Mughan and Gunther's edited collection is a substantial contribution to these more identifiably political science themes. With the one notable, if understandable, absence of the Internet, the book brings together admirably the concerns of established and establishing democracies. A strong cast of scholars analyzes the contemporary contribution of media to democracy in 10 countries: the United States (Thomas Patterson), Japan (Ellis Krauss), the United Kingdom (Holli Semetko), Germany (Max Kaase), The Netherlands (Cees van der Eijk), Spain (Gunther et al.), Italy (Carlo Marletti and Franca Roncarolo), Russia (Ellen Mickiewicz), Hungary (Miklós S¨kösd), and Chile (Eugenio Tironi and Guillermo Sunkel). The various individua l parts are knitted together by Mughan and Gunther's introductory and concluding chapters. They establish the organizing principle, which is to combine micro-level analysis of media impact in specific circumstances and elections with a macro focus on media systems and their inferences for the distribution of power.

Comparative volumes usually work best when a clear overarching hypothesis is tested against the experience of the individual cases (for example, David L. Swanson and Paolo Mancini, Politics, Media and Modern Democracy, 1996). This book is not so tightly connected. Its research question is more open-ended and descriptive. It says, in effect, let us bring together macro and micro evidence in key countries and then assess it for patterns in media environments and the democratic character of political regimes. This approach pays some dividends. The chapters on the newer democracies, Russia, Hungary, Chile, and Spain, are especially fascinating reading. This is less because they tell us things we did not know—research on these countries is fairly readily available elsewhere—but because they are brought together, side by side, and amount to a riveting account of the importance of media in high politics. Clear cross-country patterns do emerge in totalitarian and authoritarian control over press and television and, equally, the crucial role of media in every case to the progress of democratization. Anyone who doubts the political importance of the media would do well to read these chapters and, indeed, the ones on Italy and Berluconi's adventures and van der Eijk's provocative account of the media's leading role in the depillarization of Dutch society.

The appeal of this book rests precisely in the quality of some of its individual chapters. Equally, the chapters that work best are those rooted most firmly in politics, especially the politics of transformation. This is not quite the same thing as saying that the political impact of the media is most interesting in those countries with the most intriguing politics, although that is a tempting conclusion. Rather, it is to say that the macro-level analysis of politics and media is the most valuable contribution of this work. The idea of micro-level analysis was a useful one in theory. It is instructive to know the state of media effects research in the various countries and the consensus verdicts about their impact on public opinion and elections at key times. However, this was generally the most variable and least satisfactory contribution of the country-specific chapters. Methodological approaches to the micro either were simply not comparable from country to country, or were not well elaborated, or were indeed detailed but too time-specific to add greatly to general arguments. Despite this, the editors' conclusion makes strong, too strong, claims for the direct impact of media in individual elections and, more broadly, for the “socializing” effects of media in central European countries. To take one example, it is simply not proven, or capable of proof from the evidence as presented here, that continued support for socialism in central Europe was the direct consequence of “four decades of government censorship and propaganda.” Direct media effects are notoriously difficult to prove beyond doubt, and the editors simply did not need to overegg the evidence of their authors: There is abundant material here anyway for the political impact of the media.

One would not wish to quibble too much with editors, given the sheer amount of the work in this book and the general value of its contribution. Yet one cannot help being struck by the irony of some of their U.S.-centric assumptions, given the internationally collaborative nature of the work. One such example is their major conclusion, which is to challenge the “conventional wisdom” that the freer the media are from government, the stronger the contribution to democracy. This may be the conventional wisdom in the United States, although they offer few references in support of it. However, it is simply not the conventional wisdom in Europe, where for more than 10 years media researchers have sought to protect key elements of public service systems precisely in the interests of a democratic public sphere. The free market/democratic media equation just does not hold as strongly in Europe. In fact, the editors unwittingly admit as much with their comment that European Union commitment to public service systems suggests that it is unlikely that many democracies will copy the U.S. system. A second example is their introductory comment contrasting the current wave of pessimism about media performance with earlier decades of apparent contentment. Again, this simply ignores the European critical tradition, the Frankfurt School, and the critique of sociology-grounded media studies. With these caveats, the editors deserve congratulations for a volume that is a sure-fire certainty on all political communication reading lists.

Mughan's solo book, The Presidentialization of Parliamentary Elections, is, in contrast, far less grand in its ambition. It is a careful and rigorous study of the theme for which he is most familiar in Europe, the trend toward “presidentialization” of parliamentary politics. He confines the study to Britain, which, he argues reasonably, is a fair test-bed. If the proposition is that presidentialism is a broad phenomenon of democratic politics, it should certainly be apparent in Britain, with its strong majoritarian electoral system. A key contribution of this work is Mughan's attempt to define the phenomenon of “presidentialism'' and “presidentialization” and, more importantly, to deconstruct it into testable elements. Presidentialization is defined as a movement over time away from a system of collective to one of personalized government, within the parameters of an unchanging constitution. He tests two aspects of this. First, presentation: Has there been a move from predominantly party- to predominantly leader-focused presentation? Second, are leader effects becoming more influential on voting decisions? This is a helpful approach to a subject that is as ubiquitous as it is vague in the political communications literature. Personalization, we are told endlessly, is a defining feature of modern democratic politics. Rarely, though, are the core concepts defined and, even more rarely, carefully tested against empirical evidence. So in this sense Mughan's book is a welcome addition.

Beyond that, though, it is slightly disappointing. Mughan's test of personalization of presentation comprises primarily content analysis of Times editorials in election campaigns since the 1950s and Gallup opinion polls conducted for newspapers. There are two main difficulties with these data: one admitted, one not. The first is that since the Gallup surveys do not include a measure of party identification, there is no way of disentangling party and leader preferences. The second is the use of Times leaders as a proxy for party campaigns. It is neither uncommon nor unreasonable to suggest that campaigns have become more leader-centered, but it is not verifiable or otherwise from these data. One is left wondering quite what Mughan's content analysis adds to the existing body of research, which does indeed demonstrate increasing media attention over time to leaders. He concludes that there is a trend toward presidentialization consistent with the “threshold” model, that is, a substantial increase in personalization at threshold elections, followed by fluctuations, which never return to the prethreshold lows. However, beneath the claim for the trend, Mughan's conclusions are modest and predictable from preexisting studies: Leadership is still far less important than party and probably sufficient to swing results only in exceptionally tight contests. This is solid stuff from Mughan, but aside from his opening discussion of the core concepts, it is more safe than exciting.