Failed Democratization in Prewar Japan: Breakdown of a Hybrid Regime 9780804790741

Citation preview

Failed Democratization in Prewar Japan

T H E W A LT E R H . S H O R E N S T E I N ASIA-PACIFIC RESEARCH CENTER

Studies of the Walter H. Shorenstein Asia-Pacific Research Center Andrew G. Walder, General Editor The Walter H. Shorenstein Asia-Pacific Research Center in the Freeman Spogli Institute for International Studies at Stanford University sponsors ­interdisciplinary research on the politics, economies, and societies of contemporary Asia. This monograph series features academic and policyoriented research by Stanford faculty and other scholars associated with the Center.

also published in the shorenstein asia-pacific research center series New Challenges for Maturing Democracies in Korea and Taiwan Edited by Larry Diamond and Gi-Wook Shin (2014) Spending Without Taxation: FILP and the Politics of Public Finance in Japan Gene Park (2011) The Institutional Imperative: The Politics of Equitable Development in Southeast Asia Erik Martinez Kuhonta (2011) One Alliance, Two Lenses: U.S.-Korea Relations in a New Era Gi-Wook Shin (2010) Collective Resistance in China: Why Popular Protests Succeed or Fail Yongshun Cai (2010) The Chinese Cultural Revolution as History Edited by Joseph W. Esherick, Paul G. Pickowicz, and Andrew G. Walder (2006) Ethnic Nationalism in Korea: Genealogy, Politics, and Legacy Gi-Wook Shin (2006) Prospects for Peace in South Asia Edited by Rafiq Dossani and Henry S. Rowen (2005)

Failed Democratization in Prewar Japan breakdown of a hybrid regime

Harukata Takenaka

Stanford University Press Stanford, California

Stanford University Press Stanford, California ©2014 by the Board of Trustees of the Leland Stanford Junior University. All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or in any information storage or retrieval system without the prior written permission of Stanford University Press. Printed in the United States of America on acid-free, archival-quality paper Library of Congress Cataloging-in-Publication Data Takenaka, Harukata, 1971- author. Failed democratization in prewar Japan : breakdown of a hybrid regime / Harukata Takenaka. pages cm Includes bibliographical references and index. isbn 978-0-8047-6341-7 (cloth : alk. paper) 1. Democracy—Japan—History—20th century. 2. Japan—Politics and government—1912-1945. I. Title. jq1681.t339 2014 320.952'09043—dc23 2013038540 isbn 978-0-8047-9074-1 (electronic) Typeset by Bruce Lundquist in 11/14 Adobe Garamond

Contents

Figures and Tables

ix

Acknowledgments

xi

Introduction

1

Chapter 1—Analytical Foundations

23

Chapter 2—Defining the Regimes of Prewar Japan

45

Chapter 3—Political Institutions under the Meiji Constitution

77

Chapter 4—The Semi-Democratic Regime, 1918–1926: Containment of the Military

87

Chapter 5—The Semi-Democratic Regime, 1926–1929: The Rise of the Military and the Decline of Party Government

103

Chapter 6—The Semi-Democratic Regime, 1929–1932: Crisis and Breakdown

125

Conclusion

159

Notes

175

Selected Bibliography

207

Index

233

Figures and Tables

figures Figure 1.1

The erosion of legitimacy

38

Figure 1.2

Analytical perspective

42

Figure 2.1 Dahl’s model of democratization

71

Figure 2.2. Dahl’s model of paths of democratization

71

Figure 2.3. A new model of democratization

74

Figure 2.4. A new model of paths of democratization

74

tables Table 1.1.

Types of political systems

29

Table 5.1.

Number of tenant and labor disputes, 1913–1935

115

Table 5.2. Number of tenant unions and labor unions, 1918–1935

116

ix

Acknowledgments

This book has come a long way since its initial drafts, something for which I owe thanks to many people. To begin, I am greatly indebted to Daniel Okimoto, whose support ranged from general writing strategies and suggestions for its design to comments on individual chapters. Larry Diamond’s feedback was critical from the outset. He carefully read every chapter and commented on each extensively, which greatly improved them. I am also grateful to Peter Duus, whose advice shaped the empirical parts of this book, and David Brady, who reminded me of the importance of keeping a comparative perspective in mind. A number of people read the entire manuscript and improved it immeasurably. Mikuriya Takashi not only encouraged me to pursue publication, but also was very generous in sharing his knowledge of Japanese political history, while Masaru Kohno offered many insightful comments. I must also thank Andrew Walder for his very useful instructions on the manuscript’s revision, and for his patience as I finished it. The suggestions from the three anonymous reviewers were also very helpful. My editors of Stanford University Press, Stacy Wagner and Geoffrey Burn, were very tolerant and flexible as this book took shape over the years. Mitani Taichiro gave me important suggestions at an early stage on how to collect pieces of evidence to support my arguments. The comments of ­Gabriel Almond, the late Igarashi Takeshi, Miyazaki Ryuji, and the late ­Takahashi ­Susumu on how to approach the topic of this book were similarly vital. Kurt Gaubatz, Judy Goldstein, Mike McFaul, Terry Moe, Doug Rivers, Scott Sagan, Philippe Schimitter, and Barry Weingast all shaped my ideas in xi

xii  Acknowledgments

incalculable ways, while Jennfier Amyx, Amy Searight, and Yves Tiberghien gave me much academic inspiration. I cherish my friendships with Rebecca Bill Chavez and Mariko Yoshihara and the inspiration they have given me over the years. In Japan, I owe thanks to the faculty of the National Graduate Institute for Policy Studies, where I currently teach and do research. Shiraishi Takashi, Tsunekawa Keiichi, Kitaoka Shinichi, Iio Jun, and Masuyama Mikitaka always provide me with guidance, enlightenment, and intellectual stimulation. I must also thank my colleagues in Japan for their joint academic endeavors: Muramatsu Michio, Otake Hideo, Inoguchi Takashi, Adachi Yukio, Kabashima Ikuo, Tani Masami, Ono Koji, Tanaka Aiji, Kawato Sadafumi, Kobayashi Yoshiaki, Hiwatari Nobuhiro, Inoue Toshikazu, Kume Ikuo, Nonaka Naoto, Hirano Hiroshi, Katō Junko, Yamada Masahiro, Takenaka Yoshihiko, Karube Tadashi, Tatebayashi Masahiko, Uchiyama Yū, Kanai Toshiyuki, Makihara Izuru, Miura Mari, Takeda Okiyoshi, Horiuchi Yūsaku, Machidori Satoshi, Taniguchi Masaki, Taniguchi Naoko, Hosoya Yūichi, Takayasu Kensuke, and Fukumoto Kentarō. I thank the Ministry of Finance, and the National Personnel Authority of the Japanese Government for their financial support. Special thanks also go to Michael Armacost, Aoki Ken, Shiho Barbir, Jeanette Colyvas, Donna Fung, Greet Jaspert, Anne Marie Kodama, Zera Murphy, Mimi Nishimura, the late Michel Oksenberg, Yumi Onoyama, James Raphael, Thomas Rohlen, Henry Rowen, and Gi-Wook Shin. Thanks also to Megan Hendershott for her excellent edits. Finally, I thank my parents, Takenaka Haruhiko and Yayoi, for their genuine and enthusiastic support over the years, without which completing this book would not have been possible. Tokyo, Japan, January 6, 2014

Failed Democratization in Prewar Japan

Introduction

On May 15, 1932, Prime Minister Inukai Tsuyoshi was assassinated by naval officers in the official residence of the prime minister in Tokyo. On May 26, Admiral Saitō Makoto was appointed prime minister by the emperor, based on the recommendation of Genrō Saionji Kinmochi. Saitō’s appointment put an end to the practice of party government—in which leaders of political parties became prime ministers—and meant that the semi-democratic regime of prewar Japan had broken down. Following the breakdown of the semi-democratic regime, an authoritarian regime in which the military projected strong influence was established. This regime brought Japan to the Second World War.

Objectives and Challenges How and why does a semi-democratic regime—a regime that developed as a result of a significant degree of democratization—break down without experiencing further democratization? These are the questions I raise in this book. My answer emerges through a case study of changes in political regimes in prewar Japan. The semi-democratic regime is a subtype of hybrid regimes, which have been increasingly drawing scholarly attention in recent years.1 Hybrid regimes contain attributes of democratic regimes as well as authoritarian regimes. ­Examples of hybrid regimes include competitive authoritarianism, electoral authoritarianism, pseudo democracy, and so on. 1

  Introduction

I undertake a case study of a semi-democratic regime to advance research on hybrid regimes. It is only in recent years that students of political regimes have initiated intensive research on hybrid regimes. As a result, many interesting questions still remain to be answered. Studies on hybrid regimes have more or less focused on a particular, contemporary type, namely, those in which the nature of political competition can be characterized as competitive authoritarian or electoral authoritarian.2 In these regimes, although elections are regularly held, political competition is not fair nor free and authoritarian incumbents have a huge advantage over the opposition. Yet, hybrid regimes are not limited to those that have been intensively studied. They can be multidimensional, just like democratic regimes.3 There can be hybrid regimes in which electoral control over political offices may be different than in democratic regimes. Furthermore, hybrid regimes are not limited to the contemporary era. Historically, there are also instances of such regimes. This study on a semi-democratic regime is an effort to expand our knowledge of other types of hybrid regimes. I define a semi-democratic regime as follows: 1. Even when there is competition for political offices, and elections are held regularly to fill political offices, they are not fully free or fair. Civil rights that are necessary to make political competition and elections free and fair—such as freedom of expression and association—are not sufficiently protected. 2. Not all effective political offices, that is political offices that have power to influence policy formulation, are held accountable to the electorate through elections. In other words, even when free and competitive elections are regularly held, there remain “reserved domains”—political offices that are not accountable to the electorate but can exercise significant political power. 3. Only a portion of the population has the right to vote. Semi-democratic regimes develop into democratic regimes when they fulfill the following conditions: 1. There is competition for political offices, and the people select political officeholders through free, fair, and regularly held elections. Civil rights that are necessary to make political competition and elections free and

Introduction  3

fair, such as freedom of expression and association and equality among votes, are protected. 2. All effective political offices are held accountable, either directly or indirectly, to the electorate through elections. In other words, there are no “reserved domains,” and no political offices can project significant political power if they are not accountable to the electorate, either directly or indirectly. 3. A significant portion of the population (normally, all adults) have the right to participate in elections. Judged by these definitions, Great Britain, for example, was semi-­ democratic in the latter half of the nineteenth century. With respect to political competition, two major political parties competed against each other and elections were regularly held. Yet, they were not completely fair or free because of election fraud. With respect to control over political offices by the electorate, not all effective political offices were yet accountable, either directly or indirectly, to the electorate. This was because the House of Commons’ superiority to the House of the Lords had not yet been established, although by the 1870s the practice arose that the cabinet would be formed from the majority party in the House of Commons. On the last point—electoral participation—even after the reform of 1884, only 16 percent of the total population had the right to vote. Great Britain came to fulfill the three conditions for a democratic regime with the reform of 1911, which established the superiority of the House of Commons over the House of Lords, and the reform of 1928, which established universal suffrage. Between 1918 and 1932, Japan, with its practice of party government, also fulfilled the three conditions for a semi-democratic regime. First, there was significant political competition, but it was not fully free and fair. During this period (with the exception of a few years), two major parties competed against each other and formed cabinets. Although they were not completely free and fair, elections were regularly held and determined the distribution of power in the Diet. On the second dimension—the electorate’s control over political offices once party government had been established—the cabinet was indirectly accountable to the electorate and the Lower House was directly accountable. Not all political offices, however, were accountable to the electorate; institutions such as the genrō, the House of Peers, the Privy Council, and the military were not subject to electoral control. On the third

  Introduction

dimension—electoral participation—there were limits. With the reform of 1919 males who paid an annual tax of at least three yen could vote. The reform of 1924 implemented universal male suffrage. Although Great Britain made a transition from a semi-democratic regime to a democratic regime, Japan did not. With the May 15 Incident in 1932, party government was terminated and the semi-democratic regime collapsed. After a four-year period of transition, there emerged an authoritarian regime led by the military (see Chapter 2). In other words, Japan’s prewar democratization was aborted in the early 1930s.

Studies of Hybrid Regimes This book’s analysis of the collapse of the Japanese semi-democratic regime expects to make two contributions to the literature on comparative politics. First, it expands our understanding of hybrid regimes. A recent surge in scholarly attention to hybrid regimes obviously reflects a teleological bias in the study of the so-called third wave of democratization. 4 Guillermo O’Donnell and Philippe Schmitter’s seminal research on democratic transitions has emphasized transitions from authoritarian to other types of political regimes, and has noted that not all transitions necessarily lead to democratic regimes.5 As Thomas Carothers has pointed out, however, there has been an implicit assumption in the literature that “any country moving away from dictatorial rule can be considered a country in transition toward democracy.”6 Research on democratization has often been based on the assumption, whether implicit or explicit, that a nondemocratic (authoritarian) regime can make the transition to a democratic one within a relatively short time.7 When we look at countries with long histories of democracy, it is easy to see that they did not move quickly from being nondemocratic to democratic. Most of these countries “gradually expanded the suffrage, reduced plural voting, introduced the secret ballot, and established the responsibility of prime ministers and cabinets to parliaments.”8 Although some scholars have argued that it is necessary to “telescope such a long historical process [of democratization] into a few critical years,” 9 that many scholars have modified the term “democracy” with various adjectives such as “pseudo,” “illiberal,” and “delegative” shows that for many countries such telescoping is very difficult.10 The development of a democratic regime takes time, even in our contemporary era.

Introduction  5

Today scholars pay more attention to hybrid regimes for two reasons. First, the number of countries that cannot be classified as democratic regimes or authoritarian regimes has simply increased. According to Larry Diamond, as of the end of 2001, out of 150 countries with a population of more then one million, 55 can be categorized as hybrid regimes.11 In addition, there is growing recognition that many hybrid regimes are not making the transition to democratic regimes. They just endure as they are.12 For example, in their research on competitive authoritarianism, a sub-category of authoritarianism, which has at least one of three characteristics—unfair elections, imperfect protection of civil liberties, and immense advantage enjoyed by incumbents—Levitsky and Way argue that out of the 35 cases they examine, 19 competitive authoritarian regimes persisted for at least fifteen years.13 Certainly students of political regimes have always been aware of hybrid regimes. For example, Robert Dahl refers to the development of “near polyarchies,” “competitive oligarchies,” and “inclusive hegemonies.”14 Yet, because of the aforementioned bias in the literature on democratization, it is only in r­ ecent years that scholars have embarked on intensive research on hybrid regimes.15 Studies so far have focused on such issues as the classification of hybrid r­ egimes, political dynamics under hybrid regimes, and conditions for the endurance of hybrid regimes as well as for their development toward democratic regimes. These studies have made important contributions to our understanding of hybrid regimes. However, there is further room for expanding and deepening our research on hybrid regimes, following their lead. Previous studies have three elements in common. First, they concentrate on contemporary hybrid regimes. Second, while they are aware of diversity among hybrid regimes, they tend to focus on one subtype, variously labeled as electoral authoritarianism, competitive authoritarianism, or pseudo democracy.16 This subtype has two characteristics: first, elections are regularly held yet are neither fair nor free; second, authoritarian incumbents have a huge advantage over the opposition in elections. Lastly, while previous studies are concerned with the transition of competitive authoritarian regimes to democratic regimes as well as their persistence, they overlook the possibility that hybrid regimes may break down and turn into classic authoritarian regimes. The current state of the literature on hybrid regimes has several implications for further research. To begin with, historically there have been many instances of hybrid regimes. The experience of traditional democracies such

  Introduction

as Great Britain tells us that, historically, democratization unfolds over a very long time. After autocratic, despotic, or authoritarian regimes begin to democratize, free competition among politicians, regularly held elections, guarantees of various political rights, and control over political offices gradually emerge. These first democratic steps transform nondemocratic regimes into hybrid regimes. One important subtype of hybrid regimes deserves more attention: r­ egimes in which nonelected officials can have substantial political influence. The literature on competitive authoritarianism as well as electoral authoritarianism, after all, focuses on the nature of political competition. In this subtype, which does not encompass all hybrid regimes, some political offices, in particular the military, which are not held accountable to the electorate neither directly nor indirectly, may wield significant political power. Finally, we should be aware of the possibility that hybrid regimes can turn into classical authoritarian regimes. When hybrid regimes break down, exploring the circumstances of their collapse is likely to enhance our knowledge of the dynamics of the transformation of political regimes. Bearing the aforementioned points in mind, I follow the lead of recent studies by focusing on a semi-democratic regime, a subtype of hybrid regimes that has often been left out of current research. I emphasize that some countries may go through two steps in the process of making a transition to a democratic regime; that is, developing into a semi-democratic regime first and then developing into a democratic one—a possible course also recognized in traditional arguments on democratization.17 If a semi-democratic country is to make the transition to democracy in two steps, it is important that it stabilize before proceeding with further democratization. This raises questions about the conditions that determine whether regimes stabilize or collapse. In particular, I explore the conditions under which a semi-democratic regime breaks down—an issue not addressed in the literature—and propose an analytical framework to explain such a collapse, paying particular attention to the timing and process of the breakdown.

Prewar Japanese Political Development and Comparative Politics The second objective of this research is to enhance our understanding of prewar Japanese politics. As a result of past research, there is an extensive literature on political development in prewar Japan, particularly in the 1910s

Introduction  7

and early 1920s as well as the late 1930s. Furthermore, such political issues as the labor movement and social policy in prewar Japan have been the subject of past research. Yet, much work remains to be done to understand political development between the mid-1920s through the mid-1930s. Thus, my analysis of political development from the late 1910s to mid1930s provides what has been a missing link in the literature. I reconstruct political development during this period within the same, single framework in order to show why political parties that gained a huge amount of political power in the late 1910s lost it by the early 1930s. In discussing the period between the mid-1920s and mid-1930s, I briefly refer to the state of research on prewar Japanese politics. The literature on how democratization in Japan proceeded until the early 1920s is extensive. While this literature focuses on political parties and shows how they came to dominate politics, few works discuss what happened to political parties once they came to power. For example, Tetsuo Najita’s discussion of the political development of Japan between 1905 and 1915 focuses on the powerful leader Hara Kei and describes how he expanded the power of the Seiyūkai, the political party he led, but does not reach the year 1918, when the Seiyūkai came to dominate government and the semi-democratic regime emerged.18 Peter Duus takes a more comprehensive view of the growth of political parties in prewar Japan by drawing attention to such political parties as the Dōshikai and Kenseikai, which competed against the Seiyūkai, to demonstrate how a two-party system emerged in prewar Japan.19 Development of a two-party system was certainly an important part of prewar Japanese democratization, yet Duus’s analysis ends when the Minseitō, Seiyūkai’s rival, was formed in 1927. Duus does not discuss a series of events between the late 1920s and mid-1930s that weakened political parties and triggered the collapse of the semi-democratic regime. Murai Ryōta also traces how the practice of party government developed, focusing on the process of nominating prime ministers,20 yet he again covers only the period until 1927 and does not examine why the practice did not continue. Kitaoka Shinichi describes the history of Japan, focusing on politics between 1924 and 1941,21 yet again there is a problem with the period covered since his narrative does not deal with the initial phase of the semi-democratic regime. Gordon Berger also describes political development in prewar Japan by focusing on political parties.22 Yet, he discusses only the period between 1931 and 1941. Berger is mainly concerned with how political parties re-

  Introduction

tained power even after the year 1932 when democratization was aborted and p ­ olitical parties could not make their presidents into prime ministers.23 Banno Junji gives an account of the period between 1918 and 1932 as a part of his political history from the Meiji Restoration through the late 1930s.24 However, he emphasizes particular political events such as the signing of the London Naval Treaty by the Hamaguchi cabinet or a failed attempt to make a grand coalition between the Seiyūkai and Minseitō. In short, none of the aforementioned studies covers political development between the late 1910s and early 1930s within a single analytical framework. As a result, they cannot explain why political parties, which came to dominate politics by the early 1920s, grew politically weak from the mid1920s and ultimately why democratization failed in prewar Japan. Although several scholars have studied extensive periods of prewar ­Japan, which include the period from the late 1910s through the early 1930s, the main subject of their analysis is not failed democratization but other political issues. For example, Andrew Gordon traces the history of the labor movement in general between 1905 and 1940 while undertaking a case study of Nankatsu, an area in the eastern part of Tokyo.25 Sheldon Garon traces the formulation of social policy in the prewar period, paying particular attention to the period between 1918 and 1945.26 Gregory Kasza traces government policy on various types of mass media such as the press, film, and radio.27 Kasza has also examined the development of administered mass organizations orchestrated by the prewar state to penetrate into society.28 Richard Samuels examines the role of leadership in history by looking at important political figures in prewar Japan such as Itō Hirobumi and Hara Kei.29 Aside from these scholars, Maruyama Masao addresses the question of why the military intervened in politics and a fascist regime emerged in prewar Japan.30 Maruyama argues that military intervention became possible and a fascist regime developed because of the fragmentation of political institutions under the Meiji Constitution. Since no political actor or political institution had consolidated political power, it was very difficult to stop the military once it had begun to intervene in politics. While Maruyama captures the inherent defects of the Meiji Constitution, he cannot explain why different political regimes emerged under that constitution. In other words, it is difficult to explain why democratization was aborted after having made significant progress by just looking at the constant influence of the Meiji

Introduction  9

Constitution. Further, he does not chronologically trace how the balance of power among different political actors changed. While all these works contribute to understanding their respective subjects, they do not directly address why prewar Japanese democratization failed and why the semi-democratic regime came to break down. In the extensive literature on prewar Japanese politics there are but two major works that address why democratization in prewar Japan failed—one by Robert Scalapino and the other by Barrington Moore. Scalapino conducted a single case study to examine why democracy failed in Japan before the Second World War.31 He locates its failure in the demise of political parties and tries to explain why they could not retain their power once they had acquired it. The failure of democracy in prewar Japan, Scalapino believes, was a deviant case, because Japan was unable to develop a full-fledged democracy despite various favorable socioeconomic conditions. He argues: “In some respects, Japan resembles a ‘controlled experiment.’ The presumed prerequisites for a modern democratic society— the independent nation-state, industrialization, and mass literacy—were all attained by modern Japan. Thus, one is not dealing with a society, which lacked the elementary democratic ‘requirements.’”32 Scalapino emphasizes the roles of two structural variables to make a deterministic argument that democracy in Japan was doomed to fail. Democracy could not develop in prewar Japan, he argues, because of the inherited feudal tradition and the timing of economic development. The feudal tradition, which emphasized the family system and the importance of the group, and discouraged respect for individual rights, was a significant obstacle to the emergence of democracy. Japanese capitalism was unable to break free of this tradition and spread liberalism. In the West, by contrast, industries developed on their own. They, together with the newly created middle class, overwhelmed the feudal tradition, challenged the government, and spread liberalism, thereby establishing a strong foundation for the development of democracy. In Japan, industries themselves retained the feudal tradition. Scalapino also claims that the timing of economic development in Japan discouraged political parties and democracy from developing. In Japan, where the state took the initiative in the race to catch up with the West economically, industries depended on the state. As a result, they and the middle class could not challenge the government and push for full-fledged democracy. Instead they often colluded with the government; even when

  Introduction

they supported the parties their support was given to enhance their parochial interests and not to promote democracy. Moore argues that there are three distinct routes to modernization: “bourgeois revolution,” “revolution from above,” and “communist revolution.” To explain how the countries he examines followed different paths, he operationalizes three variables: the strength of the bourgeoisie, the mode of commercial agriculture, and the potential for peasant revolution. For Japan, he argues that a particular combination of these variables explains the rise of Japan’s fascist regime. In Japan, where the potential for peasant revolution was weak, a coalition between the commercial-industrial elites and the landed upper class with a “labor repressive” mode of production brought on the Meiji Restoration and ultimately led Japan into a fascist regime. Moore also argues that democracy in Japan was destined to fail and that the Depression was a mere catalyst because Japanese capitalists (probably identical to commercial elites) could not challenge the state since they depended on it. These same capitalists also did not distribute the benefits of capitalism across a wide range of the population, thereby failing to raise the stakes of preserving capitalist democracy among the people. Moore adds that the plight of the peasants and petty bourgeoisie under capitalist development drove the military to intervene in politics and led to the rise of a fascist regime, although the coalition between commercial elites and landlords remained. Scalapino and Moore mainly rely on structural factors, such as economic and social structures and tradition, to make deterministic arguments about the nature of political regime development. They show that structural factors imposed severe constraints on the development of democracy in prewar Japan. The problem with this approach, however, is that while structural factors do not vary in the short-to-medium term, countries often experience different types of political regimes in such a time frame. Despite the constant constraints of structural factors, different political regimes existed in prewar Japan. It is important to note that the semi-democratic regime which existed between 1918 and 1932 was replaced by a military authoritarian regime. This implies that other factors are at work in triggering changes in political regimes. In addition, their research has become dated. Moore’s research dates back more than thirty years, and Scalapino’s more than half a century. In recent years, the study of democratization has advanced, and now focuses on the process of democratization itself. It is helpful to incor-

Introduction  11

porate insights from such recent advances in order to create an analytical framework to explain the breakdown of semi-democratic regimes. Being aware of these two problems, I undertake a case study of Japanese political development and explain the collapse of the semi-democratic regime in prewar Japan. In constructing an analytical framework to explain the collapse, I use insight gained from studies in comparative politics, in particular in the field of democratization. In constructing a framework, I take note of the objective of social science: the development of theory.33 Theory has to explain why a phenomenon happens accurately, parsimoniously, and generally.34 Parsimony means that theory has to explain why a phenomenon happens with as small a number of variables as possible; generality means that it has to explain as many cases as possible. In social science, however, it is very rare that completely identical phenomena happen more than once. As each of several phenomena of the same kind has original elements of its own, tensions exist between accuracy on one hand and parsimony and generality on the other.35 That is, when a theory tries to explain one particular case as accurately as possible, it loses parsimony and generality; when it tries to explain as many cases as possible, it loses accuracy. Assuming that such a tension exists, and recognizing that a theory cannot capture all elements relevant to a phenomenon, I attach particular importance to parsimony and generality. In other words, a theory should single out several factors that are most relevant in causing a given phenomenon and, by concentrating on those factors, explain as many cases as possible. Thus, in developing an analytical framework for the breakdown of semidemocratic regimes, I will not try to capture every factor that is relevant but instead concentrate on factors that are most relevant. In this Introduction, I have so far introduced the book’s main themes and objectives. In the remaining section, I discuss in greater detail the nature of the semi-democratic regime and how it differs from other political regimes, such as democratic and authoritarian regimes. In Chapter 1, I discuss the theoretical foundations for developing an analytical framework on the breakdown of semi-democratic regimes. After introducing existing approaches that can be applied to explain the breakdown of semi-democratic regimes, I propose a new analytical framework. In Chapter 2, I show that Japan was a semi-democratic regime between 1918 and 1932 in the light of the definitions I have introduced in this Introduction. Chapters 3, 4, 5, and 6 explain the breakdown of the semi-­democratic

  Introduction

regime in prewar Japan, relying on the analytical framework presented in Chapter 1. In the Conclusion, I summarize the book’s arguments and consider their implications for research on democratization.

Defining the Semi-Democratic Regime As I have argued above, the semi-democratic regime is one that has already undergone a significant degree of democratization. It is in this sense a h­ ybrid. There have been three particularly influential definitions—two recent and one almost two-decades old—of such regimes. The first, by Schedler, describes them as “electoral authoritarian”: “Electoral authoritarian regimes play the game of multiparty elections by holding regular elections for the chief executive and a national legislative assembly. Yet they violate the liberaldemocratic principles of freedom and fairness so profoundly and systematically as to render elections instruments of authoritarian rule rather than ‘instruments of democracy.’”36 According to the second, by Levitsky and Way, such regimes are “competitive authoritarian”: “Competitive authoritarian regimes are civilian regimes in which formal democratic institutions exist and are widely viewed as the primary means of gaining power, but in which incumbents’ abuse of the state places them at a significant advantage vis-à-vis their opponents.”37 Although the terminology differs and there are nuanced differences between the two definitions, electoral authoritarian regimes and competitive authoritarian regimes have common characteristics.38 That is, electoral competition is neither fair nor free and incumbents have a huge advantage over the opposition due to their manipulation of elections and the state apparatus. The third definition is from Diamond, Linz, and Lipset: [A regime is semi-democratic when] the effective power of elected officials is so limited or political party competition so restricted, or the freedom and fairness of elections so compromised that electoral outcomes, although competitive, do not produce true popular sovereignty and accountability, or in which civil and political liberties are so uncertain that some political orientations and interests are unable to organize and express themselves peacefully, without fear.39

None of these definitions, however, captures the full gamut of hybrid regimes. As Levitsky and Way put it, “there are multiple ways to be partially democratic.”40 To encompass other subtypes of hybrid regimes I begin by

Introduction  13

examining a classic definition of the democratic regime. Here I refer to a definition of democratic regimes by Robert Dahl. He proposed the following eight conditions as requisite for democratic regimes:41 1. Freedom to form and join organizations; 2. Freedom of expression; 3. The right to vote; 4. Eligibility for public office; 5. The right of political leaders to compete for support; 6. The availability of alternative sources of information; 7. Free and fair elections; 8. Institutions for making government policies depend on votes and other ­expressions of preference.

A semi-democratic regime can be thought of as one that, as a result of democratization, satisfies these eight conditions to a significant degree. In defining a semi-democratic political regime, it is necessary to reflect on what democratization essentially means. As Dahl argues, the essence of democratization is the development of a political system in which the government responds to the preferences of its citizens, whom it considers to be political equals.42 Dahl also notes that democratization consists of two dimensions—the increase of public contestation and the expansion of participation. In other words, under a democratic regime, free and fair political competition emerges, as do people who have the right to participate in elections. Although Dahl nicely captures the essence of democratization, he fails to consider one important element—the increase in electoral control over political offices. Democratization must lead to a situation in which all political offices are accountable to the electorate, either indirectly or directly, a point often emphasized in recent research on democratization.43 This is an important point. Even when fair and free competition emerges in politics and the people have the right to participate in elections, if some political offices are not accountable to the electorate and can exercise power freely, government cannot respond to its citizens’ preferences. Therefore, it is necessary to consider the expansion of electoral control over political offices as an important facet of democratization. Thus, it is possible to consider democratization as a process consisting of the following three components: the development of free and fair political competition; an increase in electoral control over political offices; an increase in the number of people with the right to participate in elections. The

  Introduction

concept of electoral authoritarianism as well as of competitive authoritarianism acknowledges that the second and third can be dimensions of hybrid regimes but attaches greater importance to the first. Yet, if we are to enhance the scope of research on hybrid regimes in order to capture historical cases, it is necessary to avoid slighting the other two dimensions. Also the concept of semi-democracy is useful in analyzing contemporary hybrid regimes for which the electorate has limited control over political offices. The definition of semi-democratic regimes I have introduced at the beginning of this Introduction captures the three dimensions of democracy. When democratization proceeds along all three dimensions, a democratic regime will finally emerge. My definition is similar to that of Diamond, Linz, and Lipset, although they emphasize the state of electoral competition, while I attach particular importance to the second dimension. How, then, are democratic regimes defined in the way introduced at the beginning of this Introduction? Larry Diamond introduces two definitions of democracy: electoral democracy and liberal democracy.44 Electoral democracy attaches importance to competition for political offices and free and fair elections. Liberal democracy adds three more requirements. First, all political offices must be accountable to the electorate. Second, there must be horizontal accountability among political offices. Third, pluralism must be firmly established. The definition of democracy I adopt in this book can be located halfway between electoral and liberal democracy. The definition of electoral democracy is insufficient in that it does not clearly capture the dimension of putting all political offices under the control of the electorate. On the other hand, the notion of liberal democracy is susceptible to problems of its own. A definition of liberal democracy can be analytically useful for examining the conditions under which a government responds to its citizens’ preferences. However, when we seek to identify the minimum conditions for developing a political system in which a government will be responsive to its citizens, a detailed definition such as liberal democracy raises too many different issues, rendering arguments difficult and cumbersome.

