The Decline of the Death Penalty and the Discovery of Innocence. By Frank R. Baumgartner, Suzanna L. De Boef, and Amber E. Boydstun. (Cambridge University Press, 2008.)

sharing constitution and Togo did not—Norris’s puzzle highlights the potential that different types of institutional arrangements can have for processes of democratization. Norris conceptualizes power-sharing and powerconcentrating regimes on the basis of four formal institutional features—the type of electoral system, the horizontal concentration of powers in the type of executive, the vertical centralization of power in unitary or federal states, and the structure and independence of the mass media. Four detailed chapters provide descriptions of the power-sharing and power-concentrating characteristics of each institution and summarize the arguments for and against their democracy-enhancing and conflictreducing effects. Employing cross-sectional timeseries data for 191 countries for the period from 1973 to 2004, Norris tests the effects of each institution on various indicators of levels of democracy. In each instance she finds that it is the power-sharing rather than the power-concentrating versions of these institutions—e.g., proportional representation electoral systems rather than majoritarian systems, federal rather than unitary states—that are associated with higher levels of democracy. The chapter on federalism and decentralization is particularly well developed, with attention given to classifying different types of decentralization (administrative, fiscal, and political) and constitutions (federal, unitary, and mixed unions) in order better to operationalize vertical forms of power sharing. The chapter on the fourth estate, on the other hand, strikes a somewhat odd note in that Norris never makes quite clear how the media fit her definition of a power-sharing institution as one which gives ‘‘multiple political elites a stake in the decision-making process’’ (23). Working within the framework of consociational theory, Norris notes at the outset of the book that she seeks to focus on the capacity of institutional reforms to facilitate democratic consolidation and to generate lasting peace settlements in states emerging from civil wars. Although the book thoroughly addresses the first of these issues, it misses the mark where the latter is concerned. First, the empirical tests the book employs are not designed to examine the impact that institutions have on the duration of the peace. A focus on levels of democracy, while appropriate for exploring the impact institutional reforms have on democratic consolidation, says little about the success those reforms have in stabilizing the peace following civil conflict. In light of the increasing number of scholars who claim that introducing democratic institutions, particularly elections, in the immediate postconflict environment is likely to destabilize the peace, it would be of interest to examine the impact power-sharing and power-concentrating regimes have in this context. Doing so, however, requires the use of a different dependent variable. Second, as Norris herself notes in passing, additional types of power-sharing institutions may also be central to formulating a durable peace. Rules that call for the government and former armed adversaries to share military power, for example, arguably play as important a role in the construction of negotiated peace settlements now as the formal constitutional rules on which Norris focuses. Studies whose goal it is to examine the impact of power-sharing institutions on settlement stability increasingly take these less traditional forms of power-sharing institutions into account. Do power-sharing institutions work? The unambiguous answer provided by this book is ‘‘yes’’—they work to consolidate democracy. Students of consociational theory will find much to admire in this book. Although it sticks to a formal and traditional conception of institutions that play a role in the management of conflict, it deftly synthesizes the core assumptions and claims of consociational theory and identifies the limitations of previous research that has sought to test the performance of consociational arrangements on democracy. The scope of the book, combined with its methodological rigor, ensure that it will stand as an important contribution to the empirical study of democracy.