«by Dyron Keith Dabney A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy (Political Science) ...»
22 See Cox and Thies, 2000, Carlson, 2007 and the unpublished manuscript by Matthew Carlson, “Buying Votes in Japan's Lower House." Paper presented at the annual meeting of the The Midwest Political Science Association, Palmer House Hilton, Chicago, Illinois, Apr 15, 2004.
the unusual election system for Lower House elections, with medium sized, multimember, single-ballot districts. The details of this election system have been extensively analyzed by many election studies scholar cited and addressed in the previous chapter.23 Each district had three to five seats, filled by the top vote getters among the usual slate of nine to ten contestants. The most important point about MMD election system is that in order to win a majority a party had to average over two winners per district, and in fact the Liberal-Democratic party ran multiple candidates—often three or four—in nearly every district. An LDP candidate within range of gaining a seat often found it easier to pilfer votes from fellow LDP candidates than from opposition party candidates. The JSP similarly ran more than one candidate in each election district, but unlike the LDP, it found it increasingly difficult to field multiple candidates.24 In various ways, the system decreases the importance of party and increases the importance of individual candidates (including the chances for newcomers to break in to the legislative arena with little or no party support).
To many Japanese and Western political scientists intimately familiar with elections, structural causes like campaign laws and the election system are legitimate explanations for Japanese candidates’ distinctive campaign behavior. Yet, they are less convinced by the cultural explanation for candidates’ behavioral distinctiveness.25 On the other hand, most non-political scientists (in particular journalists, and many politicians 23 See Jain, 1993; Seligmann, 1997, Otake, et. al., 1998, Christensen, 1996 and 1998 and Reed, 1999.
24 In terms of party-level electoral strategies at the national level, that put a great premium on running the right number of candidates (lest they knock each other off), but we are concerned with the campaign strategies of individual candidates.
25 In the “rational choice” tradition, Mark Ramseyer and Frances Rosenbluth reject the cultural explanation. Likewise, Steven Reed rejects the cultural explanation. Gerald Curtis largely supports the structural explanation, but he does not ignore cultural factors.
little about the virtues of a parsimonious explanation and social (political) science theory.
A notable exception to these generalizations is Scott Flanagan, a distinguished and very theoretically oriented political scientist who takes cultural explanations very seriously.
In a seminal article published three years before Curtis’s first book, “Voting Behavior in Japan: The Persistence of Traditional Patterns” (1968),26 Flanagan argued that traditional attitudes and norms among Japanese voters and politicians led to an unusual emphasis on what he called (using sociologists Talcott Parsons’ famous “pattern variables”27) “ascriptive-diffuse” relationship between voter and candidate. In effect, what is at work in the campaign process is vertical, patron-client, face-to-face relationships, and by extension, social networks based on such face-to-face relationships.
Flanagan called this distinctive phenomenon “kankei voting,” using a term that means both relationships in general and “contacts” in a political context, particularly with respect to candidates’ organization and mobilization of a voter support base.28 Flanagan did not deny the importance of the campaign laws and the electoral system as explanations for campaign behavior, but he maintained that cultural factors had significant independent explanatory power--and indeed, helped account for why Japan instituted and continued such unusual structural conditions. Curtis (in his later The Japanese Way of Voting, 1988) rejected the application of the “patron-client” 26 “Voting Behavior in Japan: The Persistence of Traditional Patterns,” Comparative Political Studies, Vol.
1 (October 1968), 391- 412. Also see “The Japanese Party System in Transition,” Scott C. Flanagan, Comparative Politics, Vol. 3, No. 2 (January 1971), pp. 231-253; and Japanese Political Culture: Change and Continuity. Takeshi Ishida. (New Brunswick, N.J.: Transaction Books, 1983).
27 Parsons argued that instrumental and expressive interactions, called “pattern variables,” existed in societies.
28 It should be noted that this pattern resembles what V.O. Key called “friends-and-neighbor voting” in the American south, as depicted in the classic Southern Politics in State and Nation. (New York: Knopf, 1949).
social network relationships of a decidedly Japanese nature.
In the two decades or so after the pioneering research by Curtis and Flanagan, American political scientists explored Japanese elections from various angles. Some took implicit or explicit positions on the structure versus culture debate.29 Most interesting was a series of co-authored works by Flanagan and Bradley Richardson, because the former remained on the cultural side and the latter was essentially a structuralist.30 In the most important book on Japanese electoral behavior, the magisterial The Japanese Voter (1991), Richardson and Flanagan with Japanese colleagues tried to link analysis of social networks with the literature on electoral behavior using survey analysis.31 That task proved to be quite difficult, but the book includes solid accounts of Japanese elections from the voter’s point of view as it related to conventional tactics candidates used to establish and maintain voter support.
There has also been some attention to election campaigning per se, based on empirical observation. Curtis’s later books (1988 and 1999) present the overall framework and how it has changed very well. Satomi Tani wrote an excellent study on electoral keiretsu.32 That term (borrowed from its main usage to describe Japanese conglomerates) refers to deals between local politicians and Diet candidates, trading pork 29 In fact, this argument has pervaded American social science research on Japan: see John Creighton Campbell, “The Tasks of Social Science Research on Japan: Dispelling Stereotypes, Integrating Theory, Grappling with Culture,” background paper for the International Workshop on "Current and Future Trajectories of Social Science Research on Japan." Tokyo University, Institute of Social Science, November, 2006.
30 The work, Japanese Electoral Behavior (Thousand Oaks, CA: Sage, 1977), and particularly the book that became the standard text in the Japan politics field: Politics in Japan (Boston: Little, Brown, 1984).