The Semi-Democratic Regime and Other Political Regimes How does the semi-democratic regime differ from other political regimes? I have so far considered the differences between it and the democratic regime,

Introduction  15

and have also demonstrated how it is distinct from electoral authoritarianism and competitive authoritarianism. Below I examine differences between the semi-democratic regime and other nondemocratic regimes. My definition is different from that provided by other scholars and is used in different ways. The term “semi-democratic regime,” as defined by other scholars, is often used to highlight that elections are unfair, as in “electoral authoritarianism” and “competitive authoritarianism.” For example, according to William Case, in the semi-democratic regime “governments regularly hold elections, thus offering a snapshot of propriety on voting day. But they have limited civil liberties beforehand, thereby hindering opposition parties in contesting effectively. . . . Government candidates may make heavy use of state resources in campaigning, practices winked at by election commissions.”45 Likewise, Paul Brooker states that semi-democratic regimes are democratic regimes with serious flaws. They suffer from limitations on suffrage and sovereignty, systemic weaknesses, shirking that involves the misuse of public power to influence elections, and semi-competitive elections that are difficult to distinguish from those of disguised dictatorships.46 In Brooker’s eyes, the key characteristic of the regime he calls semi-democratic is unfair elections. One other political regime that is similar is the tutelary regime. In the words of Samuel Fitch, “in tutelary regimes, the armed forces participate in the policy process and exercise oversight over civilian authorities. The military’s share of power within such regimes may vary, although their implicit veto power is usually respected when the issue involves intense and widespread military pressure.”47 However, there is a major difference between semi-democratic regimes and tutelary regimes. Political offices in the semi-democratic regime that are not controlled by the electorate are not necessarily limited to the military. Valenzuela calls these offices “reserved domains” and refers to the monarch and the high civil service as examples.48 Members of the second chamber and of institutions that assist the monarchy are also candidates for such political offices. I would add that there can be other political offices that are not held accountable to the people. Also, the adjective “tutelary” itself has a drawback. Tutelary as an adjective signifies “serving as a protector, guarding, or ­patron.”49 It reflects that the military intervenes in politics because it perceives itself as the guardian of the country.50 Yet, the term does not really capture the reality that the military remains outside the control of the electorate.

  Introduction

As I have suggested earlier, there have been historical instances of semidemocratic regimes. Yet, semi-democratic regimes are different from traditional nondemocratic regimes such as competitive oligarchical regimes and authoritarian regimes. Dahl refers to competitive oligarchical regimes, which accurately describes Great Britain during the period when its two major party system gradually emerged.51 Although Dahl does not give a detailed definition of the competitive oligarchical regime, his discussion of democratization reveals that the competitive oligarchical regime is one in which public contestation has significantly increased but participation has not. In other words, in a competitive oligarchy, there exists a significant degree of competition in politics and political rights that are necessary to secure political competition. Even so, the number of people with voting rights is quite limited. Thus, the competitive oligarchic regime and the semi-democratic regime are similar in that significant political competition takes place, but different in that the number of people who participate electorally is much more limited in the former. Finally, we should distinguish between authoritarian regimes and semidemocratic regimes. Juan Linz, who pioneered research on authoritarian regimes, defined them as “political systems with limited, not responsible, political pluralism: without [an] elaborate and guiding ideology (but with distinctive mentalities); without intensive nor extensive political mobilization (except [at] some points in their development); and in which a leader (or occasionally a small group) exercises power within formally ill-defined limits but actually quite predictable ones.”52 Linz originally used this definition to classify regimes that are neither totalitarian nor democratic.53 Limited political pluralism in an authoritarian regime means that there is some diversity in the supporting groups of the regime. It also means that the state does not dictate the operation of all of society (which a totalitarian regime tries to do), and that some groups are independent of the state, particularly in the economic and social arenas. The political leaders in an authoritarian regime are not accountable to the citizens, but rather to the ruling class. While there are often single parties supporting such regimes, when it comes to politics the people are passive or apathetic. Even when citizens participate in politics, it is often because of the regime’s encouragement. Under an authoritarian regime, political groups with power do not ask people to demonstrate explicit support. Likewise, an authoritarian regime, unlike a totalitarian regime, does not require a commitment from the peo-

Introduction  17

ple to a certain ideology. Rather, an authoritarian regime is characterized by a diffusion of mentalities, which only vaguely stipulate various ideas from many sources, reflecting the diversity of the groups that support the regime. “Ideology” clearly refers to a set of ideas to which its believers strictly adhere, while “mentalities” vaguely describes “noncodified ways of reacting to different situations.”54 In other words, “the complex coalition of forces, interests, political traditions, and institutions—part of the limited pluralism—requires the rulers to use as symbolic referent the minimum common denominator of the coalition.”55 Though it may be modest, the predictability of exercising power in an authoritarian regime derives from respect for legal order and procedural legality. Because of the lack of an ideology, repression is often directed toward what people do against the regime and not toward what they conceive in their minds. In a totalitarian regime, ideas themselves become subject to repression. Semi-democratic and authoritarian regimes are similar in that under both regimes political offices without accountability to the people can pro­ ject significant political power. In an authoritarian regime, some political offices may be subject to control from the ruling class. However, the way political offices are controlled is very informal and not institutionalized. In a semi-democratic regime, by contrast, some offices are accountable to the people through elections, although elections may not be perfectly fair. Furthermore, political competition is much more limited under authoritarian than under semi-democratic regimes. Elections and political parties may exist in authoritarian regimes, but even when they do a single party supporting the regime holds a dominant and privileged status, which ­reduces the significance of electoral competition. In a semi-democratic regime, there is a significant degree of competition among political parties, and often power is transferred from one political party to another as a result of elections, even if elections may not be perfectly fair. While in an authoritarian regime the people only passively participate in politics, in a semidemocratic regime the people often voluntarily and actively participate.

Examples of Semi-Democratic Regimes The political regime that existed in Japan between 1918 and 1932 was semidemocratic. Although elections were regularly held and competitive, they

  Introduction

were neither perfectly free nor fair. Several offices, the Privy Council, and the House of Peers remained outside electoral control. Above all, the military, which was not subject to civilian control, wielded significant influence in politics. Lastly, only men could vote after the introduction of universal male suffrage in 1925. I describe the semi-democratic nature of the regime in more detail in Chapter 2. great britain

Among other examples of semi-democratic regimes is Great Britain from the middle of nineteenth century until the turn of the century. During this period competitive elections were regularly held. Yet, they were not free and fair. Bribery and coercion were prevalent in elections.56 It was very common for candidates to buy votes. The more money voters were offered, the more heated elections became.57 In addition, employers projected coercive influence over their employees on how they would vote.58 Moreover, an important political institution, the House of Lords, which could block legislation passed by the House of Commons, was not accountable to the electorate. And it was not until the end of 1860s to the early 1870s that civilian control over the military was established.59 Lastly, suffrage was limited. Even after the expansion of suffrage with the Reform Act of 1884 only 16 percent of the total population could participate in elections.60 I will more fully compare the semi-political regime of Japan with that in Great Britain in Chapter 2. There are also examples of semi-democratic regimes in the period after the Second World War. Here one must note that the degree of electoral participation in the postwar era in the two examples explored below—­Brazil and Thailand—was comparatively very high given that they, like most countries, had adopted universal suffrage. brazil

Brazil between 1945 and 1964 is often considered democratic—for example, Thomas Skidmore treats it as such61—but in fact Brazil during this period was semi-democratic. Three important political offices—the presidency, the Senate, and the Chamber of Deputies—were selected through elections and elections were regularly held. Yet, they were not fully fair or fair. First, the Communist Party of Brazil was banned from participating in elections after 1947 and

Introduction  19

the senators and deputies of the Communist Party elected in the election of 1947 were expelled.62 Second, there remained the tradition of “coronelismo,” as provisional oligarchies could control and give the votes of “most of those dependent upon [them] for a livelihood” to candidates of their choice.63 Control of the electorate over political offices posed another problem. This was because an important political office, namely, the military, which had a great deal of political influence, was not accountable to the electorate. The military was given prerogatives by the constitution. Article 176 stipulated that the military is “under the supreme authority of the President of the Republic, and within the limit of the law,” and Article 177 that “it is the mission of the Armed Forces to defend the Country and guarantee the constitutional powers, as well as law and order.” Thanks to these articles, the military could legally intervene in politics as a “moderating power.”64 Exploiting these prerogatives, it indeed intervened in politics a number of times during this era. It demanded President Getúlio Vargas’s resignation in August 1954, which led to Vargas committing suicide. It staged a coup d’état in November 1955 to secure the inauguration of Juscelio Kubischek de ­Oliveria, who won the presidential election in October. Some officers opposed Vice President João Goulart becoming president from August to September 1961 after the sudden resignation of President Jânio Quadros. The opposition by some military officers led to the introduction of a parliamentary system, which was intended to weaken the power of the president. Finally, the military intervention in 1964 led to the collapse of the semi-democratic regime. thailand

From 1978 to 1997 Thailand also was semi-democratic.65 First, although elections were regularly held and competitive, they were neither completely free nor fair since they were corrupted by vote buying.66 Candidates, systematically distributing money to secure votes, hired leading figures in the area as huajhanaen (often translated as canvassers) to distribute cash and attract support.67 Sometimes candidates even bribed officials counting votes.68 Second, important political offices remained outside the control of the electorate. To begin with, the prime minister did not have to be an elected politician under the 1979 and 1991 constitutions. It was only after the socalled Black May popular protests that the constitution of 1991 was amended and the prime minister came to be chosen from the members of the House of Representatives.

  Introduction

Throughout this period, two important offices were not subject to the control of the electorate: the monarchy and the military. The monarchy was not just ceremonial; it had real power. For example, under the constitution of 1979 and 1991 it was the king who appointed prime ministers and ministers. The power of appointment was not nominal. He sometimes picked prime ministers of his choice. For example, in 1980, he selected General Prem Tinsulanonda and had him succeed then unpopular Prime Minister Kriengsak Chamanan.69 After Prime Minister Suchinda Kraprayoon, the leader of a coup d’état of 1991, resigned in the face of strong public opposition in May 1992, the king reappointed Anand Punyarachun, the prime minister before Suchinda, as the prime minister.70 The king often influenced the content of laws as well as activities of the government,71 and when the country was faced with a crisis, he often intervened as a mediator. When the public demonstrations demanding ­Suchinda’s resignation spread between April and May 1992, the military tried to contain the demonstrations by force, which led to bloodshed. The king called on Suchinda and Chamlong Srimuang, the leader of an opposition party, the Palang Dharma Party, to end the confrontation. As for the military during the period from 1978 to 1997, it was politically powerful although its power gradually declined. Initially, it secured the position of the head of the government. When General Prem Tinsulanonda became prime minister, he was still commander in chief and held this position until 1981.72 In addition, the 1979 constitution had legally secured the military’s influence in politics. The constitution adopted a bicameral system in which the members of the House of Representative were chosen through elections, and the members of the Senate were appointed by the prime minister. The Senate served as the base for the military to project its influence. The constitution did not permit civil servants and military officers to hold political office, and it gave only the House but not the Senate the right to entertain a motion of no confidence. Yet, the constitution contained transitional clauses, valid for four years, which allowed officials and officers to hold political offices and senators to participate in votes of no confidence as well as in votes on important bills in joint sessions with the House.73 Military officers dominated the Senate. Although the temporary clauses expired in April 1983, Prime Minister Prem made new appointments of senators just before the expiration to secure the dominance of the military.74 Even after the legal prerogatives of the military had been diminished, the

Introduction  21

military continued to influence politics and the Prem cabinet continued to rely on support from the military.75 After the coup d’état in 1991 and Black May, the military’s influence was reduced and it gradually withdrew from politics.76 But it still remained influential in the Senate. Under the 1991 constitution senators were appointed by the king and could be chosen from among public officials as well as military officers. The military dominated the Senate until 1995.77 In addition, the military kept its autonomy with regard to its structure and human resources management and, as of 1995, civilian supremacy was yet to be established.78

Conclusion Democratization as a process consists of three elements: the development of free and fair political competition; an increase in electoral control over political offices; and an increase in the number of people with the right to participate in elections. Semi-democracy is a political regime that has advanced to significant degree along these three dimensions, but is not totally democratic. Great Britain in the latter half of nineteenth century and Japan between 1918 and 1932 are examples of semi-democracy before World War II, and Brazil between 1945 and 1964 as well as Thailand from 1979 to 1997 are examples of semi-democratic regimes after the war. This book explores how the semi-democratic regime breaks down and proposes an analytical framework to explain that breakdown. This is an attempt to enhance the understanding of hybrid regimes, which have received more scholarly attention in recent studies of democratization. It is also an effort to connect the experience of Japanese political development in the prewar era with current debates in comparative politics. In the next chapter, I offer an analytical framework to explain the breakdown of semi-­ democratic regimes.

chapter 1

Analytical Foundations

The breakdown of the semi-democratic regime in prewar Japan was gradual, reflecting the progressive change in the relationship between party government and the military. In other words, the breakdown was a process: democratic forces, that is, political parties and party politicians supporting the regime, step by step gave way to nondemocratic forces—the military— which opposed the regime, thereby bringing about a shift in the balance of power from party government to the military. This was the result of changes in three areas: political institutions, regime legitimacy, and semi-loyalty.

The Breakdown of Semi-Democratic Regimes: Existing Approaches There are three approaches to explaining the breakdown of semi-democratic regimes. The first, the conditional approach, argues that such regimes break down when necessary and sufficient conditions for further democratization do not exist. The literature on democratization offers several factors as conditions for democratization; one may argue that the semi-democratic regime is destined to collapse when a country lacks these conditions. The second, the structural approach, views the breakdown of semi-democratic regimes as a result of change in the social structure. However, when these two approaches are hard-pressed to provide an adequate explanation of the collapse of semi-democratic regimes, it becomes necessary to turn to the third approach. It considers the breakdown as a phenomenon of regime change in general, but as a phenomenon in its own right. In recent years, as research on hybrid regimes has advanced, some scholars have examined conditions 23

  Analytical Foundations

that affect the development of one type of hybrid regime, competitive authoritarianism.1 Yet, few studies have attempted to explain the breakdown of semi-democratic regimes. To undertake this task, it is therefore necessary to construct a new analytical framework. In this section, I consider whether one can explain the breakdown of semi-democratic regimes using the two existing approaches, that is, the conditional approach and the structural approach. the conditional approach

In general, the literature on democratization discusses whether particular conditions are necessary for a democratic regime to emerge. Here I examine if findings on the relationship between democratization, the level of economic development, and political culture can explain the breakdown of semi-democratic regimes. Economic Conditions  One might argue that a low level of economic development provides sufficient grounds for a semi-democratic regime to collapse. This contention is based on arguments that connect the development of democratic regimes to their level of economic development. Seymour Martin Lipset, who first proposed this relationship, argued that “the more well-to-do a nation, the greater the chances that it will sustain democracy.”2 He divides the countries he studies into two groups: European and English-speaking nations, and Latin American nations, and further divides each group into more and less democratic countries. To test his assertion about this symbiotic relationship, Lipset uses four indices of economic development—wealth, industrialization, education, and urbanization—and compares the averages of each set of indices among the more and less democratic countries within each group. The results support his proposition, showing that “the average wealth, degree of industrialization and urbanization, and level of education is much higher for the more democratic countries.”3 Robert Dahl claims that the relationship between the degree of economic development and democracy has already been settled. He declares that: The higher the socioeconomic level of a country, the more likely that its regime is an inclusive or near-polyarchy [democracy]. If a regime is a polyarchy [democracy], it is more likely to exist in a country at a relatively high level of socioeconomic development than at a lower level.4

Analytical Foundations  25

Dahl also refers to the “thresholds” above or below which chances for democracy increase or decrease. He shows that in countries where the GNP per capita is over about $700–800 in 1957 U.S. dollars, democracy is likely to emerge, whereas in countries where the GNP per capita is below about $100–200, democracy is unlikely to develop. He does not think that there is a “simple and one-directional” causal relationship between the level of economic development and the development of democracy. There are, however, many deviant cases such as the former Soviet Union, where industrialization did not lead to democracy, and the United States, where democracy developed before industrialization.5 Samuel Huntington, whose work is more recent than that of Dahl, also mentions thresholds as related to prospects for democratization. He sets the upper threshold at $3,000 GNP per capita and the lower threshold at $1,000 in 1976 U.S. dollars, calling the range between $1,000 and $3,000 the “transition zone.”6 That is, many countries that fell in this range in the mid-1970s experienced democratization by 1989.7 Larry Diamond performs his own quantitative analysis to show a strong correlation between the level of economic development and democracy. He also gives an extensive overview of the research on the subject to date, and reaches the following conclusion: It is important to emphasize here the extraordinary consistency with which the central premise of Lipset’s thesis has stood up through all manner of tests. Although different studies and research designs yield different angles of inference and interpretation, they virtually all demonstrate a consistent and strong positive relationship between the level of economic development and democracy.8

Despite these findings, it is difficult to rely on previous research on the relationship between economic development and development of democratic regimes to explain the collapse of semi-democratic regimes. While Lipset, Dahl, and Diamond all agree that such a relationship exists, and agree that greater economic development contributes to sustaining and developing democracies, none claims that greater economic development is a necessary and sufficient condition for democratic development. Thus, it is difficult to ascribe the development or nondevelopment of democratic regimes solely to the level of their economic development. The breakdown of a semi-democratic regime and the development of a semi-democratic regime into a democratic one are two different phenomena. One could argue that a semi-democratic regime does not develop into a democratic regime because it has not yet reached a certain level of economic

  Analytical Foundations

development. However, semi-democratic regimes can endure with few or no changes in the level of their economic development. This fact prevents us from using this theory to explain why the semi-democratic regime collapses. Political Culture  Many scholars have attached particular importance to the role of political culture. Gabriel Almond and Sidney Verba define political culture as “attitudes toward the political system and its various parts, and attitudes toward the role of the self in the system.”9 Almond and Verba argue that a certain political culture, in particular the one developed in Great Britain and the United States, which they call “civic culture,” is favorable to the development of stable democracy. They define civic culture as “a pluralistic culture based on communication and persuasion, a culture of consensus and diversity, a culture that permitted change but moderated it.”10 Almond and Verba classify political culture into three general types: the parochial, the subject, and the participant. They base these classifications on the attitudes individuals take toward the political system in general, the policy formulation process (“input aspect”), and the actual policies formulated (“output aspect”), as well as how individuals conceive of themselves as political actors. In the parochial political culture, individuals have no expectations about the political system, are indifferent to both the input and output aspects of the political system, and do not conceive of themselves as active participants in politics. In this case, individuals’ political orientations are not distinguished from their religious and social orientations. In the subject political culture, individuals are oriented toward the political system in general and to output aspects of the political system. However, they are indifferent to input aspects of the political system and do not consider themselves to be active participants in the political process. In the participant political culture, individuals are oriented toward the political system in general, as well as to both the input and output aspects of the political system. Not surprisingly, they are also active participants in politics. Almond and Verba consider civic culture to be a mixture of the three political cultures articulated above. In civic culture, “individuals become participants in the political process, but they do not give up their orientations as subjects or as parochials.”11 Such mixed orientations contribute to the development of stable democracy because they mitigate tensions caused by the contradictory objectives that leaders in a democratic regime have to fulfill.

Analytical Foundations  27

While leaders in a democratic regime must be responsive to the members of the regime, they have to be powerful enough to implement policy effectively. Thus, political leaders face a dilemma: when they become too responsive to the members of the political system, they cannot effectively rule. On the other hand, when they use too much power to govern effectively, they lose responsiveness to the members of the system. The mixed orientations of the members of the political system under civic culture resolve this dilemma. As active participants in the political process, individuals encourage leaders to be responsive to their demands and expectations. However, their orientations as subjects and parochials modify their participation, preventing politics from becoming polarized and unstable and enabling the political leaders to govern and make political decisions. In short, each individual’s behavior as a “potential active citizen” forces leaders to be responsive while also allowing them to rule. In this way, civic culture promotes stable democracy. Larry Diamond also emphasizes the importance of political culture. He argues that “moderation, cooperation, bargaining, and accommodation” among the leaders, which is indispensable for developing and maintaining democracy, can be maintained in the long run only if a political culture supporting these behaviors is cultivated among the leaders as well as the masses.12 Though political culture may be important in developing and maintaining democracy, two fundamental questions arise: Is a political culture favorable to democracy likely to emerge anywhere in the world? And if so, how does such a political culture emerge? Lucian Pye gives a negative response to the first question. He does not believe that the political cultures in the world will eventually converge into one favorable to democracy, a type of regime that has appeared so far mainly in the West. Pye believes there are irreconcilable differences between the political cultures of the West and Asia.13 He contrasts Asian and Western political c­ ulture and explains the implications of these differences for democracy. While Western political culture emphasizes individualism, pluralism, and active participation to influence actual policy outcomes, Asian political culture favors the collectivity, makes no room for pluralism, and views participation as a means to give individuals a sense of a belonging, rather than concrete political rewards. He argues that such a political culture makes the prospects grim for democratic development in Asia. Given that Japan has sustained a democratic regime since the Second World War, and that South Korea and Taiwan have democratized, it is im-

  Analytical Foundations

possible to accept Pye’s claim that democracy cannot develop in Asia.14 Even if we accept a theory that democracy does not develop where some particular political culture exists, the collapse of the semi-democratic regime remains hard to explain. Although one could argue that a semi-democratic regime does not democratize further because of a particular political culture, once such a regime has developed to a certain point, it is impossible to ascribe its breakdown to political culture. Given that political culture is unlikely to change in a short time, it is logically inconsistent to argue that a semi-democratic regime that has developed under a particular political culture (which may be unfavorable for the development of democracy) collapses because of this same political culture. In short, it is impossible to ascribe the breakdown of semi-democratic regimes to either their level of economic development or their particular political culture. The reasons for the collapse of a semi-democratic regime and its advancement are two different issues. Various arguments on democratization deal with the necessary conditions or contributing factors for a democratic regime to develop, but one cannot reverse these arguments to explain the breakdown of semi-democratic regimes. A semi-democratic regime that collapses must be seen as a regime that has failed, and not as one that has not fully developed into a democracy. The following section considers the extent to which such collapses may be explained by applying a general theory of regime change with a focus on structural change. the structural approach

In his seminal research on regime change, Samuel Huntington examines the relationship between changes in the structure of the economy and society caused by modernization and political order and argues that modernization often leads to political instability, which in turn invites military intervention.15 According to Huntington, modernization breaks down traditional customs and increases people’s aspirations. However, as the pace of economic development lags and fails to meet people’s changing aspirations, social frustration spreads. This frustration can be ameliorated in two ways: through opportunities for social and economic mobility or through the development of political institutions to accommodate it. In modernizing countries, social and economic mobility is generally limited except for the flow of the

Analytical Foundations  29

population from rural to urban areas, and so political institutions remain the only way to solve social frustration. Thus “social frustration leads to demands on the government and the expansion of political participation to enforce these demands.”16 Interestingly, in many modernizing countries where political institutions have not been sufficiently developed to convey these demands to the government, expanded political participation often leads to political instability. Whether a country remains politically stable or not depends on whether political institutionalization can keep pace with the expansion of political participation.17 Huntington distinguishes political regimes according to their level of political participation and the ratio of political institutionalization to political participation (see Table 1.1).18 He labels a political regime a civic regime when its political institutions can accommodate the actual level of political participation in the system and a praetorian regime when its political institutions cannot accommodate political participation and the people must resort to extra-legal means to express their political demands. A political regime often turns from a civic regime into a praetorian regime when political institutions cannot sufficiently develop to accommodate expanded political participation. Huntington argues that in praetorian regimes the military often intervenes in politics in order to restore the political stability lost when political institutions fail to accommodate expanded political participation. The military intervenes in order to carry out reforms in what Huntington calls oligarchical praetorianism, to restore political stability in radical praetorianism, and to protect the existing order in the face of unregulated mass mobilization in mass praetorianism.19 The semi-democratic regime can be considered a type of Whig regime. That is to say, a significant number of people have the right to participate in the electoral process and politics is stable. Thus, if we follow the logic of table 1.1 Types of political systems ratio of institutionalization to participation Political participation

High: civic

Low: praetorian

Low: traditional

Organic

Oligarchical

Medium: transitional

Whig

Radical

High: modern

Participant

Mass

source: Samuel Huntington, Political Order in Changing Societies (New Haven, CT: Yale University Press, 1968), 80.

  Analytical Foundations

Huntington’s theory, the semi-democratic regime collapses and transforms into a radical praetorian regime because the existing political institutions cannot accommodate expanded political participation and politics consequently becomes very unstable. Huntington’s theory, however, is hard-pressed to explain the collapse of the semi-democratic regime because of two large, inherent theoretical shortcomings. First, it is difficult to demonstrate causal relationships among a series of events (stipulated in his model) that lead to the collapse of political regimes. In other words, although it can be shown that a series of events takes place at the same time, it is difficult to prove that these events indeed cause one another. The second shortcoming lies in the vagueness of the concepts Huntington uses in his theory, such as expansion of political participation, political instability, and political institutionalization. In particular, it is problematic that he does not specify the degree of political institutionalization necessary to accommodate expanded political participation that will in turn maintain political stability. No political institution is immune from defects. Thus, if certain events take place—demonstrations, strikes, or riots, for example—that suggest loss of political stability, they can be evidence of defects in the existing political institutions and one can ascribe the loss of political stability to those defects on a post hoc basis. In other words, arguments based on this theory are unfalsifiable, which makes it unable to explain the breakdown of semidemocratic regimes.

