Despite the many calls for bridge building between the fields of International Law and International Relations, genuinely integrative studies are few and far between. Lawyers leaven their writings with a dash of real politic here and utility maximizing there; International Relations scholars enlist the authority of legal interpretation and harvest insights into legal reasoning. But these are seldom exercises in genuine dialog, aimed at producing new theoretical perspectives, views that are more than the sum of their parts, which promise to advance understanding in both fields. Legitimacy and Legality in International Law is refreshing in this regard. Brunnee and Toope mine two complimentary strands of international legal and international relations theory to generate an ‘interactional’ theory of international law. They dig deep enough to grasp the complexities of each strand, and produce an artfully integrated amalgam of Lon Fuller's approach to law (transplanted into the international arena) and constructivist international relations.
The resulting interactional theory has much to commend it. First, it makes a decisive break with command theories of law, which reduce law to the edicts of sovereign authorities backed by sanctions. Brunnée and Toope shift the focus away from the origins of law – its authoritative sources – to its qualitative characteristics. ‘[W]hat distinguishes legal norms from other types of social norms’, they argue, ‘is not form or pedigree, but adherence to specific criteria of legality’ (Brunnée and Toope Reference Brunnée and Toope2011, 307). Whatever the limitations of this approach, it gives us another way of thinking sensibly about how law might function at the international level, as legal norms, from this perspective, can emerge in decentralized ‘communities of practice’ lacking central authority. Second, it treats international law as a practice grounded in social understandings. All too often it is treated as a reified institution, a set of codified norms, ‘legislated’ by states, open to more or less authoritative interpretation. For Brunnée and Toope, international law is inextricably embedded within the wider social order, its meanings are generated by context-specific discourses, it is reproduced and transformed by everyday social practices, and, most importantly, it ‘lives’ in world comprising not merely states but peoples, NGOs (non-governmental organizations), corporations, international organizations, etc. This has the great virtue of giving center stage to issues of legitimacy and reciprocity, and accommodates curious features of the contemporary politics of international law. It helps us understand, for example, why the international legitimacy of the 2004 Iraq War was shaped as much, if not more, by the ‘folk’ legal interpretations of global protesters (‘No war without a Security Council mandate’) than the learned opinions of government lawyers and legal publicists.
In the following pages, I engage one of the central features of Brunnée and Toope's interactional theory – their account of international legal obligation. Understanding such obligation is essential if we wish to comprehend why states or other actors observe or fail to observe international law, and Brunnée and Toope place it right at the heart of their theory, arguing ‘that law's distinctiveness rests in the concept and operation in practice of legal obligation’ (Brunnée and Toope Reference Brunnée and Toope2011, 307). Existing theories attribute such obligation to a variety of factors, including the existence of sanctions, the act of consent, the legitimacy and fairness of rules, and processes of deliberation. Brunnée and Toope take a different path, however. Obligation is generated through participation in particular kinds of practices, practices that instantiate social norms that meet certain ‘criteria of legality’. When states and other actors interact with one another in the performance of such practices they develop a certain ‘fidelity’ to the law, they come to see its precepts and processes as legitimate.
There is little doubt that certain kinds of interactions can generate feelings of obligation, and in the modern international system, it may well be the case that states that interact through practices that meet the criteria of legality have heightened senses of legal obligation. It is not clear to me, however, that this simple, if important, observation amounts to more than a partial theory of international legal obligation, even in the elaborated form presented by Brunnée and Toope. Elsewhere, I argue that existing theories of international legal obligation suffer from the problem of interiority; they attribute obligations to a particular aspect of a given legal order, but they lack the theoretical resources to account for that aspect or the wider order of which it is part (Reus-Smit Reference Reus-Smit2003, 593). While distancing themselves from such accounts, Brunnée and Toope inadvertently fall into the same trap. Obligations are said to emerge from practices that meet the criteria of legality, yet these practices not only go unexplained, they are naturalized, treated as though they are the only kind of legal practices states might reasonably adopt. This comes slam up against history, though. The practices to which they refer – those that meet Fuller's criteria of legality – are modern artifacts, products of a 19th century revolution. Prior to this a different set of international legal practices were privileged, instantiating different criteria of legality. Brunnée and Toope insist that legal practices are framed and informed by wider social understandings, but their reading of these is decidedly thin.
