Zeev Maoz was one of the first international relations scholars to take up the network banner and carry it into the fray of international relations. This volume is his order of battle, detailing what he means by networks, what the important characteristics of networks are, how these facets of networks can be observed, and the implications of different kinds of network structures. From this, he describes the network structure of the international system and develops and evaluates a theory of network(ed) international politics.

For Maoz, network analysis has the potential to provide a framework for the systematic study of indirect relations among states, whereby the fate of China–U.S. relations affects, for example, the future of Japan. At the same time, network analysis provides, for Maoz, a bridge across the levels-of-analysis chasm that has plagued international relations scholarship since the early 1950s. Moreover, by using the bridge of network analysis, one can putatively observe new levels of analysis, ignored by past paradigms. Herein, for Maoz, networks come close to being a Morton Kaplan type of self-reinforcing international system, which largely determines the putatively sovereign behavior of individual nation-states (see Morton Kaplan, System and Process in International Politics, 1957). And finally, network analysis can provide a whole new array of concepts, such as structural holes, to the study of many problems that have perplexed international relations analysts for decades. Thus, networks provide a lens on how states behave strategically toward one another (classic realism), how their actions have externalities that may be beneficial at a distance (liberalism), and how ideas that are nurtured in the network collectivity also have purchase on world politics (constructivism).

The major building blocks of the network structure are developed mostly from the sociological literature known as social network analysis (SNA). Maoz starts with the basic mathematical representation in SNA and other network perspectives: an n × n matrix connecting n nodes to each other, with entries reflecting the existence or strength of connections between each pair of actors. This is known as an adjacency matrix or (weighted) graph in some contexts, but in SNA it is called the sociomatrix. This matrix is interpreted as a mathematical representation of the network, focused on the strength of ties among nodes or, for Maoz, countries.

These network structures have a number of characteristics, but most importantly those that inure to the node (e.g., how many links flow toward a given country) and others that exist at the network level (e.g., how dense is the network; i.e., how many links are non-zero out of all possible links). Centrality is the most important concept in SNA. How central to the network is a given node? How many connections are necessary to link a given node to all others? How central is a node in the secondary or indirect ties (e.g., neighbors of neighbors) that exist in the network, compared to the direct ties? Beyond these kinds of measures, one finds network characteristics of nodes to be broadly important. For example, how similar are the network roles (of, say, brokerage) of two different actors? The idea of structural equivalence is an important concept that has considerable purchase in the international relations context. Clique overlap, block portioning, network polarization, and other concepts from SNA get adapted and customized to serve Maoz's theory of network relations in international relations. The description of the toolbox is done quite well, and can serve as a very brief primer to some of the basic concepts in network science for those interested in international relations.

Maoz describes the structure of the international system using the basic tools of SNA, as well as some he develops specifically. He shows that things change, but not necessarily in immediately obvious ways. Density of the international network generally goes down over time in trade, for example, while international organization membership of network density grows substantially. Transitivities are another important network characteristic that evolves over time. High transitivity implies consistent cooperation patterns throughout the network, but dips in these measures signal potential trouble in the international system. Finally, polarization is also assessed and shown in these analyses not to be supportive of the general informal impressions about the link between rising polarization and political conflict.

The theory of network formation and evolution perhaps comprises the most unique contribution of Maoz's book. Herein, he tackles the issues of why (or how) states choose to cooperate, whether there is spillover in cooperation across domains, and what structural consequences are bound to this cooperation via the network evolution. This theory is developed against a backdrop of realism based on suspicion of others and common interests. The liberal lighting given the realist backdrop suggests that states are not the only actors and that different structures can affect outcomes, including the emergence of institutions. Finally, security is not the main driving force, but may actually be subsidiary to other interests.

To this stage, Maoz adds a constructivist score that is similar in many respects to the ideas of Kaplan; namely, the system is itself a causal actor. Maoz's actual theory assumes a large set of antecedents: security under anarchy, power maximization, suspicion of others, strong national identity effects, and the importance of common interests, identity, and prior history. This leads to a number of implications at different stages of the evolution of the network. Different things happen at the beginning, as opposed to what occurs in a stable international system. A big force in all of this is the modified expectations that accumulate in individual countries based on observing what has been occurring in the network writ large.

There is a lot that is interesting in this very precise development of a theory, spanning realism, liberalism, and constructivism. The network story is not a simple collection of unrelated hypotheses, and as a result it is difficult to summarize. In short, the theory “postulates a process that characterizes the alliance choices of states and the implications of these choices for other types of networks” (p. 200). Maoz is suggesting that the network itself is the main context for the individual behavior of nation-states, not their attributes alone, nor even the attributes and behaviors of their partners taken one at a time. Rather than a seriatim unveiling, he is positing that networks are the “whole Monty.”

The conclusions of the empirical analysis are notable, and I summarize them briefly only in one area. Security and nonsecurity concerns seem to be correlated with alliance formation. At the same time, democratization, trade, and cultural similarity within a network tend to reduce treat perceptions by states. Furthermore, cooperation within the relevant potential partners does not appear to dampen states' drive to acquire alliance partners. Stated this way, these do not seem very network related, but seem to be statements that come directly from a long line of research on alliances, trade, and militarized interstate disputes. However, Maoz gets to these conventional kinds of conclusions with conventional hypothesis tests, using variables that emerged from his use of network theory.

This is a very important book, bringing together realism, liberalism, and constructivism through the use of SNA. I hope it is a widely used volume in international relations seminars and that it inspires others to use network analyses to learn more about the international system. This book will be criticized for focusing too much on states and ignoring the many other network actors playing important roles in the world (e.g., the World Bank, Taliban, intergovernmental organizations, nongovernmental organizations). The next iteration of this research will leap outside of the network of nation-states to one that includes a wider array of actors that we now know to be increasingly important. It would be easy to dismiss this book because it deals only with networks of nation-states, in an epoch in which many now believe that asymmetric networks are more salient. This thread leads to valid criticisms but may overlook the main contribution of Networks of Nations, which is to signal a way forward in its approach, one that will resonate with traditional IR scholarship that focuses on the accumulated data we have on nation-states and their dyadic interactions.