Key Factors Since neither the conditional approach nor the structural approach can adequately explain the breakdown of semi-democratic regimes, a third approach is needed, one that considers collapse not as one phase of general regime change but as a distinct phenomenon. To explain the collapse of the semi-democratic regime, it is necessary to take into account the regime’s very nature. As a result of a significant degree of democratization, under the semi-democratic regime political forces that are subject to electoral control (democratic forces) coexist with political forces that remain independent of electoral control (nondemocratic forces). The semi-democratic regime breaks down when the balance of power

Analytical Foundations  31

b­ etween democratic and nondemocratic forces gradually changes, and the latter overthrow the former. How, then, can one construct an analytical framework to explain this process? I seek to do so by focusing on primary factors that are relevant to the breakdown of the semi-democratic regime so that this framework can be applied to various cases. The first important factor is political institutions.20 The political process and political behavior usually unfold under constraints from political institutions. Thus, when we analyze the political process under which a regime collapses, it is necessary to examine the effects that political institutions project over the process of the collapse. However, regime change can occur even when political institutions remain the same. This means that political institutions are not the only important source of change. Other factors are also at work because political institutions exercise a constant effect over the political process and over political actors. Thus, it is necessary to explore other factors that are relevant in addition to political institutions to the collapse of the semidemocratic regime. Research by Juan Linz and others on the breakdown of the democratic regime in Europe and Latin America offers invaluable insight on this question.21 Although the democratic regime and the semidemocratic regime are different, their breakdown is similar in that, in both cases, nondemocratic forces topple democratic ones. In his general argument about how democratic regimes break down, Linz states that it is necessary to look into the political processes actually taking place in order to explain a breakdown. On that basis, he builds a “descriptive model” of the process of breakdown.22 He points out many important “­elements” relevant to a breakdown, including the legitimacy of the democratic regime, party systems, disloyal and semi-loyal opposition, violence, insoluble problems, crises, and the initial installation and consolidation of democratic regimes. Furthermore, he suggests that when a democratic regime cannot effectively solve pressing problems, the people begin to withdraw their support of the system, thereby undercutting its legitimacy, and disloyal opposition increases its attacks on the regime. As various political developments—such as political violence by disloyal parties, ambivalent attitudes toward violence of the forces supporting the regime, implicit support of disloyal parties by some parties who actually support the regime, and party fragmentation—further undermine the legitimacy of the regime, disloyal p­ arties increase their power

  Analytical Foundations

and their capacity to mobilize people and to use force as they garner increased popular support. Given this situation, and even as they try to co-opt disloyal parties into the ruling coalition, leaders within the regime are often deceived by the ambivalent attitudes of the leaders of disloyal parties, who imply that they might cooperate with the regime-­supporting ­parties. However, this leads to a transfer of power to the disloyal parties, further enabling their “legal revolution” and bringing about the regime’s breakdown. Although Linz has made an enormous contribution to the literature on the breakdown of democratic regimes, a subject that has been understudied, his “descriptive” model includes too many factors in the breakdown of the democratic regime, making it difficult to construct a parsimonious analytical framework. While Linz and others limit the number of relevant factors in the breakdown of the democratic regimes compared with historical descriptions, they also admit that their research does not make a sufficient number of generalizations.23 This is probably why Linz himself has never claimed that his work and that of others is an effort to construct a theory of the breakdown of the democratic regime. When we scrutinize the factors that Linz lists, it becomes clear that many are indeed relevant to the collapse of democratic regimes because they affect legitimacy. Crises and intractable problems undermine legitimacy if a regime cannot deal with them effectively. Likewise, the manner in which a democratic regime is initially founded and consolidated has a lasting impact on its legitimacy. As for the party system, Linz only argues that an extreme multiparty system is likely to contribute to the breakdown of democratic regimes but does not discuss how it relates to legitimacy. This factor, however, is related to legitimacy since an extreme party system makes it difficult to deal with urgent problems, thereby contributing to the erosion of legitimacy. Although Linz mentions violence, he argues that its role is marginal. Among the various factors listed by Linz, then, legitimacy, a disloyal opposition, and semi-loyalty play important roles in the breakdown of the democratic regime. Among the remaining factors, legitimacy is a key component in sustaining any kind of political regime, and should therefore be included in an analytical framework to explain the collapse of semi-­democratic regimes. Semi-loyalty should also be included. According to Linz, semi-­loyalty is “a willingness to encourage, tolerate, cover up, treat leniently, excuse, or justify” challenges to a democratic regime.24 Semi-democratic regimes are precariously balanced between democratic and nondemocratic forces. If

Analytical Foundations  33

some democratic forces become semi-loyal to a regime and adopt an ambiguous attitude toward challenges to the regime from nondemocratic forces, the balance of power changes and the regime’s collapse becomes increasingly likely. Although Linz emphasizes the role of disloyal opposition in the collapse of democratic regimes I would argue that this factor should be the objective—not merely an element of—that analysis. It is therefore possible to separate legitimacy and semi-loyalty from other factors Linz identifies as important in the collapse of semi-democratic regimes. Here I present a caveat. Although Linz describes in detail the concept of semiloyalty, he does not explicitly show how it is different from legitimacy. When political actors act semi-loyally to a regime, they, after all, do not entirely accept the legitimacy of the regime. Thus, the concept of semi-loyalty is related to the concept of legitimacy. In constructing my analytical framework later in the chapter, I will say more about the differences between the two concepts. Thus far, I have argued that the breakdown of the semi-democratic regime can be analyzed as a process in which the balance of power between democratic and nondemocratic forces changes. I have also discussed how to construct a framework for analyzing that breakdown and have argued that political institutions, legitimacy, and semi-loyalty are important factors that must be included in any analysis. In the following section, I will discuss the role of these factors in more detail and flesh out my framework. political institutions

Insofar as political institutions set rules in politics, they affect the power of political actors and the strategies they pursue. Political institutions in a semi-democratic regime include both democratic and nondemocratic elements and affect the power and strategies of democratic as well as nondemocratic forces. In a semi-democratic regime, nondemocratic forces can exploit nondemocratic elements that exist in political institutions to expand their power and challenge the regime. Since nondemocratic forces employ existing political institutions in launching a challenge, it is difficult for democratic forces to resist and contain them. legitimacy

Seymour Martin Lipset calls Max Weber “the fountainhead of legitimacy theory,” and indeed we can trace discussions of the role of legitimacy in politics back to Weber.25 Weber argues that in addition to relying on “material

  Analytical Foundations

or affectual or ideal motives,” a political system also relies on the belief in its legitimacy to sustain its control.26 He identifies three types of legitimacy in particular: rational, traditional, and charismatic.27 Since Weber, legitimacy has been regarded as important because of its role in the endurance of political systems; today, many scholars accept that “legitimacy is an idea of central importance in political thought and ­science.”28 Nonetheless, theoretical discussions of legitimacy have not yet fully converged and many debates continue on a number of theoretical fronts related to this concept: the definition of legitimacy, the importance of a particular group’s belief in legitimacy, factors that affect legitimacy, and how to measure legitimacy. What Is Political Legitimacy?  Scholars have defined legitimacy in several different ways. Juan Linz offers a minimum definition: “the belief that in spite of shortcomings and failures, the existing political institutions are better than any others that might be established.”29 Seymour Martin Lipset has described how “legitimacy involves the capacity of the system to engender and maintain the belief that the existing political institutions are the most appropriate ones for the society.”30 Richard Merelman has defined political legitimacy as the “quality of ‘oughtness’ that is perceived by the public to inhere in a political regime.”31 Today legitimacy as a concept is closely related to the concept of democracy, inasmuch as various scholars have examined its role in such political issues as the collapse of democracies and the consolidation of newly founded democracies, as well as its current status in Western industrialized countries.32 Thus, many scholars have given definitions of “democratic legitimacy” and the “legitimacy of democracy.” For example, Peter Merkl defines “democratic legitimacy” as consisting of three elements: “the presence of a community or nation,” “a solemnly and widely accepted legal and constitutional order of democratic character,” and “an elective government responsive to the expressed needs of the people.”33 Some scholars have even gone so far as to incorporate democratic values into the definition of legitimacy itself. Leonardo Morlino and José Montero define legitimacy as “a set of positive attitudes of society towards its democratic institutions, which are considered as the most appropriate form of government.”34 In view of today’s widespread conviction that democracy is the most appropriate political system, it may be natural that there is a tendency to

Analytical Foundations  35

consider democracy as the only polity in the contemporary world that can acquire legitimacy. However, drawing a close connection between legitimacy and democracy is only a recent trend, reflecting the continuous spread of democracy in the latter half of the twentieth century. Historically, and even today, examples abound in which polities other than democracies have acquired legitimacy. This is why Weber conceived of three types of legitimacy. Indeed, Lipset explicitly states that “legitimacy, in and of itself, may be associated with many forms of political organization, including oppressive ones.”35 Since legitimacy can be associated with any kind of political regime, its definition should not be tied to any specific form of political regime. In this study I use Linz’s definition of legitimacy, since it meets this requirement. To reiterate: legitimacy is “the belief that in spite of shortcomings and failures, the existing political institutions are better than any others that might be established.” Whose Belief in Legitimacy Is Important?  With legitimacy thus defined, there remain two questions concerning legitimacy. First, is it necessary to make a distinction between belief in the legitimacy of the regime at the elite level and at the mass level? If it is necessary to do so, an additional question arises: Is it necessary to focus on the belief in legitimacy held by a particular group or not? To begin with the first question, some scholars, such as Peter McDonough, Samuel Barnes, and Antonio López Pina, and Leonardo Morlino and José R. Montero, do not make distinctions among groups in examining changes in belief in the legitimacy of the regime.36 However, scholars have often distinguished between the beliefs in legitimacy held at the elite and mass levels and have tended to argue that the former are more important. For example, Robert Dahl argues that “as with other beliefs, the views held by activists and leaders are likely to be more crucial than those of other people.”37 Likewise, Linz argues that the way organized groups in a regime evaluate the regime’s performance is more likely to affect the regime’s legitimacy.38 However, these scholars by no means consider the beliefs at the mass level irrelevant. For example, Diamond, Linz, and Lipset state that “theories of democracy stress that democratic stability requires widespread belief among elites and masses in the legitimacy of the democratic system.”39 Dahl also argues that “since even inactive or excluded strata may sometimes be mobilized, their beliefs are very far from irrelevant.”40

  Analytical Foundations

Normally, the elite and organized groups in society have more influence than the masses because they have many sources of power, such as the ability to set agendas in ongoing politics, direct access to financial resources, and easier access to information. Given the elite’s and organized groups’ greater influence in politics, their belief in the legitimacy of a regime is more relevant in sustaining democracies than the beliefs of the masses. The masses, however, are not completely powerless. They have the potential to mobilize and protest against authority, and their opinions exert influence on the elite through various connections. Above all, in democracies the masses have the power to vote. Thus, their beliefs about legitimacy are also important in sustaining democracies. How would this conclusion be different in the case of semi-democratic regimes? While legitimacy among the elite is more important than that among the masses even under a democratic regime, this is even truer in a semi-democratic regime. The elite in semi-democratic regimes are more powerful than their counterparts in democratic regimes. This is because the power the masses can exercise through voting is limited since many important offices are not subject to electoral control. This does not mean, however, that beliefs at the mass level are irrelevant, although they may be less relevant than under a democratic regime. Under semi-democratic regimes, the masses elect the leaders of some political offices and, as in democracies, influence the elite through various channels. Furthermore, they have the potential to mobilize themselves, which often plays a significant role in transforming an authoritarian regime into a semi-democratic regime. Turning to the second question, given that the elite’s beliefs are more important, do we need to focus on a particular group among the elite? I focus on several groups among elite, namely, the military, close aides to the emperor, and the bureaucracy as well as newspaper journalists and intellectuals with influence over the public. There are two reasons for attaching significance to these groups. First, as they were exogenous to democratic forces, such as political parties and party politicians who supported party government, their attitude toward the semi-democratic regime was crucial for its sustainability. Second, while in addition to these groups, political parties and party politicians of course projected significant influence over the fate of the regime, in this book I treat the attitudes of political parties and party politicians not in terms of legitimacy but of semi-loyalty as I make a distinction between legitimacy and semi-loyalty, which I describe shortly.

Analytical Foundations  37

Factors Affecting Regime Legitimacy  Let us look closer at the factors that affect the legitimacy of a political regime. I begin with factors that affect the legitimacy of a democratic regime and then extend the discussion to include semi-democratic regimes. A central concern affecting the legitimacy of democratic regimes relates to the poor performance of particular governments: Will it erode the legitimacy of the regime itself? Put another way, can people distinguish the performance of particular governments under a democratic regime from the regime’s actual legitimacy? If people can make such a distinction, the poor performance of a particular government will not erode its legitimacy. On the other hand, if people cannot make these distinctions, the poor performance of particular governments will not only undercut the regime’s legitimacy, but also enhance the possibility of collapse. Thus, in the long run, regime performance affects the legitimacy of democratic regimes: effective performance is likely to contribute to building legitimacy, poor performance to undercutting it. Thus, for example, in order to convince people that the existing regime is better than any other conceivable type, the regime must show that it is better at expanding people’s welfare. Whether poor performance in the short run will diminish legitimacy is another matter. Two major conditions must be met in order for poor short-run performance to lead to the erosion of legitimacy. First, if regime legitimacy has been fully established, people can distinguish it from regime performance under particular governments and will not allow poor regime performance to affect their views of regime legitimacy. Such a regime can rely on “a large reservoir of legitimacy” even when it performs poorly.41 On the other hand, “regimes that lack deep legitimacy depend more precariously on current performance and are vulnerable to collapse in periods of economic and social distress.” 42 Linz ascribes one of the major reasons of the collapse of the Weimar Republic in the middle of economic crisis to “the prior weak legitimation of the political system.”43 Even when a democratic regime has not yet established firm legitimacy, however, a second condition must be met in order for poor performance to lead to the erosion of legitimacy. Legitimacy is by definition judged comparatively. Defining legitimacy as the belief that the existing political institutions are better than others that might be established means that legitimacy cannot be sustained once another political regime begins to appear more appealing. Even weak regime performance will not erode legiti-

  Analytical Foundations

macy if no alternative regime appeals to the people. Morlino and Montero state that “the lack of a clearly perceived alternative to democracy” explains why the legitimacy of Spain’s newly founded democracy did not erode despite the accumulation of various problems in the country during the 1980s.44 By contrast, Linz and Stepan argue that economic crises in the Weimar Republic led to the regime’s breakdown because there were alternative types of regimes that strongly appealed to the people.45 In sum, while poor regime performance over the long run is likely to under­mine regime legitimacy, it is less likely to lead to the erosion of legitimacy in the short run if the regime has legitimized itself for a significant time or if there is no appealing alternative regime to replace it (see Figure 1.1). All this leads to the question of the elements accounting for the performance of a democratic regime. One can look at performance from two different angles. The first is to consider regime performance from the procedural aspect of policy formulation. Linz discusses two factors affecting regime legitimacy—the efficacy and the effectiveness of the regime—that can be representative of regime performance when examined from a procedural standpoint. As Linz puts it, efficacy is “the capacity of a regime to find solutions to the basic problems facing any political system that are perceived as more satisfactory than unsatisfactory by aware citizens” and effectiveness “the capacity actually to implement the policies formulated, with the desired results.”46 Following these definitions, efficacy refers to the capacity to draw up policies, while effectiveness refers to the capacity to implement them. In

Is the government’s performance poor?

Yes

Is the regime’s legitimacy yet to be established?

Yes

Are there any alternatives to the current form of regieme?

No

No

No

No erosion of legitimacy

No erosion of legitimacy

Legitimacy by default

figure 1.1. The erosion of legitimacy

Yes Erosion of legitimacy

Analytical Foundations  39

order to satisfy the people’s demands, a regime must do both. Even when a regime has devised an excellent policy, if it lacks the capacity to implement it, it cannot increase the people’s welfare. Likewise, even when a regime enjoys an outstanding system for implementing any kind of policy, if it lacks policies to deal with current problems, it cannot solve them. Thus, making this distinction between regime efficacy and effectiveness allows us to pinpoint the source of the poor performance that weakens regime legitimacy. The other approach to studying regime performance is to examine it according to areas a regime tries to address through its policies. One can, for example, distinguish between the social-economic and political spheres.47 In the social-economic sphere, regime performance can be judged by economic factors such as growth, unemployment, per capita income, the sense of equality among the people, and so on. In the political sphere, regime performance can be evaluated by such factors as the spread of political violence, corruption, and the transparency of the political process. While the erosion of legitimacy is often associated with poor performance in the economic sphere, one must bear in mind that evaluating regime performance includes evaluating performance in other areas as well.48 For example, discussing the effect of economic reforms on legitimacy, Diamond argues that “what damages . . . political legitimacy . . . are the perceptions that a few are benefiting while many stagnate and suffer . . . as a result of political connections and corruption.”49 The impact poor performance in one sphere has on legitimacy only heightens the impact it has in other spheres. Although the arguments presented so far have been developed mainly in the context of examining the effect of poor regime performance on the legitimacy of democratic regimes, they can also be applied to the effect poor regime performance has on the legitimacy of semi-democratic regimes. In semi-democratic regimes, legitimacy is more likely to vanish because of an awareness that alternative regimes exist. In a democracy, any alternative regime will most likely be authoritarian. In a semi-democracy, a democratic regime is an alternative, in addition to an authoritarian option. Those who support authoritarian regimes attack democratic elements in the semi-­democratic regime, while those who uphold full democracy criticize authoritarian components in the semi-democratic regime. While the democratic regime has to prepare for criticisms from just one side, the semi-democratic regime has to anticipate opposition from two sides, making its legitimacy more difficult to sustain.

  Analytical Foundations

Measuring Legitimacy  Many scholars admit that despite the importance of legitimacy, it is hard to use or operationalize it as a variable.50 Difficulties lie in distinguishing the popularity of particular governments from the legitimacy of the larger regime, and demonstrating that people’s satisfaction has led to an accumulation of legitimacy.51 Today, most scholars trying to measure the legitimacy of a regime rely on opinion polls asking questions about whether respondents trust government, whether they prefer democracy or authoritarianism, and so on. They trace, often via statistical methods, how the answers shift over time.52 There are two problems with this approach, however. First, when no opinion polls exist, it is impossible to measure changes in the level of legitimacy. Second, this approach does not necessarily reflect current arguments on the role of legitimacy. Measuring legitimacy through opinion polls means measuring legitimacy at the mass level and not at the elite level, thus missing the views of the segment of society that many scholars have identified as crucial. I seek to measure legitimacy in a different way. In this book, I assess the level of legitimacy by relying on various materials, such as newspaper editorials, opinions of intellectuals expressed in newspapers and journals, diaries of political leaders, voter turnout rates, the number of labor strikes, and the number of peasant conflicts with landlords, as evidence. While available evidence dictates that the assessment of regime legitimacy will focus on the elite, and on opinion leaders in particular, this is unlikely to be a serious problem in assessing changes in the level of legitimacy. As I have already indicated, under a semi-democratic regime, regime legitimacy among the elite is more important than among the masses. In addition, some members of the elite— opinion leaders—have a wide segment of the population as their audience and thus are likely to affect views on regime legitimacy among the masses. semi-loyalty

Linz explains the breakdown of democratic regimes as the result of “semi-­loyalty,” a concept that can also provide insight into the collapse of semi-democratic regimes.53 To understand semi-loyalty, we first have to understand the meaning of loyalty and disloyalty to a regime. When political actors are loyal to a regime, they follow the rules of the game under the regime, which specify the way to govern and the way to acquire power for governance. Even when political actors oppose the policies of a particular

Analytical Foundations  41

government in the regime, they oppose the policies according to the rules of the game when, for example, opposition parties challenge the ruling party in debates on the floor of parliament. Disloyal political actors, on the contrary, overtly challenge and refuse to follow the rules of the game. They defy policies designed by the government and resort to physical force to replace it. It is also important to understand semi-loyalty is a concept involving a particular group of political actors. Here I am referring to a set of attitudes and behaviors taken by so-called system-oriented political actors.54 Political actors who have an interest in sustaining a democratic regime weaken it by acting semi-loyally. It appears that semi-loyalty, as the term suggests, falls between loyalty and disloyalty. It is not an overt challenge to the regime. Rather it is, as Linz describes it, “a willingness to encourage, tolerate, cover up, treat leniently, excuse, or justify” disloyalty to the regime.55 Furthermore, “political violence, assassination, conspiracies, failed military coups and unsuccessful revolutionary attempts provide the test situations for semiloyalty.”56 For example, when a coup d’état is revealed, political actors who oppose severe punishment of those involved in the coup demonstrate semi-loyalty to the regime. Yet, to reiterate, semi-loyalty as a concept is related to the concept of ­legitimacy. It is the degree political actors on the side of the regime accept the legitimacy of the regime. When political actors act semi-loyally to a regime, in essence they do not entirely accept or support its legitimacy. In other words, semi-loyalty relates to belief in the legitimacy of a democratic regime held by elected politicians and political parties who have vested interests in the stability of the regime. The concept of semi-loyalty under a semi-democratic regime can be treated in the same way. Semi-loyalty refers to a set of attitudes and behaviors taken by democratic forces, namely, political parties and party politicians supporting the regime. When they act semi-loyally, they are in effect not fully endorsing the legitimacy of the regime. Democratic forces have an interest in sustaining the semi-democratic regime. A part of the democratic forces, however, may become semi-loyal to the semi-democratic regime. They often do so because they seek to enhance their own power or short-term interests without clearly realizing the longterm consequences of their actions. They do not intend to overthrow the regime nor do they wish to see its collapse.

  Analytical Foundations

The semi-loyalty of important political actors has a negative effect on the fate of the regime, however, because it can greatly affect the balance of power between the democratic forces loyal to the regime and nondemocratic forces that are challenging it. Semi-loyal groups strengthen nondemocratic forces by justifying or encouraging their conduct. They weaken democratic forces by refusing to stand by them and create a united front against the nondemocratic opposition. As a result, the power of democratic forces diminishes in the face of challenges from nondemocratic forces, and ipso facto the semi-democratic regime weakens. Here it is important to note the influence of political institutions on semi-loyalty. Since they have influence over the behavior of political actors, we should bear in mind that political institutions may tempt some political actors into semi-loyalty to the semi-democratic regime.

Analytical Framework As discussed above, democratic and nondemocratic forces coexist under the semi-democratic regime and the relationship between them is influenced by three factors: political institutions, legitimacy, and semi-loyalty (Figure 1.2). First, political institutions determine the legal rights that democratic and nondemocratic forces can exercise and strongly influence their choice of action. The legal rights of nondemocratic forces affect how they may threaten

Legal rights held by democratic forces

+ Legitimacy of the regime

+

Semi-loyalty

–

Legal rights held by nondemocratic forces

Balance of power between democratic forces and nondemocratic forces

– + Positive effects on power held by democratic forces – Negative effects on power held by nondemocratic forces

figure 1.2. Analytical perspective

Analytical Foundations  43

the semi-democratic regime and challenge democratic forces, while the legal rights of democratic forces affect how they deal with challenges from nondemocratic forces. Moreover, political institutions influence legitimacy and semi-loyalty through their role in the actions taken by democratic and nondemocratic forces. Second, the level of legitimacy of a semi-democratic regime affects the power democratic forces can project against challenges from nondemocratic forces. When the level of regime legitimacy is high, democratic forces can contain nondemocratic ones. As the level of the regime legitimacy declines, the capacity of the democratic forces to resist challenges from nondemocratic forces diminishes. Third, the semi-loyalty displayed by some of the democratic forces diminishes the power of democratic forces, as it increases that of nondemocratic forces. Ultimately, when regime legitimacy is lost and some democratic forces become semi-loyal to the semi-democratic regime—simultaneously refusing to cooperate with other democratic forces to resist nondemocratic ones—the semi-democratic regime breaks down. Two cautionary remarks in anticipation of criticisms of my framework: first, I can imagine that some may point out that this framework contains too many factors, making the analysis less explanatory than descriptive. Such criticisms are likely to be based on the fact that one key variable, legitimacy, is subject to variation because of influences from other factors, such as economic crises and corruption. I do not deny that legitimacy changes as a result of various incidents in many arenas. However, this is not tantamount to including all these incidents as relevant factors in the framework. When an analytical framework includes all incidents related to the incident it tries to explain, it becomes difficult to make it broadly applicable. In order to develop an analytical framework that can be applied to explain various cases of the collapse of the semi-democratic regime, it is necessary to concentrate on factors that are defined in general terms, such as legitimacy or semi-loyalty. Second, some may be skeptical of the way I assess the level of legitimacy. The difficulties of measurement should not prevent us from examining a relevant variable. Here, it is worth referring to the way the philosopher Imre Lakatos articulates how to judge whether an existing theory is falsified.57 He argues that the fact that some cases cannot be explained by an existing theory does not necessarily lead to the theory’s falsification. An existing theory is only falsified when a new theory is proposed that not only explains issues that eluded the existing theory, but also explains everything addressed

  Analytical Foundations

by the existing theory. In short, an existing theory cannot be falsified unless a better theory appears. This formulation can be applied to methods of measurement. Mere defects in an existing measurement method should not be cause for abandoning the method itself. A method should only be abandoned when a new, better method appears. Thus, the defects in the method I propose for measuring legitimacy should not necessarily lead to the wholesale rejection of my method. Even today, there is a mismatch between the theory of legitimacy and the means of measuring the level of legitimacy. Although theory emphasizes the belief in legitimacy among the elite, the method of measuring the level of legitimacy, with its heavy reliance on opinion polls, is biased toward measuring views among the masses. The method I have proposed will contribute to diminishing this mismatch, since it can be used to measure belief in legitimacy at the elite level. Relying on this framework, I explain the breakdown of the semi-­democratic regime in prewar Japan as a gradual shift in the balance of power between party government and the military. Democratic forces, namely, political parties and party politicians supporting the regime, slowly lost power while the military opposing the regime gradually expanded its power, ultimately succeeding in bringing down the regime. It is to this subject we now turn. In the next chapter I examine the nature of the different political regimes in Japan between 1889 and 1945 and show that the regime between 1918 and 1932 was semi-democratic.

chapter 2

Defining the Regimes of Prewar Japan

Japan in the years from 1918 to 1932 was ruled by a semi-democratic regime, one that was markedly different from the regimes that preceded and followed it. From 1889 to 1945 in fact Japan had three, distinct regimes: the first, from 1889 to 1918, was a competitive oligarchy; the second, from 1918 to 1932 was semi-democratic; and the third, from 1936 to 1945, was military authoritarian. Using the three criteria I have identified to define democracy and semi-democracy—the nature of political competition and elections, the extent of electoral control over political offices, and the extent of electoral participation—this chapter compares the three regimes and elucidates their respective distinctive characteristics. To illustrate the special nature of prewar Japanese democratization and the semi-democratic regime in prewar Japan, this chapter also compares democratization in Britain and Japan. In doing so, it examines the implications of the differences between British and Japanese democratization for the study of democratization as a whole.

The Pre-1889 Regimes The regime that ruled Japan before the rise of the Hara cabinet in 1918 did not satisfy any of the three criteria for democratic and semi-democratic regimes noted above. Instead, it was a competitive oligarchy. In order to characterize clearly the nature of this regime, it will be helpful to step back and examine regimes that existed even earlier, beginning with the Tokugawa Shogunate. 45

  Defining the Regimes of Prewar Japan

The Tokugawa Shogunate, established in the year 1600, was an oligarchical regime. The shoguns had to come from the Tokugawa family, while hanshu—the governors of autonomous feudal domains—usually came from the ruling family in each han—autonomous feudal domain under the Tokugawa Shogunate. For example, the governor of the Satsuma han was from the Shimazu family. The ruling class was limited to warriors called shizoku, who served as officials at various levels of government. The shoguns and hanshu normally served for life or until they decided to retire. Under this system, there was no formal accountability of political leaders nor was there any formal institution to reflect the people’s opinions in the selection of political offices. The oligarchical regime of the Tokugawa Shogunate collapsed in 1867 with the Meiji Restoration. The regime that followed, which was founded with the Imperial Proclamation of the Restoration in 1867, was a transitional oligarchical regime, which lasted until the adoption of the Meiji Constitution in 1889. This regime was oligarchical in that a small number of the people in the ruling class, based on blood and affiliation with social entities—namely, hanbatsu, clan cliques with a few traditional aristocrats—controlled political power. The hanbatsu cliques were those belonging to han, such as the Chōshu and Satsuma han, that played a principal role in the Meiji Restoration, overthrowing the Tokugawa Shogunate and establishing the new political system. This oligarchical regime was transitional inasmuch as the role of legal authority as the source of legitimacy increased over time. At the outset, the regime relied on legitimacy derived from the emperor’s traditional authority. When the new government announced its fundamental policy principles in 1868, for example, it released them in the form of the emperor’s oath to the gods. Nonetheless, modern political institutions such as a bureaucracy, administrative organizations, and the cabinet system developed, increasing the role of legal authority as a source of legitimacy. Certainly, the traditional authority of the emperor continued to play an important role in sustaining that legitimacy, as demonstrated by the fact that the Meiji Constitution promulgated in 1889 was in form of a “gift” from the emperor to his subjects. Nevertheless, the adoption of the Meiji Constitution, which codified Japan’s development of a series of basic legal rules for governance, symbolized the growth of legal authority as a major source of legitimacy. Thus, the adoption of the Meiji Constitution marked the beginning of a new regime, a competitive oligarchical regime, which lasted until 1918. It

Defining the Regimes of Prewar Japan  47

also marked a watershed in Japanese political history. First, it symbolized the increase of legal authority as a source of legitimacy, which transformed Japan from a traditional political regime to a modern political one. Second, it founded the Diet, opening the way to accommodate people’s political participation through formal political institutions, although the degree of accommodation was small in the beginning.

The Competitive Oligarchical Regime, 1889–1918 The regime that emerged after the adoption of the Meiji Constitution in 1889 was a competitive oligarchy. As we will see in the following sections, by its nature this regime prepared the way for the semi-democratic regime that followed. political competition and elections

The competitive oligarchical regime was characterized by fierce political competition.1 Competition was intense both horizontally within the hanbatsu cliques and vertically between the hanbatsu cliques and political parties. Political competition was further complicated because of alliances between some of the leaders of the hanbatsu cliques and of the political parties. Political competition among the hanbatsu political leaders emerged because of factional rivalry between the Satsuma clique and the Chōshu clique. Even in the same clique one finds strong personal rivalries, as exemplified by the contention between Itō Hirobumi and Yamagata Aritōmo, both of whom belonged to the Chōshu clique. While Itō was favorably disposed toward political parties and even became the president of the Seiyūkai, Yamagata opposed them. In addition, in terms of Japan’s foreign policy after the Sino-Japanese War (1894–1895), while Yamagata supported the formation of an Anglo-Japanese alliance, Itō opposed it and supported cooperation with Russia instead.2 Not only was there was tough political competition among hanbatsu political leaders, but hanbatsu political leaders and the political parties also competed for power. Soon after the establishment of the Diet, as political parties often sought to undermine the power of hanbatsu political leaders by trying to reduce budgets, the hanbatsu political leaders tried to diminish the power of political parties by dissolving the Lower House a number of times. Even though a stable relationship between the hanbatsu political leaders and the

  Defining the Regimes of Prewar Japan

political parties gradually developed after the Sino-Japanese War, their relationship was always shadowed by a struggle for power, and they often fought each other over the compensation for cooperation. In particular, on the surface a stable cooperative relationship developed after the Russo-­Japanese War (1904–1905) between a hanbatsu leader, Katsura Tarō, and Saionji Kinmochi, the president of the Seiyūkai, because they alternately formed the cabinet. Still, competition between these two groups persisted, even behind the alternating premiership,3 which merely reflected the balance of power between them. In other words, as neither side could overwhelm the other, they had to strike a bargain, thus creating a façade of stable cooperation. Competition among hanbatsu political leaders and between hanbatsu political leaders and political parties was further complicated by rivalry among political parties and by a series of alliances between hanbatsu political leaders and political parties. Put simply, party competition was dominated by an ongoing rivalry between the political parties derived from the Jiyutō and those derived from the Kaishintō, even though these political parties embraced a number of mergers with smaller political parties and small factions split from these parties to form even smaller parties. The Jiyutō formed the Kenseitō with the Shimpotō, and after the split of the Kenseitō, former members of the Jiyutō formed the Seiyūkai. The path of the Kaishintō was more complicated. The Kaishintō joined with other political parties to establish the Shimpotō, which soon joined together with the Jiyutō to found the Kenseitō. When it split, the former members of the Kaishintō formed the Kensei Hontō. Then, the Kensei Hontō merged with small parties to launch the Kokumintō, from which most of the members fled to join the Dōshikai formed by Katsura Tarō. Then the Dōshikai fused with other small parties to set up the Kenseikai. Several examples demonstrate how alliances between the hanbatsu political leaders and the political parties shifted under this regime. While the second cabinet led by Itō, a leader of the Chōshu clique, formed an alliance with the Jiyutō in 1895, the next cabinet, in 1896, was formed by Matsukata, a leader of the Satsuma clique, who struck a deal with the Shimpotō. After the second Yamagata cabinet, the alliance between the Kenseitō and its successor, the Seiyūkai, and the Chōshu clique dominated politics. In particular, Katsura Tarō, a leader of the Chōshu clique, and Saionji ­Kinmochi, the Seiyūkai president, often bargained with each other during their premierships. Yet their relationship was actually far from stable, as they often

Defining the Regimes of Prewar Japan  49

attempted to form alliances with other political groups. For example, the second Katsura cabinet explored the possibility of an alliance with the Kensei Hontō and other political parties and the second Saionji cabinet attempted to ally itself with the Satsuma clique. Indeed, Katsura’s efforts to form an alliance with the Kensei Hontō and other political parties materialized in the formation of the Dōshikai under the third Katsura cabinet in 1912, although his efforts to sustain his cabinet failed. On the other hand, the alliance between the Seiyūkai and the Satsuma clique became a reality under the first Yamamoto cabinet, which succeeded the third Katsura cabinet. In this way, under the political regime that existed between 1889 and 1918, political competition persisted on various fronts. electoral control over political offices

Electoral control formally occurred only over the Lower House and in practice did not reach the government. In other words, the electorate exerted very limited control over political offices. electoral participation

The regime between 1889 and 1918 was characterized by the electorate’s growing, if still limited, political participation. Institutionally, this regime allowed part of the population to elect the members of the Lower House, which had significant power in approving the budget and drafting bills. The regime initially gave the vote to about 451,000 people, which amounted to 2.0 percent of the adult population (over the age of twenty) at the time of the first general election in 1890.4 It expanded the franchise in 1900. At the time of the seventh general election in 1902, about 983,000 people, or 3.9 percent of the adult population, had the right to vote.5 This made it possible to have the interests of some parts of the population reflected in politics and policy. Representation grew throughout this regime as the political parties representing their interests progressively expanded their political power and eventually succeeded in participating in the government itself. comparison with authoritarian regimes

The competitive oligarchical regime differed significantly from an authoritarian regime. The former allowed more political pluralism and conferred more power on such representative institutions as the Lower House and political parties, which often could challenge the government.6 Although

