In this paper we provide evidence from Japan that bears on a general theory of agenda power in legislatures. By agenda power we mean the power to determine: (a) which bills are considered in the plenary session of the legislature and (b) restrictions on debate and amendment to these bills, when they are considered. While a substantial amount of work has focused on the second category of agenda power, including studies of special rules in the US House (e.g., Sinclair forthcoming), closure in the UK House of Commons (e.g., Cox, 1987; Dion, 1997), and the guillotine in the French National Assembly (e.g., Huber, 1996), there is very little on the first and arguably more fundamental sort of agenda power. This agenda power – the power to decide which bills will actually be considered on the floor of the legislature – is our focus here, and henceforth when we refer to ‘agenda power’ we shall mean this narrower conception.

This article employs a simple model of sophisticated voting under incomplete information and explores the strategic contexts of the vote on political reform bills in Japan. The government-sponsored political reform bills were voted down by the defection of government coalition members in the House of Councillors before a final compromise was reached in the joint committee of both houses and passed subsequently. In contrast to the accepted view that the defectors were short-sighted sincere voters, I show that Japan's institutional arrangements created an uncertainty about the agenda in the legislative process and led to the sophisticated voting behavior of pivotal voters whose preferences were different from the party leadership. The analysis underscores the importance of sophisticated voting for the empirical study of Japanese legislative politics.

Japan constitutes a very interesting and paradoxical case from the point of view of the place of trust in the processes of institution building and institutional dynamics. This problem has, of course, been the basic thrust of Durkheim's emphasis on the importance of precontractual elements for the fulfillment of contracts seemingly dealing with purely ‘utilitarian’ considerations. But this crucial insight – and problematic – has not been systematically followed up in the social science literature. Only lately it has been again taken up – initially, perhaps paradoxically – from within various rational choice approaches which have come to recognize that continuity of patterns of social interaction and of institutional frameworks cannot be explained by purely rational-utilitarian considerations (Braithwaite and Levi, 1998; Kramer and Tyler, 1993). At the same time the more recent analyses have also pointed to some of the complexities, paradoxes; and problems of the construction of trust in social interaction and institution building.

Japanese society is often said to be one with a high premium on social capital. Two major theses have been put forward with regard to social capital in the last few years. One, advanced by Putnam (1993), is that social capital enables democracy to work. In other words, the historically acquired and accumulated social capital in terms of the propensity of individuals to engage with others in community and associational life facilitates the task of democratically working out the resolution of conflicts of interest and collectively producing good public policy. The other, advanced by Fukuyama (1995), postulates that social capital allows the creation of prosperity. In other words, a high level of social capital enables business firms to take risks and stretch networks fully in the creation of wealth on a large scale for a prolonged period of time.

Developments in both China and Russia are a challenge to political science, and more particularly to theories of political culture. Both countries are engaged in profound processes of transition involving the abandonment of totalitarianism and the adoption of market-based economies. It is, however, far from clear what form their political systems will eventually take. They are currently following strikingly different paths. Are the differences a reflection of their distinctive cultures? Or, are the differences more structural, a manifestation of their respective stages of economic and social development? Or, are they merely the consequences of the idiosyncratic choices and policy decisions of the two leaderships?

With the 1989 eclipse of communist ideology and power in Central and Eastern Europe, the political order of democracy has, on the one hand, proved to be the superior way of organizing a society where in politics the pluralist interests of the people can be articulated and represented freely without fear of repression through competitive elections and otherwise, and where particularly through the operation of market mechanisms citizens are furnished with reasonably satisfactory economic circumstances to conduct their everyday lives. On the other hand, quite different from what many contemporary observers had anticipated, liberal democracy has been subjected to closer and closer critical internal scrutiny, and with this also alternate conceptions of how to organize a democratic polity are now more than before a matter of debate and controversy.

Prior to its publication, Gary Cox's Making Votes Count was widely and eagerly anticipated. (Indeed, some years ago, I received a referee report dismissing my submission as unnecessary because superior analysis would eventually appear in Cox's then forthcoming manuscript.) Upon its release in 1998, the book was instantly lauded: it collected multiple awards, including the prestigious Woodrow Wilson prize for best book published on government, politics, or international affairs. This acclaim was scarcely surprising – Cox has been one of the foremost scholars of elections and legislatures for the whole of his professional career. He is responsible for an impressive body of work spanning multiple research topics, nations, and methods of analysis. This book is testimony to his breadth, as it catalogs the key electoral features of virtually the whole set of modern democracies (more than 70), and makes frequent forays into diverse nations in search of empirical support for novel theoretical findings. In brief, Cox's project is to bring together a large formal, deductive literature on voting rules and social choice with an equally voluminous empirical, inductive literature on elections and party systems in the world's democracies. The prizes it garnered are one measure of the book's success at this merger; a large boost in Cox's swelling citation count will doubtless follow, reiterating the judgement that this is the major work on electoral law and voting to date.