state law-making only. Unless states are conceived of as officials and as trustees in international law-making processes, this lack of congruence between international law-makers and legal subjects cannot be bridged.
These considerations about the role of states as international law-makers have important normative consequences. First, states do not make international law just for themselves as free, rational agents, but as officials for their respective populations, other states and IOs. Their role as officials constrains their competence not only in terms of internal accountability, but at the international level itself. States are bound by the rule of international law, ie the set of values and principles associated with the idea of international legality. Second, when acting as officials, whether in the law-making or law-enforcement process, states have to co-ordinate among themselves, the way officials in a democratic state would. However, this kind of co-ordination among officials differs from the domestic context. States act as law-making officials, but also, as will be shown, as proxy-subjects. As a result, both the co-ordination among officials and that between subjects are often merged and cannot easily be dissociated.
Of course, in cases where the normative requirements stemming from states’ role as officials in international law-making are not respected, states can still enter into normative albeit non-authoritative relationships. This is the case with contract-like treaties, for instance. The difference is, however, that they do not act as officials in such cases and cannot in principle bind as a public authority would. A ready objection is the lack of legal security this would imply in international relations. If states can only act as official authorities in international law-making when they are democratic, this drastically reduces the number of states which can produce binding international laws. Following Rawls, one may consider that some non-democratic societies — such as decent hierarchical societies — are capable of ensuring sufficient accountability to protect the value of self-determination. Of course, this does not prejudge states’ ability to bind themselves as subjects only, as we will see in the next section.
Second, international organisations are an increasingly important international law-maker. This is the case at the United Nations or in other international and supranational IOs such as the WTO or the EU. Their unilateral law-making processes themselves involve mostly states, but also sometimes individuals. IOs provide an attractive institutional model for the present account of international law’s authority as they resemble national law-making institutions where subjects organise themselves as official law-makers and constitute different powers that intervene in the process. It therefore provides a platform or screen of officialdom that separates officials from legal subjects. This is the division states have traditionally eluded to in their interstate dealings and which, I have argued, can undermine the legitimate authority of the legal norms they produce, unless the conditions discussed before are respected.
Finally, individuals are increasingly involved in international law-making processes as well, albeit never on their own. They are usually associated to states in IOs’ institutionalised law-making processes, but can also be represented qua citizens or at least as peoples. Examples of parliamentary assemblies in IOs abound. Interestingly, in those cases, the involvement of individuals complements that of a state in a necessary way. States represent the political community as an entity at the international level, but this kind of representation may not always protect individuals in a non-majoritarian way, thus making it important to ensure double representation at the international level that can at least ensure that different majorities will arise. Interestingly, individuals have also become international law-makers in a third way, through the role of national parliaments in international deliberations. When individuals participate or are represented in the international law-making process, however, it is important they do so in a public capacity and not as individuals. In this sense, although the consultation of civil society in international law-making is important in epistemic terms, it serves a very different purpose.
The second part of the identification of the subjects of authority in international law pertains to those subjected to the duties to obey international law. Again, those can be manifold and range from states to IOs and individuals. Different law-making processes will produce laws that apply to different subjects, but the same law-making process may provide different subjects with authoritative rules. Thus, a treaty may bind states primarily, but it may also eventually bind an IO when it takes over the obligations of its member states or even independently from a state succession, and may by extension bind individuals in those states.
First, states are the primary subjects of binding international norms. Most duties stemming from international legal norms directly constrain the action of states. Following the analogy between states and individuals entering private contracts discussed before, states are generally held as being able to bind themselves as free, rational agents. This approach is misleading in this context as well, however. The service conception of authority adopted in this paper contends that authority can only be justified if it facilitates its subjects’ conformity with the (objective) reasons that already apply to them and hence respects their autonomy. Its application therefore has a pre-requisite: the subject bound by a legal norm needs to be an autonomous subject, as it is only so that its freedom to choose from a range of options can be furthered by an authoritative directive. If it has no autonomy, it cannot be subject to legitimate authority in an interesting way.
The analogy between authority for states and individuals presupposes therefore that the value of autonomy extends to the choices and actions of states. At first sight, it seems plausible that it does, given the value of shared membership in a national political community and, as a result, of the collective self-determination of such communities. The problem is that the value of state autonomy can only be explained in terms of the autonomy of the individuals constituting it. States are quite unlike individuals when it comes to the value of their autonomy. Their autonomy cannot simply be equated with that of any of their domestic legal subjects, but is the product of those subjects’ autonomy as a political entity.
As a result, states can only be bound by international legal norms when they represent those subjects as officials and hence can bind them as proxy subjects to international law. When a state is morally bound by a norm of international law, the duties imposed on it will require action that burdens individuals either indirectly, through international state action that is costly to national resources, or directly through the duty to enact domestic laws in order to transpose international law into domestic law (or implement the latter directly in the domestic sphere). This affects individuals’ balance of reasons as a result and explains why the autonomy of states and its ability to be bound depends on its constituencies and hence on its ability to represent the latter.
Of course, states remain free, rational (albeit artificial) agents and as such they can enter into binding agreements the way an individual would enter into a contract. This can be the case for many contract-like treaties and other international agreements, although consent does not necessarily bind in all cases. The opposite view would simply strip from states their right to bind themselves and hence from any of the meaningful implications of their quality as primary international legal subjects. Further, states’ international legal obligations to obey would remain in place even if they are illegitimate, as they are often backed-up by legal sanctions. And so would states’ moral obligations to abide by morally correct directives which bind individuals (and states for them collectively) in any case. But populations unrepresented by those states would not be morally bound by those legal directives qua law. Nor could those states be bound in that way as a result. True, this constitutes a risk and it could eventually weaken those states’ credentials at the international level. However, it could also provide an important incentive to democratise states from within, thus eventually leading to an increase in the overall legitimacy of international law itself.
Second, international organisations can also be subject to duties to abide by international legal norms. Again, their autonomy only exists in a meaningful way if they can be said to represent states and individuals, and hence be bound as proxy subjects to international law. This has important consequences in terms of IOs’ duties under international law, especially when their member states are bound but do not comply and where individuals need to be protected. In those circumstances, the double degree of representation makes for a complex allocation of duties.
Finally, individuals may also be bound by international law, with or without their own state. They can be said to be bound directly, as when they bear international legal duties independently from their transposition in national law, but also indirectly through their state as their future national and international actions may be constrained by their state’s own duties. Even in the former case, individuals can only rarely be bound without their state being bound at the same time. States will usually have duties to implement direct individual duties under international law within their domestic legal orders. This is a particularly sensitive question in the human rights context, where the co-existence of direct human rights duties of individuals or groups of individuals with those of the state raises difficult conceptual issues — about the nature and justification of human rights in particular. In cases where both states and individuals are bound by the same legal norm, however, their duties may actually be different in content.
B. Justifications of Authority
(i) International co-ordination and Authority
The next question one ought to address in any account of legal authority pertains to the justification of authority, and more precisely to how autonomy can be said to be enhanced by a given authority. If the democratic co-ordination-based justification of law’s legitimate authority I described before constitutes a convincing justification of the law’s authority in the domestic context, it is even more so at the international level where it matches the reality of international law-making. This has to do with the sources of international law, their relationships, their subjects and finally, their subject matter.
First, the horizontal nature of most relationships between subjects of international law makes justifications of authority based on a vertical ruling relationship between an authority and governed subjects less probable. The legitimacy of major international law-making processes, such as customary law-making, multilateral treaty-making or international institutional law-making, is therefore well accounted for by a co-ordinative justification of the authority of international law. As a matter of fact, enhancing those sources in a co-ordination-abling manner can increase their legitimacy. For instance, developing current codification processes even further can raise the chances of co-ordination around a clear canonical set of rules. Further, the multilateralisation of law-making processes enhances the transparency and information about a salient set of rules.