The problem of legal obligation
When actors observe legal rules out of a sense of obligation, they do so because they consider the rules in question, or the legal system of which they are a part, to be legitimate. In other words, they observe the law because it is ‘right’ to do so. Narrow self-interests might lead them in the same direction, but the thing about obligation is that it counsels rule observance even when this conflicts with such interests. Indeed, arguably this is the social function of obligations. But while obligation can encourage rule observance, obligation and compliance are not synonyms. Obligation is one source of rule observance, compliance is the fact of such observance. This having been said, obligation is a crucial source. Scholars disagree about how much compliance can be attributed to feelings of legal obligation, particularly in international relations. We know two things, though: that obligation is a comparatively cheap source of compliance (requiring neither the maintenance of a regime of sanctions or the ongoing satisfaction of vagarious self-interests), and that legal orders characterized by high levels of obligation will, in all likelihood, display higher levels of rule compliance. Understanding the roots of legal obligation is of some importance, therefore, especially in a decentralized international system lacking systematic or effective sanctions.
Yet such understanding has proven allusive. For legal positivists, actors are obliged to obey the law because they will suffer sanctions if they do not. And because such sanctions are underdeveloped at the international level, realists hold that international legal obligation is either weak or non-existent. But as many scholars have observed, this puts the cart before the horse. Law is not obligatory because it is enforced; it is enforced because it is obligatory (Fitzmaurice Reference Fitzmaurice1956, 2). For rationalists, consent, not sanctions, is the source of legal obligation – actors incur legal obligations when they expressly, or tacitly, agree. Not surprisingly, this is the default understanding of most students of international relations, resonating as it does with the prevailing view of states as self-interested, atomistic actors operating in a world without central authority. It is vulnerable to a devastating critique, however, articulated most famously by H.L.A. Hart. Acts of consent are only obligating because there exists a prior norm that mandates that promises to observe legal rules are binding, and because this norm gives consent its normative standing, consent itself cannot be the source of that prior norm's obligatory force (Hart Reference Hart1994, 225). The roots of legal obligation must lie elsewhere. For Thomas Franck, they lie in the perceived ‘fairness’ of a given legal order. This is partly dependent on how the order, and the rules it generates, distributes benefits and burdens. It is mostly determined, however, by two other things: the ‘intrinsic properties’ of the rules themselves (e.g. whether ‘they treat likes cases alike and are uniform in application’), and the wider procedural legitimacy of the system. The problem is, however, that this latter legitimacy rests on the sanctity of the system's operating principles: ‘(1) that states are sovereign and equal; (2) that their sovereignty can only be restricted by consent; (3) that consent binds; and (4) that states, joining the international community, are bound by the rules of that community’ (Franck Reference Franck1995, 29). We return, therefore, to consent as the root of international legal obligation, with all its attendant limitations.
Practices make perfect
In casting their interactional theory of international law as a theory of legal obligation, Brunnée and Toope promise to resolve this central problematic. Their argument is complex and multilayered, and merits unpacking. As noted in their introduction to this symposium, ‘shared understandings’, ‘criteria of legality’, and a ‘practice of legality’ ‘are crucial to generating distinctive legal legitimacy and a sense of commitment among those to whom law is addressed’ (Brunnée and Toope Reference Brunnée and Toope2011, 308). Let me address each of these elements, though in reverse order.
For Brunnée and Toope, interactional international law is not a set of treaties, nascent judicial bodies, or underdeveloped enforcement mechanisms – it is a ‘community of practice’ (Brunnée and Toope Reference Brunnée and Toope2010, 27). Legal obligation is an ‘internalized commitment’, a ‘feeling’ actors have about the legitimacy of a legal order and its attendant rules (Brunnée and Toope Reference Brunnée and Toope2010, 45). These feelings are not internally generated, however; they are socially constructed. But instead of adopting a simple norm diffusion argument – in which ‘external’ norms are ‘internalized’ by actors, giving form to their identities and interests – Brunnée and Toope see complex processes of structuration at work. Social norms do indeed condition actors’ beliefs about things like legal obligation, but these norms only exist and are reproduced because actors construct and enact them. Practices thus mediate the relation between agents and social structures, constituting both agential identities and interests and intersubjective systems of meaning. The implications for a theory of legal obligation are clear – only through social interaction, and the participation in shared practices, can actors develop an ‘internal commitment’ to observe the law. Fostering a community of practice that can engender such feelings is thus essential to the viability of any legal order, not the least the international.
