Identifying Threats and Threatening Identities: The Social Construction of Realism and Liberalism. By David L. Rousseau, Stanford: Stanford University Press, 2006. 296 pp., $55.00 (ISBN: 0-80475415-2). David Rousseau has accomplished a feat of grand proportions in his book, Identifying Threats and Threatening Identities: The Social Construction of Realism and Liberalism . In eight succinct chapters and just under 300 pages, Rousseau presents a model of identity formation that explains threat perception and employs experimentation, simulation, and case study approaches to test the model. The effort is more ambitious than the book's title implies. Identifying Threats and Threatening Identities is a clear and definitive statement of a social psychological theory of international relations that is worthy of treatment alongside the works of Kenneth Waltz, Robert Keohane, and Alexander Wendt. Even though it has many limits and leaves a number of questions unanswered, this book is an important theoretical and empirical work for international relations theorists and security scholars alike. In Identifying Threats and Threatening Identities , Rousseau addresses four sets of questions: (1) how do individuals define members of the “ingroup,” and how do they treat “outgroups”; (2) how do individuals construct beliefs and opinions; (3) how do ideas spread throughout a population, and when do they become hegemonic; and (4) how do societies differ in the construction of threats? Rousseau reviews the threat perception literature from realist, liberal, and constructivist standpoints (Chapters 2 and 3). He then introduces a “construction of threat model” aimed at answering the four sets of questions and demonstrating when realist and liberal predictions are correct. Rousseau's model is premised on psychological insights regarding ingroup-outgroup dynamics, which suggest that threat perception is a function of whether the other is seen as similar to you or different. Rousseau hypothesizes that the construction of “self” and “other” is based on prior values and beliefs, but that such constructs are subject to manipulation. He hypothesizes that identities are changed through social interaction. He offers three …