The co-ordinative authority model accounts particularly well for the making of valid customary law and hence for its ability to claim legitimate authority. It can, for instance, help lift the so-called paradox of customary law-making (ie that it requires mistakenly believing that something which is about to become law through being practised as such is already law). In a co-ordination-based approach to law, participants have to believe that the salient point of co-ordination, ie the law, is a point over which others will co-ordinate qua law, before it actually becomes law and co-ordination can take place. The opinio juris corresponds, in other words, to the conviction that others will co-ordinate if one does, rather than to the later belief in one’s duty to obey an existing legal norm and hence to co-ordinate over it for the future. Of course, not all co-ordinative outcomes, and in this case norms of customary law, will necessarily be vested with legitimate authority depending on whether the conditions of legitimacy have been respected in each case, but they can have a serious claim to legitimacy.
Second, the lack of centralised and hierarchical law-making processes in international law confirms the need for various levels of co-ordination within the same regime, but also across regimes. Co-ordination can ensure coherence between sets of norms and across regimes, without hierarchy. It can also account for the existence of a public authority and multilateral and multi-level law-making community despite the absence of a centralised law-making process and institution. There must be a way, indeed, of distinguishing the authority of international law from that of the Pope, for instance, when giving directives about global peace or sharing the world’s natural resources. This distinction is made difficult in the international context given the dispersed nature of public authority. Co-ordination provides that very distinctive feature and justification.
Third, the connection between subjects of domestic and international law, and the fact that international law binds individuals directly in the domestic context, make the democratic justification of the authority of international law even more potent. This is particularly important in circumstances where legal sources, regimes and orders overlap and increase the potentiality of conflicting claims to authority in comparison with the domestic context. This is even more the case as those conflicts cannot be solved by reference to simple legal rules of conflict within each legal order and even less by reverting to the conflicting background moral reasons as they would in the domestic legal order.
Finally, the democratic co-ordination-based justification of authority fits the need for international regulation on difficult matters of common concern among subjects whose diversity of views creates pervasive and persistent disagreements. This applies in cases of classic co-ordination problems such as: problems related to disease, economic instability, environmental degradation, the proliferation of weapons of mass destruction, migration movements that cannot be addressed by individual states acting alone but only through co-ordination. It also applies, however, to conflict and partial conflict co-ordination cases where there is disagreement about issues of justice and common concern, and where it is better that all co-ordinate over the same set of international norms rather than acting individually (even correctly) according to their own reasons. In a legal system characterised by deep divergence in ethical, religious and political beliefs and practices, democratic co-ordination provides one of the best justifications of authority to escape irreducible substantive controversies.
Of course, legal enforcement mechanisms and sanctions are often mentioned among the features that make law one of the best means of co-ordination. Those features, however, are not necessary for co-ordination to take place. As a result, the lack of enforcement mechanisms and sanctions, at least of centralised ones, in international law does not undermine its co-ordinative ability in any significant way. The canonical nature of most international legal norms and the development of institutional mechanisms of interpretation and implementation compensate for the lack of centralised sanctions.
(ii) International Democracy and Authority
Like in the domestic context, the co-ordination-based justification of international law’s authority is enhanced in circumstances of disagreement about justice if the co-ordination procedure and functioning is democratic. What democracy means in the international law-making context is a difficult issue and scope precludes fully addressing it here. In a nutshell, global democracy does not equate either with indirect state democracy nor with a state-like democracy at the international level. On the contrary, it groups all democratic processes that occur within and beyond the national state and which outcomes affect individuals within that state, but in ways that link national democracies among themselves and to other transnational, international or supranational democratic processes.
As I have argued elsewhere, the best account of international legitimacy is a demoi-cratic account, ie an account based on the functional and territorial inclusion (pluralistic) in national, regional and international law-making processes. Different levels in those processes (multi-level) of all states (and groups of states) and individuals (and groups of individuals) qua pluralistic subjects of the international political community (multilateral) have fundamental interests that are significantly and equally affected by the decisions made in those processes.
Importantly, global demoi-cracy can only strive if states are democratised from within and at the same time; this is actually a key feature of the legitimate authority of international law for states, as discussed before. Indirect state democracy does not replace direct individual participation or representation in international processes, however, at least from the perspective of the authority of international law over individuals.
There are many reasons for this. To start with, their interests might diverge from those of their national polity, because they are members of a minority at the national level. In fact, even in a democratic state, the international or external interests of that state might actually differ from those of the sum or the majority of that state’s citizens, given the primary and increasing role of the executive in foreign policy and international negotiations. Furthermore, even when individual interests match their government’s, the rule of international democracy, and in particular majoritarian rule in multilateral law-making (but also veto rights in the case of unanimous voting), cannot always guarantee international results that are in line with individual national will, or at least respect the principle of political equality in terms of the proportionality of votes to the size of the populations represented. Thus, indirect international democracy models face the famous dilemma between states defending their citizens’ interests at the expense of other states and their citizens on the one hand, and following the rules of international democracy at the expense of their own citizens’ interests, on the other. Moreover, not all individuals affected by international law are citizens of democratic states, and hence have a say in national democratic processes pertaining to international issues, or are represented by democratically elected representatives in international fora. This creates an inequality in legitimacy.
This account of global demoi-cracy may usefully be complemented with the democratic principles of broad accountability developed by Buchanan in his account of legitimate international law-making institutions. By the latter, he means that,
[t]hese institutions must cooperate with external epistemic actors — individuals and groups outside the institution, in particular transnational civil society organizations [sic] — to create conditions under which the goals and processes of the institution as well as the current terms of institutional accountability, can be contested and critically revised over time, and in a manner that helps to ensure an increasingly inclusive consideration of legitimate interests, through largely transparent deliberative processes.
Of course, democracy requires minimal guarantees of human rights to function properly and these should therefore be respected within international law-making processes. However, since the legitimacy of international human rights norms arguably follows a specific kind of justification, understanding their co-originality with democratic self-determination is even more important at the international level than in the domestic context. In those circumstances, it would be vain to look for a foundation for democratic legitimacy in international human rights law. Rather, international law-making should be organised so as to both provide those rights with deliberative and inclusive legitimacy and to bolster that deliberative process by protecting minimal rights.
In any case, other justifications for the authority of international legal norms may also be provided when they are not already covered by the democratic co-ordination justification. This is the case of the volitional, decisive and cognitive abilities of certain norms of international law. In most cases, however, as in the domestic context, the democratic co-ordination justification of authority encompasses a lot of other justifying grounds and it constitutes the most important justification of international law’s authority. As a matter of fact, democratic co-ordination captures precisely those other grounds that compensate for the predominant circumstances of epistemic limitations. This is likely to be even more pertinent in international law given the diversity of the objects of international law, of its subjects and of their relationships.
Like in the domestic context, however, it may of course happen that some international legal norms are not legitimate on any of those grounds. Democracy, even defined as it is in this article, is still limited and largely incremental in the international context. As discussed before, this hiatus between the general scope of the claim to legitimacy laid by international law and the effective scope of its legitimacy in practice is likely to be more significant than in the domestic context, in the absence of a single centralised set of law-making institutions and processes. This need not be a source of concern, however, in view of the widespread de facto authority of international law and of the co-existence of other reasons to respect international law.
C. Piecemeal Authority and Other Reasons for Respect
As it is the case in the domestic context, it is important to emphasise other reasons for respect and recognition of the authority of international law which ought not be confused with authoritative reasons to obey but can complement them. These reasons might, as a matter of fact, explain how international law’s claim to general legitimacy may seem entirely granted in practice, although it does not and cannot have general legitimate authority over all its subjects at one given time.
One of those complementary reasons for respect (and not for obedience) is consent. Interestingly, consent theory is the dominant account of legitimacy among international lawyers and this is traditionally captured by the general principle of international law pacta sunt servanda. The popularity of consent in accounts of the legitimacy of international law may partly be explained by the widespread failure to distinguish between the normative and sociological senses of legitimacy: at least on the basis of an individual analogy, there seems to be some empirical connection between believing a norm is binding and having previously consented to it.
In normative terms, however, as we have seen before, consent fails to provide independent authoritative reasons to obey the law. It can strengthen legitimate authority and can express trust that facilitates co-ordination, but it cannot generate it. In international law, there are further specific reasons why (democratic) state consent is neither a sufficient nor a necessary condition of legitimacy.