Brunnée and Toope insist, however, that this must be a particular kind of community of practice, one in which prevailing practices instantiate specific ‘criteria of legality’. Practices deemed ‘legal’ must be general, officially promulgated, prospective, clear, non-contradictory, realistic, constant, and congruent (the actions of officials must be consistent with legal norms) (Brunnée and Toope Reference Brunnée and Toope2010, 26). Two things are noteworthy here. First, it is these criteria that distinguish legal from social norms – the former satisfy them, the latter do not. Second, it is by enacting practices that meet the criteria of legality that actors develop feelings of legal obligation. Indeed, ‘[o]nly when the conditions of legality are met, and embraced by a community of practice, can we imagine agents feeling obliged to shape their behavior in the light of the promulgated rules’ (Brunnée and Toope Reference Brunnée and Toope2010, 41). Legal obligation is thus the product of legal practice, and because of this interactional law can be said to have its own ‘internal morality’ (Brunnée and Toope Reference Brunnée and Toope2010, 30).
To this point, it appears as though the realm of law – that realm comprising practices that meet the criteria of legality – is a discrete field of human experience and practice, the internal characteristics of which generate legal obligation. Yet Brunnée and Toope are clear that the development of such a realm is dependent upon the existence of wider ‘shared understandings’. These operate at three ‘layers’ in their account. The broadest are those social understandings that ‘are relevant to law's intelligibility and to perceptions of reasonableness’ (Brunnée and Toope Reference Brunnée and Toope2010, 68). Although underspecified by Brunnee and Toope, it is these understandings that are said to ‘determine when and how much law is possible at any given time’ (Brunnée and Toope Reference Brunnée and Toope2010, 68–69). The second layer consists of shared understandings about legality itself. This seems to be where understandings about the criteria of legality exist, as ‘participants in a legal system must build and maintain a practice guided by the requirements of legality’ (Brunnée and Toope Reference Brunnée and Toope2010, 69). The final layer consists of substantive understandings about things like human rights, environmental protection, etc. While an interactional legal system cannot exist without a background web of shared understandings, Brunnée and Toope argue that this need not be very thick, and that a community of legal practice can evolve where only the most basic first and second layer understandings exist. This has obvious implications for the development of international law, where shared substantive understandings between states might be scarce or non-existent.
Naturalization trumps understandings
The idea that legal practices are embedded within, and constituted by, layers of nested social understandings is a significant step toward overcoming the limitations of existing approaches to international legal obligation, all of which attribute obligation to an internal feature of a legal system but lack the resources to explain that feature's existence or normativity. A focus on historically contingent understandings promises insights into why these characteristics of a legal system emerge and why they have normative veracity for historically and culturally located actors. Yet Brunnée and Toope fail to grasp either the potential or significance of this idea, underspecifying the shared understandings they categorize, and, in the end, nullifying them by naturalizing their ‘criteria of legality’.
As legal practices have varied from one historical system of states to another, any account of how shared understandings condition legal practices needs to clearly specify those understandings. But in Brunnée and Toope's account these are decidedly opaque. Of their three layers of understandings, the first and second can be expected to do most constitutive work – the first affecting law's ‘intelligibility’ and perceived ‘reasonableness’, the second cognizance of ‘legality’ itself. Brunnée and Toope place a considerable burden on readers to work out precisely what these categories are and what differentiates them. It appears, though, that the former consists of the shared knowledge and values that enable certain practices to be recognized and deemed socially acceptable. The second seems to comprise understandings of the internal norms of the legal system itself, norms cataloged in Brunnée and Toope's ‘criteria of legality’. Whether or not this is an accurate rendering of these categories, a second problem concerns what goes in them. Brunnée and Toope do give examples, but it is never clear where these really fit in the overall schema. In the discussion surrounding their articulation of the three categories, they talk about how social values about human rights to life and dignity affect the capacity of governments to legislate on the death penalty (Brunnée and Toope Reference Brunnée and Toope2010, 66–69). But these values fit most readily in their third layer of substantive understandings, not in either of the two deeper layers. At another point, they argue that ‘legal interactions have been built upon a shared understanding of state sovereignty and around the customary principles of sovereign equality and non-intervention’ (Brunnée and Toope Reference Brunnée and Toope2010, 71). One can only assume that these belong in the first, deepest, layer of understandings, as they are clearly not about ‘legality’ per se. Yet this is highly problematic. To begin with, in a variety of historical contexts states have shared understandings about sovereignty, but the meanings contained within these have varied considerably. For instance, only in the last century have understandings of sovereignty been conjoined with customary principles of sovereign equality and non-intervention. Furthermore, understandings of sovereignty have historically coincided with very different international legal orders – the ‘naturalist’ order that prevailed until the early 19th century was, for example, very different to the ‘positivist’ order that evolved thereafter. Most importantly, the criteria of legality emphasized by Brunnée and Toope became ‘intelligible’ only in this latter era (Absolutist legal norms of hierarchy, particularism, and non-reciprocity sit uneasily with these criteria).