  Defining the Regimes of Prewar Japan

only a limited number of people and groups had the opportunity to exercise political power, no single group or person could dominate. Besides, political participation by the people was much more intense than in an authoritarian regime. Although the franchise was limited, one of the two chambers in the Diet, the Lower House, was directly selected in elections in which political parties competed against one another. In an authoritarian regime, the roles of representative institutions, such as a national assembly and political parties, are much more restricted, and the government, controlled by a single person or group, can project greater power. Additionally, political participation by the people in an authoritarian regime is more restricted and passive. A comparison of Japan’s competitive oligarchical regime during this period with the regime in Spain under Franco, a typical authoritarian regime, clearly illustrates these points.7 As for the role of representative institutions, it seems that formal constraints on the national assemblies were greater in Spain under Franco. The power of the Spanish Cortes was heavily restricted.8 The government did not have to submit all bills to the Cortes but could circumvent it by “‘sanctioning . . . decree-laws for the regulation of matters’ which were explicitly within the jurisdiction of the Cortes.”9 In the competitive oligarchical regime of Japan, however, the government had to submit all bills to the Diet. And the Diet often rejected bills and budgets submitted by the government, while it could initiate and draft its own bills. Certainly, the Japanese Diet was also constrained. The government could adopt two kinds of decrees, emergency and independent, which had the same validity as laws. It also could implement the budget of the preceding fiscal year even when the Diet refused to approve the budget. Yet, when the government issued an emergency decree, it had to ask the next Diet to authorize it. Should the Diet refuse to meet government request, the decree lost its validity. Furthermore, the government could not amend existing laws via an independent decree, nor issue an independent decree on matters for which the constitution required the adoption of laws. With respect to the budget, the government could only implement a budget with the same components as that of the preceding year. In other words, it was impossible for the government to freely form a new budget. In Franco’s Spain, the people’s representation in the national assembly was restricted. According to Juan Linz’s discussion of the complicated composition of the Cortes, even after the introduction of “family representation” in 1967, only a tiny fraction of the 548 members were directly selected from the

Defining the Regimes of Prewar Japan  51

people, who had a limited franchise, while the government and Franco had great power in appointing a large number of its members.10 In the Japanese Diet on the other hand, though the House of Peers was not a representative institution, since the government appointed a significant number of its members, the Lower House was a representative institution. Although the franchise was limited, all members of the Lower House were elected by the people through regularly held elections, making it impossible for the government to control the Lower House. The Spanish government, in contrast, was able to bridle the Cortes through its power to appoint its members.11 Perhaps the largest difference between Spain under Franco and Japan’s competitive oligarchy was the role of political parties. In Spain, all political parties were illegal except the Falange (later Movimiento Nacional), while in Japan there were a number of political parties in the Lower House and they all had at least a chance to share in political power. 12 Moreover, the Falange “failed to exert a significant influence in the formulation of public policy.”13 The political parties in Japan, on the other hand, significantly influenced policy formulation, not only through debates in the Lower House with the government but also by participating in the government. Differences in the role of the national assemblies and the political parties also distinguished the nature of political participation in these two distinct regimes. The Lower House and the rise of political parties in Japan made it possible to reflect the people’s political disposition in politics to some degree, but in Franco’s Spain, political participation remained “overwhelmingly passive.”14 democratization under the competitive oligarchical regime

It is important to note that democratization proceeded under Japan’s competitive oligarchical regime. First, it progressed along the lines of electoral participation. Suffrage was expanded as a result of the electoral reform in 1900, which gave the vote to all men over twenty-five who paid at least ten yen in direct national tax. As a result, the ratio of the number of the people with suffrage increased from 2.0 percent to 3.9 percent of the adult population. Second, and more importantly, democratization proceeded in terms of the electorate’s control over political offices, as political parties and popular movements increasingly influenced the formation and dissolution of the cabinet. Certainly, there are instances in which the cabinet’s tenure did not reflect the will of the electorate. For example, under pressure from the hanbatsu cliques, the first Saionji cabinet fell in 1908 in spite of the Seiyūkai’s victory in the

  Defining the Regimes of Prewar Japan

general election two months before. Confrontation with the Lower House or openly expressed public dissatisfaction with the government, however, often led to the downfall of governments. Accordingly, there gradually emerged in this regime a de facto responsibility of the government for its policies to the Lower House and the public because the behavior of political parties in the Diet, which were often supported by public movements outside the Diet, could have a decisive influence over the fate of the government. The Lower House’s rejection of government policy often led to the collapse of cabinets. More importantly, at a later stage of the regime, popular opposition to the government tended to affect its fate. These tendencies often induced the government to take de facto responsibility for its policies. I use “de facto” because the Meiji Constitution did not formally assume the government was responsible to the Diet. For example, the second Matsukata cabinet dissolved the Lower House and resigned in response to a vote of no confidence in 1897. It resigned because it could not win the support of the political parties controlling the Lower House on the issue of an increase in the land tax.15 The third Itō cabinet also was forced to resign after its landtax increase bill was overwhelmingly rejected by the Lower House in 1898. After the Russo-Japanese War, the cabinet’s tenure came to be affected by popular movements in addition to its relationship with the political parties dominating the Lower House. The resignation of the first Katsura cabinet in 1906 was the first case that demonstrates how popular movements contributed to making the government responsible for its political conduct. Prime Minister Katsura’s concern about the reaction of the public to the peace treaty that ended the Russo-Japanese War played a major role in the resignation of his first cabinet and the formation of the first Saionji cabinet. Currently available materials demonstrate that Katsura worried about public opposition to the terms of the treaty even before signing and, expecting difficulties in handling the forthcoming Diet session, decided to resign.16 The first Katsura cabinet did not resign in response to an actual confrontation with the Lower House, such as the approval of a no-confidence motion by the Lower House against the cabinet. Yet Katsura’s preoccupation with public opinion and the physical expression of the public’s resentment of the government in the form of riots indicate that his resignation reflected his taking responsibility for various policies related to the Russo-Japanese War, including the peace treaty itself. The collapse of the third Katsura cabinet in 1913 is a more obvious case of accepting responsibility to the Lower House and the public. A popular

Defining the Regimes of Prewar Japan  53

movement against the third Katsura cabinet surged after the second Saionji cabinet was forced to resign in 1912 because of what the public saw as the hanbatsu cliques’ arbitrary exercise of power. This movement drove most of the political parties in the Diet, except the Dōshikai in the Lower House, to support a vote of no confidence against the cabinet. In the face of the vote and enraged citizens, who went so far as to surround the Diet for several days, the Katsura cabinet simply had to resign, even abandoning the idea of dissolving the Lower House. To introduce further examples, the first Yamamoto cabinet had to resign in 1914 in the face of public protest against the government for corrupt transactions with foreign companies involving the bribery of navy officers, and the Terauchi cabinet resigned over its responsibility for the Rice Riots of 1918. In short, under this regime the government was more disposed to acknowledge responsibility for its conduct to the Lower House and the public. It found it very difficult to adopt policies that were contrary to the will of the Lower House and the public. the significance of the competitive oligarchical regime

The competitive oligarchical regime occupied a significant place in Japan’s political development. Under this regime, the country’s political leaders became accustomed to political competition and public opinion gradually came to be reflected in politics. This regime paved the way for the semidemocratic regime.

The Semi-Democratic Regime, 1918–1932 In this section, I examine the nature of the political regime that was inaugurated with the formation of the Hara cabinet in 1917 and broke down with the end of the Inukai cabinet in 1932. I show that this regime was semi-democratic. political competition and elections

The period between 1918 and 1932 was characterized by intense political competition between the two major parties, the Seiyūkai and the Kenseikai (later the Minseitō), although between 1924 and 1927 a tripolar party system existed due to the split of the Seiyūkai into the Seiyūkai and Seiyu Hontō. The Kenseikai (later the Minseitō) and the Seiyūkai dominated politics, and

  Defining the Regimes of Prewar Japan

took turns forming the government during most of the period between 1918 and 1932, except for two years between 1922 and 1924. Apart from the 1924 election, in which three large parties competed, the two major parties won more than 80 percent of the seats in general elections held during these years. By this time, although personal leadership was quite influential in these two parties, they had developed nationwide organizations with regional branches in each prefecture.17 Consequently, the two parties dominated regional politics, penetrating the prefectural assemblies. By the end of the 1920s, approximately 90 percent of the members of the prefectural assemblies belonged to one or the other of the two political parties.18 Between 1918 and 1932, the Seiyūkai and the Kenseikai competed against one another, advocating different policies and agendas.19 When universal suffrage became a political issue under the Hara cabinet, the Seiyūkai argued that adopting it was premature, while the Kenseikai advocated its immediate introduction.20 The two parties put forward different policies on how to manage the economy, with the Seiyūkai persistently supporting an expansionist fiscal policy and the Kenseikai (later the Minseitō) t­oeing a tighter fiscal line. The differences were revealed most conspicuously when the return to the gold standard became an issue. The Minseitō cabinet under Prime Minister Hamaguchi reintroduced the gold standard and maintained it, while the Seiyūkai opposed the return. The parties differed over foreign policy as well, as demonstrated when China began to be unified under the Guomin­dang in the late 1920s and moved to reclaim various interests held by foreign countries in China. The Minseitō considered it important to maintain a cooperative relationship with the United States and Great Britain and tried to secure Japanese interests in China through diplomatic means, whereas the Seiyūkai emphasized securing Japanese interests, even if this meant increasing tensions with the United States and Great Britain. The Seiyūkai supported an aggressive approach in China, including the use of armed force to promote Japanese interests there. In the general elections held between 1918 and 1932 (except in 1924 when the two political parties cooperated to defeat the ­Kiyoura cabinet), the Seiyūkai and the Kenseikai (later Minseitō) competed against each other to gain support for their respective policies. Despite this intense competition between the two major parties, elections were neither perfectly free nor fair, because of such factors as preferential government treatment of the ruling parties in election campaigns and restrictions imposed by the Peace Preservation Law. The political parties ap-

Defining the Regimes of Prewar Japan  55

pointed their members as prefecture governors and influenced the personnel policy of the Ministry of Interior when they were in power, and thus were often in a position to force prefecture governors and officials of the Ministry of Interior to regulate election campaigns arbitrarily. Violations of the electoral regulations that were punished were primarily those committed by opposing parties. However, it is difficult to determine the actual effect of all this on electoral outcomes, as indicated by the fact that the Seiyūkai could not win a majority in the 1928 election even though the Seiyūkai government gave significant preferential treatment to Seiyūkai candidates. Furthermore, it is generally thought that the government’s preferential treatment of the ruling party was more conspicuous under the Seiyūkai government than the Minseitō government.21 In addition, under the Peace Preservation Law adopted in 1925 the government could punish those who organized or joined organizations in order to change the political system or challenge the private property system. Under this law, aimed at excluding communists from politics, the communist movement was severely suppressed, even during the period of party government. In 1928 and 1929, the government arrested members of the Communist Party, greatly weakening it as an organization. Between 1918 and 1932, therefore, there was significant competition among political parties, although elections were not fully free or fair because of government intervention and the exclusion of people with particular ideologies from participation. electoral control over political offices

To what extent were political offices subject to electoral control by the people during the regime from 1918 through 1932? A major characteristic of this regime was the existence of de facto parliamentary government, in which two major political institutions, the Lower House and the government, were subject to electoral control. De Facto Parliamentary Government  The Lower House was the only political institution under the Meiji Constitution that was formally subject to the people’s control through elections. The Lower House had the right to approve the bills and the budgets proposed by the government and the power to propose bills in its own right. Under the Meiji Constitution, the government was not formally responsible to the Lower House but to the emperor,

  Defining the Regimes of Prewar Japan

which meant that the people could not formally have indirect control over the government through their control over the Lower House. From 1918 to 1932, however, except for the period between 1922 and 1924, political parties controlled the government with their presidents as prime ministers, and in retrospect we can observe several rules that guided the nominations of these officials. When the prime minister resigned due to illness or died, the person who succeeded him as party president was appointed as the next prime minister. However, when the prime minister resigned with his cabinet due to a political deadlock, the president of the first opposition party in the Lower House was appointed the next prime minister. During this period, the electorate consequently had de facto indirect control over the government. If this custom had remained in practice and if Japan had experienced a watershed incident similar to the dismissal of Viscount Melbourne in nineteenth-century Britain (which made it clear that the cabinet was responsible to the House of Commons and not to the crown), Japan might have succeeded in firmly establishing a parliamentary government system. However, this practice was abandoned in the face of the rising influence of the military. After de facto parliamentary government had been established, however, political offices not subject to electoral control retained power over their various and respective political domains. Four major political institutions were not subject to electoral control: the genrō, the House of Peers, the Privy Council, and the military.22 Genrō  Genrō was a name given to the leaders in the hanbatsu cliques who led the country after the Meiji Restoration and whom the emperor asked to advise him on political issues. After the collapse of the third Katsura cabinet in 1913, genrō retained the power to recommend to the emperor candidates for the post of prime minister. As the emperor normally appointed prime ministers based on their recommendations, they had virtual power to appoint whom they chose. After the death of genrō Matsukata Masayoshi in July 1924, Saionji Kinmochi became the sole remaining genrō, and except for Hara ­Takashi and Katō Tomozaburō, all prime ministers were appointed based on his recommendations. Thus even though the practice of party government prevailed between 1918 and 1932, it was an informal custom dependent on a single person. The House of Peers  The House of Peers was another major institution not subject to electoral control. Its members were selected by mutual election among peers or by government appointment. While the House of Lords in

Defining the Regimes of Prewar Japan  57

Great Britain continually lost political power to the House of Commons, Japan’s House of Peers retained equal legal power with the Lower House. The House of Peers, which was very conservative, became an obstacle to implementing various reforms initiated in the Lower House, such as the adoption of universal suffrage and the legalization of labor unions. The first Katō Takaaki cabinet tried to reduce the power of the House of Peers in 1925, but was unable to bring off dramatic reform. Accordingly, the power of the Lower House—which was directly subject to electoral control—was severely constrained by an institution outside electoral control. The Privy Council  The Privy Council was established to “deliberate upon important matters of State” and to advise the emperor.23 The government had to get the approval of the Privy Council on a number of issues, including the ratification of treaties, amendments to the constitution, emergency decrees, and provisions on the organization of the cabinet and the government. In other words, the government had to secure the Privy Council’s approval on everything but the budget and bills.24 As happened with the emergency decree to solve the financial crisis in 1929 under the first ­Wakatsuki cabinet, the Privy Council often impeded the implementation of the government’s policies. In short, the Privy Council wielded significant political power, but since the emperor appointed its members, it was not at all accountable to the people. The Military  The most powerful office not subject to electoral control was the military.25 According to the two articles in the Meiji Constitution relevant to control over the military, “the Emperor has the supreme command of the Army and the Navy” (Article 9) and “the Emperor determines the organization and peace standing of the Army and Navy” (Article 12).26 As for command the over military, in practice Article 11 came to be interpreted as stipulating that the military was responsible directly to the emperor and was not subject to any control by the government, which guaranteed the military’s unfettered autonomy. Even though the government interpreted Article 12 to mean that the government was responsible for the organization of the military, the military believed that it at least shared responsibility with the government, given the relevance of military organization to military operations. Furthermore, the “Cabinet-Official Organization,” an imperial ordinance introduced in 1889, stipulated in Article 7 that “matters which pertain to

  Defining the Regimes of Prewar Japan

military secrecy or military orders and are presented to the Emperor, except for those items which the Emperor has seen fit to entrust to the Cabinet, are to be reported by the Minister of War and the Minister of the Navy to the Cabinet Prime Minister.”27 This article came to be interpreted as meaning that the chiefs of the General Staff of the army and navy as well as the army and navy ministers could directly report to the emperor on military affairs without prior consultation with the prime minister. These two legal prerogatives enjoyed by the military gave it ample room to maneuver without respecting government policy. In sum, though two important political institutions, the Lower House and the cabinet, fell under the electorate’s direct and indirect control respectively, the electorate had no jurisdiction over several political institutions that wielded substantial political power. Consequently, not all effective political offices were accountable to the electorate in the Japanese regime between 1918 and 1932. electoral participation

The franchise was expanded under the regime that existed between 1918 and 1932. Soon after its formation, in 1919 the Hara cabinet lowered the eligibility requirement for the franchise to an annual tax payment of at least three yen. This change doubled the number of people with the franchise from around 1.4 million in 1917 (4.8 percent of the adult population) to about 3.1 million in 1920 (10.2 percent of the adult population), when the first election was held under the new law. Universal male suffrage was realized in 1925, giving the franchise to adult men over twenty-five. Under universal male suffrage, 12.4 million people (19.1 percent of the population and 37.3 percent of the adult population) had the right to vote during the 1928 general election.28 The universal male suffrage enacted in 1925 franchised enough people to qualify the regime as semi-democratic. How, then, can one evaluate the results of the expanded suffrage in 1920? Was 10.2 percent of the adult population sufficient to cross the threshold to a semi-democratic regime? Although this is a matter of judgment, the answer seems to be yes, for the following reasons. If an electorate possessing a limited franchise had homogenous interests clearly distinguished from those of the rest of the population, and no interests other than theirs were reflected in politics, a regime with only 10.2 percent of the adult population franchised might not deserve to be

Defining the Regimes of Prewar Japan  59

called semi-democratic. Certainly, an electorate under a limited franchise is likely to be more homogeneous than the one under universal suffrage and probably supports policies protecting its privileges. This may be why the Seiyūkai won the 1920 general election, even though it contended that introducing universal male suffrage was premature. The enfranchised portion of the electorate simply did not have a large stake in universal male suffrage. One can conceive of issues around which the opinions of the entire electorate might have converged, but it seems more natural to argue that there were varied opinions on many issues, which reflected society’s overall diversity. For example, in the 1924 election, public opinion surged against the Kiyoura cabinet, which consisted mainly of members of the House of Peers. In the election, the voters chose the Seiyūkai, the Kenseikai, and the Kakushinkurabu (which opposed the Kiyoura cabinet) over the Seiyu Hontō (which supported the Kiyoura cabinet), bringing about the Kiyoura cabinet’s resignation. None of the three requirements for democracy—free and fair political competition and elections, and protection of political freedoms to make political competition and elections free and fair; control by the electorate over the significant political offices; and full participation by the population in elections—was completely fulfilled by the regime in power from 1918 and 1932. However, it did fulfill each of the three criteria to a substantial degree, thereby meeting the definition of semi-democracy. From 1922 to 1924, three consecutive nonparty governments ruled Japan. For several reasons, this period can be treated as part of the same regime (1918–1932). First, these governments lasted only for two years. The Kiyoura cabinet in particular—which was forced to dissolve the Lower House about twenty days after its formation due to opposition from political parties demanding party government—survived as a mere caretaker government for the rest of its governance. Second, these governments clearly differed from those in the competitive oligarchical regime in that the prime ministers were no longer appointed as protégés of the hanbatsu cliques. Third, at least two of the three requirements for semi-democracy, namely, the nature of political competition and the extent of electoral participation, were satisfied. In addition, that the political parties demanding the formation of a party government could so easily trigger the collapse of the Kiyoura cabinet and win the general election demonstrated that party government had already become the norm in that period.

  Defining the Regimes of Prewar Japan

The Military Authoritarian Regime, 1936–1945 It took several years for a new regime to emerge after the semi-democratic regime broke down in 1932. The period between 1932 and 1936 was transitional in that the political parties still retained significant power, but could no longer form cabinets. After this, a political regime dominated by the military emerged in 1936 and lasted until 1945.29 It is possible to define this regime as military authoritarian, as I will show below. political competition and elections

Under this regime, the ruling class consisted of a coalition of the army, the navy, different groups in the bureaucracy, some party politicians, and a few aristocrats. Political competition was not free or fair because the suppression of various civil rights necessary to secure political competition and free and fair elections intensified during the transitional phase of this regime. For example, in 1935 the Okada cabinet banned Minobe Tatsukichi’s interpretation of the Meiji Constitution. Minobe, a prominent scholar of constitutional law at the Univetsity of Tokyo, had argued that although the emperor was the supreme institution in the state, his power, like that of other state institutions, was constrained by the constitution. This interpretation was crucial because by denying that the emperor had unrestricted power as an absolute monarch, and recognized that other political institutions could share in governing the state, it provided the theoretical underpinning for party government.30 Under the Okada cabinet, Minobe’s constitutional interpretation, the dominant constitutional theory at the time, came under attack as irreverent to the emperor and was finally banned by the government. The Communist Party and its members were likewise suppressed a number of times during this period and were driven underground in 1935.31 Suppression was not limited to certain ideologies; the military strongly constrained expressions of anti-military opinion in the Diet, as demonstrated by the resignation of the Hirota cabinet in January 1937 and the removal of Saitō Takao from the Lower House in February 1940. In the seventieth Diet, starting in December 1936, Hamada Kumimatsu of the Seiyūkai attacked the military’s intervention in politics. In response, Army Minister Terauchi Juichi, infuriated by Hamada’s speech, pressured the cabinet to dissolve the Lower House in retaliation, although the cabinet as a whole could not agree on whether to do so. Faced with this challenge, Prime Minister Hirota Kōki

Defining the Regimes of Prewar Japan  61

decided that the cabinet as a whole should resign.32 In the seventy-fifth Diet, starting in December 1939, Saitō from the Minseitō condemned the government and the army’s policy in dealing with the “China Incident”—Japan’s military confrontation with China since 1937. In response, the army and the government demanded that the Lower House remove Saitō. It acquiesced and in March 1940 Saitō was expelled.33 The government, under pressure from the military, intervened in electoral campaigns as well.34 The Ministry of Interior restricted criticisms of the military in the electoral campaigns.35 Political competition under the regime diminished to the point that political parties vanished in 1940. To understand the political parties’ ultimate fate, it is helpful to look briefly at the formation of the Taisei Yokusan Kai, the Imperial Rule Assistance Association, founded in October 1940 by Prime Minister Konoe, an aristocrat from a family close to the imperial family.36 Under this regime, no single political power including the military could coordinate the various political actors and formulate coherent policies, and in recognition of this fact Konoe tried to set up an organization that could integrate the Diet and the government, gathering participants from various sectors of society: cabinet ministers, members of the Lower House, members of the House of Peers, rightist groups, journalists, party politicians, and businessmen.37 As for party politicians, they dissolved their parties to join the Imperial Rule Assistance Association, expecting to recover their political power.38 Since the views of participants in the new organization were so diverse, however, the Imperial Rule Assistance Association was stillborn, and never fulfilled Konoe’s original objective.39 Even though basic political freedoms necessary for political competition and fair and free elections were sharply curtailed and political parties ceased to exist, political competition did not completely disappear. As implied earlier, there was limited political pluralism in the form of competition within the ruling class. The army held the dominant position in the regime’s ruling class, strongly influencing the composition of the cabinet and government policy, and made numerous attempts to concentrate all political power in its hands. The army, however, could not fully monopolize political power because of the resistance of various other groups in the ruling circles. Other groups thwarted the army’s efforts to establish complete hegemony. ­Although the army intervened in cabinet formation and policy formulation, the leading aristocrats retained the power to recommend prime ministers.40

  Defining the Regimes of Prewar Japan

Most importantly, the army could not bring all of the government, namely other ministries, under its control. Specifically, the army made a number of attempts to set up institutions in the cabinet which would have had vast power over domestic economic affairs and legal authority to regulate other ministries.41 But because of resistance from the navy and the bureaucracy, the institutions that were established were never given such authority. In short, the army had to share power with other institutions. electoral control over political offices

Not surprisingly, electoral control over political offices diminished greatly under the authoritarian regime. The de facto parliamentary cabinet system was now gone and the electorate no longer held sway over the government. Under this regime, the military, and the army in particular, influenced the composition of cabinets and policy. Its power was fully demonstrated upon the formation of the Hirota cabinet, when it forced Prime Minister Hirota to agree to its demands concerning the appointment of ministers and the design of government policies. Under the Hirota cabinet, the military acquired the ultimate institutional prerogative to dictate the composition of cabinets and policies.42 It succeeded in amending the Regulation on the ­Organization of the Army and Navy Ministries so that only officers in service could become army and navy ministers.43 This meant that if the military did not like the prime minister or disagreed with cabinet policy, it could abort the formation of a cabinet or bring down the cabinet either through a refusal to appoint military ministers or through their resignation. For example, in addition to interfering in the appointment of ministers to the Hirota cabinet in 1935, in 1937 the army drove Ugaki ­Kazushige, who was appointed prime minister by the emperor, to decline the appointment by refusing to send an army minister to the cabinet.44 It triggered the collapse of the Yonai cabinet in July 1940 through the resignation of the army minister and a subsequent refusal to send a succeeding minister.45 Evidence of the army’s commanding influenced in policy formation is evident in the formation of the first Konoe cabinet in May 1937, when Konoe Fumimaro agreed to the terms set by the army on basic policy principles.46 In particular, the first Konoe cabinet adopted the army’s plan to strengthen the Japanese economy.47 After the second Konoe cabinet was formed in July 1940—at the army’s insistence—the army went on to design

Defining the Regimes of Prewar Japan  63

the basic principles of the government’s foreign policy. According to these principles, which were devised in response to the opening of the war in Europe, the government was to seek opportunities to expand into European colonies in Southeast Asia while strengthening its ties with Germany and Italy.48 When the tension between the United States and Japan heightened between 1940 and 1941 and the United States demanded Japanese troops withdraw from Manchuria as one of the conditions for easing the tension, the army stubbornly refused. As the army gained political power, Japan’s political parties continued to lose it. This culminated in their dissolution in 1940 as described in the preceding section, a drastic blow to the importance of the Lower House. Since some party politicians (as well as former party politicians) continued to be appointed as cabinet ministers and the Diet retained its formal constitutional power, it is possible to view the political parties and the Diet as retaining some power.49 But to uncover the real power of these entities, it is necessary to look more deeply into the state of political parties and the Diet. Party politicians were expected to have a pro-military attitude, but even so normally they were appointed to ministerial posts of marginal influence.50 Under the authoritarian regime, to make important decisions many cabinets set up inner cabinets composed of the prime minister and the army, navy, foreign, and finance ministers.51 The record of legislation under this regime, presented by Awaya Kentarō,52 illustrates the marginalization of the Lower House. According to his record, 790 of 880 bills submitted by the government were approved by the Lower House from the sixty-ninth Diet (May 1936) to the eightyseventh Diet (June 1945), while only three bills submitted by members of the Lower House passed—“the legislative authority of the Diet became a mere ­formality.”53 At the same time, the Diet, including the Lower House, did not engage in substantive debates over the budget after the February Twenty-Sixth Incident, but instead complaisantly agreed to enormous increases in military spending.54 The military budget jumped from 1.4 billion yen in 1936, representing 49.2 percent of the overall budget, to 12.5 billion yen in 1941, or 75.6 percent, when the war with the United States began.55 Furthermore, severe constraints were imposed on the Diet with the adoption of the National Mobilization Law in 1938, which gave the government the authority to regulate virtually all wartime economic activities not by law but by decree.56 As Gordon Berger claims, the government promised that

  Defining the Regimes of Prewar Japan

“the Mobilization Law was only to be invoked under wartime conditions, and not to deal with the China Incident.”57 The government reneged on this promise after the Diet approved the law, however, and began to invoke the law one month after its promulgation.58 Berger may be right in arguing that “the Diet retained the right to veto legislation not covered by the National General Mobilization Law.”59 But this was merely a nominal power, and the adoption of the National Mobilization Law only confirmed the already established practice by which the government could exercise political power without any constraint from the Lower House. To summarize, the electorate might have controlled the Lower House most of the time, even under this authoritarian regime, since the government only intervened seriously to get pro-government candidates elected during the election of 1942. And yet, as we have seen, the electorate had no control over other political offices, and thus could not influence government policy formulation. Nor could party politicians play a significant role in the government. On the other hand, the Lower House, the only institution over which the electorate could exercise control, ceased to function as an institution for initiating and drafting bills and checking the government through debate. In the final analysis, under the authoritarian regime electoral control over political offices was at best marginal.

The Prewar Regimes Compared To demonstrate that a semi-democratic regime existed between 1918 and 1932, it is necessary to distinguish it from the regimes that preceded and followed it. The following sections take up this task. comparing the semi-democratic regime with the competitive oligarchical regime

I begin by comparing the semi-democratic regime with the competitive oligarchical regime, showing that they were different in the three aspects, namely, political competition and elections, electoral control over political offices, and electoral participation. Political Competition and Elections  In Japan’s competitive oligarchical regime, political competition centered on the hanbatsu cliques and political parties, in particular the Seiyūkai. Under the semi-democratic regime, after the fall of the cabinet of Terauchi Masatake (the last protégé of the hanbatsu cliques) and

Defining the Regimes of Prewar Japan  65

the formation of the Hara cabinet, the rivalry between political parties and the hanbatsu cliques came to an end. Political competition shifted to the two ­major political parties, the Seiyūkai and the Kenseikai (later the Minseitō). To be sure, competition between political parties also existed under the competitive oligarchical regime. After the formation of the Seiyūkai, however, it held a dominant place in the regime and other political parties could not really challenge it, except when the Dōshikai came to power under the second Okuma cabinet. In short, under the competitive oligarchical regime, competition was one-sided, in favor of the Seiyūkai, whereas competition under the semi-democratic regime was between two equally powerful political parties. It would be fair to point out that there was competition between political parties and nonparty governments even under the semi-democratic regime, as seen most clearly in the rivalry between the Kiyoura cabinet and the political parties. However, the Kiyoura cabinet was not a creature of the hanbatsu cliques. As the hanbatsu political leaders aged, symbolized by ­Yamagata’s death in February 1922, the hanbatsu cliques could no longer act as a coherent political force. There remained only a handful of individual institutions outside electoral control, such as the Privy Council, the military, and the House of Peers, which had been coordinated by the hanbatsu leaders. Lacking a coordinating organ, these institutions could not act cooperatively to challenge the political parties. This was probably why the latter found it so easy to overthrow the Kiyoura cabinet. Electoral Control over Political Offices  Under the competitive oligarchical regime between 1889 and 1918, the government was not subject to the indirect control of the electorate. Though political parties participated in the government, the three cabinets led by party presidents emerged sporadically and none of the cabinets was formed consecutively. Between 1918 and 1932, by contrast (except for the period between 1922 and 1924), political parties ceaselessly formed cabinets. Though apparent only in retrospect, there was a custom that implicitly regulated the succession of cabinets. In addition, political parties’ control over the government was much stronger under the semi-democratic regime than under its competitive oligarchical predecessor. In the latter, the Seiyūkai came to power because of the nature of its presidents: its founder, Itō Hirobumi, was a prominent leader in the hanbatsu cliques, and its second leader, Saionji Kinmochi, was an aristocrat.60

  Defining the Regimes of Prewar Japan

­ oreover, only a few party politicians were ministers in the two Saionji M cabinets.61 Under the semi-democratic regime, party presidents were appointed as prime ministers because they led the political parties, and most cabinet ministers were party politicians. Electoral Participation  The number of people who could participate in politics through elections was much smaller under the competitive oligarchical regime than under its successor. Before the electoral reform of 1919, the number of people with the franchise was 1.42 million, only 4.8 percent of the adult population. As noted above, the portion of the adult population able to participate in politics increased enormously under the semi-democratic regime, first to 10.2 percent with the electoral reform of 1919, and then to 37.3 percent with the reform of 1924. After the introduction of universal male suffrage in 1924, people’s movements, which had characterized the latter part of the competitive oligarchical regime, faded away, indicating yet another change in the nature of political participation by the people. comparing the semi-democratic regime with the authoritarian regime

Thus far I have demonstrated that the regime in Japan from 1918 to 1932 can be considered semi-democratic, with a nature clearly distinct from that of the competitive oligarchical regime that preceded it. To round out that argument, I will show the semi-democratic regime was distinct from the military authoritarian regime that followed. Political Competition and Elections  The nature of political competition in the semi-democratic regime was different from that in the military authoritarian regime. Under the semi-democratic regime, the heart of political competition was the rivalry among political parties. They fiercely competed over the control the government. Under the authoritarian regime, competition among political parties continued until their demise in 1940. The political significance of competition among political parties, however, diminished as they could no longer control the government and were marginalized. Under the military authoritarian regime, the focus of competition shifted to the one among the military, the bureaucrats, party politicians, and the close aides to the emperor. Although the army was the dominant force, it could not control all sectors of the government.