The reasons why state consent is not a necessary condition for the legitimacy of international law are well-known. To start with, states are no longer the only law-makers in international law. IOs and individuals have become participants in unilateral international law-making and multilateral treaty-making processes, and legitimising their role by reference to state consent would be making at best, too diluted a connection. Further, even when states are still the main international law-makers, their consent is increasingly less necessary — unless the claim is that increasing portions of international law are becoming illegitimate. As examples, one may mention the difficulties of objecting to customary law and the development of multilateral law-making treaties whose norms have customary value for states which have not ratified them and in the development of which majority voting is often used instead of unanimity.
Furthermore, even when state consent is given, it is not sufficient for legitimacy to obtain it. To start with, there is great disparity of power among states and this puts the voluntariness of a state’s agreement into question. Furthermore, many states do not represent all of their people, or at least do not do so in a reasonable way. Finally, the connection between state consent and international (or worse, supranational) institutions is becoming increasingly loose both diachronically and synchronically.
Despite similar limitations qua source of authority to those applying in the domestic context, consent is even more important as a source of recognition and respect of the authority of international law than in the domestic context.
To start with, due to the plurality of subjects involved in international law-making and their manifold roles, as is particularly clear from the role of states qua subjects and officials, consent can clarify the existence of justifications for the authority of international law for specific subjects. Furthermore, consent can often be used as a normative strengthener when states consent to abide by duties of justice or to authoritative international legal norms. This is particularly important in international law, as consent can ease co-ordination by clarifying participants’ intentions to co-ordinate and their trust in public authority, and hence can further legitimate authority.
Finally, the pre-eminence in traditional international law, but also in certain areas of international law today, of contract-like promises between states, and the difficulty in distinguishing obligations stemming from those promises from obligations to obey the law qua law, explains how consent could usually account for widespread cases of de facto authority of international law. For instance, qua rational collective agents, states can bind themselves legally through consent and promises under certain conditions, even when they cannot be deemed authoritatively and morally bound by those obligations because they do not act as officials for their constituency.
Of course, there are conditions to be respected before consent can be deemed to strengthen independent authoritative reasons — conditions which may themselves overlap with justifications of authority itself. First, consent cannot be used to bind oneself to immoral acts. Second, state consent ought to be free and unconstrained, on the one hand, and duly informed, on the other.
D. Service Authority and Sovereignty
One of the main challenges to the legitimacy of international law is that it allegedly fails to respect the autonomy of states, intruding upon domains in which they should be free to make their own decisions. State sovereignty is often understood in international law as a competence and in particular as the power to make autonomous choices (so-called sovereign autonomy). The legitimate authority of international law is as a result often opposed to state sovereignty.
Based on the service conception of authority discussed before, authority can only be justified if it facilitates its subjects’ conformity with the (objective) reasons that already apply to them and hence respects their autonomy. Regarding some matters, however, it is more important that a person reaches and acts on his or her own decision, rather than take a putative authority’s directives as binding, even if doing the latter would result in decisions that, in other respects, better conform to reason. This is what Raz has recently referred to as the independence condition. As argued by Tasioulas, it is difficult, however, to distinguish those cases from cases where legitimate authority can apply, the incompatibility being at the most contingent and relative to certain circumstances.
This is even more clearly the case in international law. If states are deemed as officials both qua law-makers and qua proxy-subjects of authority in the international legal order, their autonomy cannot simply be equated with that of any of their domestic legal subjects. It is the product of those subjects’ autonomy as a political entity. Considered in both its internal and external dimensions, a state’s sovereign autonomy is a purely legal construct, not something which value is to be assumed as a first principle of normative analysis. In its internal dimension, the state works as a legal organisation — it is the outcome of organising certain rules of public life in a particular way. Its sovereignty is artificial and it is legally constructed for the benefit of those whose internal interests it protects. In its external dimension, the sovereignty and the sovereign autonomy of the individual state are equally artefacts of international law. There can be no international legal order without sovereign states, but equally there can be no sovereign states without international law. What a state’s sovereignty is and what it amounts to is not given as a matter of the intrinsic value of its individuality, but determined by the rules of the international legal order. Those rules define state sovereignty so as to protect the internal and external interests of the political community qua sovereign equal to others, but also to protect the interests of other subjects of international law.
Hart, in his chapter on international law in The Concept of Law, argued that states’ sovereignty and hence autonomy are defined by the limits of international law,
[i]f in fact we find that there exists among states a given form of international authority, the sovereignty of states is to that extent limited, and it has just that extent which the rules allow. Hence we can only know which states are sovereign, and what the extent of their sovereignty is, when we know what the rules are. (emphasis added)
As a result, provided states act qua officials and proxy-subjects when making international law and the other conditions of authority discussed before are fulfilled, state sovereignty is necessarily compatible with international legal authority. There is simply no self-sufficiency of states outside the international legal order and without co-ordination with other states. If this is correct, the incompatibility between service authority and freedom is likely to be even more contingent in the international legal order than in the domestic context given the prima facie coherence of a legal order. Since international law defines the sovereign autonomy of states, the relationship between the legitimacy of new norms of international law and state sovereignty is a legal one that ought to be solved as a conflict of legitimate authority within international law itself.
As a matter of fact, since the subjects of international and national law can be equated as the single referent in terms of autonomy and those legal orders compared in terms of service to their autonomy, the relationship between national and international law in terms of legitimacy is facilitated. Indeed, those relationships have to be organised around a principle of subsidiarity in the protection of individual autonomy, thus limiting international constraints on state sovereignty to cases where there are insufficient national constraints of that kind. This explains why, for instance, international legal duties stemming from human rights guarantees are minimal and why domestic human rights norms usually benefit from the principle of favour (as exemplified by Article 53 of the 1950 European Convention on Human Rights). This is the case because of the decoupling of popular sovereignty from national sovereignty in certain cases where international law-making is more inclusive of all those whose fundamental interests are significantly affected. This applies to a state’s population’s own interests, but also to those of other states in a system where sovereignty and hence national autonomy can only be respected if all other sovereigns are treated equally and their populations are considered with equal respect.
Of course, the absence of independent state autonomy distinct from that of the sum of its legal constituents does not mean that states cannot be deemed as collective agents and hence relate as free, rational agents. In that context, however, as discussed earlier, states do not act to promote their constituency’s autonomy and cannot therefore be deemed as authoritatively bound by their actions in moral terms. This does not preclude legal obligations, of course, as well as moral reasons to abide by morally correct directives. Furthermore, as discussed before, complementary sources of non-authoritative reasons for action such as consent or trust usefully complement those legal duties. To that extent, their sovereignty may have to be protected, but not against the authority of international law.
4. Exclusions and Exceptions to International Legal Authority
Even if one deems the authority of international law as justified in certain cases and compatible with state sovereignty, its legitimate authority has been questioned on two major counts in recent times. The first is parochialism and the alleged absence of legitimate authority of international law over certain people and societies, and the second is disobedience to legitimate authority and the alleged exception to the legitimate authority of international law in certain circumstances. Whereas the former implies a lack of authority on the part of international law, the latter aims at justifying exceptions to that authority.
A. Relativism and Exclusions of International Legal Authority
A well-known challenge to the legitimacy of international law is based on a brand of moral relativism. One main objection is that international law embodies a ‘parochial’ set of values that it unjustifiably imposes on people and societies who do not share it. It is useful to distinguish strict parochialism from a sub-group of parochial complaints that are also known as ‘exceptionalist’. According to exceptionalists, and in particular American exceptionalists, although international law may possess legitimate authority over all other subjects of international law, it does not have authority, or does not have it to anything like the same extent, over the US. This is because the US is more likely to conform with the reasons that apply to it, if it is unconstrained by those norms.
If one refers to the conditions justifying authority in the Razian conception discussed before, the claim that is made by parochialism is that international law does not have legitimate authority over certain subjects of international law. The parochialist complaint can be read as denying that international law facilitates conformity with objective reasons, as opposed to the ‘reasons’ asserted by certain dominant groups. In other words, parochialism denounces the legitimacy of international law for disregarding the DC. By contrast, exceptionalism considers the reasons reflected by international law as applying to the putative subject, but claims that even if they do, the latter does not conform better with them by treating international law as authoritative. Exceptionalists reject the legitimacy of international law for violation of the NJC, presumably on all its potential grounds including co-ordination.
