History

We accept the critique of Reus-Smit that the theory we propose is historically, and therefore conceptually, time bound. We do not contend that our understanding of legal obligation, of what makes it possible, is true for all time. In fact, we are quite clear that our goal is to explain the contemporary practice of international law. This modesty of aspiration should not be surprising. The state today, which remains a dominating force in the creation, maintenance, and destruction of international law, is not the same as the pre-modern state. Conceptions of citizen engagement, as well as the roles of civil society, and of individuals and corporations, have changed dramatically over the last century.

If law is embedded in underlying shared social understandings and is continuously contingent on a practice of legality, as we argue that it is, then any claim of temporal universality would be nonsensical. Social understandings do change over time, and we accept that these social understandings have an influence upon the criteria of legality that make law possible. We describe the criteria that Fuller proposed, and that we adopt, as ‘largely uncontroversial’ in the context of an analysis that is grounded in the shared understandings of the early twenty-first century, although we will see in a moment that context does not dictate all conclusions, even if it strongly shapes them. Furthermore, we readily admit that our criteria of legality are connected to a set of weak liberal commitments to relative autonomy balanced against the fundamental human need to communicate.

Our point is that although all theorizing (perhaps especially jurisprudential theorizing) is at all times historically contingent, and here we agree with Dunoff, we attempt to limit the necessary points of agreement on shared ends to a bare minimum. This is why we do not spend much effort attempting to specify the underlying shared understandings that support the basic self-ordering of a society; they are not much stronger than a commitment to seeing law as necessary to shape certain types of social interaction. At the next level identified by Reus-Smit, the required shared understandings are indeed more specific and historically contingent. We entirely agree that the conceptions of sovereign equality and non-interference in internal affairs that we see in the UN Charter, and which definitely shaped the interactions of states in the latter part of the twentieth century, are not parallel to the imperial understandings of statehood that marked the nineteenth century. Again, all this tells us is that ideas, similar to all social understandings, change over time. So does law, and even the conception of law.

Dunoff's concern about the role of history in our interactional theory of international law is pragmatic. He gently suggests that because legal theory is inevitably both descriptive and normative, a view to which we gladly subscribe, the choices of emphasis that we make in our interactional account may be misdirected at issues that are not truly central to the evolution of contemporary international law.

As Dunoff himself notes, choices of case studies are inevitably both selective and incomplete. We could indeed have chosen to focus on the problem of Iranian nuclear capabilities or of North Korean sabre rattling. Our choice of specific topics was not, in fact, predicated on a desire to focus on the role of the temporarily ‘single’ superpower. Instead, we decided to look at three issues that had played defining roles in the evolution of international law in the twentieth century, and remained highly salient in the twenty-first: controlling the use of force (core to the aims of the UN Charter and central in the evaluation of power relations in international society); human rights, through the vehicle of the anti-torture norm (because of the enormous influence of human rights discourse and the complexities of cultural difference); and climate change (one of the great challenges of our time, and a classic problem of the global commons). It is true that these were also issues over which American exceptionalism asserted itself for a time. But a central point of Fuller's argument, with which we agree, is that it is a mistake to see law as purely or even primarily instrumental. This is why the ‘criteria of legality’ are so important: they create just the sort of internal critique of law that Koskenniemi describes; they impose requirements that may actually make it hard to pursue some ends.

It is of course true that attempts are often made to use law to bolster ‘political’ preferences, but the criteria of legality tend to de-instrumentalize law by requiring conditions that make political exploitation harder. Our chapter on the anti-torture norm reaffirms this point in stark terms. Although our personal political commitments would strongly drive us to reaffirm the centrality of the Convention against Torture, with its absolute prohibition, our interactional analysis forced us to conclude that the rule against torture was weakly supported by substantive shared understandings, and that it may fail to meet the criteria of clarity and congruence of official action with the norm. It is all too evident that state practice does not reaffirm (as we would wish it to) the absolute nature of the rule. In other words, although all theorizing is history bound, and one might have expected that twentieth century liberalism would have affirmed a rule outlawing torture, it turns out that the criteria of legality are able to force a clear-headed assessment of the posited rule.