Defining the Regimes of Prewar Japan  67

The political freedoms necessary for political competition—and free and fair elections—were much more restricted under the authoritarian regime. Certainly, political competition was not perfectly fair under the semi-democratic regime as the government often intervened in electoral campaigns in favor of ruling parties and the communist movement was crushed. Under the authoritarian regime, however, there were severe restrictions on freedom of speech and thought. As mentioned in the preceding section, liberal political ideas were much more restricted than in the semi-democratic regime as demonstrated by the ban on the Minobe’s constitutional interpretation. Furthermore, anti-military opinion was rigorously suppressed. Electoral Control over Political Offices  When we compare the semi-democratic regime with the military authoritarian regime, we see there was a significant difference in the degree of electoral control over political offices. Under the de facto parliamentary government of the semi-democratic regime electoral control extended beyond the Lower House to the government. Control diminished under the military authoritarian regime. Party government was no longer the custom. Military officers, aristocrats, and bureaucrats shared important positions in the government including the position of the prime minister. While party politicians who were members of the Lower House were assigned ministerial positions until demise of the parties, those positions were marginal. After the passage of Kokka Sōdōin Hō, the National Mobilization Law, in 1938, the Diet’s legislative power was reduced.

The Pattern of Development of the Semi-Democratic Regime in Japan and Implications for the Pattern of Democratization To demonstrate several distinctive features in the pattern of the rise and fall of the semi-democratic regime in Japan, the following discussion compares it with the democratization in Great Britain after the Glorious Revolution from 1688 to 1689. I have referred to the state of political competition in nineteenth-century Britain in the Introduction. Here I discuss in greater detail changes in the state of electoral control over political offices as well as the state of electoral participation. semi-democracy in nineteenth-century great britain

Over a very long time Great Britain developed first into a semi-democracy and then a democracy. Even after the Glorious Revolution, the govern-

  Defining the Regimes of Prewar Japan

ment was not parliamentary nor had Parliament yet established supremacy over the crown. Democracy in Great Britain came about as parliamentary government gradually developed and suffrage was expanded. The crown, along with other nonelected political institutions, continued to lose its political power, and parliamentary government was continuously solidified through legal reforms and an accumulation of political precedents. Suffrage continued to expand and culminated in the adoption of universal suffrage in 1928. Even after the Glorious Revolution, the crown could still veto bills passed by Parliament and “dismiss even the most popular or politically powerful minister.”62 After the reign of Queen Anne, which ended in 1714, no crown vetoed bills passed by Parliament.63 By the beginning of the nineteenth century, two political groups, the Tories (later the Conservatives) and the Whigs (later the Liberals) gathered around two competing sets of ideas about electoral reform, religion and the like, although they were not yet organized in the modern sense.64 Also in the same period it became clear that the government was responsible to the House of Commons, not to the crown. In 1834, William IV dismissed Prime Minister Viscount Melbourne, who led the government formed by the Whigs, and appointed Sir Robert Peel, who subsequently formed a minority government of the Tories. Peel dissolved the House of Commons in January 1835, expecting to get a Tory majority. The electorate, however, viewing Melbourne’s dismissal as a “royal coup d’état,” returned the Whigs as a majority to the Commons, which eventually resulted in Peel’s resignation and Melbourne’s appointment as prime minister.65 This series of events marked “a constitutional watershed,” as “never again would a British monarch dismiss a ministry.”66 The year 1841, when Melbourne resigned after the Whigs’ defeat in the general election, marked the first time in which the choice of the electorate brought about a change in the government.67 Another three decades passed, however, before the two major political parties would form cabinets that reflected the results of the general election. The political parties that preceded the Reform Act of 1867 were groups tied together mainly by personal connections, with limited internal cohesion; differences between them were often vague.68 After the Reform Act, the modern two-party system began to develop in parallel with the growth of party organizations in the countryside.69 From the 1870s on, with some exceptions, the party controlling the House of Commons would form the government and a change in

Defining the Regimes of Prewar Japan  69

the dominant party in the House of Commons through a general election usually brought about a corresponding change in the government. By the beginning of the twentieth century, the modern two-party system of the Conservatives and the Liberals (later replaced by Labour) had been firmly established, with nationwide organizations and strong discipline over party members as demonstrated by party voting.70 As the crown’s political power was gradually reduced, so too was the political power of other nonelected institutions, such as the military and the House of Lords. As for the military, from the end of the eighteenth century, while “the War Office administered the civilian affairs of the army,” “the Commander in Chief, directly under the sovereign, was responsible for military command and discipline.”71 However, Edward Cardwell, secretary of state for war in the Gladstone government, which presided from 1868 to 1874, completed the shift to firm civilian control over the military by putting three major military offices—the commander in chief, the surveyor general of the ordnance, and the financial secretary—under the control of the secretary of state for war.72 It took much longer to curtail the prerogatives still held by the House of Lords, namely the power to veto bills passed by the House of Commons. 73 In 1911, the government led by Lloyd George finally succeeded in abolishing the prerogatives of the House of Lords.74 In the meantime, the franchise was expanded in 1832, 1867, and 1884, and the secret ballot was introduced in 1872.75 The Reform Act of 1884 basically adopted universal household suffrage while giving suffrage also to “lodgers who occupied lodgings worth 10 [pounds] a year,” thus increasing the share of people with voting rights to about 16 percent of the total population.76 In 1918, all men over twenty-one years of age were granted suffrage, along with women over thirty who were either taxpayers or wives of taxpayers. In 1928 universal suffrage was finally enacted, giving suffrage to everyone over twenty-one. Nonetheless, elections in the second half of the nineteenth century were hardly free and fair. Vote buying was prevalent and employees were often coerced to vote as their employers wished.77 When we look back on political development in Great Britain in the light of the definition of the semi-democratic regime presented in this book, it is clear that Britain was a semi-democracy in the latter half of the nineteenth century. It was not until the beginning of the last century that it became a full democracy.

  Defining the Regimes of Prewar Japan comparison of the development of semi-democracy in japan and democracy in great britain

In Dahl’s model of democratization (see Figures 2.1 and 2.2), Japan and Great Britain are similar in that the development of political competition (public contestation) preceded the expansion of inclusiveness. In his model, Japan and Great Britain followed path I in Figure 2.2. Although competition in elections was not perfectly fair as late as the late nineteenth century, Dahl argues that in Great Britain conditions for political competition (public contestation) had been developed by the end of the eighteenth century.78 In Japan, although there were some constraints on political ideas (as demonstrated by the continuous suppression of the communists), and the fairness of elections was often undermined by government intervention in favor of parties that supported it, political competition gradually matured after the Diet was established in 1889. Indeed, had political competition been significantly curtailed, it would have been inconceivable for political parties to criticize the government freely in the Diet, to submit multiple motions of no confidence, and to disapprove the budget submitted by the government, which often drove the government to an impasse. The two countries were also similar in the ways political parties had developed and had become increasingly powerful. Political parties in both countries did not emerge as full-fledged modern organized parties but rather as political groups with low internal cohesion that gathered around individual leaders. Yet there were also some differences. By the beginning of the twentieth century, British political parties had developed into modern political entities with strong internal discipline and national party organizations. In Japan, although clear distinctions between the basic policies supported by the major parties had appeared by the 1920s, their regional branches in the countryside still depended on personal connections.79 In this sense, Japanese political parties stopped halfway along the path of development into fully modern party organizations. The two countries traveled a similar path in the growth of electoral control over the government. Great Britain gradually developed a parliamentary government system in which the head of the major party led the cabinet. In the 1830s the crown lost the power to dismiss the prime minister at will, and from the 1870s a practice matured in which the party controlling the House of Commons usually formed the cabinet. In Japan, the emperor never dismissed the prime minister on his own volition (with one exception,

Competitive oligarchies

Polyarchies

III I Liberalization (public contestation)

II

Closed hegemonies

Inclusive hegemonies

Inclusiveness (participation)

figure 2.1. Dahl’s model of democratization source: Robert Dahl, Polyarchy: Participation and Opposition (New Haven, CT: Yale University Press, 1971), 7.

Liberalization (public contestation)

I

B

A

III

B

A

II

Inclusiveness (participation)

figure 2.2. Dahl’s model of paths of democratization source: Robert Dahl, Polyarchy: Participation and Opposition (New Haven, CT: Yale University Press, 1971), 34.

  Defining the Regimes of Prewar Japan

in 1928). Under the semi-democratic regime in Japan, the parties supporting the government normally held a majority of cabinet posts. During this period, there was no parliamentary government in Japan in today’s sense that the political party winning the largest number of seats in the general election forms the cabinet. In Japan, the result of general elections did not determine the makeup of cabinets except for the first Katō Takaaki cabinet in 1924. In fact the Hamaguchi and Inukai cabinets, which were minority governments at their outset, gained the majority in the Lower House through general elections held after their formation. Yet, in Great Britain, the practice whereby changes in the dominant political party in the House of Commons resulted in a change in government did not spring full blown. About thirty years passed between the first time an election led to a change in the government (in 1841, when the Peel cabinet was formed) and the next occurrence (in 1868, when the Disraeli cabinet was established). Given this pattern, one can regard Japan during this time as being in the process of developing such a political custom. Despite their many similarities, there were, of course, great differences between Great Britain and Japan. Great Britain eventually succeeded in bringing most political offices under electoral control, directly or indirectly, although it took a very long time. Civilian control over the military was established under the Gladstone government of 1868 to 1874, and the House of Lords lost its legislative veto power in 1911. Even before the formal establishment of civilian supremacy over the military, cases in which it influenced politics were rare. Japan, however, was unable to place such political offices as the military, the Privy Council, and the House of Peers under electoral control. Certainly, there was significant progress until the 1920s in diminishing so-called reserved domains. For example, the first Yamamoto cabinet made it possible in 1913 to appoint officers in reserve as military ministers, who before then had to be officers in service. In February 1923, Prime Minister Katō Tomozaburō even acknowledged in the Diet that he would support the idea of changing the regulations on military ministers to allow civilians to be military ministers, although he did not specify a concrete time schedule.80 Aside from military matters, in 1925 the first Katō Takaaki cabinet moderately reformed the House of Peers for the first time. Until then, the number of aristocratic members exceeded the number of government-appointed members. Through the reform, the number of members appointed by the government equaled the number of aristocratic members.81

Defining the Regimes of Prewar Japan  73

However, Japan never succeeded in establishing electoral control over the military, the House of Peers, or the Privy Council. Choosing civilians as military ministers was never institutionalized, the military kept its prerogatives, the House of Peers retained the same legal power as the Lower House, and the Privy Council remained intact. These “reserved domains” not only became serious obstacles to the semi-democratic regime’s development into full democracy but also, more importantly, contributed greatly to the collapse of the semi-democratic regime itself. The preceding discussion implies that expansion of electoral control over effective political offices, or the reduction of “reserved domains,” is an important aspect of democratization. Dahl’s model overlooks electoral control and includes only two dimensions, political competition (public contestation) and political participation. This is probably because the country on which he based his model of democratization, Great Britain, did not have serious problems in reducing “reserved domains.” Introducing a third component to Dahl’s model—the reduction of “reserved domains,” or electoral control over all significant political offices—will contribute to better understanding the path of democratization. This new dimension helps us understand, for example, why democratization proceeded smoothly in Japan after the end of the Second World War. As I have demonstrated, the semi-democratic regime in prewar Japan fulfilled to a substantial degree the requirements for democracy in terms of political competition and electoral participation. The largest obstacle impeding the semi-democratic regime’s further development into a fully democratic regime was the ability of so many political institutions that did not fall under electoral control to influence politics. The defeat of Japan in the Second World War led to the dissolution of these institutions.82 The military was fully dissolved, the Privy Council was abolished, and the upper house came to be selected through elections. This means that the obstacles prohibiting the semi-democratic regime from developing further toward democracy were either taken away or fell under electoral control. In other words, the key requirement of electoral control over political offices was fulfilled after the war. This final component also has implications for contemporary democratization and can help us understand the problems faced by currently democratizing countries. In contemporary cases of democratization, political participation is quickly fulfilled. Universal suffrage has been introduced

Electoralisms

Democracies

Semidemocracies

Political competition

Semidemocracies

Semi-competitive civilian regimes

Authoritarianisms

Electoral control over political offices

figure 2.3. A new model of democratization note: Suggestions from Larry Diamond contributed to the typologies of regimes in the upper left and lower right corners in the model. On electoralism, see Terry Lynn Karl, “Imposing Consent? Electoralism vs. Democratization in El Salvador,” in Elections and Democratization in Latin America, 1980–1985, ed. Paul W. Drake and Eduardo Silva (San Diego: Center for Iberian and Latin American Studies, University of California, San Diego, 1986), 9–36.

Political competition

I

B

A

III

B

A

II

Electoral control over political offices

figure 2.4. A new model of paths of democratization

Defining the Regimes of Prewar Japan  75

soon after democratization starts, as was the case with Spain, or it has already been introduced by prior regimes, as was the case in the former Soviet Union. What remain are the guarantee of political competition and the reduction of “reserved domains.” This situation enables us to create a new two-dimensional model of contemporary democratization—democratization in “the third wave”—as well as a new two-dimensional model for paths of contemporary democratization (see Figures 2.3 and 2.4). In the new model for paths of democratization, there are three paths to democracy. The first path is to have full political competition first and then to establish electoral control over political offices (I). The second path is to establish electoral control over all political offices at first, and then to move toward full political competition (II). The third path is to establish electoral control over all political offices and to have full political competition simultaneously (III). This might explain the difference in the smoothness of democratization between Spain and Latin American countries. Spain could take the second path because civilian control over the military had been more or less established by the previous regime, which made it relatively easy to develop democracy.83 By contrast, several Latin American countries in which the military controlled the preceding regime were forced to take the third path, which made the development of democracy more difficult.84

chapter 3

Political Institutions under the Meiji Constitution

In Japan’s semi-democratic regime, the legal system underpinned by the Meiji Constitution strongly influenced the relationship between the party government and the military, imposing many constraints on the former while granting many prerogatives to the latter. On the one hand, constraints imposed by the constitution not only limited the party government’s legal authority over the military but also circumscribed its capacity to make policies that would meet the public’s expectations. On the other hand, the military could employ prerogatives granted by the legal system under the Meiji Constitution to challenge the government. These two issues are the subject of this chapter.

An Institutionally Weak Party Government Party government in Japan’s semi-democratic regime was weak for two major reasons. First, it was unable to maintain internal cohesiveness; second, its political power was fragmented under the legal system based on the Meiji Constitution. The government had difficulty maintaining internal cohesiveness as a collective body because of its fundamental organization. Under the Meiji Constitution, ministers were responsible to the emperor, and the prime minister did not have the right to discharge a minister.1 ­Because in common practice government decisions had to be unanimous, any disagreement within the government could lead to the collapse of the government itself.2 For example, the second Wakatsuki cabinet was driven to resign because of disagreements between Prime Minister Wakatsuki Reijirō 77

  Political Institutions under the Meiji Constitution

and Minister of Interior Adachi Kenzō in December 1931.3 Adachi proposed to Wakatsuki that they form a coalition government with the Seiyūkai to overcome the political crisis triggered by the ever deepening economic depression and the Manchurian Incident. When Wakatsuki, who at first supported the idea, finally decided not to adopt Adachi’s proposal in the face of opposition from other ministers, Adachi refused both to attend cabinet meetings and to resign. As Prime Minister Wakatsuki did not have the right to remove him, Adachi’s behavior paralyzed government operations, and the Wakatsuki cabinet was driven to resign. The party government’s weakness also stemmed from the fragmentation of its political power under Japan’s legal system. Although under the Meiji Constitution the emperor formally was sovereign, in practice he did not exercise political power himself. Various institutions—such as the Lower House, the House of Peers, the government (cabinet), the Privy Council, the genrō, and the military—gave their “consent” and “advice” to the emperor on the exercise of power and thus shared real political power. In fact, as Mitani Taichirō emphasizes, the Meiji Constitution assigned no political institution the role of coordination, thereby preserving the independence of all of them.4 Under the party government coordination between the government and the Lower House became easier than before as political parties functioned as the bond between the two institutions. Thus, the party government provided one solution to the fragmentation of political power under the constitution.5 However, the party government could not exercise power over other political institutions because the constitution fragmented political power. Various institutions, such as the House of Peers, the Privy Council, the genrō, and the military, remained independent of the party government and often impeded its policies. the house of peers

The House of Peers had rights virtually identical to those of the Lower House. It could approve bills and budgets submitted by the government, in addition to enacting bills on its own. Some scholars, such as Satogami Ryōhei and Miyazaki Ryūji, argue that in the latter half of the 1920s the House of Peers’ power to constrain the government diminished.6 However, as they also admit, the House of Peers’ formal power was little affected by the reforms forced on the house in 1925.

Political Institutions under the Meiji Constitution  79

Contrary to these scholars’ arguments, the House of Peers often thwarted the party government’s efforts to put its policies in action by declining to pass bills. In the fifty-sixth Diet, for example, it killed many important bills (such as for tax reform and the Labor Accident Assistance Law) submitted by the Tanaka cabinet by tabling them, even though the Lower House passed these bills.7 The tabling of major bills led a newspaper to report the completion of this session of the fifty-sixth Diet with the following headline: “The Fifty-Sixth Diet Has Ended, Demonstrating No Confidence in the Government: Most Important Bills Finally Dead in the House of Peers.”8 Likewise, in the fifty-ninth Diet, the House of Peers failed to pass the Labor Union Law and the Tenancy Law, both of which had been passed by the Lower House, thereby depriving the party government of important means to deal with labor and tenant farmer unrest. In a word, under the semi-democratic regime, the House of Peers continued to constrain the party government. the privy council

The Privy Council, appointed by the emperor, was established “in order for the emperor to ask for advice on important matters of State.”9 It had the right to approve a wide range of actions, such as the ratification of treaties, amendments to the constitution, emergency decrees, and changes in the organization of major political institutions. This meant that the government had to garner approval from the Privy Council for virtually everything except laws and the budget,10 and even had to secure the Privy Council’s approval of laws related to governing institutions, such as the Law of Houses and the Law on the Organization of the Court.11 Even under the semidemocratic regime, no changes were imposed on the formal power held by the Privy Council. Using its veto power, the Privy Council severely curtailed the power of the party government. For example, it kept the Tanaka cabinet from amending the Law on the Organization of the Court,12 a law related to governing institutions under the Meiji Constitution. The government submitted the bills to the Privy Council to gain its approval before introducing them to the Diet, but the council declined to approve them, and instead recommended that they be withdrawn from consideration. Accordingly, the government withdrew the bills and gave up hope of amending the law. The Privy Council even caused a cabinet to collapse.13 When the first Wakatsuki cabinet tried to deal with a financial crisis in April 1927 through

  Political Institutions under the Meiji Constitution

an emergency decree that would have provided a government guarantee for an emergency loan by the Bank of Japan, the Privy Council refused to approve the decree. This drove the Wakatsuki cabinet into an impasse and led it to resign. the genrō

After the fall of the third Katsura cabinet in 1913, genrō retained the power to recommend candidates for appointment as prime minister. As the emperor normally appointed the prime minister based on the recommendations of the genrō, they had virtual power of appointment. Even under the semi-democratic regime with its de facto parliamentary government, all prime ministers were appointed by the emperor upon the recommendation of the genrō. After genrō Matsukata passed away in July 1924, Saionji alone, as the sole surviving genrō, could exercise this power. The possibility that Saionji could appoint someone who was not a member of a party as prime minister posed a potential threat to party government. The practice of party government was not yet firmly consolidated. In other words, political parties relied on the personal judgments of Saionji to retain control over the cabinet. unstable succession of cabinets

The party government was not a parliamentary government. This raised a major political problem: succession. In parliamentary systems it is firmly established that the politician, normally the leader of a political party, who obtains the support of the majority of the parliament becomes prime minister. He/she stays in power as long as he/she can keep such support. Usually it is general elections that give opposition parties the chance to come to power. There are two ways for a general election to be held. One is when a vote of no confidence passes, in which case the cabinet can call for a general election. Or, the cabinet can hold a general election at its discretion even when a vote of no confidence has not passed. In any case, after a general election the leader of a political party who can command the support of the majority of the parliament becomes the prime minister. The Meiji Constitution did not have any clause on the prime minister and cabinet much less on the nomination of the prime minister. Thus, the party government did not have a firmly established rule on who would become the prime minister. By convention it was, as I have mentioned, left to the genrō. The convention that the general election determines which

Political Institutions under the Meiji Constitution  81

political party holds power had not been firmly established. Thus, there always remained the possibility that opposition parties, even when they did not have a majority in the Diet, could have their leader nominated as prime minister. Likewise, there was no rule on cabinet resignations. In practice, under the semi-democratic regime cabinets resigned when prime ministers became ill or died or when they faced a political deadlock.14 This aspect of the Meiji Constitution often seduced politicians of opposition parties to drive the cabinet into political difficulties even by relying on nondemocratic forces, including the military, expecting their leader would become the next prime minister. The party government might have been able to set up customary rules on cabinet succession if a cabinet had chosen to dissolve the Lower House and call a general election when it was driven into a political impasse, but no cabinet made such an attempt.

Party Government under the Semi-Democratic Regime Versus the British Parliamentary Government The weakness of Japan’s party government in the semi-democratic regime is obvious when compared with the British parliamentary government at this time. First, the British parliamentary government did not suffer from weak internal cohesiveness; the prime minister’s strong leadership of the government was established in the period from the end of the nineteenth century to the beginning of the twentieth.15 Above all, although the Japanese prime minister did not have the right to remove other ministers, in the British system this right had been firmly established by the end of the 1880s.16 Second, while the British parliamentary government ensured coordination between the House of Commons and the government, it did not face significant political challenges from other political institutions, unlike Japan’s party government in the semi-democratic regime. The House of Commons had achieved complete dominance over the House of Lords by the beginning of the twentieth century and by the end of the nineteenth century its control over the military was also firmly established.17 In other words, the British parliamentary government succeeded in consolidating political power, something that Japan’s party government was never able to achieve before the war. Third, the British parliamentary government didn't have to be ­concerned about cabinet succession since the rule on cabinet succession had been firmly established in 1870s.

  Political Institutions under the Meiji Constitution

An Institutionally Strong Military As we have seen, the party government under the semi-democratic regime was institutionally weak. The military, on the other hand, retained great political power under the legal system based on the Meiji Constitution. It not only enjoyed significant autonomy from the party government through its legal prerogatives, but it could potentially influence the policy of the party government through its ministers in the government and even formulate and implement policies on its own in such areas as military operations, a power fully demonstrated in a series of conspiracies the army spearheaded in Manchuria. To appreciate the prerogatives of the military and the extent of its autonomy from the party government, it is necessary to understand how issues relevant to the military could be handled and which military institution had jurisdiction over these issues.18 Military issues were divided into two large domains. The first involved military operations, such as the formulation of strategies for warfare or delivery of orders in military campaigns. The second involved military organization, such as the quantity of armaments and the structure of the military. The army and the navy each comprised two major branches, chiefs of the General Staff and ministers. While the chiefs of the General Staff were responsible for operational issues, the ministers looked after organizational matters. issues related to military operations

Article 11 of the Meiji Constitution, as noted in Chapter 2, stipulated that “the Emperor has the supreme command of the Army and Navy,”19 meaning that he had direct command over military operations. The Meiji Constitution itself contained no article that secured the military’s autonomy from the government. However, due to practices established even before the constitution was introduced, Article 11 was interpreted as giving the military autonomy over its operations.20 Procedurally, the chiefs of the General Staff of the army and the navy had the right to report directly to and get approval from the emperor with only ex post facto reports to the prime minister. 21 This practice was legally affirmed by Article 6 of “On the Offices of Cabinet Ministers,” adopted in 1885 as an order by the Chancellor of the Realm, and Article 7 of the “Cabinet-Official Organization,” adopted in 1889 as an imperial ordinance.22

Political Institutions under the Meiji Constitution  83 issues related to military organization

Article 12 of the Meiji Constitution stipulated that “the Emperor determines the organization and peace standing of the Army and Navy.”23 However, politicians such as Itō Hirobumi and Itō Miyoji, who were involved in drafting the Meiji Constitution, asserted that, in contrast to the military’s autonomy in issues related to operations, the government was responsible for matters relating to the military’s organization.24 In reality, however, the government’s control over “the organization and peace standing of the Army and Navy” was ambiguous. For one thing, as has been described, army and navy ministers also had the right to report directly to the emperor and have their policies approved without prior consultation with the government and with only an ex post facto report to the prime minister under Article 7 of the “Cabinet-Official Organization.” The chiefs of the General Staff of the army and the navy shared this privilege.25 This was an exception to the practice of other ministers in reporting to the emperor. Although the Meiji Constitution dictated that each minister was individually responsible to the emperor, individual ministers did not have the right to report directly to the emperor to get approval for policies within their jurisdiction.26 Rather, the prime minister reported to and received approval from the emperor on their behalf, securing the integrity of the government.27 Using their prerogatives, the army and navy ministers often (though not always) reported directly to and got approval from the emperor on the organizational affairs of the military without prior consultation with the government.28 Certainly, the army and navy ministers were members of the government. Therefore, as long as they acted as members of the government, this prerogative was not likely to have caused serious problems. But if they chose to act on behalf of the military, they could potentially curtail the government’s ability to control the military, even with respect to military organization. A second and even larger problem was posed by the fact that it was hard to draw a clear line between military operations and organizational issues. This became an obstacle to the government exercising control over organizational affairs.29 The General Staff of the military often felt that the emperor as supreme commander had authority not only over operational issues but also over some organizational ones, in particular the amount of armaments, thus expanding the range of issues that fell within the emperor’s purview.30 Therefore, the military considered it a violation of the emperor’s supreme com-

  Political Institutions under the Meiji Constitution

mand for the government to make decisions on military organizational affairs without the military’s prior consent.31 As we will see in Chapter 6, this gave rise to a serious problem over the signing of the London Naval Treaty in 1930. officers as military ministers: two faces of military ministers

In 1913, the first Yamamoto cabinet expanded the eligibility of military ministers so that officers in the first and second reserve, in addition to officers in service, were eligible to serve in the cabinet. This reform potentially curtailed the military’s capacity to topple the government by declining to send its officers as ministers. Under the semi-democratic regime, however, the army and navy ministers continued, in practice, to be officers in service, creating a complicated situation. On the one hand, military ministers could act as members of the government, with accompanying responsibility for the military’s organizational affairs. On the other, military ministers could act as military officers, representing the interests of the military as a whole in the government. Often the role of military ministers shifted; when they decided to act as members of the government, they could contribute to containing the General Staffs’ claim to authority over organizational affairs and to implementing the government’s policy. Also when military ministers decided to act on behalf of the government, they could even contribute to the virtual subordination of the supreme command to the government through their personal influence. But when military ministers decided to represent the military’s interests in the government, the government’s capacity to control the military was constrained and the military’s autonomy was buttressed. Under these circumstances, the military could project its influence in policy formulation through its ministers.32 Given that government decisions had to be made unanimously and that the prime minister could not discharge other ministers, the military retained a virtual veto power over the party government when its ministers acted on behalf of the military.

Party Government’s Weakness and the Military’s Strength: Political Implications The party government’s institutional weakness and the military’s institutional strength obviously put the party government in an inferior position. Moreover, the party government found it difficult to sustain unity against

Political Institutions under the Meiji Constitution  85

the military. At the same time, constraints imposed by other major political institutions curtailed its ability to implement policies to address social problems, which in turn eroded the regime’s legitimacy. This weakening of legitimacy indirectly deprived the party government of the power to confront the military and maintain the regime. The historical prerogatives that secured the military’s autonomy enabled it to defy the policies of the party government and to act on its own initiative. In doing so, the military justified its actions as falling under the emperor’s supreme command. Moreover, even when the government wanted to modify actions initiated by the military, it found it hard to do so because it lacked formal authority to discipline the military. The ambiguity of jurisdiction over military organizational affairs also gave the military an opportunity to attack the party government. The party government’s weakness and the military’s institutional strength under the Meiji Constitution’s legal system remained basically unchanged during the course of the semi-democratic regime. Nevertheless, in 1932, the semi-democratic regime broke down, implying that forces other than institutional ones were at work in determining its fate. Here it is necessary to introduce other factors, such as legitimacy and semi-loyalty, to explain how the balance of power between the military and the party government decisively shifted toward the military, culminating in the breakdown of the semi-democratic regime. In the following chapters, we will see how changes in certain dynamic variables—specifically, regime legitimacy and the semiloyalty of political parties and party politicians—triggered the breakdown of the semi-democratic regime.

chapter 4

The Semi-Democratic Regime, 1918–1926 Containment of the Military

Despite the military’s institutional prerogatives, the party government succeeded in constraining it from 1918 to 1926. This was because the semidemocratic regime enjoyed a high and sustained level of legitimacy among a broad swath of the population.