I am assuming here that neither of those objections based on a sceptical view of morality. Of course, adopting an objective view of morality does not equate with adhering to a monist conception of morality: the background to the following analysis is an objective albeit pluralist account of morality that can accommodate conflicts of values and different orderings between them. Further, it should be emphasised that holding to moral objectivity does not mean denying the importance of the contextualisation of moral values recognised by international law at the domestic level, nor the possibility of the historical national localisation of objective values recognised by international law and of historical changes in that localisation over the course of time. Finally, one may legitimately contend that the institutional dialogue and mutual adjustment promoted by democratic co-ordination in international law-making, and international decision-making generally, pays sufficient attention to the issue of cultural diversity when adopting or applying international law. As a result, the difficulties raised by cultural or social pluralism in both objections should have been addressed.
Scope precludes addressing both objections with respect to the implications of moral pluralism for the legitimacy of international law in full in this article. It is important, however, to address them briefly to the extent that they affect the plausibility of cross-cultural legitimacy of international law and cannot simply be put at rest by reference to the piecemeal nature of the legitimacy of international law. While the proposed account of the authority of international law might be able to accommodate some degree of cultural and moral pluralism, these objections’ exclusions from the legitimacy of international law are too broad even for a piecemeal account of the authority of international law. If successful, they would preclude a whole set of international legal norms from applying to a superpower or exempt a whole range of cultures (regions and countries) from the scope of application of some important international legal norms such as human rights in particular. As a result, one needs to prove that neither the dependence nor the normal justification theses can be interpreted as the parochialism and exceptionalism objections claim they can.
Despite appearances, the moral pluralism objection made by parochialism can be met. It relies on the absence of correspondence between the reasons or, more often, their orderings given by certain international legal norms (mostly human rights norms), and those applying to international legal subjects, such as states in particular. True, the DC can be defeated in the absence of a set of abstract reasons matching those of a legal subject. However, as discussed before, once those abstract reasons match, dependence does not exclude the possibility of an imperfect match between those values in a concrete situation or, more importantly, a different ordering of the same abstract reasons. This is an essential feature of the co-ordination-based re-interpretation of the law’s legitimacy — the DC in circumstances of reasonable disagreement about justice and other matters of common concern is fulfilled by democratic co-ordination over one of the orderings of values at stake. As a result, the need to decide and act upon a single ordering of reasons on certain matters of common concern, and hence the co-ordination on the basis of legal norms in those areas, may make it — all things considered — acceptable to privilege one ordering of values in interpreting those norms over alternative eligible orderings favoured by other societies; provided, of course, those abstract reasons or values themselves are shared. In fact, moral pluralism and the plurality of orderings of conflicting reasons is part and parcel of law-making at the domestic level, and there are no qualitative differences in this respect at the international level.
Generally speaking it is worth noting, however, that international law often only entails incompletely theorised agreements and highly indeterminate norms in any case, thus leaving sufficient leeway for further co-ordination and decision-making on the concrete ordering of values at the domestic level. It accommodates moral pluralism, in other words, by not forcing complete orderings of the same values over its ultimate subjects, ie individuals. Moreover, in the case of human rights, as Tasioulas elegantly argues, complete justifications need not be given by international human rights law or its application, which only protect fundamental interests but leave the specific balancing between those interests and their concretisation to national legal orders. Of course, human rights norms exist both at the international and at the domestic levels. The role of international human rights is different from that of domestic human rights, however. International human rights are abstract rights. They do not affect the specific political role and the justification of human rights in democratic politics. This in turn weakens the intensity of the duties placed by international human rights norms on states and the need for international co-ordination beyond a minimal threshold.
As to the exceptionalist objection, a similar rebuttal may be put forward. The argument made before relative to the non-epistemic nature of the NJC in its co-ordination-based re-interpretation explains how the exceptionalist’s argument cannot hold in an international context. In that context, indeed, most matters are of common interest and become questions of common concern when a single decision has to be made. In those circumstances, international legal subjects’ actions necessarily influence each other and they need to co-ordinate accordingly. Even in cases where they might be able to comply with their own reasons better on their own, and arguably to make things better for others as well, the NJC ought to be interpreted as preventing subjects from taking the law into their own hands.
In any case, disobedience to prima facie authoritative reasons given by international law is not absolutely prohibited in the proposed account of authority. It may be justified in certain cases, thus undermining the ‘all or nothing’ flavour of the exceptionalist objection.
B. Disobedience and Exceptions to International Legal Authority
In the Razian account of authority used here, legitimate reasons for action are understood as prima facie reasons that may be precluded by weightier reasons of justice in certain circumstances. These circumstances are those captured by the ideas of conscientious objection, when disobedience is private and does not aim at expressing discontent, and civil disobedience, when disobedience is public and aims at changing the law.
Scope precludes addressing the concept and conditions of civil disobedience in full detail in this context. It suffices to say that in a democratic co-ordination-based account of authority, conditions related to the exhaustion of democratic channels need to be fulfilled before civil disobedience can be accepted as a justification for the violation of one’s duty to obey the law. Accordingly, in the current conditions of international law-making, the lack of complete implementation of the democratic requirements in the law-making process make the case for civil disobedience stronger, provided the other conditions are fulfilled. In a nutshell, the latter pertain to the political motivation, the publicity, the non-violent nature and the proportionality of the act of international disobedience.
This justification of disobedience to legitimate international legal norms should not be conflated, however, with some authors’ defence of illegal reform of international law. According to that idea, a breach of illegitimate international law may be justified if it helps making international law more legitimate. True, in the context of civil disobedience, the intentional violation of existing international legal norms is part of an effort to create new and more just ones. However, the difference with civil disobedience is that the international legal norm that is disobeyed is not legitimate and that there is no prima facie moral obligation to abide by that legal norm. The breach of international law only violates the legal duty to obey international law.
All the same, given that states are legally created entities whose freedom is inherently law-constrained, one may argue that illegal state action is even less justifiable than it is in the case of an individual disobeying the law. Furthermore, even though no issues of legitimacy arise in this case and the conditions for the justification of such a breach of legality may be less strict as a result, illegal reform of international law can be questioned on grounds pertaining to the idea of the rule of law.
The rule of law celebrates features of a well-functioning system of government such as, among others: publicity and transparency in public administration, the generality and prospectivity of the norms that are enforced in society, the predictability of the social environment that these norms help to shape, the procedural fairness involved in their administration, the independence and incorruptibility of the judiciary, and so on. More precisely, it identifies a society where those in power exercise it within a constraining framework of public rules rather than on the basis of their own preferences, their own ideology, or their own individual sense of right and wrong. As a result, legality is also a matter of the quality of the law’s sources. The law-making processes by which we identify valid legal norms should themselves be such as to satisfy the requirements associated with the rule of law. In other words, the rule of law requirements on the law-making process reflect the idea that those subject to the law should be able to identify the law and conform to it.
The same can be said about the legality of international law. International law-making processes should be such as to satisfy some of the requirements associated with the rule of international law and in particular the requirements of clarity, publicity, certainty, equality, transparency and fairness. In the long run, and despite the occurrence of forms of illegal law-making in the current circumstances of international law, international law’s legality will only be able to consolidate itself if its law-making processes are organised so as to reflect the very values inherent in the rule of international law. This is even more important in international law as legality and legitimacy overlap less often than they do at the domestic level, thus making efforts to enhance the plausibility of any international legal norm’s claim to legitimacy even more important. If that plausibility relies on the ability to co-ordinate democratically, as I have argued it should, then international law-making processes should be as democratic and respectful of the rule of law as possible.
Large parts of international law benefit from de facto authority; its norms are generally obeyed in practice, even if it is for various instrumental and non-instrumental reasons, albeit not necessarily legitimate ones. Its de facto authority is actually reinforced by the particular importance of consent in a pluralistic, multilateral and multi-level international community. As a matter of fact, international law claims much more legitimate authority than it can ever have in practice. Moreover, even when it is legitimate, that legitimacy is piecemeal and varies depending on the legal subjects.