The chapter on torture and its very uncomfortable conclusions also belie Koskenniemi's reprimand that the interactional theory we advance cannot resist the blandishments of twentieth and twenty-first century technocracy. He is right that we suggest that one of our goals is to help practitioners understand how legal obligation is created, sustained, and destroyed. But this goal does not imply that we are supporting a functionalist or an instrumentalist approach. We hold firmly to the view that law is indeed a ‘standard of criticism’; this is precisely why the criteria of legality are central in distinguishing social norms from legal norms. They tend to insulate law from the underlying shared understandings in which law is grounded. For this reason, the interactional theory is not circular. It does not test the existence of law merely by reference to the ideas already held in society or to the compliance of society with the law. Law is relatively autonomous, although it can be progressively undermined if practice fails to support the criteria of legality. This is because law will come to be seen as pure hypocrisy if it is meaningless in practice.

In suggesting that interactional law could become a form of managerialism, Koskenniemi focuses on our chapter devoted to climate change. He wonders whether we have turned the law into ‘a strategic instrument in the hands of the advocates of the climate change project’. This suggestion is surprising given that in this chapter we use the criteria of legality to cast doubt on the existence of any substantive rules effectively addressing climate change. The most we assert is that through learning associated with the continuous interaction of states, scientists, and civil society, some procedural norms have evolved that could continue to advance a more substantive legal regime. Our point is actually similar to the idea that Koskenniemi links to Fuller, Kant, and Weber: law is fundamentally about ‘how members in a community should relate to each other’. It is not about a precise set of pre-determined goals. In this sense, law is both historical and a-historical.

Mystery

In reading Koskenniemi's comments, focusing as they do on the ‘mystery’ of legal obligation, we were reminded of a marvelous sequence in the film Shakespeare in Love, where theatre manager Philip Henslowe, played by Geoffrey Rush, is asked how it is that a play can go on when everything seems so chaotic at the dress rehearsal. His response ‘It is a mystery’. But just as theatre can still be a subject of analysis, even in the face of that mystery, legal obligation can be analyzed, and its mysteries unraveled, at least in part.

Indeed, we think that Fuller helped to explain one of the central elements of law's mystery by pointing to an internalized understanding of obligation, what Fuller called ‘fidelity to law’. The idea that agents often comply with law because they have been encouraged through the processes by which law is created to believe in its legitimacy is, in a sense, ‘mysterious’. This is precisely why Fuller's approach is often connected to quasi-naturalism, as Koskenniemi himself suggests. But Fuller's naturalism, and ours, is distinctly ‘light’. We emphasize that the criteria of legality blend limited notions of autonomy with the fundamental human desire to communicate. When further supported by a continuing practice of legality, they produce a sense of legitimacy. It is important that Fuller's criteria, and ours, do not require any further commitment to specific substantive values, such as western-style democracy or free markets. We hasten to add that the legitimacy we address is a distinctly legal legitimacy, and is not a political form that might allow actors to justify disregarding inconvenient legal requirements.

The further mystery is that as interactions increase within a framework of legality, it may be possible to build up stronger shared substantive commitments. This is why procedural commitments in law are so important; in the context of climate change, we suggest that the continuing interactions provided for through the UN-focused climate regime just might produce stronger substantive commitments in the longer term. We are not shilling for the UN per se, nor do we imagine that the framework created was ideal, nor that it was the only framework imaginable. All we argue is that the climate change regime's procedural legality has proven to be surprisingly resilient; it continues to focus energies and it still has a chance to enable substantive agreements.

In a very different context, the use of force, we argue similarly that the decision-making framework of the Security Council, although under pressure, remains the only procedural track that has the capacity to produce a sense of legitimacy in the application of collective force. In this case, the accretion of practice over more than 60 years has helped to reaffirm substantive commitments such as the limitations on claims to self-defense.

We agree in part with Koskenniemi's suggestion that the mystery of obligation lies in law's disciplinary autonomy. Law is not coextensive with, nor is it reducible to, economic analysis, game theoretic approaches, sociology, or anthropology. We emphasize that our approach is rooted in a theory of law, not of ‘governance’. However, we are less nervous than Koskenniemi about trying to gain insights from cognate disciplines such as International Relations (IR). In particular, IR debates over the generation of shared understandings help to place law within a social context.

Reus-Smit does not believe that we have drawn enough from the IR literature on shared understandings; implicit is the idea that perhaps we have kept the law too mysterious, not grounded firmly enough in historically specific understandings of state and sovereignty. Our use of Fuller's criteria of legality ‘naturalizes’ what is contingent. But just as law is not reducible to economics or sociology, it is not reducible to history either. There is some ‘mystery’, affected and shaped by historically contingent ideas, yes, but trying to grasp a unique sense of what makes law different from other disciplines. In our time, we think that the criteria of legality functioning alongside continuing practices of legality are enough to generate the mysterious fidelity to law. In our experience, the strongest complaints by IR practitioners against international lawyers are always methodological. IR may still be caught in the desire to fit politics within a paradigm of scientific analysis where there is no mystery – only supposed causality.