In the end, though, much of this is moot. Brunnée and Toope argue that legal practices are grounded in deeper social understandings, yet time and again they naturalize practices that instantiate their favored criteria of legality. Nowhere is this clearer than in their discussion of the minimal conditions needed for interactional law to evolve between states. States need not share common purposes or values, all that is needed is ‘very limited shared understandings that there is a need for law in shaping international communication and interaction’ (Brunnée and Toope Reference Brunnée and Toope2010, 81). But what is it that states imagine ‘law’ to be? The answer implied by Brunnée and Toope is that it is a set of practices that embody their criteria of legality. This implication is reinforced by their claim that these criteria ‘are largely uncontroversial’ (Brunnée and Toope 2011, 311). It is also evident in how they deal with the obvious liberal roots of their criteria. They readily admit these roots, acknowledging that their conception is ‘grounded in a western liberal tradition that upholds legal rationality’ (Brunnée and Toope Reference Brunnée and Toope2010, 81). At the same time, however, they try to limit the significance of this, arguing that the only commitment they take from liberalism is ‘to a symbolic understanding of autonomy and communication’ (Brunnée and Toope Reference Brunnée and Toope2010, 81). These are not, however, taken for what they are – values specific to modernity. Instead, they are treated as universals. Autonomy, they write, ‘is a cognate for diverse needs and aspirations but is inextricably bound to the human need to communicate’ (Brunnée and Toope Reference Brunnée and Toope2010, 81). When it all comes down to it, therefore, Brunnée and Toope end up in a position not dissimilar to that of English School pluralists – international society is a practical association bound together by interests in coexistence not substantive values, and international law emerges to meet these practical interests. The problem is, however, that they, like their pluralist counterparts, struggle to explain why states have adopted different legal practices in different historical contexts (Reus-Smit Reference Reus-Smit1999). More specifically, they struggle to explain why states might attach normative value to practices that instantiate their criteria of legality; an important step, one would assume, in understanding why they might feel obliged to enact such practices and observe the rules they generate.
One might respond, of course, that even if Brunnée and Toope have misspecified and naturalized their criteria of legality, the shared understandings that undergird the practices that generate feelings of legal obligation must include something like criteria of legality. If not, historical actors would have no way of recognizing their practices as ‘legal’, and scholars would have no way of distinguishing the international legal practices of different historical epochs, however much they might have varied. These are difficult issues, conceptually and analytically. My own, very tentative, responses are twofold. First, I am not denying that some kind of criteria of legality form part the shared understandings that inform obligation generating legal practices. Rather, my position is that these are always historically and contextually contingent – there are no historically transcendant, universal criteria of legality on which a general theory of international legal obligation can be constructed. Second, if this is true, then there are no generalizable criteria of legality that we can use to identify what constitutes legal practices in different historical (or indeed cultural) contexts. This leaves, of course, the difficult question of what we should use to identify such practices. There is no scope here to address this quesiton in any detail, but it is difficult to escape the conclusion that the ‘local’ understandings of historical actors must play a key role in the identification of ‘their law’.
Obligations of choice, obligations of practice
By way of conclusion, let me distinguish between two kinds of international legal obligations. The first are those emphasized by Brunnée and Toope, the ones that derive from the shared, routinized enactment of a given set of legal practices – obligations generated by ‘doing’. The second are those derived from the valuing of an idea; more specifically, from valuing a particular idea of law and its attendant form of rules and practices – obligations grounded in ‘choosing’. Such valuing has been most explicit at great moments of international institutional change. For instance, the Hague Conferences of 1899 and 1907 were important not just because of their contribution to the development of modern laws of war, but because participants explicitly sought to transpose liberal ideals of domestic law into the international arena. In well-developed legal orders, these two kinds of obligations are likely to be deeply entwined; partly because constitutive ideas recede into the background and are most apparent in the practices they license, and partly because, consciously or unconsciously, as actors perform certain practices they internalize ‘folk’ versions of these constitutive ideas. Yet two facts recommend that these forms of legal obligation be treated as relatively autonomous and, as such, analytically distinct. First, as the Hague example demonstrates, at key moments in international history states have made normative choices in favor of particular kinds of legal practices, and these normative choices have conditioned the subsequent development of distinctive realms of practice. Second, in different historical contexts, states have made different normative choices, licensing different kinds of legal orders. A holistic theory of international legal obligation thus needs to grapple with both types of obligation, their respective sources, and their complex interrelation. Brunnée and Toope take us some way along this path, but the full potential of their account is undermined by their rudimentary understanding of the social understandings that constitute international legal practices, and by their tendency to treat Fuller's criteria of legality as though they were the criteria of legality.