If claiming more authority than one can have and only having piecemeal authority are common features of domestic law, it is even more common in the international realm given the multiplicity of law-making processes, law-makers and legal subjects. Although the legitimate authority of international law need not be co-extensive with the law’s claim to legitimacy for international law to be valid, legitimacy matters for the legality of international law to be sustained in the long run, but also to justify its increasing impact on individuals.
In that context, I have argued that a co-ordination-based re-interpretation of the Razian model of authority, and a democratic one in particular, provides the best account of the legitimacy of international law. Its autonomy-based approach constitutes a convincing account of the way to identify the subjects of the authority of international law and capture in particular the relationship between states and individuals in that context. This approach justifies the authority of international law, in particular with respect to its horizontal and decentralised law-making processes, explains the role of additional reasons to respect international law and in particular the role of state consent, and, finally, reconciles international law’s authority with state sovereignty. Furthermore, common objections to the legitimacy of international law deriving from moral or cultural relativism can also be met in that framework.
The main gist of the autonomy-based approach to the authority of international law developed in this article is that a complete understanding of that authority requires looking beyond, or more exactly, within the state as the primary subject of international law and hence of authority. While it is true that states are the primary law-makers and primary legal subjects in international law, it would be wrong to understand them as individuals entering into mutual promises and contracts. Even though international law-making is still clearly horizontal and mostly implicates the same subjects at each end of the process, states are only primary law-makers qua officials for their national constituency. And they are only primary legal subjects qua proxies for their own subjects.
As a matter of fact, understanding the legitimacy of international law in relationship to its ultimate subjects, ie individuals, provides potential answers to recent developments in international governance. This is particularly relevant when facing the contemporary challenges raised by internal and external legal pluralism. In a legal pluralist environment where competing claims to legitimacy are made over the same people by national, regional and international legal orders, this approach to legal authority reveals the necessary connections among the justifications for the legitimacy of domestic, regional and international law. One may actually venture that it is because international law is becoming a direct source of obligations for individuals that the boundaries between legal orders ought to be eroding and that international legal pluralism is here to stay. It is not so much international law that has changed, but its subjects. Strictly speaking, one may even say that there is not so much a legitimacy crisis as a legality crisis in current international law. It is important as a result that international legal practice and international law-making processes adapt to those subjects’ moral situation and the need to justify international law obligations with respect to their situation.
Interestingly, one main result of emphasising the importance of justifying the authority of international law to individuals is that we can see just how much of it depends on the conception of the state in its current form and its relationship to us all. Lifting the state veil is what Julius Stone had started in his search for justice in the structure of international law and is a topic that ought to be addressed anew in the burgeoning field of the philosophy of international law.
[∗] Professor of Public International Law and European Law, University of Fribourg (Switzerland). This is a revised version of the Julius Stone Address I gave on 19 August 2008 in Sydney. I would like to thank Wojciech Sadurski for the kind invitation and hospitality and Kevin Walton for the excellent co-ordination before and after the lecture. Many thanks also to the Sydney Law School faculty members who attended the lecture and the seminar on the next day for their helpful comments and feedback. Further thanks are due to the audience at the University of Frankfurt’s Cluster of Excellence on the Emergence of Normative Orders’ opening conference on 14 and 15 November 2008 for their remarks and critiques. Last but not least, I would not have been able to prepare my article in better conditions than those provided by the Adam Smith Research Foundation, University of Glasgow, in July and August 2008 and would like to thank Adam Tomkins for inviting me there. More generally, I am grateful to John Tasioulas and Allen Buchanan for lively discussions of their respective papers on the legitimacy of international law (see Buchanan, below n1 and Tasioulas, below n20), as those discussions have helped me clarify my own ideas in the course of our disagreements.
 Allen Buchanan, ‘The Legitimacy of International Law’ in Samantha Besson and John Tasioulas (eds), The Philosophy of International Law (2010, forthcoming).
 The article focuses on legal authority in contrast to private authority, but also to political authority in general. On the distinction, see Joseph Raz, The Authority of Law (1979) (‘Raz 1979’); Joseph Raz, The Morality of Freedom (1986) 23, 38 and 70 (‘Raz 1986’); Joseph Raz, Ethics in the Public Domain (1995) 210, 341 and 355 (‘Raz 1995’); Joseph Raz, ‘Comments and Responses’ in Lukas Meyer, Stanley Paulson and Thomas Pogge (eds), Rights, Culture and the Law: Themes from the legal and political philosophy of Joseph Raz (2003) 253–78 (‘Raz 2003’); Joseph Raz, ‘The Problem of Authority: Revisiting the Service Conception’ (2006) 90 Minnesota Law Review 1003, 1004–5 (‘Raz 2006’). Interestingly, some authors, influenced by international relations literature, focus on political authority only, and not legal authority in order to grasp the specificity of law-making by international organisations: see, for example Daniel Bodansky, ‘Legitimacy’ in Daniel Bodansky, Jutta Brunnee and Ellen Hey (eds), Oxford Handbook of International Environmental Law (2007) 704–23; Ian Clark, Legitimacy in International Society (2005); Jean-Marc Coicaud and Veijo Heiskanen (eds), The Legitimacy of International Organizations (2001). This approach faces the risks of: being too broad (by covering political duties distinct from duties to obey the law), of begging the question of the existence of an international political community and, finally, of conflating too quickly law and law-making institutions. See also section 3a, below.
 In what follows, I will use ‘authority’ to mean legitimate authority. For the same use of the term, see Raz 1986, above n2; Raz 1995, above n2; Raz 2006, above n2.
 See Raz 1986, above n2, 23.
 See Raz 1986, above n2; Raz 1995, above n2; Raz 2006, above n2 for the distinction between theoretical and practical authority.
 See Herbert Hart, The Concept of Law (Rev ed, 1994) 227–30 on the difference between international legal norms and morality. Of course, Hart (230–2) then rejects the necessary existence of a moral duty (or conviction thereof) to obey international law qua morally correct law or even qua law tout court, but this has to do with his own account of law’s authority rather than with a specificity of international law itself.
 See Raz 1995, above n2, 342–3.
 See Raz 2006, above n2, 1006–7. See also Allen Buchanan, Justice, Legitimacy and Self-Determination: Moral foundations for international law (2004) 16–19; Buchanan, above n1. See, however, Thomas Franck, ‘Legitimacy in the International System’ (1988) 82 American Journal of International Law 705, 706 (‘Franck 1988’); Thomas Franck, ‘The Power of Legitimacy and the Legitimacy of Power: International law in an age of power disequilibrium’ (2006) 100 American Journal of International Law 88, 91 and 93 (‘Franck 2006’) for a conflation of the sociological and normative approaches to legitimacy.
 On de facto authority, see Raz 1986, above n2, 65; Raz 2006, above n2, 1005–6. De facto authority implies a claim to legitimate authority, albeit not necessarily a justified one. However, legitimate authority does not necessarily imply de facto authority, even though they are likely to be connected. As a result, international legal norms can be legitimate without being effectively complied with and vice-versa; the question of practical compliance with international law (and of its motivation) is an altogether different question. On that question, see Mary Ellen O’Connell, The Power and Purpose of International Law (2008).
 See Hart, above n6, 220.
 See Raz 1986, above n2, 27–8.
 This view contrasts with that of Mattias Kumm, ‘The Legitimacy of International Law: A constitutionalist framework of analysis’ (2004) 15 European Journal of International Law 907, 917. International legality cannot therefore be used as a presumption of legitimacy. See Samantha Besson, ‘Theorizing the Sources of International Law’ in Samantha Besson, and John Tasioulas (eds), The Philosophy of International Law (2010, forthcoming).
 Legitimate authority ought not therefore be understood as excluding power-play in international relations. Scope precludes, however, addressing this topic here. See Bodansky, above n2, 707. Note that de facto authority implies the exercise of actual power, albeit not necessarily coercive power.
 See Raz 1995, above n2, 80; et seq Raz 1986, above n2; Raz 2006, above n2 on consent as a source of simple voluntary obligations and as a source of respect or recognition of an institution, but according to whom consent can only be accepted as a source of authoritative obligations if the other conditions of legitimacy are fulfilled independently.
