The constructivist turn in the study of world politics provides new impetus to studies of the political deliberations of human agents. The co-constitution of agents and structures implies the non-acceptability of accounts that fail to consider the interpretations human agents provide to structural conditions. But neither can we accept the reverse. Studies of the interpretations of political agents should adequately account for the structural constraints on those interpretations. This paper illustrates how easily agency studies can underestimate structural constraints by reference to a most serious and scholarly account of agency in the Vietnam War commitment decisions, Yuen Foong Khong's Analogies at War: Korea, Munich Dien Bien Phu and the Vietnam Decisions of 1965. The argumentative burden resides with those who offer accounts that hold or imply that agents acted from non-structural motives.

The Republic of Korea (Korea hereinafter) has been widely regarded as one of the most vigorous and analytically interesting third-wave democracies (Diamond and Shin, 2000: 1). During the first decade of democratic rule, Korea has successfully carried out a large number of electoral and other reforms to transform the institutions and procedures of military-authoritarian rule into those of a representative democracy. Unlike many of its counterparts in Latin America and elsewhere, Korea has fully restored civilian rule by extricating the military from power. As is the case in established democracies of North America and Western Europe, free and competitive elections have been regularly held at all the different levels of the government. In the most recent presidential election, held in December 1997, Korea also established itself as a mature electoral democracy by elevating an opposition party to political power. In Korea today, there is general agreement that electoral politics has become the only possible political game in town.

The central theme of Japanese subnational governmental study has been center–local relations. Researchers looked into the question of whether Japanese subnational governments have sufficient autonomy to pursue their own preferences. In other words, the policymaking process of subnational governments in Japan has long been examined in relation to the national government. It is true that, under the centralized governmental system, the national government of Japan exerts a substantial influence on subnational policymaking. Students of subnational governments cannot avoid this issue. However, one cannot understand the reality of subnational policymaking only by looking at the center–local relations and the administrative institutions that define the relations. Indeed, one needs to ask what determines subnational policy outcomes, to what degree the national influence affects the determinants, and under what conditions these determinants function autonomously. The present paper directly examines the subnational policymaking process itself, not the center–local relations, and tries to understand its mechanism. The paper proposes a research framework using a statistical method that has been developed in the area of diffusion study. In the framework, national influence is integrated into an analysis of subnational policymaking as one of the factors that affect policy determinants. Using the framework, the paper analyzes the process of four regional policies, which will lead to a better understanding of subnational policymaking.

Patrick Fournier and Masaru Kohno (2000) have considered some apparent differences between my view of strategic voting in Japan (Cox 1997) and Steve Reed's (1990). I think they succeed in showing these differences to be small in most instances. Along the way, they note some problems with the use of the S–F ratio. In this note, I comment briefly on this latter issue.

Five major claims are made in our paper on strategic voting within the context of Japan's multimember single non-transferable vote (SNTV) electoral system (Fournier and Kohno, 2000). Two claims deal with the reconciliation of Steven Reed's (1990) and Gary Cox's (1997) important work on extending Duverger's law to the Japanese case, and three claims deal with the informational effects of partisan labels on strategic voting.

The most important election held in 2001 was that to the House of Councillors. Here, however, I will report on several surprising gubernatorial elections and the shocking LDP party presidential election. Each of these elections sent a similar message from the voters: ‘it is time for a change’. Powerful political machines using tried and true campaign techniques were repeatedly defeated by novices whose primary attraction was that they were not part of the political establishment.

Koizumi Junichiro's sweeping victory in the presidential primary of the Liberal Democratic Party (LDP) represents a palpable change in the national mood that may substantially affect the parliamentary conditions, although its impact on legislation is yet to be seen. In this essay, I briefly review the Diet in the past one year. In particular, the review deals with the legislative records in the 150–152 Diet sessions.

Political developments in Japan from the latter part of 2000 to the year 2001 were highlighted by the executive transition from unpopular Yoshiro Mori to exceptionally popular Junichiro Koizumu as the nation's leader.