In 1990, following the Iran-Contra affair, the late US Senator Daniel P. Moynihan published On the Law of Nations, a short but passionate book in which he decried the emergence in the United States in the 1980s of the view that belief in international law was associated with weakness in foreign policy. In that climate, he wrote, ‘real men did not cite Grotius’ (Moynihan, 1990, p. 7). Moynihan blamed the Iran-Contra affair and the US invasion of Grenada in part on this view. Nevertheless, Moynihan concluded optimistically that a commitment to international law would prevail over self-interest.

He was too optimistic. One short generation later, we have once more lived through a decade in which ‘real men did not cite Grotius’. This has been manifested, particularly in the United States, by the (re-)emergence of a rational choice theory of international law. This theory is epitomized by Jack L. Goldsmith's and Eric A. Posner's The Limits of International Law (Reference Goldsmith and Posner2005), in which the authors argue that the power and significance of international law is exaggerated: states comply with international law when it is in their self-interest to do so. While they deny claiming that customary international law does not exist, Goldsmith and Posner's argument that customary international law is not an exogenous influence on state behaviour leads inexorably to precisely that conclusion: if states comply with international law, or at least customary international law, only when it is in their self-interest to do so, it may be argued the law does not exist independently of the states' self-interests. This would mean, of course, that states are not constrained in pursuing their self-interest. Moynihan quoted Wordsworth: ‘the good old rule/Sufficeth them, the simple plan/That they should take, who have the power/And they should keep who can’ (p. 4).

Not surprisingly, there has been a strong reaction to this argument.Footnote ¹The Economic Structure of International Law is perhaps the most comprehensive reaction to date. Joel Trachtman presents a painstaking review of economic approaches to understanding international relations theory and international law. Trachtman is careful not to oversell the merits of economic approaches to law. He quotes Keynes: economics ‘is a method rather than a doctrine, an apparatus of the mind, a technique for thinking, which helps its possessor to draw correct conclusions’ (p. 5). Trachtman uses the methods of economics not to reject the rational choice theory, but to argue that a rationalist analysis leads to different conclusions regarding state behaviour and the binding nature of customary international law than Goldsmith and Posner.

Trachtman explains that the central assumptions of economics are that individuals seek to maximize the achievement of their preferences and that normatively the only valid source of preferences is individuals. Accordingly, ‘the central theory suggested by the economic approach to international law is that states use and design international transactions (including all rules of international law) or institutions to maximize the participants’ net gains' (p. 15). At the level of international law, Trachtman considers that states exist to aggregate individual preferences. Thus, Trachtman's approach is based on ‘the idea that international law is produced in order to allow states to achieve their preferences with greater effectiveness through exchanges of authority: through transactions in jurisdiction’ (p. 11). It should be noted that this approach can be traced back even to the uncitable Grotius: ‘just as the laws of each state have in view the advantage of that state, so by mutual consent it has become possible that certain laws should originate as between all states, or a great many states; and it is apparent that the laws thus originating had in view the advantage, not of particular states, but of the great society of states. And this is what is called the law of nations, whenever we distinguish that term from the law of nature’ (as quoted in Kelly, 1992, p. 242). For this reviewer, however, Trachtman's view of states as aggregating individual preferences and designing international law to advance those preferences does not fully respond to H. L. A. Hart's interpretation of sovereignty as being negative in force: ‘a sovereign state is one not subject to certain types of control, and its sovereignty is that area of conduct in which it is autonomous’ (Hart, Reference Hart1994, p. 223).

At the core of his book, Trachtman uses game theory to develop a model to determine when states would be more likely to develop multilateral equilibria of compliance with a rule of customary international law (CIL). Trachtman uses this model as the basis for a general theory of the binding nature of international law and of the capacity of CIL to affect state behaviour.

Trachtman is realistic about the limits of game theory and indeed economic theory generally, noting that such theories merely represent assumptions about behaviour. Under the basic prisoner's dilemma, which was the cornerstone of Goldsmith's and Posner's approach, it may be predicted that individuals (and states, when they are the players in the game) will have difficulty in coordinating. In the international law context, this would mean that binding rules of CIL would not be formed. Trachtman notes that contrary to this prediction, the evidence suggests that in reality such cooperative behaviour is widespread. Trachtman points out that much of the evidence is obtained in circumstances in which the assumptions of the prisoner's dilemma are violated – individuals can communicate, enter into agreements, and repeat play. Of course, if real life does not conform to the game, the game must be changed in order for it to be useful as a model to predict real-life behaviour. This is, in effect, what Trachtman attempts to do.