 See, for example Andrew Guzman, How International Law Works: A rational choice theory (2007); Jack Goldsmith and Eric Posner, The Limits of International Law (2005) 185 et seq for a rational choice account of states’ compliance with international law. See critiques by Allen Buchanan, ‘Democracy and the Commitment to International Law’ (2006) 34 Georgia Journal of International and Comparative Law 305; Franck 2006, above n8.
 See, for example John Rawls, A Theory of Justice (1971) 114–17 and 333–7; Jeremy Waldron, ‘Special Ties and Natural Duties’ (1993) 22 Philosophy and Public Affairs 3. Of course, traditional critiques against the justice-based justification of authority are less incisive when applied to international law. Many general principles of international law such as equality, human rights or good faith, may actually also be principles of justice, and international subjects may therefore have independent or additional duties of justice to abide by those principles. Scope precludes, however, addressing this topic in the present article, and in particular the difficult issue of the allocation of duties of global justice to states, and of the nature and extent of those duties (see Thomas Pogge, World Poverty and Human Rights: Cosmopolitan responsibilities and reforms (2002); Liam Murphy, ‘International Responsibility’ in Samantha Besson and John Tasioulas (eds), The Philosophy of International Law (2010, forthcoming)). In any case, the justification of the law’s moral right to rule qua law remains essential in an international community ridden by cultural and moral disagreements and needs to be addressed as such.
 See, for example Thomas Franck, Fairness in International Law and Institutions (1995). Scope precludes addressing this topic in the present article.
 See, for example Buchanan, above n15, 315–16; Buchanan, above n1; Jeremy Waldron, ‘The Rule of International Law’ (2006) 30 Harvard Journal of Law and Public Policy 15 on the relationship between the legitimacy of international law and the rule of international law. See section 4b, below.
 See, for example Raz 1986, above n2, 66; et seq Raz 1995, above n2; Raz 2006, above n2 on the distinction. See also Buchanan, above n1.
 On these reasons, see, for example John Tasioulas, ‘The Legitimacy of International Law’ in Samantha Besson and John Tasioulas (eds), The Philosophy of International Law (2010, forthcoming).
 See Tasioulas, above n20.
 See, for example Kumm, above n12, 909–17.
 See, for example Franck 1988, above n8; Franck, above n17; Fernando Teson, A Philosophy of International Law (1998) (‘Teson 1998’); Daniel Bodansky, ‘The Legitimacy of International Governance: A coming challenge for international environmental law’ (1999) 93 American Journal of International Law 596; Buchanan, above n8; Kumm, above n12; Fernando Teson, Humanitarian Intervention: An inquiry into law and morality (3rd ed, 2005) (‘Teson 2005’); Goldsmith and Posner, above n15; Allen Buchanan and Robert Keohane, ‘The Legitimacy of Global Governance Institutions’ (2006) 20 Ethics and International Affairs 405; Buchanan, above n15; Bodansky, above n2; Rüdiger Wolfrum, ‘Legitimacy in International Law’ in August Reinisch and Ursula Kriebaum (eds), The Law of International Relations – Liber Amicorum Hanspeter Neuhold (2007) 471–82; Daniel Bodansky, ‘The Concept of Legitimacy in International Law’ in Rüdiger Wolfrum and Volker Röben (eds), Legitimacy in International Law (2008); Buchanan, above n1; Tasioulas, above n20.
 See, for example Hart, above n6, 213 et seq.
 See Hart, above n6, 213–15.
 See, for example Franck 1988, above n8, 91; Franck, above n17, 4–8; Tasioulas, above n20. But see also Goldsmith and Posner, above n15, 186–9 for a rebuttal of non-instrumental reasons for states to abide by international legal norms.
 See Raz 1986, above n2, 215; Raz 1995, above n2, 230–7 on the sources thesis and its justification based on the law’s claim to legitimate authority. See also Besson, above n12.
 See Tasioulas, above n20.
 See Franck, above n17, 6; Franck 2006, above n8, 91.
 See, for example Besson, above n12; David Lefkowitz, ‘The Sources of International Law: Some philosophical reflections’ in Samantha Besson and John Tasioulas (eds), The Philosophy of International Law (2010, forthcoming).
 See the discussion in Raz 1986, above n2, 62. See especially Raz 2006, above n2, 1005–6, 1007–10.
 See Buchanan, above n1.
 Of course, some international legal norms are backed up by coercive sanctions (eg the prohibition of the use of force) and some international institutions claim to have exclusive centralised jurisdiction over certain matters (eg the UN Security Council over the use of force). However, these qualities cannot be generalised, contrary to what is the case in the domestic legal order. In any case, the absence of sanctions is not a constitutive element of the legality of international law, and it is even less a requirement of its legitimate authority: see Hart, above n6, 216–20.
 See the discussion of the changes in the sources of international law in Besson, above n12; Wolfrum, in Reinisch and Kriebaum above n23; Kumm, above n12, 909–17. Most of those changes are usually interpreted as bringing international law closer to domestic law with respect to its nature, but also to its legitimacy and justifications for that legitimacy.
 See on relative normativity, Prosper Weil, ‘Towards Relative Normativity in International Law?’ (1983) 77 American Journal of International Law 413; John Tasioulas, ‘In Defence of Relative Normativity: Communitarian values and the Nicaragua case’ (1996) 16 Oxford Journal of Legal Studies 85. Due to constraints of space, the section of the lecture addressing questions related to the relative normativity of international law and the authority of jus cogens norms or soft law was excised: see, however, Besson, above n12.
 See the discussion in Tasioulas, above n20.
 See the discussion of Buchanan, above n1 in Tasioulas, above n20.
 See, for example Kumm, above n12, 909–17.
 Of course, sharing the same concept of legitimacy does not exclude providing different justifications of the legitimacy of national and of international law, as we will see. See Tasioulas, above n20.
 One may give different examples, such as international criminal law or EU law. Of course, international law differs from EU law in that it is not an integrated legal order whose subjects are always both states and individuals. See Samantha Besson, ‘How International is the European Legal Order? Retracing Tuori’s Steps in the Exploration of European Legal Pluralism’ (2008) 4 No Foundations Journal of Extreme Legal Positivism <http://www.helsinki.fi/nofo/> at 29 June 2009, on integrated legal orders.
 The reference to ‘citizens’ covers more than those individuals actively taking part in the political life of the state, and is a shorthand for all individuals residing in a state and whose fundamental interests are affected by the decisions taken in that state.
 See Kumm, above n12, 910; Murphy, above n16.
 See Raz 1986, above n2; Raz 1995, above n2; Raz 2006, above n2.
 See also Tasioulas, above n20.
 See Raz 1986, above n2, 35 et seq; Raz 2006, above n2, 1012–20.
 See Raz 1986, above n2, 57 et seq; Raz 2006, above n2, 1012–20.
 See Raz 1986, above n2, 42 et seq; Raz 2006, above n2, 1012–20.
 See Raz 1986, above n2, 53; Raz 2006, above n2, 1014.
 See Raz 2006, above n2, 1012 et seq.
 See James Griffin, On Human Rights (2008) 33.
 See Raz 1986, above n2, 75 et seq.
 See Raz 1986, above n2, 88 et seq; Raz 2006, above n2, 1028–9, 1037–40. See also Raz 1995, above n2, 80–94.
 For a general discussion of the shortcomings of the consent-based justification of political authority, see Alan John Simmons, Moral Principles and Political Obligations (1979).
 See Raz 1986, above n2, 89; Raz 1995, above n2, 355–69. See the discussion in Scott Hershovitz, ‘Legitimacy, Democracy and Razian Authority’ (2003) 9 Legal Theory 201, 215.
 See Raz 1986, above n2, 90, 93; Raz 1995, above n2, 368–9.
 See, for co-ordination-based accounts of legal authority founded on Razian authority, Jeremy Waldron, ‘Authority for Officials’ in Lukas Meyer, Stanley Paulson and Thomas Pogge (eds), Rights, Culture and the Law: Themes from the legal and political philosophy of Joseph Raz (2003) 45; Samantha Besson, The Morality of Conflict (2005) 161 et seq, 459 et seq and 503 et seq.