The Relationship between the Party Government and the Military As we have seen, the party government was fundamentally inferior to the military because of its institutional weakness and the institutional prerogatives guaranteed the military by the Meiji Constitution. Yet between 1918 and 1926, the party government did not face any serious threat from the military; quite the contrary, it succeeded in containing the military. During this period, the relationship between these two institutions was characterized by challenges the party government launched against the military.1 Even during the period when nonparty governments ruled the country, pressure from the political parties contained the military’s power. the hara cabinet

The Hara cabinet curbed the military in three ways: through de facto subordination of the supreme command to the government over military operations in Siberia; by appointing civilians to act as agents of military ministers in their absence; and by enabling civilians to be appointed as governors of Japanese colonies.2 87

  The Semi-Democratic Regime, 1918–1926

The Hara cabinet subordinated the emperor’s supreme command over the military to the government on a de facto basis in the case of army operations in Siberia.3 After the number of troops sent to Siberia had reached as many as seventy-two thousand at one point after August 1918, the Hara cabinet gradually reduced that number and shrank the front line.4 It devised these policies without consulting the General Staff of the army; rather, it notified the General Staff after gaining the emperor’s approval.5 This was possible because Army Minister Tanaka Giichi supported these policies and contained the General Staff of the army.6 The General Staff of the army did not have to follow the orders of the army minister, who was formally a member of the cabinet, but was bound to adhere to the orders of the chief of the army’s General Staff. Nonetheless, the General Staff followed the government’s directions. Therefore, it can be said that the Hara cabinet achieved de facto subordination of the supreme command over the military to the party government, at least in affairs related to the withdrawal of troops from Siberia. The Hara cabinet further succeeded in enhancing the power of the party government when Japan, the United States, Great Britain, and other major powers convened the Washington Conference in November 1921 to discuss the reduction in naval armaments and other affairs related to the East Asia and the Pacific. Navy Minister Katō Tomozaburō was sent to Washington as one of the Japanese delegates, which raised the question of how to manage the affairs of the navy in his absence. Despite the army’s strong opposition, the Hara cabinet entrusted Prime Minister Hara with affairs related to the navy in Katō’s absence.7 As Hara stated, this was an improvement over the then-current regulation; it set a precedent in which a civilian handled military affairs, paving the way for the future appointment of civilians as military ministers.8 Until the Hara cabinet, governors of the Japanese colonies—Korea, ­Taiwan, and the Kwantung Leased Territory—were required to be military officers, a significant military prerogative. The Hara cabinet broadened the eligibility for these posts to include civilians.9 These were important posts since governors had significant influence over the management of economic resources and, more importantly, over Japanese foreign policy toward China, given the strategic location of the colonies.10 The expansion of eligibility “reduced the political significance and power of the military, which thus far had held exclusive control over the governance of the colonies.”11

The Semi-Democratic Regime, 1918–1926  89 after the hara cabinet

Challenges to the military continued after the demise of the Hara cabinet in November 1921. Political parties increasingly demanded civilian military ministers, and in March 1922 the Lower House passed a resolution requesting that civilians be appointed as military ministers.12 As described in Chapter 2, seeing an increased demand to appoint civilians as military ministers, Prime Minister Katō Tomozaburō expressed in February 1923 his support for amending the regulations on military ministers to pave the way for civilian military ministers, although he did not refer to any concrete time schedule in the Diet.13 A series of arms reductions was also imposed on both the army and the navy, though not all reductions took place under the party government. Some were implemented by the Katō Tomozaburō cabinet, which was not a party government, even though the Seiyūkai acted as a supporting party of the cabinet. In February 1922, the Takahashi cabinet, the successor to the Hara cabinet, signed the Washington Naval Treaty, even though it would reduce armaments more drastically than the navy had initially expected based on discussions at the Washington Conference. Under the treaty, fourteen major warships were scrapped and construction of six others, already begun, was terminated.14 Following the reduction in naval armaments, there were increasing demands that the army’s armaments also be reduced. In March 1922, under the Takahashi cabinet, the Lower House passed a resolution requesting a reduction in the size of the army.15 In response, the army began to design plans to do so. Under the next cabinet, led by Katō Tomozaburō, the army reduced its forces in August 1922, releasing 2,000 officers and 57,000 noncommissioned officers and soldiers.16 Shrinking the size of the army was also on the agenda of the first Katō Takaaki cabinet. In 1925, it disbanded four divisions, releasing 1,200 officers and 36,000 noncommissioned officers.17 In sum, the relationship between the party government and the military between 1918 and 1926 was characterized by continuous challenges to the military from the party government. Certainly, the party government was not fully successful. The military remained formally independent from the government in operational affairs, civilians continued to be ineligible to be military ministers, and, despite a huge reduction in manpower, the army cutbacks in 1925 did not significantly reduce the military budget, because the army used the money freed up by the reductions to modernize.18 Yet

  The Semi-Democratic Regime, 1918–1926

the party government did gain the upper hand in its relationship with the military, providing little room for the military to topple it.

Analysis: The Legitimacy of the Semi-Democratic Regime During this period, the semi-democratic regime was stable. The party government got the upper hand against the military because the semi-­democratic regime had a high degree of legitimacy. As Juan Linz writes, “it seems unlikely that military leaders would turn their arms against the government unless they felt that a significant segment of the society shared their lack of belief and that others were at least indifferent to the conflicting claims for allegiance.”19 Likewise, on the effect of regime legitimacy on the civilmilitary relationship, Felipe Agüero argues, “a military will find it harder to push for nondemocratic prerogatives and to resist government policies when the government is visibly backed by a wide array of electorally strong political forces.”20 In short, when a regime enjoys legitimacy among a wide segment of the population, the military obeys the government and does not challenge it. Such was the situation in Japan between 1918 and 1926. The semi-democratic regime could claim legitimacy because it provided legal opportunities for a large segment of the population to participate in politics and because the party government was formed by political parties chosen by the electorate. Bolstered by this legitimacy, the party government was able to overcome its institutional weakness and confront the military. the popularity of the hara cabinet

The formation of the Hara cabinet was applauded because it was viewed as a party government that reflected the interests of a wide segment of the population, not because it possessed elements unique to the cabinet or to Prime Minister Hara. This demonstrates that the semi-democratic regime had legitimacy at the outset. Legitimacy among the Elite  The mass media and intellectual elite welcomed the foundation of the Hara cabinet because it was a party cabinet. For example, Chūō Kōron, a leading magazine at the time, praised the cabinet’s foundation: We . . . have held firm to the belief that the trend of the times would ultimately lead to the rise of the Hara cabinet and have been expecting its rise at the earliest

The Semi-Democratic Regime, 1918–1926  91 moment possible. This is not because we trust Mr. Hara or the Seiyūkai. This is because allowing the political party with a plurality in the Lower House to manage politics is the only way to bring the people and politics close to each other today. Without bringing the people and politics close to each other, there will be no sound progress in politics based on the constitution. When people become politically conscious, political parties cannot become plural parties without respecting the people. Therefore, the people’s disposition will be reflected through plural parties. . . . We believe that the emergence of this cabinet has a very significant meaning in our political history.21

Likewise, the Tōkyō Asahi Shimbun, a leading newspaper, embraced Hara Takashi’s appointment as prime minister the day after his appointment: From the perspective of correct practice in politics based on the constitution as well as the general tide in current politics, the person who would form the cabinet after the Terauchi cabinet had to be Mr. Hara Takashi, president of the Seiyūkai. . . . If Mr. Hara accepts the appointment and forms the cabinet, the cabinet will of course be a party cabinet. . . . I cannot help but be pleased that this is the result of the power of the political parties finally being recognized in society. In particular, I am very content that, for the sake of progress in politics based on the constitution, Mr. Hara, who is not a peer, will form the cabinet.22

Yoshino Sakuzō, a professor in the Faculty of Law of the University of Tokyo, long a strong advocate of party government, wrote: The citizens have loathed the predominance of the bureaucratic and military cliques for a long time. . . . What the citizens sincerely desired was . . . a cabinet that had its basis in the citizenry. This desire indeed gave rise to the Hara cabinet.23

In the same vein, Yoshino, under the penname of Furukawa Gakuto, ascribed the rise of the Hara cabinet to the wishes of the people: It was the trend of the times and the citizens’ desire that drove [the genrō to recommend Hara Takashi as the next prime minister]. We can say that the citizens removed Terauchi and invited [Hara]. This means that the times have established a principle that we should not entrust government to the discretion of a few bureaucrats and have gradually worked on the citizens, making them understand that they have to have a strong government that is based in the masses.24

Yamakawa Hitoshi, a socialist ideologue, also welcomed the Hara cabinet as “an inevitable stage in the political development of Japan.”25 Available evidence shows that even traditional oligarchic leaders endorsed the legitimacy of the regime. Matsumoto Gōkichi, Yamagata’s secretary,

  The Semi-Democratic Regime, 1918–1926

wrote in his diary that Yamagata had told him that “while the polity has to adopt constitutional monarchy, politics must be based on minponshugi.”26 Minponshugi, which can be translated as the “principle of the people as the base,” was an idea associated at the time with both the party government system and universal suffrage.27 Legitimacy among the Public  Although it is difficult to collect evidence to assess the semi-democratic regime’s initial level of legitimacy among the public, the following evidence suggests that such legitimacy did exist. Maeda Renzan, a contemporary journalist, described later how ordinary people welcomed the emergence of party government: Since the establishment of the cabinet system up to now, the emergence of no other cabinet has deeply impressed the people of the country the way the Hara cabinet has. Celebrations and commemorations organized by private organizations have continued over a number of days and evenings.28

Likewise, Suzuki Bunji, a labor leader and the president of the Yūaikai trade union, enthusiastically greeted the Hara cabinet because Prime Minister Hara was a common man, not a peer, and his cabinet was a party government.29 In view of Suzuki’s eminence, his opinion can be taken as representative of the position of labor at the time. the unpopularity of the hara cabinet

Although the Hara cabinet was popular when it emerged, as many historians have pointed out it became unpopular toward the end of its tenure.30 As discussed in Chapter 1, when a regime is established but its legitimacy has not yet been solidified, its performance will have a direct influence on its legitimacy. Since the semi-democratic regime had just emerged in 1918, its legitimacy could not have been firmly established during the Hara cabinet’s tenure. Therefore, if the Hara cabinet became unpopular toward the end of its tenure, one must examine whether the regime’s performance under the Hara cabinet actually diminished its legitimacy. The Hara cabinet faced a crucial and controversial issue: should there be a limited expansion of suffrage or, instead, should universal male suffrage be introduced? The cabinet came down in favor of a limited expansion, generating a storm of criticism. When it introduced a bill to expand the franchise in 1919, Ishibashi Tanzan, a leading journalist (who became prime minister after World War II), criticized the cabinet for merely ex-

The Semi-Democratic Regime, 1918–1926  93

panding suffrage instead of introducing universal male suffrage. He judged the expansion insufficient to accommodate citizens’ entry into politics. 31 The Tōkyō Asahi Shimbun also repeatedly criticized the Hara cabinet for not introducing universal male suffrage.32 Some even argued that the absolute majority the Seiyūkai won in the Lower House in the general election of 1920 did not actually represent a victory of the citizenry because of the limited franchise and the newly introduced small electoral district system, which was advantageous to the Seiyūkai in particular.33 Despite such criticism, the Hara cabinet’s handling of the issue of suffrage probably did not erode the regime’s legitimacy. When the Hara cabinet emerged, many people expected it to expand the franchise.34 However, and despite Ishibashi’s criticisms, there was not yet a true surge of demand for universal male suffrage. Even opposition parties, such as the Kenseikai, had not begun advocating the introduction of universal male suffrage. ­Matsuo Takayoshi argues that the mood reflected in the mass media indicated support for the expansion of suffrage, but not the introduction of universal suffrage.35 It therefore seems fair to say that in 1919 the Hara cabinet responded swiftly to this demand, expanding the franchise in the first regular session of the Diet in 1919, and thereby sustaining the regime’s legitimacy. With respect to the issue of universal male suffrage, I base my arguments on Banno Junji’s careful examination of how the Hara cabinet tackled this challenge.36 When the Kenseikai and the Kokumintō submitted a bill to introduce universal male suffrage in 1920, the Hara cabinet dissolved the Lower House to let the electorate determine whether universal male suffrage should be introduced. This involved choosing between the Seiyūkai, which opposed the immediate introduction of universal male suffrage, and opposition parties, which supported its immediate adoption. 37 This decision was problematic and controversial because it involved asking the already enfranchised part of the electorate to decide whether to extend suffrage to other segments of the population. Examining how intellectuals at the time, such as Ishibashi Tanzan and Yoshino Sakuzō, viewed this decision, Banno concludes that it is difficult to criticize the Hara cabinet’s political maneuver because it was the only legal means then available to incorporate the greatest number of the population in decision making.38 In the general election of 1920, the Seiyūkai won an absolute majority in the Lower House, which justified the Seiyūkai’s argument that universal male suffrage was ­premature. The Hara cabinet probably could have

  The Semi-Democratic Regime, 1918–1926

s­ trengthened the legitimacy of the semi-democratic regime by introducing universal male suffrage. That is not to say, however, that the legitimacy of the semi-­democratic regime declined under the Hara cabinet because it was reluctant to introduce universal male suffrage. The process it embraced— namely, leaving the ultimate decision to the electorate—maximized the incorporation of public opinion in politics. The question of suffrage aside, the Hara cabinet often implemented policies aimed at expanding the Seiyūkai’s base of support. It manipulated personnel policies in the bureaucracy and engaged in pork-barrel politics in order to develop infrastructure, such as railroads.39 However, the way the Hara cabinet managed the Lower House after the victory of the Seiyūkai in May 1920 was high-handed, and a series of corruption scandals involving leaders of the Seiyūkai was revealed toward the end of its tenure.40 These policies and incidents understandably undermined the cabinet’s reputation.41 Some might argue that the Hara cabinet’s ultimate unpopularity diminished the legitimacy of the semi-democratic regime. While this is plausible, I would suggest, however, that the semi-democratic regime retained its legitimacy, not least because the public considered it better than the regimes that Japan had experienced up until then. Regime legitimacy is by definition comparative. Even when a regime’s legitimacy has not been firmly established, poor regime performance may not lead to erosion of legitimacy if the public strongly rejects alternative regimes. In the years between 1918 and 1926, the alternative regime was the former competitive oligarchical regime, which many disliked. To borrow Juan Linz’s words, the public considered the semi-democratic regime during this period to be “better than any others that might be established . . . in spite of [its] shortcomings and failures.”42 Yoshino Sakuzō evaluated the regime in this very context in March 1920, noting “it goes without saying that even today’s imperfect constitutional politics with many evil practices is much better than that in the past with its much more restricted franchise.”43 the first katō takaaki cabinet

In October 1921, Hara Takashi was assassinated, which would seem to underscore the regime’s increasing unpopularity and declining legitimacy. Public reaction to the Kiyoura cabinet at the beginning of 1924 demonstrated, however, that the semi-democratic regime could claim legitimacy, even if that legitimacy sprang from public belief that the existing govern-

The Semi-Democratic Regime, 1918–1926  95

ment was the best among poor options. When the second Yamamoto cabinet resigned at the end of December 1923, the genrō Saionji Kinmochi and Matsukata Masayoshi recommended that the emperor nominate Kiyoura Keigo, the chairman of the Privy Council, as the next prime minister. The Kiyoura cabinet emerged as the third nonparty government in a row with initial support only from the House of Peers, which could have brought Japan back to the former competitive oligarchical regime.44 Because of this, the Kiyoura cabinet was subjected to overwhelming public criticism when it came into being. After Kiyoura was appointed prime minister, the Ōsaka Asahi Shimbun denounced his appointment and reported on January 5, 1924, that “it is extremely anachronistic that Mr. Kiyoura ignores public opinion, reverses the direction of political development, and forms a House of Peers cabinet without any basis in political parties.”45 The Tōkyō Nichinichi Shimbun likewise condemned his appointment, and roundly criticized the nonparty cabinet: All recent cabinets formed on the basis of the support of the House of Peers have been impotent and inexperienced. They not only failed to respond to the expectations of the citizens but also delayed the formulation of various policies necessary for progress. . . . The public is fully aware that the incompetence and lack of principle of the two [past] cabinets [the Katō Tomozaburō cabinet and ­Yamamoto cabinet] were due to the fact that they had their base in the House of Peers without any close relationship to political parties. In light of this, now that the Yamamoto cabinet has collapsed, it is a matter of course and the view of the public that the forthcoming new cabinet should have a solid foundation in the Lower House and have the capacity to forcefully implement its principles and policies. However, to ignore this fact and to form [the Kiyoura cabinet] not only ignores the reality of politics but fully defies the demands and interests of the citizens.46

The criticisms of the mass media all converged on the point that the ­ i­youra cabinet could not reflect the citizens’ interests and opinions in polK icy making because it was not a cabinet based on political parties. In other words, behind the criticism of the Kiyoura cabinet was public rejection of the former competitive oligarchical regime and support for the semi-­democratic regime. Criticism was so strong that even Prime Minister Kiyoura was forced to make the unusual announcement that “I usually believe that a cabinet based on political parties is the most appropriate for conducting politics based on the constitution.”47

  The Semi-Democratic Regime, 1918–1926

The results of the general election in May 1924 and the tone of the mass media after the election further suggest that the semi-democratic regime had legitimacy, due to the total rejection of the former competitive oligarchical ­regime. After the Kiyoura cabinet was formed, the Kenseikai, the Seiyūkai, and the Kakushinkurabu agreed to join forces to form a party government and reject the cabinet, starting so-called the Second Pro-Constitution Movement.48 In return, the Kiyoura cabinet dissolved the Lower House and held a general election in May 1924. The three parties, which urged support for party government, won the election over the Seiyū Hontō, which had split from the Seiyūkai to support the Kiyoura cabinet in the Lower House. The election was fought over whether a party government or a nonparty government was to rule. The result was clear. The electorate demonstrated its firm preference for party government and thus for the semi-democratic regime. The three political parties won 285 out of 464 seats in the Lower House, as opposed to 110 seats won by the Seiyū Hontō.49 It is significant that the three parties were able to win overwhelmingly despite the direct and indirect intervention and coercion that the government marshaled against them in the general election.50 After the general election, the Tōkyō Asahi Shimbun wrote, “the citizens demonstrated their will that the Kiyoura cabinet should collapse and a party government . . . should be established.”51 The preference for party government over nonparty government was so strong that many newspapers emphasized that the Kiyoura cabinet had to be the last nonparty cabinet.52 The Yomiuri Shimbun even cynically wrote that the only good thing the Kiyoura cabinet did was to mark the real end of nonparty government.53 This evidence demonstrates that the semi-democratic regime had legitimacy, if only based on a rejection of its alternative. The following evidence suggests that around the time the Kiyoura cabinet resigned and the first Katō Takaaki cabinet took office, the semi-democratic regime also had legitimacy in its own right. This was because its policies reflected the interests of a wide segment of the population. The Ōsaka Asahi Shimbun revealed its support for the party government system for its own sake on January 10, 1924: It goes without saying that the essence of politics based on the constitution [should be the following]. The political parties should design national policies for the citizens as a whole and ask citizens [to judge] whether their policies are right or wrong. Then, the political party that has principles that match the will of the citizens should form the cabinet. If its principles no longer agree with the

The Semi-Democratic Regime, 1918–1926  97 will of the citizens [during its time in office], it should transfer the government to an opposition party.54

Minobe Tatsukichi, a professor in the Faculty of Law at the University of Tokyo, compared the formation of the Katō cabinet with that of the Hara cabinet, suggesting that he, too, valued the semi-democratic regime for its own sake: The replacement of the Kiyoura government by the Katō government has made me feel very comfortable, as if a long rainy season has ended and the sun, though weak, has begun to shine. This feeling is similar to the one I had when the ­Terauchi cabinet collapsed and the Hara cabinet replaced it several years ago. . . . Because of its members, the new government is much more worthy of people’s trust than the previous government.55

The way the first Katō Takaaki cabinet emerged very likely helped to enhance the legitimacy of the semi-democratic regime. The will of the electorate, as expressed in the general election, led to the formation of the cabinet. This point was highly valued by opinion leaders. For example, on June 10, 1925, Tōkyō Asahi Shimbun reported, “in the eyes of the people when [the three parties supporting party government and universal suffrage] won the election and the Kenseikai became the ruling party in the Diet, it had already become a matter of course that Mr. Katō [the president of Kenseikai] would be appointed the prime minister.”56 It also lauded his appointment as being in accord with the expectations of the citizens and considered it “progress in politics” because “[in the past] the transfer of the government had not necessarily taken place in the way which had been expected, and the expectations of the citizens had often been betrayed.”57 Yamakawa Hitoshi praised the Katō cabinet in the same vein: The change in the government this time had [a unique] characteristic . . . [namely,] ordinary citizens could anticipate who would form the government. And the government changed as they had expected. The change in the government was boring. There was little excitement among “the audience” because there was none of the wrangling, farce, drama, or sideshow that has normally taken place. Because of this, however, the change of government this time must have given some idea to the citizens that they were not a mere “audience.” The emergence of a government “trying to protect the constitution” [i.e., party government] is significant in that the process of change in government became simple; that is, it became clear that a direct and clear relationship exists between the vote of the electorate and who will form the government.58

  The Semi-Democratic Regime, 1918–1926

Saionji’s shift from recommending Kiyoura Keigo as prime minister in January 1924 to supporting Katō Takaaki as the next prime minister in June 1924 further suggests that the semi-democratic regime enjoyed legitimacy. No primary source material directly reveals Saionji’s motives for appointing Kiyoura to form a nonparty government, yet it has been widely speculated that he appointed Kiyoura to organize a caretaker cabinet and have the cabinet conduct the general election, which was scheduled for May 1924, when the Lower House’s term expired.59 It is also believed that Saionji had already decided to let the plural party form the new cabinet, replacing the Kiyoura cabinet after the general election. We should also note that Saionji recommended Katō Takaaki despite his personal disposition against him, which underscores Saionji’s evident respect for the will of the electorate as expressed in the general election. Saionji personally lacked confidence in Katō, which led him to refrain from recommending Katō to succeed Prime Minister Takahashi when the Takahashi cabinet resigned in 1922.60 Backed by the legitimacy of the semi-democratic regime, the first Katō Takaaki cabinet overcame the resistance of the House of Peers and succeeded in introducing universal male suffrage. Since the beginning of the 1920s, a large part of the population had favored the introduction of universal male suffrage, which quite likely contributed to further strengthening the regime’s legitimacy. Evidence suggests that even the military recognized the regime’s legitimacy during this period. For example, at some point between the beginning of November and the end of December 1924, Army Minister Ugaki wrote in his diary: “[We] cannot ignore the [demand for arms reductions] in managing the current situation. Rather, it is necessary to deal with [the demand] with significant respect and appropriate honesty. I have to work with patience and forbearance.”61 Around July 1925, Ugaki further wrote that “aristocratic politics is not appropriate while party politics is not without negative side effects. Despite problems and troubles that accompany party politics, I must say that party politics is the trend and major current in the world.”62 In a letter dated May 26, 1924, when the Kiyoura cabinet was about to resign following the general election, Hoshino Shōzaburō, the commander of the Ninth Division, wrote that General Tanaka Giichi, one of the most powerful leaders of the army at the time, told newspaper reporters that “[the next government] of course would be a party government.”63 In a letter

The Semi-Democratic Regime, 1918–1926  99

dated June 12, shortly after the Katō cabinet took office, Hoshino himself wrote, “I am very happy with the emergence of the new government and the stability of politics.”64 Warning against army involvement in politics, he commented “today we should not mix the army and politics.”65

Seeds of the Semi-Democratic Regime’s Breakdown Although the semi-democratic regime that emerged in 1918 demonstrated its strength in 1924 when party government was revived, it was then that the seeds of its breakdown were planted. Party government failed, except for the introduction of universal male suffrage, to undertake fundamental institutional reforms of the legal structure, thereby forfeiting an opportunity to strengthen its legitimacy, which would in turn have strengthened party government. By failing to implement legal reforms that would have extended electoral control over political institutions not under the electorate’s control, the party government left untouched the wide-ranging power of the Privy Council and the House of Peers. More importantly, because of the party government’s inaction, the military formally retained all of its political power; its autonomy in operational affairs likewise remained intact. The ambiguity over which institution—the government or the military—was responsible for organizational affairs persisted during this period, as at the beginning of the semi-democratic regime. The Privy Council and the House of Peers used their powers to obstruct the party government in formulating and implementing policies dealing with various social and economic problems; this diminished the semi-­ democratic regime’s legitimacy. The prerogatives of the military that were left intact created a firm foundation upon which the military was able to challenge party government in later periods.66 The failure to introduce universal male suffrage at an earlier point represented a missed chance to strengthen the legitimacy of the regime. Banno Junji addresses this issue in his article “Seitō Seiji no Kakuritsu” (The ­Establishment of Party Politics). It was difficult for the people to criticize the dissolution of the Lower House in 1920 on the pretext of “resorting to the judgment of the citizens” on whether to introduce universal male suffrage.67 Banno nonetheless feels that the general election of 1920 was “a deadly political mistake for the development of party politics in prewar

  The Semi-Democratic Regime, 1918–1926

Japan.”68 It delayed the introduction of universal male suffrage “by putting the Seiyūkai in a position of not being able to support any draft [introducing] universal male suffrage” once it had opposed the introduction of universal male suffrage under the pretext of “national opinion.”69 While he clearly assumes that the delay in introducing universal male suffrage undermined the development of party politics in Japan, Banno does not explain why this is so. An early introduction of universal male suffrage not only might have strengthened the semi-democratic regime’s legitimacy, but also could have allowed the party government to embark on other reforms to enhance the electorate’s control over other political institutions, particularly the House of Peers. When the Katō cabinet was formed, one of the major tasks it faced was the introduction of universal male suffrage. While it also attempted to reform the House of Peers to diminish its power in response to pressure from the public, this attempt was half-hearted. In order to introduce universal suffrage, the cabinet had to gain approval from the House of Peers, which had traditionally been reluctant to implement it. In other words, the House of Peers held universal male suffrage hostage to prevent the drastic reform of its own workings. Had universal male suffrage been introduced before the first Katō Takaaki cabinet was established, the Katō cabinet could have concentrated more on reforming the House of Peers. And, given the support it enjoyed when established, the Katō cabinet probably could have curtailed the political power of the House of Peers. As it was, the House of Peers continued to impede the party government in its efforts to carry out various policies, which affected the efficacy and effectiveness of the semi-democratic regime, thereby eroding its legitimacy. Reform of the House of Peers would have made it easier for the party government to implement various policies dealing with social and economic problems and could have prevented the erosion of regime legitimacy. Therefore, the delay in introducing universal male suffrage indirectly contributed to the erosion of the regime’s legitimacy, and in turn weakened the party government’s political power. By extending suffrage to citizens without property, the party government also implemented policies that discouraged those without property from gaining political representation—and alienated them as a result. The introduction of the Peace Preservation Law in 1925 which outlawed communism was a conspicuous example. Such practices simultaneously forfeited the chance to strengthen legitimacy and eroded it.

The Semi-Democratic Regime, 1918–1926  101

During the period between 1918 and 1926, the party government was on the offensive against the military. The party government could put itself in a position superior to the military because the regime could claim legitimacy among a wide segment of the population, including the military. To be sure, the level of legitimacy diminished toward the end of the Hara cabinet, but legitimacy generally ran high when both the Hara and the first Katō Takaaki cabinets were in power. At the time, the military followed the policies of the party government. As a result, the party government did not feel it necessary to revise the institutional prerogatives of the military. This, however, meant that the party government forfeited the chance to reduce military prerogatives, which would have strengthened its position and regime legitimacy. In the face of these missed opportunities, the military endured the party government’s challenges and succeeded in retaining all of its power, which it would later use to attack party government. Both the party government’s failures and the military’s successes would have serious repercussions for their relationship after 1926.

chapter 5

The Semi-Democratic Regime, 1926–1929

The Rise of the Military and the Decline of Party Government

Although the semi-democratic regime was vigorous between 1918 and 1926—from the Hara cabinet to the Katō Takaaki cabinet—it gradually weakened between 1926 and 1929 under the first Wakatsuki cabinet and the Tanaka cabinet. This period was transitional in many senses. The balance of power between party government and the military began to tilt toward the military due to a decline in the regime’s legitimacy. During the regime’s final collapse, several incidents also decisively affected the relationship between party government and the military. The shift in the balance of power was clearly demonstrated by the assassination of Chang Tso-lin and—in the face of opposition from the army—the party government’s failure to punish the officers involved in the crime.

The Relationship between the Party Government and the Military In June 1928, Japanese army colonel Kōmoto Daisaku assassinated Chang Tso-lin, a Chinese warlord who controlled Manchuria, by blowing up the train on which Chang was returning from Beijing to Manchuria.1 This incident was a two-pronged challenge of the party government by the military. First, Chang’s assassination defied the government’s decision to let him withdraw from Beijing and to maintain Japanese interests in Manchuria by cooperating with him. Formally, Colonel Kōmoto’s act was a conspiracy undertaken on his own initiative; he had received no orders from the Kwantung Army or from the General Staff of the army in Tokyo. However, in fact his act was close to an organizational conspiracy by the Kwantung Army. He 103

  The Semi-Democratic Regime, 1926–1929

tried to do what the Kwantung Army had asked the government to approve, a proposal endorsed by the General Staff of the army in Tokyo—that is, to strip Chang Tso-lin of power and place Manchuria under Japanese control. When it became certain that the Guomindang would capture Beijing in May 1928, the Kwantung Army persistently requested, through the General Staff of the army, that the government deliver an order to dispatch its troops to disarm Chang’s troops on their way back to Manchuria.2 Furthermore, it was Muraoka Chōtarō, the commander in chief of the Kwantung Army, who had originally planned, when the government refused these requests, to assassinate Chang, using Japanese troops stationed in Beijing and Tianjin. Learning of the plan, Kōmoto decided to assassinate Chang himself.3 The military challenged the party government not only by defying its policy but also by thwarting Prime Minister Tanaka’s efforts to punish the conspirators and to publicly announce the truth in order to discipline the military as whole. After discovering the truth behind Chang’s assassination in October 1928, Tanaka, influenced by a recommendation from Saionji, was at first committed to publicly announcing the facts and court-martialing the officers involved.4 In December, Tanaka reported to the emperor that he had discovered that it was Colonel Kōmoto who had assassinated Chang and that he would punish him alone, in order to maintain discipline within the army.5 The army, however, opposed punishing the conspirators and publicly announcing the facts behind the incident. The highest officers in the army were already committed in November to hiding the truth, although they left some room for the possibility of a court-martial.6 However, faced with pressure from middle-ranking officers also opposed to a court-martial of Kōmoto and others, the army decided that it would oppose not only announcing the truth publicly but also punishing the officers involved via a court-martial.7 Prime Minister Tanaka acquiesced and agreed to settle the issue by punishing Commander Muraoka, Colonel Kōmoto, and others only lightly, under the pretext of that they neglected their duty to guard the South Manchurian Railway.8 The party government was thus unable to penalize the military for defying its policy, marking a significant shift in the balance of power between the two. Chang’s assassination had already implied that the balance of power between party government and the military had begun to tilt toward the military, in that the party government could not compel the military to adhere to its policy. Yet, at the moment of the incident itself, the new bal-

The Semi-Democratic Regime, 1926–1929  105

ance of power, in which the military was in a stronger position than the party government, was still not fixed. The seesaw between the military and the party government was still moving. Whether the military could gain the upper hand in the future depended on whether the party government would be able to punish Kōmoto and discipline the army. Its ability to rein in the army would establish a precedent; any future challenge of the party government by the military would be futile and would be punished. Since the party government ultimately failed to discipline the army, however, it became clear that the military could defy the party government at will, and get away with whatever it chose to in the future.