Trachtman notes that CIL is more likely to emerge and be sustainable when the returns to cooperation are high relative to those to defection and when the returns to unilateral defection are low. CIL is also more likely to be formed between states that are patient and where the probability of continued interaction between the participating states is high. Trachtman concludes that there is a significant set of cases in which it is possible to form CIL, the likelihood of which depends on a number of factors, including the relative value of cooperation versus defection, the number of states involved, and the extent to which the increasing number of states involved increases the value of cooperation, including whether the issue is a public good or a commons problem, the information available regarding cooperation and defection, the relative patience of states to realize the benefits of long-term cooperation compared to short-term defection, the expected duration of the interaction, the frequency of interaction, and the existence of other bilateral or multilateral relationships between the states involved.

Directly in response to Goldsmith and Posner, Trachtman argues that his model indicates that the multilateral prisoner's dilemma can be resolved in favour of cooperation and, therefore, that CIL may affect state behaviour. The basic prisoner's dilemma game depends on a single rational player with a monolithic self-interest. As Trachtman points out, states act rationally only in theory.Footnote ² In addition, it is questionable that states' self-interest is necessarily either monolithic or always entirely self-centered. Trachtman observes that with respect to CIL, the self-interest at issue is the ‘broader, or potentially longer-term, self-interest that flows from making and achieving compliance with a rule or even rules generally’ (p. 113). Thus, ‘no-one would argue that for domestic law to qualify as law, compliance must be motivated by something other than self-interest’ (p. 113). He describes Goldsmith's and Posner's main point as the application of an assumption of the rationalist model: that agents care only about their own utility and would not follow a rule of CIL for intrinsic reasons. However, a state's broader or long-term self-interest may make it worthwhile to comply with CIL even when to do so may be against its short-term interest. Again, Grotius also made this point: ‘just as the national, who violates the law of his country in order to obtain an immediate advantage, breaks down that by which the advantages of himself and his posterity are for all time assured, so the state which transgresses the laws of nature and of nations cuts away also the bulwarks which safeguard its own future peace’ (as quoted in Kelly, 1992, p. 242). Similarly, Moynihan warned that ‘a state that finds itself tempted by self-interest to erode traditional norms may in time regret its conduct. Indeed, it may not take very long’ (p. 149).

Trachtman notes that Goldsmith's and Posner's argument could be read as simply an attempt to refute the natural law theory that states comply with CIL out of a sense of moral or legal obligation. Trachtman's economic analysis sympathizes to a degree with that narrow reading. But he rejects any broader argument that self-interest and legal obligation are mutually exclusive, claiming that ‘there is no question that law can affect behaviour through self-interest’ (p. 113). Trachtman also rejects Goldsmith's and Posner's analysis of various case studies in which they claim that states' compliance with international law was based on self-interest, pointing out that the evidence merely indicates that, in certain cases, there were plausible non-CIL reasons for the behaviour. We do not know whether, in those cases, CIL was not a contributing or alternative cause of behaviour.

The history of international law theory consists of a debate in which the pendulum has swung between the natural law theorists and the positivists. Trachtman's achievement is to show that even on the terms of positivists or rational choice theorists, the notion that states act according to their self-interest is not sufficient to establish that customary international law is not binding on states so acting: self-interest cannot justify everything. This is important. It is noteworthy that all of the writers cited above as supporting that concept wrote after periods of conflict in international affairs. Grotius wrote after the outbreak of the Thirty Years' War in Germany, Hart wrote after the horrors of the two world wars, Moynihan wrote in the context of questionable US actions in Nicaragua and Grenada, and now Trachtman writes at the end of a turbulent decade. It may be hoped that Trachtman's work will help weigh the pendulum more heavily in favour of compliance with international law.

There are many more interesting insights in this book. Trachtman applies economic analysis to great effect to questions such as the role of international organizations and the question of fragmentation of international law. Of particular interest is his view of the debate between natural law and positivism as a debate as to the scope of the authority conferred on international arbitrators: are they free to exercise moral judgment founded in a natural law perspective, or have treaty writers curtailed in advance the moral perspective that the arbitrators may apply? While this book is written very much for an academic audience, it will also reward patient study by a broader readership interested in systemic approaches to international law in general and in the specific topics discussed in the book.

Trachtman concludes by saying that his theory serves as a call to empiricism. In light of the depth and thoroughness of his theoretical work in this book, one can only hope that Trachtman himself answers this call.