 According to Waldron, above n56, 49, a question is of common concern among a group of people if it is better for a single answer to be accepted among them than for each person to deal with the question on their own, the best they can.
 See Besson, above n56, 164 et seq, 465 et seq; Waldron, above n16, 14–15; Jeremy Waldron, Law and Disagreement (1999) 101–13. See also John Finnis, ‘The Authority of Law in the Predicament of Contemporary Social Theory’ (1984) 1 Notre Dame Journal of Law, Ethics and Public Policy 115.
 See Besson, above n56, 459, 503; Waldron, above n58, 101–13.
 On the difference between consent and co-ordination, see Besson, above n56, 473–5; Waldron, above n16, 25–7. Consent can enhance co-ordination, but is not necessary for co-ordination to take place.
 See, for example Raz 2006, above n2, 1031–2.
 See Waldron, above n58, 101–13; Besson, above n56, 459 et seq on democratic co-ordinative authority.
 See, for example Hershovitz’s critique of Raz’s account of legal authority, above n54, 209–10.
 See Raz 2006, above n2, 1031 fn 20, 1037–40 for an insufficiently charitable reading of the democratic conception of authority, a reading that fails to accommodate the circumstances of pervasive and persistent reasonable disagreement about issues of justice and common moral concern, and the need to address that disagreement in current political conditions.
 In contrast to what is often said (see, for example Tasioulas, above n20) and presumably derived from a skewed idea of participatory practices in a democracy, individual consent ought not therefore be conflated with democracy as a justification of authority. See Hershovitz, above n54, 215.
 Waldron, above n56, 67–9.
 See Raz 2006, above n2, 1031.
 See Waldron, above n56, 66 on the contrast between purely individual and co-ordination-based individual reasons to obey the directives of a public authority.
 See Waldron, above n56; Besson, above n56, 490–8; Samantha Besson, ‘Review Article: Democracy, law and authority’ (2005) 2 Journal of Moral Philosophy 89.
 See the replies by Raz 2003, above n2; Raz 2006, above n2, 1040–4. See Besson, above n69.
 See Waldron, above n56, 59–61. See also Buchanan and Keohane, above n23, 408 for a similar concern in international law.
 See Raz 2003, above n2. See also Raz 2006, above n2, 1025 et seq.
 See Waldron, above n56, 61–3. See also Besson, above n56, 497–8.
 Raz 2003, above n2, 260.
 See Raz 1986, above n2, 75–6.
 Waldron, above n56, 66.
 See, for example, Hershovitz, above n54, 216 et seq.
 See Besson, above n56, 496–8; Waldron, above n58, 101–13. This view contrasts with that of Raz 1995, above n2, 347. This is particularly important in the context of international law as its moral authority is often the mere reflection of its underlying moral values (see discussion above, n16).
 See Waldron, above n56, 66. Even though it is a different sort of reason, it affects the way the NJC is satisfied individually – in conditions of reasonable disagreement over matters of common concern, the existence of a co-ordinative authority around which people know they might be able to co-ordinate necessarily affects the ways in which the NJC is satisfied individually.
 See Raz 1995, above n2, 117 on the NJC and the epistemic qualities of democratic authority.
 This view differs from that of Hershovitz, above n54, 212 et seq.
 Tasioulas, above n20. Tasioulas also argues that procedural requirements in the rule of law (such as publicity or transparency) are ‘certainly relevant to the fulfilment of the NJC’, without elaborating further.
 On this argument, see Besson, above n56, 505–6; Waldron, above n58, 102 et seq.
 See Tasioulas, above n20. Contrast with Buchanan, above n1.
 See, for example Raz 1986, above n2, 70.
 See, for example Hershovitz, above n54. See, however Raz 2006, above n2, 1031–2.
 See, for example Waldron, above n56. See, however Raz 2006, above n2, 1031–2.
 See Alan Boyle and Christine Chinkin, The Making of International Law (2007); Antonio Cassese and Joseph Weiler (eds), Change and Stability in International Law-Making (1988); Vaughan Lowe, ‘The Politics of Law-Making: Are the method and character of norm creation changing?’ in Michael Byers (ed), The Role of Law in International Politics (2000) 207–26.
 See, for example José Alvarez, International Organizations as Law-makers (2006); Robert McCorquodale, ‘An Inclusive International Legal System’ (2004) 17(3) Leiden Journal of International Law 477.
 On the difference, see Abram Chayes, ‘A Common Lawyer looks at International Law’ (1965) 78 Harvard Law Review 1396, 1410.
 This hiatus, and the matching traditional conception of international law, is presumably what lies behind certain authors’ focus on states, as opposed to individuals, as the main or only subjects of the authority of international law.
 See, for example Buchanan, above n1 who discusses ‘international law-making institutions’; Lefkowitz, above n30 who discusses ‘international actors’.
 See, for example Franck 1988, above n8; Franck, above n17; Franck 2006, above n8; Tasioulas, above n20. Of course, this may be explained by the use of international human rights law as the main example, where the law binds primarily states towards other states and (directly or indirectly) individuals, and only very rarely binds individuals directly.
 See, for example Kumm, above n12.
 See, for example Buchanan and Keohane, above n23.
 See, for example Buchanan, above n8; Buchanan, above n1 (although he mostly addresses the case of states’ duties to obey international law); Waldron, above n18 (although he focuses on the rule of law and not the authority of law).
 I owe this expression to Murphy, above n16. Hart, above n6, 231–2 already hinted at the issue, although he did not address it: ‘Precisely whose motives, thoughts and feelings on such matters of moral conviction are to be attributed to the state is a question which need not detain us here.’
 See Murphy, above n16. See also, albeit in the domestic context, Raz 1986, above n2, 72.
 Buchanan, above n1. See, however, Raz 2006, above n2, 1004 on the possibility of illegitimate institutions producing legitimate laws.
 See Waldron, above n18, 23–5.
 Ibid 23 et seq.
 See Besson, above n12.
 This is why participation in treaty-making and the use of consent therein does not necessarily imply that consent is the source of the authority of international law. This view contrasts with that of Kumm, above n12, 914. Of course, consent may still be used as one of the criteria of legal validity, but that is a different matter.
 See, for example Besson, above n12.
 See Chayes, above n90, 1410.
 See Waldron, above n18, 15; Raz 1979, above n2, 212–19; Rawls, above n16, 236–9.
 See section 2b, above. See Waldron, above n56, 67–9.
 See also Tasioulas, above n20.
 See Besson, above n12.
 See Tasioulas, above n20.
 This view differs from that of Timothy Endicott, ‘The Logic of Freedom and Power’, in Samantha Besson and John Tasioulas (eds), The Philosophy of International Law (2010, forthcoming); and Tasioulas, above n20.
 See Waldron, above n18, 21.
 Besides the direct burdens imposed by new international legal norms, state responsibility in cases of violation of those morally binding norms will trigger further (moral and not only legal) duties to cease the violation and to remedy it, thus shifting new burdens onto individuals. The question of authority (and primary obligation) is more sensitive in terms of burden-imposition, however, than that of international responsibility for a wrongful act (and secondary obligation), where the representation of its constituent individuals and their duties by the state can account for shifting part of the burden of reparation onto those individuals: see the discussion in James Crawford and Jeremy Watkins, ‘International Responsibility’ in Samantha Besson and John Tasioulas (eds), The Philosophy of International Law (2010, forthcoming); and in Murphy, above n16.
 On the difficulties this raises in the context of loans contracted out through treaties by corrupt governments, see Murphy, above n16. See also Pogge, above n16 on the ‘international borrowing privilege’.
 See Raz 1986, above n2, 87–8. See also section 3c, below.
 See Buchanan, above n1, 53 on the ‘Vanishing Subject Matter Problem’.
 Of course, this does not exclude correctives within the legal system itself from protecting states against themselves and in particular against corrupt governments.
