In his influential book, Theory of International Politics, Kenneth Waltz claimed that international politics takes place in a social system characterized by what he called ‘anarchy’, and a theory of international politics must focus on the effects of anarchy on the behavior of states. Domestic politics, he claimed, takes place in a different kind of social system, whose main property is what he called ‘hierarchy’. In subsequent years, disagreements about the properties of anarchy and its consequences among students of international politics in the United States have led to an unproductive debate among competing schools of thought. In Chapter 1 of my book, War and the State: The Theory of International Politics (Wagner, Reference Wagner2007), I reconstruct and evaluate the main arguments of those competing schools, and show that they are all invalid. Subsequent chapters argue that a resolution of the issues in this debate requires that the problem posed by Waltz be reformulated.

Waltz’s writings have been one of the most influential formulations of the central tenets of a school of thought commonly called ‘Realism’. For realists the anarchic nature of the international system is the most fundamental explanation of the recurrence of interstate wars. Even many of the critics of realism agree that interstate wars can only be prevented by global political institutions, and thus debates between realists and their critics often focus on whether that is a realistic prospect or not.

However, wars do not require states, they merely require competing organizations capable of killing and destruction, and states are among the possible consequences of wars. Wars can occur within states, between them, or across the boundaries that separate them, and the settlements that end wars can lead to more states, fewer states, or states with different boundaries. Thus, a world of independent states is not a world without a global order, as the term anarchy suggests. Rather the independent states are the global order, and the central issue raised by the controversy about realism is why peace settlements that define the boundaries among states cannot be as lasting as the settlements that provide for the establishment of governments within them.

To answer this question, we need to understand the relation between organized violence and political order. I argue in Chapter 2 that this was the main subject debated by writers in the intellectual tradition from which modern realism was derived, which is often called raison d’état, or reason of state. It is a puzzling and complex subject, in part because political order in complex societies is organized violence. Thus, both interstate warfare and domestic repression are possible in such a system, and neither can be understood in isolation from the other. The relation between them is one of the central questions debated by early modern European political theorists as the European state system developed. To think productively about it, one must understand both violence and how people organize themselves to engage in it. Considered separately, each is an extremely complex phenomenon, and together they constitute a very difficult subject of study. That is the subject of Chapter 3.

Violence is complicated in part because it can be used by some people to influence the behavior of others, and therefore can be part of a bargaining process. Violent bargaining has several properties that make it especially complex. First, variations in what is called the disagreement outcome in the bargaining literature will influence bargainers’ relative bargaining power, and the disagreement outcome can not only be manipulated by the bargainers, but also change as a result of exogenous factors. Second, any agreements that might be reached must be self-enforcing. Finally, the extent to which people are organized will influence how much bargaining power they have.

Understanding how human beings manage to create large organizations is a complex subject in its own right. Organizing for violence is especially complex, because it introduces the possibility of violent conflict between or among organizations seeking to profit from the threat of violence, as well as among the members of such organizations over how to divide the gains from it.

Given the complexity of these phenomena, it is not surprising that people would try to ignore some components of them in order to analyze the effects of others, and for many purposes this is the only sensible way to proceed. Treating states as though they were persons bargaining with each other is an example of such a simplification, and much can be said about international politics in those terms.

If we want to evaluate the competing claims that have been made about what might be the institutional basis for a peaceful global order, however, then such simplifications inhibit our understanding of the problem, since the modern state is at once the main example that we have of how violence can be organized to serve the common good, and the basic institutional building block out of which a global order might be constructed. One of the questions to be answered is whether peace requires some institution at the global level that resembles the modern state, or whether properly constructed states alone would be a sufficient institutional foundation for it.

A state is constituted by the contracts that define the organization of a government and its relation to the people it governs and to other governments, and the collection of all such contracts defines the institutional structure of the global order. Contrary to what much of the literature about international politics assumes, force can be used to renegotiate any of them; an attempted coup d’état, for example, can lead to a popular revolt, which can lead in turn to a war with another government. Moreover, each set of contracts acts as a constraint on the others. For example, the contracts that define the structure of a government and its relation to its subjects will partially determine both the goals and the capabilities of the government in its bargaining with other governments. Thus, the problem of constructing a global institutional order in which peace would be an equilibrium is like a giant jigsaw puzzle with many pieces that must fit together properly.