Analysis As the balance of power shifted during this period, two factors were at work: the legitimacy of the regime and the semi-loyalty of party politicians. Above all, political institutions under the Meiji Constitution hindered the party government from constraining the military. Thus, as is evident in Chang Tso-lin’s assassination, it was impossible for the prime minister to discipline the army if he could not get the support of the military ministers. Yet the party government did sometimes succeed in constraining the military, as we have seen in the preceding chapter in the case of the Hara cabinet. Political institutions alone do not explain these differences. Why, then, did the balance of power between the party government and the military tilt toward the military during this period? The first reason was the erosion of legitimacy, which began after the establishment of the first Wakatsuki cabinet, and which lessened the power of the party government to restrain the military. The second reason is related to the semi-loyalty of political actors, who prevented the party government from penalizing the conspirators in the Chang assassination and publicly announcing the truth. legitimacy

From 1926 to 1929, under the first Wakatsuki and Tanaka cabinets, the semi-democratic regime sustained its legitimacy at the elite level. When the first Wakatsuki cabinet resigned in April 1927, Makino Nobuaki, the Lord Keeper of the Privy Seal, felt that Tanaka Giichi, president of the Seiyūkai, should be appointed the next prime minister in view of “the common practice of politics based on the constitution”—a phrase that at the

  The Semi-Democratic Regime, 1926–1929

time meant rule by party government.9 In addition, even after the downfall of the Tanaka cabinet, the mass media welcomed the establishment of the Hamaguchi cabinet because it was a party government. The Tōkyō Asahi Shimbun cited the smooth changeover from the Tanaka to the Hamaguchi cabinet as evidence that the practice of party government was becoming the political norm.10 During this period, however, there were also clear signs that regime legitimacy had begun to erode, both among opinion leaders and political leaders. While opinion leaders repeatedly expressed their distrust of party government and political parties, political leaders had already begun to discuss nonparty government as a credible alternative. For its part, the military perceived the decline in the regime’s legitimacy and consequently did what it could to weaken party government’s power to restrain its actions. Legitimacy among the Mass Media and Intellectuals: Regime Performance in the Political Sphere  Under the first Wakatsuki cabinet leaders of the political parties were suspected of involvement in a series of scandals. These included the Matsushima Club Scandal; doubts that Tanaka Giichi, president of the Seiyūkai, usurped secret funds of the army; and the Boku Retsu Scandal.11 The Matsushima Scandal, which unfolded in January 1926 and continued to stir up politics until 1927, led to the prosecution of eminent leaders of all three major political parties, the Kenseikai, the Seiyūkai, and the Seiyū Hontō, for bribery.12 Prime Minister Wakatsuki was himself suspected of involvement in the scandal. Specifically, he was accused of giving false testimony by Iwasaki Isao, a leader of the Seiyūkai, and Minoura Katsundo, a leader of the Kenseikai, both of whom had already been prosecuted. Furthermore, in March 1926, in the fifty-first Diet, Nakano Seigo, a representative of the Kenseikai, revealed that there were misgivings that Tanaka Giichi had usurped three million yen from secret army funds when he joined the Seiyūkai and asked for an official investigation by the Lower House.13 In July 1926, the Seiyūkai and the Seiyū Hontō seized upon the Boku Retsu Scandal to attack the cabinet.14 Pak Yŏl (Boku Retsu) was sentenced to death in March 1926 for allegedly conspiring to assassinate the prince regent in 1923, although his sentence was later reduced to life imprisonment. After the trial, rightists circulated a photograph of Pak with his wife sitting on his lap, which was said to have been taken during the examination by the prosecutor. The Seiyūkai and the Seiyū Hontō denounced the cabinet for

The Semi-Democratic Regime, 1926–1929  107

treating leniently a criminal who tried to assassinate the prince regent and condemned the cabinet for reducing the punishment. Under the first Wakatsuki cabinet, the Seiyūkai, often along with the Seiyū Hontō, tried to force the cabinet to resign by publicizing scandals in which the party government and the Kenseikai were involved. The Kenseikai counterattacked by pointing to the involvement of leaders of opposition parties in these scandals, which intensified the wrangling among the political parties. The parties concentrated most of their energy on condemning one another, neglecting policy issues. Although the Seiyūkai and the Seiyū Hontō wanted to come to power, they did not want to force the government to dissolve the Lower House, as they could not predict the result of a general election under the new electoral system, in which the electorate had quadrupled.15 Thus, they tried to trigger a government collapse through political scandals, over which they judged it would be difficult for the government to dissolve the Lower House and hold a general election. Regardless of the political parties’ motivations, the scandals and disputes among them damaged their credibility with opinion leaders. For example, during the fifty-first Diet in 1926, Ishibashi Tanzan attacked the political parities for devoting themselves solely to exposing scandals, declaring that “this session of the Diet was a great accomplishment for citizens . . . in that [it demonstrated] that the existing parties—the Kenseikai, the Seiyūkai, and the Seiyū Hontō—perhaps all were ugly bands, which were no longer qualified to be responsible for national politics.”16 The Ōsaka Asahi Shimbun also deplored the behavior of the political parties in the Lower House: After two-thirds of this session, the fifty-first Diet has come to expose its ugliness all the more. . . . It seems that it is a fundamental duty of the members of the Diet to expose scandals of the members of the opposing parties rather than concentrating on important issues related to fiscal and foreign policies. This state of affairs, which is deplorable and lamentable, reveals that the current members of the political parties are dirty, that the chambers consist of those who do not understand constitutional politics, and that these dirty members of the existing political parties compose today’s legislature. . . . The bribery involving Iwasaki, Minoura, Miki, and Tanaka in the Matsushima Club Scandal, the Research Commission on Ogawa, Koizumi, Hatoyama, Akita, and others regarding the Tanaka Giichi scandal, what a state of exposure of scandals! . . . The state of corruption of the existing political parties must be more severe than can be seen from the outside.17

  The Semi-Democratic Regime, 1926–1929

The Tōkyō Nichinichi Shimbun criticized the political parties in a similar vein: Recently, the Lower House has been continually displaying behaviors that go beyond our imagination. . . . [The political parties] have almost completely neglected to engage in debate and examination of national policies and instead make frantic efforts to expose scandals and look into the secrets of their enemies, inviting the suspicions of the leaders of the political parties. The names of the people which came to the surface as being involved in a series of political scandals such as the . . . Matsushima Club Scandal . . . and the usurpation of the army’s secret funds are bosses of the factions [in the political parties]. We must say that [the political parties] will end up not only toppling their enemies but also defeating themselves . . . [by using scandals] as a weapon to [attack their enemies].18

While the political parties were busy exposing political scandals, they often were engaged in bargaining behind the scenes to acquire and secure power, which also undermined their credibility. During this period, the Kenseikai, the Seiyūkai, and the Seiyū Hontō were the three major political parties in the Lower House, which meant that cooperation between two of the three would give the two a majority in the Lower House.19 As a result, the three political parties were often engaged in coalition building. Although cooperation between the Kenseikai and the Seiyūkai never took place, the Seiyū Hontō swung between the Kenseikai and the Seiyūkai. It first cooperated with the Kenseikai and then joined the Seiyūkai to topple the Wakatsuki cabinet, only to switch allegiance again to the Kenseikai. Such changes in the relationships among the parties made opinion leaders wonder whether party governments and political parties truly represented the interests of the people. For example, the Tōkyō Asahi Shimbun criticized the ­Kenseikai and the Seiyū Hontō when they agreed to cooperate in March 1927, speculating that the two might have agreed to soon transfer the government to the Seiyū Hontō.20 It questioned whether the Seiyū Hontō, a mere third party in the Lower House, had truly been entrusted by the citizens to lead the government. Moreover, it condemned the two political parties as “hungry demons” in search of political power, pointing out that they took opposing positions on universal suffrage in the general election of 1924.21 Besides the polarization in the Lower House, the fear of a general election was another reason to engage in secret deals. As the next election would be the first with universal male suffrage, there was much uncertainty about the result and consequently the political parties wanted to delay the vot-

The Semi-Democratic Regime, 1926–1929  109

ing as long as possible. The most severe criticism of behind-the-scenes bargaining came at a summit meeting of the presidents of the three parties in January 1927. Earlier that month, in the face of growing political discord over a series of scandals, the Seiyūkai and the Seiyū Hontō had submitted a motion of no confidence to the government. In response, Prime Minister Wakatsuki invited the presidents of the Seiyūkai and the Seiyū Hontō to a summit, at which they agreed, out of respect for the beginning of a new emperor’s reign, to refrain from fomenting a political clash.22 As a result, the Seiyūkai and the Seiyū Hontō withdrew their motion of no confidence. The agreement took place because the three parties were reluctant to have general elections and because Prime Minister Wakatsuki implied that he would resign at some point after the end of the ongoing session of the Diet.23 But this compromise flew in the face of the public’s expectation of a general election, which had been repeatedly expressed through the mass media. As a result, criticism of the political parties flooded the media after the summit meeting and the withdrawal of the motion of no confidence. The Tōkyō Asahi Shimbun, for example, labeled the dealings among the parties a case of “rogues trying to outwit each other” and condemned the government for not dissolving the Diet.24 Asserting that the Diet did not represent the opinions of the citizens, it called upon the public to push for its dissolution.25 It also expressed frank distrust of the three major parties and implored citizens “not to vote for the Kenseikai, the Seiyūkai, and the Seiyū Hontō in the next general election.”26 The Ōsaka Asahi Shimbun argued that the dealings revealed that so-called party politics at the time was nothing more than oligopoly. It condemned the three presidents for postponing the general election, which, the paper proclaimed, the citizens so earnestly desired.27 The Ōsaka Mainichi Shimbun also eloquently expressed its distrust of the political parties: From every viewpoint, it is truly ugly politics . . . that the presidents of the three political parties met secretly, ignoring the chamber and the members of the Diet, which resulted in the withdrawal of the vote of no confidence submitted by the Seiyūkai and the Seiyū Hontō. . . . Confronted with the fact that our political parties have become so deeply corrupted that there seems to be no remedy, the citizens have already lost the courage to become infuriated and are just in a state of disgust. . . . It is astonishing to see how unfavorably the dealings were perceived by the public and how political parties lost even more of the citizens’ trust.28

  The Semi-Democratic Regime, 1926–1929

Although disputes over political scandals faded under the Tanaka cabinet, distrust of party government and political parties continued to grow among opinion leaders. The way the Tanaka cabinet was founded and sustained gave rise to a fundamental question: did the will of the public have any role to play in determining the fate of party government? The formation of the Tanaka cabinet implied that a party government could be sustained even when there was no manifest support for it among the citizens. The Tanaka cabinet was born of an incident in which the will of the citizens played no role. In April 1927, the Privy Council’s rejection of an emergency decree designed to solve the financial crisis triggered the resignation of the Wakatsuki cabinet and ushered in the Tanaka cabinet. Soon after its foundation, Ishibashi Tanzan argued that “the Tanaka cabinet was born without asking the confidence of the citizens” and demanded that a general election be held immediately to see whether the cabinet had the support of the public.29 The Tōkyō Asahi Shimbun also demanded the dissolution of the Lower House to verify the citizens’ confidence, arguing that it was created “because of a mere accident” and did not spring from the citizens’ support for the Seiyūkai.30 Indeed, when the Tanaka cabinet was formed, it was a minority government. Its position vis-à-vis opposing parties became even weaker when the Kenseikai and the Seiyū Hontō merged to form the Minseitō in June 1927.31 The cabinet waited until February 1928 to call a general election, betraying what opinion leaders perceived to be the public’s expectations. The government’s intervention in political campaigns as well as the result of the general election posed serious questions about the Tanaka cabinet’s support among the electorate. Expecting to win an absolute majority in the Lower House, the Tanaka cabinet held the first general election under universal male suffrage in February 1928. The Seiyūkai failed to win a majority of the Lower House, however, even though the government intervened heavily in the election campaign in favor of Seiyūkai candidates.32 In fact, the Minseitō defeated the Seiyūkai in the popular vote, although it won only 216 seats against the Seiyūkai’s 217 seats. Nonetheless, the cabinet continued to rule. It expanded its base of support in the Lower House by winning over members of the neutral parties and the Minseitō, which in turn led opinion leaders to question the meaning of the general election. Opinion leaders did not hide their distrust of the Tanaka cabinet, calling for it to resign because it had lost the people’s confidence. The Ōsaka Asahi

The Semi-Democratic Regime, 1926–1929  111

Shimbun pointed out that Prime Minister Tanaka had published a statement announcing that the government would appeal to the nation with respect to its policies immediately after the dissolution of the Lower House.33 To prove that the public no longer had confidence in the Tanaka Seiyūkai cabinet, the newspaper also pointed out that the number of votes won by the opposition parties had exceeded the number garnered by the ruling party.34 Demanding the cabinet’s immediate resignation, it also warned the government not to bribe neutral members of the Diet in order to reach a majority in the Diet.35 The Ōsaka Mainichi Shimbun likewise called on the government to resign because “the ruling party clearly lost in the face of the majority won by all opposition parties including the Minseitō” even though “the government and the ruling party made full use of their advantageous position . . . through intervention and pressure.”36 The Tanaka cabinet defied the demands for its resignation. Opinion leaders’ concerns increased as the government tried hard to win over members of the neutral and opposition parties to expand its support base in the Lower House; they warned that this “ignored the outcome of the general election.”37 In July 1928, Tokonami Takejirō, the former president of the Seiyū Hontō, split from the Minseitō, and in August founded the Shintōkurabu with thirty of his fellow Diet members. Further, in September, several other members of the Minseitō formed the Kensei Isshinkai along with other members of smaller parties.38 To engineer these splits the Seiyūkai maneuvered both indirectly and directly; these new political parties acted as supporters of the Seiyūkai.39 As a result, distrust of the political parties increased. In the face of the Tanaka cabinet’s maneuvers, the Tōkyō Asahi Shimbun questioned the support the government had received in the Diet, arguing that “the majority which the government has is a majority that was created ‘unnaturally’ in the first place,” which was why “the will of the Diet and the will of the citizens do not match.”40 Yoshino Sakuzō condemned the entire process by which the Tanaka cabinet continued to remain in power, defying the results of the general election and expanding its support base in the Lower House through backroom deals. It goes without saying that the dominance of the Seiyūkai today is a creation that has nothing in essence to do with public opinion. . . . People often argue that a political group can acquire unchangeable authority only when it has a foundation. “Fair elections” as well as “the majority acquired as a result” are necessary conditions for this so-called foundation among the citizens. . . . However,

  The Semi-Democratic Regime, 1926–1929 what is today’s majority held by the Seiyūkai? It passed two steps to build up this majority. In each step, has it ever thought about public opinion? This does not need extensive explanation. First, it committed various injustices to win its majority in the general election. Thus, although it has acquired a plurality exceeding the opposition parties by a little in the general election, it does not reflect the citizens’ views. Second, after the election, it succeeded in adding to its members by buying off some Diet members.41

The means by which the Tanaka cabinet stayed in power led the mass media to criticize not only the cabinet itself, but also the quality of politics as a whole. The Ōsaka Mainichi Shimbun candidly deplored the state of party politics: Japanese party politics has not yet reached a stage in which the rise and fall of political parties in the Diet depends solely upon the will of the electorate. Manufacturing of party members not based on elections takes place frequently. . . . [Party politicians] do not feel guilty about changing their affiliation to opposition parties after elections. Politics is in such a state that Tokonami Takejirō, who had organized the Minseitō, merging the Kenseikai and Seiyū Hontō, and then had been elected [as a member of the Minseitō], can stay calm even when he betrays the expectations of the electorate by suddenly leaving his party.42

In addition to retaining power in a way that appeared illegitimate to opinion leaders, the Tanaka cabinet amended the Peace Preservation Law through machinations that increased the general distrust of party government. In amending the law in the fifty-fifth Diet, the cabinet attempted to increase the severity of the penalty for “the establishment of associations aiming to change the polity of Japan [that is, the emperor system]” from ten years in prison to death, and tried to curb activities that would serve the objectives of such associations.43 The bill to amend the law ended up being tabled and discarded in the fifty-fifth Diet, however, after strong opposition from opposing parties. Although the Diet did not pass the bill to amend the law, the Tanaka cabinet succeeded in amending it through an emergency degree in June 1928, after the Diet had adjourned. The government alleged that communist activities were spreading and thus amending the law was urgent. It was clear, however, that the cabinet ignored the Diet and relied on the emergency decree because it had not built up a solid base of support in the Lower House, which made the prospects dim for amending the law via a bill in the Diet. The Tanaka cabinet’s disregard of the Diet provoked fury

The Semi-Democratic Regime, 1926–1929  113

in the mass media. The Ōsaka Mainichi Shimbun denounced the government, calling it “the gang against politics based on the constitution,” and complained that “if it is possible to make a bill that has been rejected by the Diet into a law by a decree, the Diet will lose its raison d’être.”44 The Ōsaka Asahi Shimbun similarly condemned the government several times for violating the constitution and ignoring the Diet.45 Legitimacy among the Mass Media and Intellectuals: Regime Performance in the Socioeconomic Sphere  During this period, the semi-democratic regime encountered economic problems, such as the financial crisis of 1927 and intensified labor disputes, which it could not effectively solve. Its ineffectiveness was often linked to regime performance in the political sphere. Opinion leaders frequently ascribed the regime’s shortcomings to the fact that the party government and the political parties remained aloof from the public, concerned solely with how to acquire political power. The largest economic problem the regime faced was turmoil in the financial system in 1927.46 By the beginning of that year, the Japanese economy was under threat from nonperforming loans held by many banks. These originated from a number of factors, such as loose lending in the economic boom during the First World War, the depression of 1920, and economic confusion following the Kanto Great Earthquake in 1923. The first Wakatsuki cabinet was determined to come to grips with the problem of nonperforming loans and introduced three bills to address it in the fifty-second Diet.47 Although these bills were passed in the Diet, debates over their content drew attention to the financial condition of banks awash with bad loans. Amid these increasing concerns, Finance Minister Kataoka announced that the Tokyo Watanabe Bank had suspended operations, triggering a crisis of the Japanese financial system in March. People worried about the management of banks scrambled to withdraw their deposits, which drove many banks into insolvency. Although the financial system recovered stability by the end of March, another crisis hit the Japanese financial system in April, when the Bank of Taiwan came close to bankruptcy because of its bad loans to the Suzuki Shōten, a large trading company. The first Wakatsuki cabinet tried to save the Bank of Taiwan by issuing an emergency decree that would give a governmental guarantee to emergency loans made by the Bank of Japan to the Bank of Taiwan. However, the Privy Council refused to approve the degree. In response, the Wakatsuki

  The Semi-Democratic Regime, 1926–1929

cabinet decided to resign. The series of events brought about yet another crisis in the Japanese financial system; the Bank of Taiwan was driven to suspend its operations temporarily, and wide scale withdrawal of deposits again drove many banks into bankruptcy. The Tanaka cabinet, which succeeded the Wakatsuki cabinet, was faced with the task of dealing with the crisis in the financial system. Thanks to initiatives by Finance Minister Takahashi Korekiyo, by mid-May the cabinet succeeded in bringing the crisis to an end and restoring stability. It received approval from the Privy Council to issue an emergency decree introducing a three-week moratorium on payments from banks and managed, in a special session of the Diet, to pass the bill providing governmental guarantees to emergency loans by the Bank of Japan. In dealing with these crises, the regime’s efficacy was not in question; it was able to find solutions to the challenges it faced, since the Tanaka cabinet basically adopted the same—ultimately successful—policy as the Wakatsuki cabinet. The regime’s effectiveness, however, was less clear. The series of crises revealed that its capacity to actually implement policies to deal with the financial crises was limited. Its ineffectiveness increased the government’s expenditures, which ultimately became a burden on the citizens. Two major political institutions involved in these crises, the Privy Council and the party government, were criticized by the mass media for their intransigence. The Privy Council was denounced for thwarting the policies of the party government. The Ōsaka Mainichi Shimbun censured it for rejecting a decision by the party government, which represented the citizenry.48 The Tōkyō Asahi Shimbun also condemned the Privy Council for “intervening in policy formulation” and “destroying the foundation of a responsible cabinet system.”49 Opinion leaders’ critique of party government, in the meantime, was twofold. They criticized the government for relying on emergency decrees rather than convening a special session of the Diet, which represented the public, to pass bills to deal with the crises.50 They also criticized it for refusing to be “loyal to the citizens” to the very end.51 The essence of this argument was that, of the political institutions under the Meiji Constitution, party government represented the citizens. Instead of acquiescing to the Privy Council, if the Wakatsuki cabinet were to remain loyal to the citizens it should have confronted the Privy Council, and let the emperor himself decide between adopting the Privy Council’s recommendation on the government decision or the government decision itself.

The Semi-Democratic Regime, 1926–1929  115

In dealing with the ongoing labor movement during this period, the r­ egime demonstrated the same lack of effectiveness it displayed in handling financial turmoil. The labor movement had become particularly prominent in Japan toward the end of the First World War (see Tables 5.1 and 5.2). Before the war, the greatest number of labor disputes was sixty, a number recorded in 1907. During the First World War, the number jumped: 398 were recorded in 1917.52 After remaining in the two hundreds in the first half of the 1920s, the number radically increased at the beginning of the 1930s, approaching table 5.1 Number of tenant and labor disputes, 1913–1935 Tenant disputes 1913 1914 1915 1916 1917

n.a. n.a. n.a. n.a. 85

1918 1919 1920 1921 1922 1923 1924 1925

256 326 408 1,680 1,578 1,917 1,532 2,206

1926 1927 1928 1929 1930 1931 1932

2,751 2,052 1,866 2,434 2,478 3,419 3,414

1933 1934 1935

4,000 5,828 6,824

Labor disputes

Average

85

47 50 64 108 398

133

1,238

417 497 282 246 250 270 333 293

324

2,630

495 383 397 576 906 998 893

664

5,551

610 626 590

609

Average

source: Nihon Tōkei Kenkyūjo, ed., Nihon Keizai Tōkeishū: Meiji, Taishō, Shōwa (Tokyo: Nihon Hyōron Shinsha, 1958), 296–299.

  The Semi-Democratic Regime, 1926–1929 table 5.2 Number of tenant unions and labor unions, 1918–1935 tenant unions

Number of unions 1918 1919 1920 1921 1922 1923 1924 1925

n.a. n.a. n.a. 681 1,114 1,530 2,337 3,496

1926 1927 1928 1929 1930 1931 1932

3,926 4,582 4,353 4,156 4,208 4,414 4,650

1933 1934 1935

4,810 4,390 4,011

Average

Number of people in unions (1,000)

1,832

n.a. n.a. n.a. n.a. 132 164 232 307

4,327

347 365 330 316 301 306 297

4,404

303 276 242

labor unions

Average

Number of unions

209

107 187 273 300 387 432 469 457

323

488 505 501 630 712 818 932

274

942 965 993

Average

Number of people in unions (1,000)

Average

327

n.a. n.a. n.a. 103 137 126 228 254

170

655

285 309 309 331 354 369 378

334

967

385 388 409

394

source: Nihon Tōkei Kenkyūjo, ed., Nihon Keizai Tōkeishū: Meiji, Taishō, Shōwa (Tokyo: Nihon Hyōron Shinsha, 1958), 298–299.

nearly one thousand. Likewise, the number of labor unions and the number of members of labor unions constantly increased from the 1920s to the 1930s. While the labor movement sought various material benefits, such as wage increases, an eight-hour workday, and recognition of collective bargaining, it also demanded that workers be recognized as individuals with rights and that they not be discriminated against.53 Despite the surge in the labor movement, the first Wakatsuki cabinet and the Tanaka cabinet did not legalize labor unions. Had they done so, the movement might have been mitigated. In 1926 and 1927, the first Wakatsuki cabinet submitted two bills to legalize labor unions but could not get either bill passed in the Lower House. The Tanaka cabinet was even more defiant

The Semi-Democratic Regime, 1926–1929  117

of the movement, as it did not introduce any bill to legalize labor unions, and actually repressed the labor movement by banning the Japan Labor Union Council and the Labor and Farmers Party in April 1928. Those who demanded the legalization of labor unions thought that the bills designed by the first Wakatsuki cabinet contained too many restrictions. For example, the Tōkyō Asahi Shimbun criticized the second bill for failing to impose any penalties on employers for firing workers for being members of a labor union and for allowing the government to issue orders to dissolve labor unions that might “disturb stability and order or violate the public interest.”54 Party government was incompetent in dealing with the labor movement, some argued, becauses political parties were preoccupied with struggling among themselves for political power. For example, Suehiro Gantarō, professor in the Faculty of Law at the University of Tokyo, denounced the first bill submitted in 1926 by the Wakatsuki cabinet, calling it a “law to restrict labor unions,” and railed against party government: If I realistically analyze the process by which this bill to restrict labor unions was drafted . . . it is possible to say that [this law was drafted because] the ministers of the cabinet followed blindly and gulped down the inappropriate resolution made by the Research Council [which was entrusted with drafting the bill]. The political parties and the cabinet supported by the political parties devote themselves solely to political competition day and night. As a result, everything that is irrelevant to the interests of the political parties tends to be left to the discretion of the bureaucracy, and the Diet also tends to swallow [the decisions of the bureaucracy] without any criticism.55

Legitimacy among Political Leaders  The available evidence suggests that the regime’s legitimacy was eroding even among political leaders, mainly over the struggle for power among political parties. The diary of Matsumoto Gōkichi, a close aide to Saionji, reveals that because of the damage that scandals had inflicted on the political parties, political leaders during the first Wakatsuki cabinet seriously considered a nonparty government. For example, in September 1926, when political disputes over scandals were intensifying, Makino Nobuaki, the Lord Keeper of the Privy Seal, demonstrated great reluctance to see Tanaka Giichi succeed Prime Minister Wakatsuki because of the scandal concerning Tanaka’s alleged usurpation of army funds. Makino felt it was acceptable for a nonparty government to succeed the Wakatsuki cabinet.56 During the same period, even Saionji was aware that a nonparty government might be seen as a possible successor to the

  The Semi-Democratic Regime, 1926–1929

Wakatsuki cabinet because of the political scandals in which the Seiyūkai was involved.57 Indeed, at one point Saionji himself entertained the possibility of a nonparty government, with a particular person in mind to lead the cabinet. He had in mind Den Kenjirō, a member of the Privy Council, to be a possible candidate to form a nonparty government to succeed the Wakatsuki cabinet.58 The struggle for political power continued to wear away the regime’s legitimacy even under the Tanaka cabinet. Itō Takashi reveals that a group led by Hiranuma Kiichirō, the vice president of the Privy Council, was conspiring to set up a nonparty government to succeed the Tanaka cabinet.59 Hiranuma’s group considered it necessary to suspend the practice of party government because political parties continued to engage in “mudslinging,” which it feared would give rise to “an extremely disgusting” situation. 60 They suspected that the Minseitō would arrest politicians of the Seiyūkai in order to inflict damage on the Seiyūkai once it came to form the government after the Tanaka cabinet.61 Legitimacy among the Public  Given that aggregate data are limited, it is hard to make a conclusive argument about the level of legitimacy among the public during this period. However, based on the evidence, it seems that the regime managed to sustain its legitimacy among the public even though the level of that legitimacy began to decline. On one hand, data on the general election of 1928 suggest that the regime at least maintained its legitimacy with the public. In that election, the electorate demonstrated a high turnout rate—90 percent—and the existing parties gained more than 90 percent of the total votes. This seems to indicate that ordinary people supported the national election and political parties, which were core institutions of the regime. However, it is difficult to draw a definite conclusion about the level of regime legitimacy from data on a single election. On the other hand, other evidence, such as data on the prefectural elections and on labor and tenant farmer disputes, is consistent with the decline in the level of regime legitimacy. Hayashi Yūichi has examined the prefectural election in 1927 and pointed out that the turnout rate had precipitously declined compared to the preceding election.62 In the urban districts, turnout decreased from 74.6 percent in 1923 to 60.7 percent in 1928, while in the rural districts it fell from 84.9 percent to 76.5 percent. In addition, he draws attention to the fact that turnout rates in large cities were ex-

The Semi-Democratic Regime, 1926–1929  119

tremely low. For example, the rate in the urban districts of Tokyo, Osaka, and Kyoto prefectures fell from 72.6 percent to 52.5 percent, 80.9 percent to 46.8 percent, and 75.1 percent to 49.2 percent, respectively. This change in voter turnout can be interpreted as a sign of increasing political apathy among a large proportion of the population.63 However, it is also likely that the decrease in the turnout rate, even in prefectural elections, was partly brought about by the introduction of universal male suffrage. Thus, the fall in voter turnout does not offer distinct support for the notion that regime legitimacy was completely lost. By the same token, the number of labor and tenant farmer disputes during the period, which was higher than in the years 1918 to 1925, also suggests an erosion of regime legitimacy (see Table 5.1). The growing number of disputes implies that discontent among workers and tenant farmers increased or at least remained at the same level as in the preceding period. The manner in which the cabinets dealt with these movements very likely contributed to the increase in or the persistence of frustration among workers and tenant farmers. In addition to denying legal recognition of labor unions, the two cabinets failed to give legal protection to tenant farmers, although the individuals and organizations leading the labor and tenant farmer movements had repeatedly called for the passage of the Labor Union Law and the Tenancy Law.64 Yet, it is difficult to draw a definitive conclusion that the regime’s legitimacy evaporated from the mere fact that the number of labor and tenant farmer disputes increased. The Decline in the Legitimacy and Weakening of Party Government  Various evidence thus suggests that, while the regime managed to sustain its legitimacy among the elite, the level of this legitimacy began to decline under the first Wakatsuki cabinet and the Tanaka cabinet. Several pieces of evidence of the level of regime legitimacy among the population at large are consistent with the assessment of the degree of legitimacy among the elite. What was the degree of decline from the preceding period? I argued in the Chapter 4 that it seems plausible that, at the beginning of the Katō Takaaki cabinet in 1924, the regime enjoyed a high level of legitimacy in that it was supported for its own positive value. The evidence, however, indicates that between 1926 and 1929 legitimacy had already declined to such a point that the regime was not judged positively on its own merits but only because of the unpopularity of its alternative: a regime centered

  The Semi-Democratic Regime, 1926–1929

on a nonparty government. In other words, it seems that the regime was considered legitimate only by default. Despite their unrelenting criticism of party government and the political parties, the mass media supported the formation of a party government after the first Wakatsuki cabinet and the Tanaka cabinet had resigned. Yet whenever the media declared support for party government, most of the newspaper editorials simultaneously rejected the possibility of a nonparty government and referred to a supposed conspiracy for the emergence of a nonparty government.65 The implication is that rejection of nonparty government was a major factor in the mass media’s support for party government. There is similar evidence concerning political leaders, who continually expressed their distrust of party government and political parties and had already begun examining a nonparty government as a plausible alternative. Ultimately, however, they seem to have refrained from endorsing a nonparty government; despite all its shortcomings, they still considered party government better than nonparty government. In August 1926, for example, after mentioning possible candidates to head a nonparty government, Saionji said, “Although this is not to say that I am afraid of the Kensei Yogokai [the Association to Protect Politics Based on the Constitution], but it seems the public thinks the political parties are better after all.”66 Furthermore, in a meeting in September 1926 with General Ugaki Kazushige (the army minister under the first Wakatsuki cabinet), Kiyoura Keigo, the former prime minister, admitted that nonparty government could be an acceptable alternative, and added: [But,] judging from the candidates [to lead a nonparty government] about whom there are public rumors, I cannot say that it is better [to have them lead the government] than to let the political parties form the government. In order to accept a nonparty government, we must be able to anticipate that it will demonstrate better and stronger governance than today. [If we allow a nonparty government] without such prospects, it will just invite confusion in politics and will be fruitless.67

