In his book, Warring Friends, Jeremy Pressman draws attention to the bargaining that happens among allied states as they attempt to influence one another's foreign policy. Pressman argues that while alliances are certainly intended to respond to threats from outside states, another motivation for formalizing alliance relationships may be the creation of institutionalized mechanisms for controlling an ally's behavior, and he finds evidence to support this claim. Pressman then attempts to determine what factors influence whether states are successful at stopping their allies from pursuing courses of action that they do not support. He finds that powerful states can restrain their partners, but only under certain conditions.

Pressman's work moves well beyond the traditional emphasis on alliances as mechanisms for capability aggregation or for signaling interests and intentions to adversaries, and, following Paul Schroeder and Glenn Snyder, focuses instead on the internal management of alliances and the role of institutions in that process. Yet, in no way does Pressman abandon past scholarship. Instead, he links the motivation to ally caused by common threat and the ability to influence an ally's policy.

Pressman begins by addressing the following question: Do states form alliances in order to restrain their partners? Pressman makes it clear that he does not see external threat and internal restraint as competing motivations; instead, he argues that the existence of threat creates an interest in coordinating policy, and thus a desire to influence what the alliance partner does. Given threat, an additional motivation for formalizing military ties between states is to create infrastructure through which the states can monitor and influence one another's military actions. Pressman finds six cases of alliance formation that he believes may have been motivated, at least in part, by goals of restraint—Germany-Austria (1879), Britain-Japan (1902), NATO-West Germany (1949), U.S.-South Korea (1953), U.S.-Taiwan (1954), and Egypt-Syria (1964). He selects these cases not because they are hard ones for his theory, but rather because they provide a plausibility probe—these are the cases most likely to fit this pattern. His analysis of the historical record leads him to conclude that goals of restraint played at least a minor role in each of the cases, and a particularly important role in motivating the U.S.-South Korea, U.S.-Taiwan, and Egypt-Syria alliances.

Next, Pressman moves on to consider a second question: What explains the success or failure of alliance restraint attempts? Here, his empirical work focuses on two long-term bilateral relationships: U.S.-Britain and U.S.-Israel. He examines two cases in which Britain tried to restrain the U.S., two cases in which the U.S. tried to restrain Britain, and seven cases in which the U.S. tried to restrain Israel. There is significant variance in whether the stronger ally (in each case, the U.S.) was successful at restraint or was successfully restrained. Britain succeeded in restraining the U.S. in one case but not the other, and the U.S. succeeded in restraining Britain in one case but not the other. The U.S. succeeded in restraining Israel in four cases, but failed in three.

From his case studies, Pressman concludes that stronger allies are able to restrain their partners only under particular circumstances: namely, when the powerful state has clear information about the ally's policy and actions, when the leadership of the powerful state is unified in its support of restraint, when the issue at stake is of high national security importance to the powerful state, and when the powerful state is willing and able to create a viable alternative policy that helps its ally achieve its goals in a non-militarized manner. In other words, a state that cares a great deal about an issue and has tools with which to act is likely to achieve its preferred outcome in negotiations with its ally.

These are not at all surprising findings and fit seamlessly with our more general understanding of bargaining advantage. What is perhaps most notable about some of the cases that Pressman studies is that the disputes between allies became public. The deterrent properties of alliances depend on their credibility, and the credibility of an alliance is affected by the extent to which potential adversaries believe the allied states will act in tandem. Allies have every incentive, therefore, to coordinate policy and present a united front based on shared interests to the outside world. In most cases, this coordinated policy will require communication, negotiation, and compromise among the allies, since interests are rarely identical. This coordination is one of the primary costs associated with forming alliances; effective alliances result in some loss of autonomy. The loss of autonomy need not be equal; in asymmetric alliances, the burden of policy change may fall more heavily on the weaker power. In general, however, developing joint policies among allies is a matter of frequent bargaining.

Given these incentives to resolve disputes privately, the fact that some of Pressman's cases played out very publicly raises interesting questions. Why did states engage in military actions only to withdraw after allied objections? Why did allies publicly object to their partners' behaviors if, in the end, their objections would be ineffective? In how many other instances was restraint invisible, as states anticipated allied objections and thus did not even consider pursuing policies they might have preferred? A reader might question whether this sample of cases—where restraint attempts are publicly visible—is actually a very unusual sample.

Pressman should be commended for taking on an important and understudied topic. While alliances have been studied intensely for years, the vast majority of analyses concentrate on the relationships between allies and adversaries. It is comparatively rare to find studies of relations among allies, and particularly of how allies bargain with one another and what determines joint policy outcomes. The evidence presented in this book suggests that standard bargaining theories are likely to prove useful in understanding allied relations. This remains a first step, however, and I hope that this work will lead other scholars to study intra-allied bargaining and the effects of alliance institutions on this process more generally, both theoretically and empirically. Further study may help us understand why we occasionally witness such public disputes among allies, and how these public disputes differ from the general policy coordination that must occur regularly to make an alliance effective.