31 Scott C. Flanagan, Shinsaku Kohei, Ichiro Miyake, and Bradley M. Richardson. (New Haven: Yale University Press).
32 Overviews are offered in their papers delivered at the Annual Meeting of the Association for Asian Studies, March 25-27, 1994
Haruhiro Fukui and Shigeko Fukai traced clientelist networks of politicians in both urban and rural electoral settings.33 It is fair to say, however, that the study of Japanese election campaign strategies per se in these years was largely a matter of filling in the picture drawn by Curtis.
However, the question of how candidates got elected in Japan was brought front and center by two developments, one theoretical (i.e., intellectual) and one in the practical, real world in nature. The intellectual development was the rise of what is loosely called the “rational-choice” approach, or more specifically principal-agent analysis, in the fields of American and comparative politics. These ideas were forcefully brought to the Japan field by Mark Ramseyer and Frances Rosenbluth in Japan’s Political Marketplace.34 They argued that nearly everything important about Japanese politics stemmed from its electoral system, which was maintained by the LDP to stay in power. This argument brought renewed interest to exploring the dynamics of MMD elections, Grofman, et. al. (1999) being a representative example. Indeed, a series of articles by the combinations of Gary Cox, Mathew McCubbins and other non-Japan specialists, and Japan-specialists like Frances Rosenbluth and Michael Theis coincided with real world developments in Japan’s electoral politics.
The trend in the real world was a growing debate about reforming the electoral system, which culminated in the law passed in 1994 that shifted the electoral system in the Lower House from MMD election system to the Mixed Member Majoritarian 33 See “Campaigning for the Japanese Diet,” in Elections in Japan, Korea, and Taiwan under the single non-transferable vote : the comparative study of an embedded institution, edited by Bernard Grofman, et.
al. (Ann Arbor: University of Michigan Press, 1999).
34 Cambridge MA: Harvard University Press, 1993.
This new election system was the most significant modification to the Public Officers Elections Law in decades. Substantial modifications also occurred to the Political Funds Control Law. For example, a public subsidy was provided to political parties upon meeting the funding conditions, corporations were banned from making direct contributions to kouenkai, and lower disclosure limits were mandated for political contributions. Instead of 1 million yen ceiling on donations before reform, the names of donors are now disclosed at 50,000 yen.35 The election reform legislation and how it came about has been extensively analyzed by political scientists (Jain, 1993; Babb, 1996;
Seligmann, 1997; Otake, et. al., 1998; Christensen, 1996 and 1998, Reed, 1999, etc.).36 A watershed of new scholarship on Japanese election campaigns occurred in response to the passage of election reform legislation in 1994.37 Whether and how the election reform signed into law by its architects would affect Japanese election campaigning became a major focus.
Evaluations of Election Reform Some politicians and journalists believed that Japan would finally attain the idolized, but elusive Western model of electoral politics—issue-oriented and partycentered elections—under the MMM election system. Yet, the vast majority of political science observers attentively deconstructing reform legislation and calculating the 35 For comprehensive details on the electoral laws see, Hrebenar, 1986 and Carlson, 2007.
36 Much of the literature on Japanese election campaigning in the last decade has been on the election reform, the election results and the consequences of election reform on political party strength. Much less has been written on candidates’ actual campaign strategies and tactics.
37 Great attention was given to the immediate, short-range impact of election reform on the election campaign. Thus, many works on the 1996 election campaign are available. Long-range analyses of campaign behavior have not been as abundant. The concentration of interest in subsequent elections waned as it became infinitely clear that evidence post-reform changes, if any, to Japanese election campaigning would take a series of general elections.
science scholars argued that reform enthusiasts were using an unrealistic and outdated standard of measurement for advanced democracies in the twenty-first century (Curtis, 1999; Reed, 2003). Issue- and party-oriented campaigning represented a paradigm that political candidates waxed poetic about, but in reality is virtually unattainable by any democratic polity. Western democracies even failed to live up to these standards. Upon investigating the first Lower House election under the new election system in 1996, Park, Otake, Isao and Masaaki (1998) claimed that excessive faith was placed in the SMD and PR election system to instantly produce policy issue- and party identification-oriented campaign behavior. Yet, they concluded that Japan would require several additional general elections before politicians (and voters) could comfortably appropriate the goals of election reform and cast aside the old campaign habits associated with the MMD election system. The conclusions they reached about the impact of reform underscore one of two outlooks about the direction of campaign behavior under the new election system; one outlook clearly being more positive than the other.
Otake, Isao, Park and Masaaki appeared mixed on the short and long term effectiveness of reform, but most political scientists argued that the change in the electoral system would eventually lead to more issue-oriented and party-centered election campaign behavior, and indirectly to more general changes: party factions and the powers of factional leaders would lose strength, and money politics subsequently would diminish. The key is that all of this probably will occur incrementally over future elections. For example, Reed (1997) maintained that the SMD and PR system will eventually lead to a two-party system (and subsequently party-centered campaigning),
could provide candidates with enough election security to forfeit a candidate-centered campaign behavior for party-centered and issue-attentive campaign behavior. In his book covering the 1993 to 2000 elections, Reed (2003) suggested that campaign behavior did not change rapidly because candidates were still adapting to the new election system.
Christensen (1998) was somewhat more positive-minded: he agreed that issue-centered and party-oriented campaign behavior among candidates will take several more elections, but he did find that some candidates already were conducting issue-oriented and partycentered campaigns in the 1996 election.38 His study, unfortunately, did not make clear whether such changes occurred with incumbents and political challengers—a not so insignificant distinction to make in light of the conventional wisdom on campaign behavior for incumbent and challenger behavior.