 One may argue that the inclusion of individuals in democratic decision-making at the international level (see section 3b, below) may compensate for the lack of representation of states participating in those processes. This view underestimates, however, the importance of the role of democratic political communities as full members of the international political community besides individuals: see Samantha Besson, ‘Ubi Ius, Ibi Civitas: A republican account of the international community’ in Samantha Besson and José Luis Martí (eds), Legal Republicanism: National and international perspectives (2009) 205, 217–19; Jean Cohen, ‘Rethinking Human Rights, Democracy and Sovereignty in the Age of Globalization’ (2008) 36(4) Political Theory 578.
 For instance, the obligation of states to prevent genocide is different in its content from that of individuals. Compare the International Court of Justice’s decision in The Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia and Herzegovina v Serbia and Montenegro) (Judgment)  ICJ Rep 91 with that of the International Criminal Tribunal for the former Yougoslavia, ICTY, Appeals Chamber, Prosecutor v Dusko Tadić, 15 July 1999 (Case no. IT-94-1-A).
 See Besson, above n56, 197–203.
 See John Tasioulas, ‘Customary International Law and the Quest for Global Justice’ in Amanda Perreau-Saussine and James Murphy (eds), The Nature of Customary Law: Legal, historical and philosophical perspectives (2006) 307, 320–4.
 See Besson, above n12 on a co-ordination-based account of international customary law and the difference between the co-ordinative rules of creation and change (secondary rules), and actual customary rules. See also section 2c, above.
 See Besson, above n56, 192–5; Waldron, above n58, 105–6.
 See Waldron’s, above n56 critique of Raz’s account of legal authority at the domestic level.
 See Raz 2006, above n2, 1020–1, 1023–5. Due to the constraints of space, the section of the lecture pertaining to conflicts of claims to primacy and the competing authority of norms stemming from different legal regimes of international law and between legal orders in conditions of (internal and external) legal pluralism was excised. See, however, the discussion in Samantha Besson, ‘Whose Constitution(s)? International Law, Democracy and Constitutionalism’ in Jeff Dunoff and Joel Trachtman (eds), Ruling the World: Constitutionalism, international law and global governance (2009) 381.
 This view differs from that of Tasioulas, above n20 who, like Raz 2006, above n2, fn 20, concentrates only on classic co-ordination problems.
 Tasioulas, above n20 describes this move (which he understands as an attempt to provide an exclusive justification of the authority of international law) as a ‘serious overreaction’, without clearly arguing against that development.
 See Hart, above n6, 219–20; Tasioulas, above n20. Contrast with Buchanan, above n1.
 See Buchanan, above n1.
 See Besson, above n118, 213 et seq. In this sense, I agree with Buchanan and Keohane, above n23 and Buchanan, above n1.
 See Samantha Besson, ‘Deliberative Demoi-cracy in the European Union: Towards the deterritorialization of democracy’ in Samantha Besson and José Luis Martí (eds), Deliberative Democracy and its Discontents (2006) 181; Samantha Besson, ‘Institutionalizing Global Demoi-cracy’ in Lukas Meyer (ed), Justice, Legitimacy and Public International Law (2009) 58.
 See section 3a, above.
 This is when the identification of the law-makers and legal subjects at stake in the argument matters.
 See Besson, above n118, 216; Robert McCorquodale, ‘International Community and State Sovereignty: An uneasy symbiotic relationship’ in Colin Warbrick and Steven Tierney (eds), Towards an International Legal Community (2006) 241.
 Daniele Archibugi, ‘The Reform of the UN and Cosmopolitan Democracy: A critical review’ (1993) 30(3) Journal of Peace Research 301.
 See Thomas Christiano, ‘International Institutions and Democracy’ in Samantha Besson and John Tasioulas (eds), The Philosophy of International Law (2010, forthcoming).
 Buchanan, above n1. See also Buchanan and Keohane, above n23.
 See Besson, above n56, 319–323.
 See, for example Teson 2005, above n23; Teson 1998, above n23; Robert Goodin, ‘Enfranchising All Affected Interests, and Its Alternatives’ (2007) 35(1) Philosophy and Public Affairs 40.
 See Besson, above n12; Allen Buchanan, ‘Human Rights and the Legitimacy of the International Order’ (2008) 14(1) Legal Theory 39; Buchanan, above n1; Buchanan and Keohane, above n23, 421–2. See also Cohen, above n118 on the human right to have rights.
 See Tasioulas, above n20 for a detailed discussion of those justifications in the context of international law.
 See Tasioulas, above n20, although my democratic co-ordination-based justification of legal authority might cover more ground overall than his different justifications.
 See Raz 2006, above n2, 1028–9.
 See Tasioulas, above n20. See also Raz 1995, above n2, 360 et seq; Raz 2006, above n2, 1037 et seq.
 See Raz 1995, above n2, 355–69. See also Hart, above n6, 224 et seq.
 On the limitations of democratic state consent, see Buchanan, above n8; Besson, above n118, 214–17. On the relationship between democracy and consent in general, see the discussion above n65.
 See Buchanan, above n8, 301–14; Buchanan, above n1.
 See Raz 1986, above n2, 87–8.
 See Lefkowitz, above n30.
 Raz 1995, above n2, 365–6; Raz 2006, above n2, 1014.
 Raz 2006, above n2, 1014.
 Tasioulas, above n20. See, however, the discussion in Raz 2006, above n2, 1015 et seq.
 See Waldron, above n18, 21–2.
 See Murphy, above n16.
 See Endicott, above n111.
 Hart, above n6, 223.
 This view contrasts with that of Kumm, above n12, 920 et seq; Mattias Kumm, ‘International Law Meets Domestic Law: Terms of engagement’ in Sujit Choudhry (ed), The Migration of Constitutional Ideas (2007) 256. Subsidiarity cannot be a ground of justification of the authority of international law. Only once that authority is justified, can subsidiarity become the principle that will demarcate conflicting claims to authority made by international and national law.
 This is a missing link in the argument by Tasioulas, above n20 who rightly distinguishes between the argument of pluralism and the argument of freedom in the human rights context, but fails to see that the key lies in the unicity of subjects between the national and the international legal orders and in the relationship between legitimate authorities in both orders.
 See Besson, above n118, 229–36.
 See in other words, Tasioulas, above n20.
 See, for example Harold Koh, ‘On American Exceptionalism’ (2003) 55 Stanford Law Review 1479.
 I owe this distinction to Tasioulas, above n20.
 See Besson, above n56, 52 et seq.
 See Bernard Williams, ‘Human Rights and Relativism’ in In the Beginning was the Deed: Realism and moralism in political argument (2005) 62, 66.
 See Buchanan, above n141; Buchanan, above n1.
 See, for example Tasioulas, above n20 for a detailed rebuttal of parochialism in a moral pluralist context, and especially in the human rights context.
 See section 2c, above.
 See John Tasioulas, ‘The Moral Reality of Human Rights’ in Thomas Pogge (ed), Freedom from Poverty as a Human Right: Who owes what to the very poor? (2007) 75; Tasioulas, above n20. See also for a modified interest-based account of European human rights and of their social contextualisation, Samantha Besson, ‘The European Union and Human Rights: Towards a new kind of post-national human rights institution’ (2006) 6(2) Human Rights Law Review 323.
 See section 2c, above.
 See Tasioulas’, above n20 response to the exceptionalism objection.
 On the distinction, see Besson, above n56, 503–4.
 See, for example Besson, above n56, 503 et seq.
 See, for example Allen Buchanan, ‘From Nuremberg to Kosovo: The morality of illegal international law’ (2001) 111(4) Ethics 673, 680; Robert Goodin, ‘Toward an International Rule of Law: Distinguishing international law-breakers from would-be law-makers’ (2005) 9 The Journal of Ethics 225.
 See Waldron, above n18, 22.
 See Jeremy Waldron, ‘The Concept and the Rule of Law’ (2009) 43(1) Georgia Law Review 1.
 See, for example Arthur Watts, ‘The International Rule of Law’ (1993) 36 German Yearbook of International Law 15; Ruti Teitel, ‘Humanity’s Law: Rule of law for the new global politics’ (2002) 35 Cornell International Law Journal 355.
 See Besson, above n12.
 Julius Stone, Human Law and Human Justice (1965).