For Dunoff, the real mystery in our book lies in our implicit decisions concerning what theorizing on law is all about. He correctly states that our kind of theorizing is not ‘purely analytical’; it contains strong elements of prescription. We agree that jurisprudence is a purposive activity, not merely a form of description. We would go further to argue that there can simply be no ‘pure theory of law’. In this sense, law is not analytically mysterious, for it is grounded in shared understandings within the society; our version of law is rooted in pragmatism, not perfectionism. This is why the practice of law is part of its very existence, not just in function but in theory.

Mastery

Our interactional theory of international law is strongly practice oriented. It is also practice generated. We want to understand what a master practitioner of international law actually does in our era. How does she or he help to create, bolster, or undermine norms? By choosing to test our ideas in three distinct issue areas, we try to identify the specifics that make up practices of legality.

Dunoff finds it surprising that we fail to discuss ‘many of the key professional tasks and practices that international lawyers routinely engage in’. Here, we may simply have a difference in judgment. We focus heavily on things international lawyers actually do. The climate change chapter offers a detailed account of negotiating practices in a multilateral framework. We discuss treaty negotiation, the role of Conferences of the Parties, and the legal evolution of national positions. International lawyers are at the heart of all of these forums and developments. In reviewing challenges to the anti-torture norm, we examine doctrinal writing, legal advice given within foreign offices and justice departments, the work of human rights NGOs (supported by international lawyers), and decisions of national courts. The use of force chapter discusses decisions of the UN Security Council and the International Court of Justice, reports of the UN Secretary-General, submissions of states to UN commissions, and contains detailed evaluations of state practice and doctrinal writing. It is true that we do not emphasize the role of international courts, but this is because their role is marginal at best in two of the issue areas we discuss, and is limited even in the context of the use force. Here, we see illustrated our observation that judicial decisions are but one element in the ongoing practices of legality in international society.

More importantly, we try to extend the practice data that need to be evaluated in determining the existence and strength of international norms. Because it is central to our argument that norms are supported by underlying shared understandings in society, we look closely at popular and high culture, and at the work of non-governmental actors such as NGOs and high-profile individuals. States remain the dominant force in the creation and maintenance of international law, but they are not the sole actors in international society. Nor are state decisions immune from the direct influence of other actors.

We find it odd that both Dunoff and Reus-Smit suggest that our book does not pay adequate attention to ‘interpretation’, a central element in any practice-based account of law. In our view, the entire book is a disquisition on interpretation, but perhaps not in the sense often associated with international law. We do not parse the separate opinions of judges of the International Court of Justice. Nor do we play out the negotiations over a specific word in a specific treaty. But the search for underlying shared understandings brings us inevitably into debates around interpretation. What is ‘torture’? Does it include ‘waterboarding’? Does ‘self-defence’ include preemptive force? Do we think that a rule limiting greenhouse gas emissions assumes differential treatment of developed and developing states?

The criteria of legality likewise demand close attention to questions of interpretation. Does a total ban on torture meet the criterion of clarity if there is no agreement on a definition of torture, as opposed to cruel and unusual punishment? Does the concept of ‘common but differentiated obligations’, as contained in the climate convention and the Kyoto Protocol, meet the criteria of clarity? By simply refusing to identify a given situation (e.g. the Rwandan genocide) as a ‘threat to international peace and security’, is the Security Council failing to meet the criteria of impossibility or constancy of practice over time? We focus on these kinds of interpretative questions throughout the book. Part of the issue for Reus-Smit seems to be that he sees ideas and practice as distinct. We do not. For us, ‘practice’ exists through the interplay of ideas and structures.

Koskenniemi would wish us to differentiate more clearly between the work of lawyers and diplomats, and the work of lawyers and IR specialists. He wants to create an almost complete autonomy of legal practice. For us, law is only relatively autonomous. Legal ‘practice’ is more complicated than the formal exercise of legal duties such as representation in court or the negotiation of a specific treaty text. International lawyers are also members of epistemic communities, and they participate in interlocking communities of practice, including communities of legal practice. We believe that the criteria of legality give to lawyers a special role that makes their contributions to social discourse relatively autonomous. The criteria of legality also provide strength to legal analysis that prevents law from becoming a mere instrument of power. Law is stronger than Koskenniemi may fear.