The barriers to the peaceful negotiation of agreements that define the relations among governments are not fundamentally different from the problems that inhibit the peaceful negotiation of agreements defining the organization of a government or its relation to the people it governs. The terms of an agreement will be influenced by the amount of force that each party can inflict on the other. But this will be influenced in turn by the nature and extent of their organization. Thus, the rise and fall of states is the product of the formation and dissolution of groups, and forcible bargaining among them. The complexity of this process is magnified further by the fact that the terms of any agreements that might be reached may influence the subsequent relative bargaining power of the parties to the agreement. An agreement defining the boundaries between two states, for example, may influence their subsequent relative military capabilities. And an agreement defining the relation between a government and its subjects will usually require the disarmament of one or more of the parties to a conflict, which might be hard to reverse if the terms of the agreement are violated later.

Thus, the main problem in resolving violent conflicts is to define the terms of such contracts in a way that will reflect the current relative bargaining power of the antagonists without altering it. One barrier to agreement may be inconsistent estimates of what the parties’ relative bargaining power actually is. Another may be the inability to craft an agreement that everyone is sufficiently confident will not place them at a disadvantage in the future. There is no guarantee that agreements that avoid both these problems and are also preferred to the continuation of a conflict by the parties to it always exist. And even if they do, they will be vulnerable to exogenous changes in incentives, expectations, or the technology of violence that can lead to subsequent renegotiation. That is why organized violence has been regarded as normal for much of human history, and the belief that political institutions of any sort that might reliably prevent it requires more justification than has been given for it.

Organized violence makes predation profitable, which in turn leads to violent contests among competing predators. Contests among predators create an incentive to mobilize more resources for warfare, which can increase the bargaining power of the prey and thus over time diminish the gains from predation. The literature on the development of the European state shows that the European state system was the product of such a process. Chapters 4 and 5 discuss why competing predators find it difficult to avoid violent conflicts even though they have a common interest in settling their conflicts peacefully, and Chapter 6 discusses whether the non-predatory states that emerge from recurring violent conflicts among predators could provide the basis for a peaceful political order at the global level. This is the central question addressed by Hobbes, Rousseau, and Kant, whose writings on this subject are now the main inspiration for controversies about the properties of anarchy among students of international politics in the United States. Kant thought that eliminating predatory rule within states would provide the foundation for peace among them. Chapter 6 discusses whether the analysis given in the preceding chapters provides the basis for a valid argument in support of that claim.

Debates about international institutions, state failure, humanitarian intervention, the democratic peace, democracy promotion, and cross-border terrorism in the aftermath of the collapse of the Soviet Union have revealed how little we understand this subject, as Chapter 6 points out. One hindrance to understanding it is its complexity, which makes it easy to make mistakes in reasoning about factors that one happens to focus on, and overlook other factors that are also important. Thus, it is important that we check the validity of arguments, and that arguments be constructed in such a way as to make that as easy to do as possible. Fortunately, the concept of equilibrium expectations developed by game theorists provides a way of accomplishing those goals, while avoiding two utopian answers to the problem of organized violence that have often been given: eliminating conflict, on the one hand, or making organized violence impossible on the other.

Another barrier to understanding is the division of intellectual labor within political science and among academic fields and subfields. The separation of the study of domestic politics from the study of international politics has meant that the states that define the boundary between them are commonly taken for granted, even though many of them may be imaginary and disagreements about what states should exist and what their boundaries should be are the main sourcs of contemporary international conflict. Debates among students of international politics on the role international institutions might play in limiting violence are often influenced by analogies with domestic political institutions, though participants in those debates usually have little understanding of how domestic political institutions prevent organized violence within states, if they do. Controversies about state failure and state building are conducted without any reference to the violent process by which the modern states whose leaders are now concerned about these problems actually emerged. And discussions of the relation between organized violence and political order are now scattered across the academic disciplines of political science, history, anthropology, sociology, and economics, and are carried on by scholars who take little notice of each other’s work. One of the theses of this book is that the relation between political order and organized violence is a coherent subject in its own right, and that game theory provides a means of unifying the many disparate discussions of it that are now scattered across the social sciences.