While the semi-democratic regime could sustain its legitimacy by default, the decline in its legitimacy had a significant impact on the political process: it undermined party government’s control over the military. If one accepts Linz’s and Agüero’s arguments about the effect of regime legitimacy on the civil-military relationship (see Chapter 4), then it is clear that when much of the public considers a regime to be legitimate, the mili-

The Semi-Democratic Regime, 1926–1929  121

tary will obey the government and not challenge it. The flip side of this argument is that when the military perceives that a regime’s legitimacy has begun to erode, it feels less restrained in challenging the government. This was exactly what happened during this period in Japan. The party government’s power to restrain the military diminished because its own legitimacy declined. General Ugaki, for example, referred repeatedly in his diary to the decline of legitimacy under way at the time. In July 1926, he wrote, “Nowadays the intellectuals in the nation have lost confidence in the vulgar struggles and policies of the existing parties.”68 In January 1928, he wrote, “The public is in an atmosphere in which they are dissatisfied with the existing political parties.”69 the semi-loyalty of party politicians

The erosion of regime legitimacy might explain why the balance of power between the party government and the military tilted to such an extent that an officer of the Kwantung Army could defy the government’s policy and assassinate Chang Tso-lin. However, the decline of regime legitimacy alone cannot explain why, after the incident, the party government was unable to recover its position against the military by punishing the conspirators and revealing the facts behind the plot. The Tanaka cabinet, as I have mentioned, could not punish Colonel Kōmoto and others to discipline the military as a whole because of opposition from the army. But there were also other reasons. Saionji demanded that the government publicly announce the truth and punish the conspirators. Currently available evidence reveals that the emperor himself also demanded that the conspirators be severely punished.70 In this case, the army could not support its opposition to revealing the truth and punishing the conspirators on legal grounds. The independence of the supreme command was also not at issue, because Colonel Kōmoto did not receive orders from the General Staff of the army. In other words, Kōmoto violated the supreme command of the emperor by acting without an order from the emperor. Even if the army had given such an order to assassinate Chang, the military could not have claimed independence from the control of the party government; such an act was considered to be related to Japan’s foreign policy, over which the government had jurisdiction. In view of the support it had from the highest political authorities at the time and the lack of any legal basis to support the army’s opposition, it is

  The Semi-Democratic Regime, 1926–1929

likely the Tanaka cabinet could have at least court-martialed the conspirators for assassinating Chang, had cabinet ministers and party politicians united to confront the military. Indeed, in December 1928, the highest authority of the army implied that it might agree to having the conspirators court-martialed.71 However, the semi-loyalty of cabinet ministers and party politicians, all of whom had a stake in preserving the party government, prevented the Tanaka cabinet from containing the military after Chang’s assassination. To discipline the army, it was necessary for the Tanaka cabinet to publicly disclose the facts behind the assassination. Such a step, however, was likely to invite criticism of the cabinet and force it to resign. The next cabinet might have been formed by the Minseitō. Many ministers and politicians of the Seiyūkai were afraid of losing power and thus were reluctant to discipline the army after the incident. In Chapter 1, I argued that political actors are considered to be “disloyal” to a regime when they refuse to follow that regime’s “rules of the game” and that political actors are considered to be “semi-loyal” to a regime when they “encourage, tolerate, cover up, treat leniently, excuse, or justify” the disloyalty of other political actors.72 In assassinating Chang, Colonel Kōmoto was disloyal to the regime because he refused to follow its rules of the game by defying the party government’s policy. Cabinet ministers and party politicians acted semi-loyally to the regime by excusing Kōmoto’s disloyalty and cooperating in covering it up. Most cabinet ministers opposed revealing the truth and punishing the conspirators because to let the world know the truth would disgrace Japan and weaken its international standing.73 They also worried that revealing the truth might lead the Seiyūkai cabinet to resign. Even party politicians in the Diet acted semi-loyally by participating in the cover-up. Although the Minseitō had learned the truth soon after the assassination, it failed to reveal the facts in the Diet, awkwardly requesting that the government reveal what it knew of “a certain incident in Manchuria” and criticizing the Tanaka cabinet for failing to properly guard the South Manchurian Railway. Besides, the resolution submitted by the Minseitō asking for the revelation of “a certain incident in Manchuria” was turned down in the Lower House.74 The semi-loyalty of cabinet ministers and party politicians made it impossible to confront the military, crippling the semi-democratic regime by setting a crucial precedent for both the military and party government.

The Semi-Democratic Regime, 1926–1929  123

Effects of Political Incidents on Political Dynamics Leading to the Final Stage of Regime Breakdown Looking beyond the beginning of the erosion of the regime’s legitimacy, it is clear that other political developments between 1926 and 1929 contributed significantly to the final breakdown of the semi-democratic regime. The first Wakatsuki cabinet and the Tanaka cabinet left two important precedents that contributed to the collapse of the regime between 1929 and 1932. The first Wakatsuki cabinet resigned when it could not get the Privy Council’s approval for an emergency decree to address Japan’s financial turmoil, setting a precedent that the government might have to resign when the Privy Council failed to approve its policies. This precedent probably exacerbated the conflict between the party government and the military over the signing of the London Naval Treaty in 1930. The conflict thus deepened because the Seiyūkai, using the violation of the independence of the supreme command to attack the Hamaguchi cabinet in signing the treaty, acted semi-loyally. It is likely that the Seiyūkai drew attention to this issue in the expectation that the Privy Council would refuse to approve the ratification of the treaty, thus leading to the resignation of the Hamaguchi cabinet and the establishment of a Seiyūkai cabinet. When the Tanaka cabinet failed to publicly disclose the truth behind Chang Tso-lin’s assassination and to penalize the military for the assassination itself, it set an important precedent that the military could get away with disobeying the policies of the party government. Learning from the Manchurian Incident that the party government could not control it, the military persistently ignored policies of the party government, acting fully on its own and thoroughly undermining the party government’s authority. With these important precedents established, and in an atmosphere of weakened regime legitimacy and increasing military politicization, the Hamaguchi cabinet was established on July 2, 1929, heralding the final breakdown of the semi-democratic regime.

chapter 6

The Semi-Democratic Regime, 1929–1932 Crisis and Breakdown

Although the balance of power between the party government and the military began to tilt toward the latter between 1926 and 1929, the military had not yet attempted to topple the party government. Between 1929 and 1932, however, under the Hamaguchi, the second Wakatsuki, and the Inukai cabinets, the balance of power between the military and the party government shifted so decisively that the military succeeded in removing the party government from power in 1932, bringing an end to the semi-democratic regime. The shift in the balance of power was demonstrated by a series of crises during this period: the signing of the London Naval Treaty in 1930, the Manchurian Incident in 1931, the March and October Incidents in 1931, and the May Fifteenth Incident in 1932. This chapter will explore how each crisis demonstrated the shift in balance of power between the party government and the military, and then explain why the shift took place, with particular reference to legitimacy and the semi-loyalty of party politicians.

The Relationship between the Party Government and the Military the london naval treaty

Overcoming the opposition of the General Staff of the navy, the Hamaguchi cabinet succeeded in signing and ratifying the London Naval Treaty in 1930. This is often regarded as an example of the party government’s successful containment of the military.1 In two senses, however, the signing also represented a fundamental change in the balance of power between the party government and the military. 125

  The Semi-Democratic Regime, 1929–1932

When we reflect on the military’s response to past attempts at arms reduction, the sea change in the relationship between party government and the military becomes clear. In previous cases of arms reduction, the army and navy did not challenge or thwart the government’s attempt to reduce the size of the military. In the case of the London Naval Treaty, however, after displaying some resistance to the signing, the General Staff of the navy openly attempted to prevent the party government from ratifying the treaty. This demonstrated the military’s increase in power relative to the party government. Second, the General Staff of the navy succeeded in increasing its legal power. As described in Chapter 3, under the Meiji Constitution it was unclear which institution—the government or the military—had jurisdiction over the organizational affairs of the military. To conciliate the General Staff of the navy, however, the navy minister signed a memorandum stipulating that the navy minister and the chief of the General Staff had to agree on any decision related to naval armaments.2 The Hamaguchi cabinet claimed this was merely an internal navy agreement and would not affect the government’s jurisdiction. The agreement meant, however, that the navy minister could no longer agree to government decisions on naval armaments if he did not have the consent of the General Staff of the navy. Since under the Meiji Constitution the prime minister did not have the power to overrule the military ministers, from then on if the navy minister opposed government plans on naval armaments—reflecting the opposition of the General Staff of the navy—the government was unable to implement such plans. The de facto joint jurisdiction over armaments between the military minister and the General Staff meant that party government had lost its exclusive control over armaments, marking a clear setback in its relations with the military. the manchurian incident

A series of crises in 1931, of which the Manchurian Incident was the most serious, demonstrated that party government had completely lost its power to restrain the military and, indeed, was on the verge of collapse. The military’s defiance of the party government in these crises was much greater than in the challenges related to Chang Tso-lin’s assassination in 1928 and the London Naval Treaty in 1930. In the earlier cases, military challenges either took the form of a personal conspiracy or occurred in response to initiatives by the party government. In 1931, by contrast, the military not only formulated policy on its own, ignoring the party government’s author-

The Semi-Democratic Regime, 1929–1932  127

ity, it also made plans to topple the party government through coups d’état, although it never carried them out. In sharp contrast to the military’s earlier challenges of the government, throughout the Manchurian Incident military units initiated and implemented policy, ignoring and eventually overcoming resistance from the party government. In addition, during the Manchurian Incident, the military as a whole endorsed and supported its units’ defiance of the government. In September 1931, the Kwantung Army began operations to occupy Manchuria. The so-called Manchurian Incident that followed consisted of a series of acts of defiance of the party government. The Kwantung Army’s operations began as a conspiracy without the government’s approval. The army pretended that it had opened fire in response to the bombing of the South Manchurian Railway by Chinese troops. Furthermore, it continued to expand the area it occupied, repeatedly ignoring the policy of the second Wakatsuki cabinet. For example, after the Kwantung Army began its military action on September 18, the Wakatsuki cabinet immediately decided in a cabinet meeting on September 19 not to expand military operations. This decision was conveyed to the Kwantung Army on the same day from Tokyo.3 The Kwantung Army ignored the order, however, and on September 21 began to advance on Jilin, extending the front line, and by the evening had occupied Jilin.4 On September 24, the Wakatsuki cabinet officially announced that Japan had no intention of occupying any territory, in Manchuria or elsewhere, nor did it intend to expand military operations.5 Again, the Kwantung Army ignored this announcement and attacked Jinzhou on October 8.6 The Chōsen Army, stationed in Korea, also flouted the government’s authority. At the time, no military action was to be undertaken abroad without government authorization. Nonetheless, when the army minister was unable to get the government to consent to sending the Chōsen Army into Manchuria, across the border with China, the Chōsen army moved anyway.7 Unable to constrain the military, the party government could do no more than give a series of post facto approvals to the military’s faits accomplis. For example, the second Wakatsuki cabinet sanctioned the intrusion of the Chōsen Army and even jettisoned its initial policy against an expansion of military operations, deciding that Japanese troops should remain in the occupied territories until “the development of Chinese regional institutions for maintaining security.”8

  The Semi-Democratic Regime, 1929–1932

The military’s defiance continued after the installation of the Inukai cabinet in November 1931. Although the Inukai cabinet opposed setting up an independent state in Manchuria, the Kwantung Army went ahead and did so, establishing Manchukuo in March 1932.9 Ten days later, the Inukai cabinet endorsed the plan to set up an independent state in Manchuria.10 Here again, the party government was forced to approve military actions that, though prohibited, had already been undertaken. the march incident and the october incident

Beginning in 1931, the military not only defied the party government by carrying out policies on its own, it also began to challenge the party government’s very existence. In March 1931, the members of the Sakurakai, an ultranationalist secret society created by young army officers led by Lieutenant Colonel Hashimoto Kingorō, plotted a coup d’état.11 They planned to establish a government headed by General Ugaki Kazushige, then the army minister in the Hamaguchi cabinet. It is said that they succeeded in securing the consent of many high-ranking army officers, including Ninomiya Harushige, the deputy chief of the General Staff of the army, Sugiyama ­Hajime, the vice minister of the army, and Koiso Kuniaki, the director of the Military Affairs Bureau.12 The coup was never carried out, allegedly because General Ugaki declined to participate.13 In October of the same year, young officers again formed a plan to stage a coup d’état, now known as the October Incident. Lieutenant Colonel Hashimoto conspired to carry out a coup d’état again with other radical members of the Sakurakai and nationalist ideologues such as Ōkawa Shūmei and Kita Ikki to establish a military government.14 His plan was leaked, however, and the conspirators were arrested on October 17. The second Wakatsuki cabinet treated the conspirators leniently, giving the leaders only a period of detention, since it was feared that severe punishment might radicalize other military officers.15 the may fifteenth incident and the collapse of the semi-democratic regime

The Manchurian Incident and the two attempted coups d’état demonstrated that the semi-democratic regime had reached the brink of collapse. The balance of power between party government and the military had tilted in favor of the military to such an extent that it had the power to imple-

The Semi-Democratic Regime, 1929–1932  129

ment policies on its own. Party government had lost its power to discipline. Moreover, by this time, the military had acquired the power to plot coups d’état which the party government was unable to counter, even after the plots were revealed. The May Fifteenth Incident and its aftermath made it clear that power had now completely gravitated to the military, bringing an end to the semidemocratic regime. The May Fifteenth Incident itself was not an attempted coup by the military but rather a series of assassinations carried out by military officers as individuals without a particular plan to replace party government.16 The military as an organization took advantage of the incident to remove the party government, however. After the Inukai cabinet’s resignation following the assassination of Prime Minister Inukai, military leaders repeatedly expressed their opposition to party government and succeeded in preventing another party cabinet from replacing the Inukai cabinet. As a result, Saionji recommended the emperor appoint Saitō Makoto, an admiral and not a leader of a political party, as prime minister. As a result, the practice of party government ended and the semi-democratic regime broke down.

Analysis Japan’s party government could not constrain the military from taking over political power for two reasons. First, the semi-democratic regime had lost its legitimacy. The party government could not effectively deal with economic problems while it was prey to a plethora of political problems. In addition, some party politicians acted semi-loyally to the regime by often justifying the military’s intervention into politics and refusing to join in a united resistance against the military when it tried to collapse the regime itself after the May Fifteenth Incident. legitimacy

Between 1929 and 1932, regime legitimacy continued to erode and disappeared completely by the end of this period at both the elite and mass levels. The primary reasons were the regime’s failure to deal effectively with economic crisis and the political parties’ continued power-seeking behavior. The following section turns to the question of how the regime lost legitimacy at both the elite level and the mass level.

  The Semi-Democratic Regime, 1929–1932

Legitimacy among the Mass Media and Intellectuals: Regime Performance in the Socioeconomic Sphere  The Hamaguchi cabinet carried out the return to the gold standard beginning in January 1930 in the expectation it would streamline Japanese industry and lead in the long run to economic recovery. At the time, however, the world had already entered the Great Depression, triggered by the collapse of the U.S. stock market in October 1929. The impact of the Great Depression and the return to the gold standard combined to produce disastrous effects on the Japanese economy, which had already experienced a series of economic depressions since the end of the First World War. Commodity prices began to tumble with the formation of the Hamaguchi cabinet and accelerated after the announcement of the return to the gold standard.17 Finally, the stock market crashed twice between March and April 1930, heralding the beginning of what has been called the Showa Great Depression.18 During the Showa Great Depression, commodity prices crashed. While wholesale prices in general declined—the wholesale price index fell 36.3 percent between June 1929 and October 1931—the prices of agricultural products and exported commodities were particularly hard hit. The price of raw silk plummeted by 66.1 percent between June 1929 and May 1932, and that of cotton yarn by 61.2 percent between June 1929 and October 1931.19 The Showa Great Depression had a devastating effect on the agricultural sector. The decline in the price of raw silk led to a fall in the price of cocoons, which many peasants relied to make a living.20 This decline, together with a drop in the price of rice, which was caused by the abundant harvest in 1930, ravaged the agricultural sector, as cocoons and rice were the two major sources of income for peasants and tenant farmers.21 The harvest was very poor in 1931, further debilitating the agricultural sector, and famine spread in some part of the country. The income of peasants and tenants dropped precipitously: the average income of peasant households dropped from 1,366 yen in 1929 to 641 yen in 1931, while the average income of tenant farmers dropped from 874 yen in 1929 to 449 yen in 1931.22 In 1930, for 58.6 percent of all peasants and 76.4 percent of all tenant farmers, living expenses exceeded income. Economic hardship caused tragedy in the villages, such as starvation and the selling of daughters, and it also increased the number of disputes between tenant farmers and landlords, from 1,866 in 1928 to 2,434 in 1929, then to 3,419 in 1931.23

The Semi-Democratic Regime, 1929–1932  131

The Showa Depression also hit the industrial sector hard. The value of industrial production fell by 26.7 percent between 1929 and 1931.24 Male real income fell by about one-tenth and female real income by about threetenths.25 Although official data on unemployment do not exist for this period, according to one estimate the number of unemployed in 1931 reached 2.5 million and the unemployment rate was about 9 percent.26 Just as disputes between tenant farmers and landlords spread during the depression, the number of labor disputes also grew, increasing from 1,420 in 1929 to 2,456 in 1931, the greatest number before the Second World War.27 Despite the deterioration of the economy, the Hamaguchi cabinet adhered to an austere fiscal policy and to the gold standard. In particular, it refused the Seiyūkai’s demands to increase expenditures by issuing government bonds to bolster domestic demand and to absorb the unemployed. And it failed to implement a social policy to deal with people’s grievances.28 It also refused to grant unemployment benefits and failed to win Diet approval for a bill to legalize labor unions. Adherence to such strict fiscal policies continued under the second Wakatsuki cabinet. Far from expanding its fiscal policy, the Wakatsuki cabinet embarked on streamlining government agencies and reducing the salaries of public officials to rein in expenditures. By the time the United Kingdom announced it would abandon the gold standard in September 1931, it had become clear that Japan could no longer sustain the gold standard. Industrial circles began to purchase enormous amounts of dollars, expecting that Japan would eventually have to abandon the gold standard. However, the cabinet still did not do so, and in response, the government sold dollars. What was more, the government raised the discount rate, which further damaged the economy. The Wakatsuki cabinet obstinately clung to the gold standard until its very last day. The way the Hamaguchi cabinet and the second Wakatsuki cabinet dealt with the economic crisis led opinion leaders to question the regime’s efficacy. Neither cabinet implemented any policies to deal with the crisis nor mitigate the problems it caused.29 Today, many criticize the Minseitō government for returning to the gold standard after the Great Depression had already begun.30 It is unlikely, however, that this in itself raised doubts about the regime’s efficacy. When Japan returned to the gold standard, it was considered the norm; most industrialized countries had already made the move by that time. Although some opinion leaders, such as Ishibashi

  The Semi-Democratic Regime, 1929–1932

Tanzan, expressed concern about the action, in general opinion leaders hailed the return to the gold standard as a great achievement.31 The Ōsaka Mainichi Shimbun praised “the sincere efforts by the current government” in bringing back the gold standard and argued that “the return to the gold standard is indispensable for a fundamental restructuring of the public and private economy.”32 Likewise, the Tōkyō Asahi Shimbun predicted that the return to the gold standard “will be quite beneficial for the economic development of our country” and “[will] return the economic fundamentals of our country to normal conditions.”33 This sort of support suggests it was not the return to the gold standard per se, but how the party government responded to the economy’s deterioration in 1930 and 1931 that undermined faith in the regime. Even after it became clear by spring 1930 that Japan had entered a depression, the Hamaguchi and the second Wakatsuki cabinet pursued the same macroeconomic policy, and economic conditions consequently deteriorated.34 While they maintained the gold standard, allowing a huge amount of money to flow out of the country, they also adhered to an austere fiscal policy. They reduced expenditures from 1.74 billion yen in fiscal year 1929 to 1.58 billion yen in fiscal year 1930, and even further to 1.48 billion yen in 1931.35 Combined with the impact of the Great Depression, this macro­ economic policy had a devastating effect on the economy: the outflow of money decreased the money supply while the decrease in expenditures reduced domestic demand. As a result, while economic growth slowed and unemployment rapidly increased, both average incomes and the price of various commodities fell. The Hamaguchi cabinet failed to come up with policies that would have mitigated the deteriorating economy’s effect on individuals.36 In other words, it failed to provide a social safety net for the ordinary citizen. It was reluctant to address the problem of unemployment, as demonstrated by the following statement by Finance Minister Inoue Junnosuke: “We will refuse to throw financial assistance to the unemployed as some people advocate. This is because we do not have the courage to repeat the same mistake in the face of clear failures by the industrial countries in Europe.”37 Even though the Showa Great Depression hit workers and farmers particularly hard and the Hamaguchi cabinet submitted bills to legalize labor unions and increase the legal protection of tenant farmers, it failed to get them passed in the House of Peers.

The Semi-Democratic Regime, 1929–1932  133

The Hamaguchi cabinet’s mismanagement of the economy exposed it to the harsh criticism of opinion leaders, such as Ishibashi: Unfortunately, the Hamaguchi cabinet is now losing the support of the people. The return to the gold standard, which the cabinet wrongly implemented, was combined with the world depression, which had incidentally taken place at the same time, to inflict severe damage on our industry. . . . However, the cabinet does not know how to deal with [the economic crisis] and has exposed its incompetence [by failing to solve the crisis]. Should the Hamaguchi cabinet fall, no matter what the direct cause may be, fundamentally it will be because the people have lost confidence in the cabinet due to its failed economic policy.38

Even the Tōkyō Asahi Shimbun, which expressed continued support for the government’s gold standard policy, criticized the Hamaguchi cabinet for failing to provide any policy solutions to the economic crisis. When the cabinet finally resigned, the Tōkyō Asahi Shimbun argued that it “demonstrated a significant degree of its incompetence in formulating social policy”—that is, for not being able to enact the bills to legalize labor unions and increase the legal protection of tenant farmers.39 It went on to point out the paucity of policies to deal with the economic crisis and claimed that none of the policies that the cabinet offered was effective. When the second Wakatsuki cabinet succeeded the Hamaguchi cabinet, the paper strongly urged it to alter the policies in place to deal with the economic crisis, contending that “it was a clear fact that the economic policy [of the Hamaguchi cabinet] was at a standstill.”40 No drastic policy change took place under the second Wakatsuki cabinet, however. By the time Great Britain suspended the gold standard in September 1931, it had become clear that the economic policy of the Minseitō governments had ended in total failure.41 Yet, the second Wakatsuki cabinet persisted in following the same macroeconomic course. It tried to maintain an austere fiscal policy in formulating the budget for the fiscal year 1932. It even planned to raise the tax rate to deal with the revenue shortfall, and when speculation in purchasing dollars took place after Great Britain suspended the gold standard, it raised the discount rate twice to maintain the gold standard, which worsened the state of the economy.42 In the face of the second Wakatsuki cabinet’s obstinacy, opinion leaders vented their fury. Ishibashi Tanzan condemned the cabinet for not taking any initiative to deal with the economic crisis. He complained the cabinet acted as if it believed “everything would somehow turn out well” in spite of

  The Semi-Democratic Regime, 1929–1932

increasing social unrest and a growing danger that “industry might totally collapse.”43 As for a planned tax increase, the Tōkyō Asahi Shimbun declared that such a move, targeted at “citizens who are in the depths of economic hardship,” would not only violate the “trust of the electorate” but also increase citizens’ hostility toward the government.44 The most thorough criticism can be found in the following editorial by the Jiji Shimpō: The developments of the past year and ten months after the return to the gold standard have contradicted predictions by the government and the depression continues to become more and more severe. . . . Although those in charge seem to remain confident, attributing [Japan’s economic depression] to the world depression, the citizens have been deceived after all. . . . Those with some knowledge of the economy will see it as a matter of course that the forced reduction in consumption by the government inevitably suppresses production and contracts the lives of the citizens in general. However, the current cabinet has dared to commit this injustice and make use of it to carry out its shortsighted policy.45

Legitimacy among the Mass Media and Intellectuals: Regime Performance in the Political Sphere  Although the regime’s inefficacy in the face of economic crisis contributed to the erosion of its legitimacy, it was not the only factor in this process. Poor regime performance in the political sphere was also at work. Political scandals and the way the party government and political parties acted toward and within the Diet made opinion leaders question whether they truly represented the opinions and interests of the Japanese citizens. Political scandals involving the Seiyūkai and the Minseitō continued to undermine the regime’s legitimacy during this period. For example, in summer 1929, after the Hamaguchi cabinet took office, two major political scandals involving the former Seiyūkai Tanaka cabinet came to light.46 In September 1929, Amaoka Naoki, the former president of the Decoration Bureau under the Tanaka cabinet, was prosecuted for seeking bribes in return for awarding decorations on the occasion of the emperor’s enthronement in 1928. In the same month, Ogawa Heikichi, the minister of railways under the Tanaka cabinet, was prosecuted for taking bribes for purchasing railways run by private companies and authorizing construction of new railways by private companies. These scandals were related to the Seiyūkai, but the Minseitō was not immune from corruption. Kobashi Ichita, the minister of education, was suspected of taking bribes from another railway

The Semi-Democratic Regime, 1929–1932  135

company in 1927 and was driven to resign in November 1929. He was prosecuted in early 1930. The trials for these crimes were reported in the mass media and continually reminded the public of the corruption of the political parties. Political scandals in the preceding period had already severely damaged the major political parties’ credentials. These newly revealed scandals further undermined the regime’s legitimacy, and many began to consider both major political parties unfit to govern the country. The Ōsaka Mainichi Shimbun, for example, expressed concern about a situation in which both of the major political parties had to be questioned about their qualification to hold office: The current situation has exposed the disgusting state of affairs of the existing political parties. The citizens have to cover their eyes in the face of the deeprooted disease of the existing parties and are disgusted with the mudslinging contests between the Seiyūkai and the Minseitō. [The political parties] are giving us more concern about the future of politics. . . . Rumors of a nonparty cabinet have begun to appear. . . . Voices expressing distrust of the existing political parties have begun to spread in all strata of society.47

When the scandal involving a minister of the Minseitō government came to light, the Tōkyō Nichinichi Shimbun aired its misgivings about the political credentials of the two major political parties, while also referring to the Seiyūkai’s earlier loss of credibility: Facing the current state of affairs, we cannot help being at a loss. Who should be in charge in the current situation and be responsible for purifying politics in the first place? Properly speaking, given such a political situation [in which a cabinet minister is suspected of involvement in a political scandal], we have to urge that the current cabinet resign, taking responsibility. If we follow the normal order [of things], the government should be transferred to the opposition party, the Seiyūkai. However, will the majority of citizens accept this unconditionally when this natural order is realized? It is not easy to erase from their memory the way the Tanaka cabinet managed the election, the way it created an unnatural majority after the election, the way it collapsed, various political scandals that emerged after its downfall, and so on. . . . In particular, many of its members were arrested in a number of political scandals. . . . We cannot say that they are more qualified than the Minseitō to be in charge of national politics.48

In the end, the paper nonetheless rejected nonparty government and concluded that the nation had no choice but to rely on the current cabinet to

  The Semi-Democratic Regime, 1929–1932

take charge. Yet the newspaper’s opinions suggest that the regime’s legitimacy had by this time already reached a critical state, as the qualifications of both major political parties began to be questioned. It is often assumed that political scandals involving party politicians frequently occurred because after universal male suffrage was introduced, political parties had to spend a large amount of money on elections.49 In particular, it was believed that the political parties were bribing voters, which again contributed to the erosion of regime legitimacy.50 Yoshino Sakuzō was disappointed that the universal male suffrage he had advocated so strongly had actually failed to clean up politics: I am one of those who have been a passionate advocate of universal male suffrage since the beginning of Taishō. I pointed to the probability that the political parties would not use money [in the elections] as one of the virtues of universal male suffrage. It was not that they would not spend money, but that they would not be able to spend money. Namely, they would not be able to afford to spend so [much money, given the increase in the size of the electorate]. Then I argued that after the disappearance of money, speeches and personality would be the only weapons in electoral campaigns. . . . My predictions turned out to be wrong.51

Yoshino went on to deplore the state of politics in which the political parties “control power to make money and use money to acquire power” and “frantically compete against one another, resorting to every conceivable means.” 52 In addition to being eroded by corruption, regime legitimacy was also undercut by the party government’s defiance of the Diet and the way the political parties acted in the Diet, leading opinion leaders to wonder whether party government and the political parties truly represented the citizens. The largest problem during this period was that the Diet was seldom convened. In the first place, the regular sessions of the Diet were only three months long, from December to March, a much shorter period than the regular Diet sessions in postwar Japan, which last at least six months starting in January.53 The government could also convene special sessions of the Diet. Under the Hamaguchi cabinet, the second Wakatsuki cabinet, and the ­Inukai cabinet, virtually only one regular session of the Diet, the fifty-ninth Diet, was held, along with two additional special sessions, the fifty-eighth and sixty-first. The first was in session for about three weeks and the second for one week.54 The Hamaguchi cabinet dissolved the fifty-seventh Diet, a regular session, shortly after its opening in January 1930, and the Inukai

The Semi-Democratic Regime, 1929–1932  137

cabinet dissolved the sixtieth Diet, another regular session, in January 1932, both times forfeiting the opportunity to engage in debates on important national policy issues. As a result, the government proposed many important policies without adequate debate in the Diet. For example, the Hamaguchi cabinet reduced the budget for fiscal year 1929, which had already been approved by the Diet, to prepare for the return to the gold standard, justifying this by claiming that reducing the budget fell within the government’s authority.55 When, in accordance with the requirements of the constitution, the government had to implement the budget drawn up for fiscal year 1929 after the fiftyseventh Diet failed to pass the budget for fiscal year 1930, it again reduced the budget to support the gold standard.56 Once more, there was no significant debate in the Diet over the return to the gold standard. That important policies were made without sufficient deliberation in the Diet raised concerns that the party government was governing without incorporating citizens’ interests and opinions. For example, Ishibashi Tanzan ascribed the attempted assassination of Prime Minister Hamaguchi to the inadequacies of the semi-democratic regime. He argued that it was not well structured to incorporate public opinion in the formulation of policy: Has the Hamaguchi cabinet ever allowed the citizens to have enough debate over carrying out the return to the gold standard? What about the London ­Naval Treaty? . . . In all instances [of policy making], even when we reach the same conclusion, the impression people get is totally different, depending on whether there have been thorough debates or not before reaching a conclusion. . . . Of course, this is not simply a crime of the Hamaguchi cabinet. Previous cabinets had become bureaucratic and despotic because our political system was originally structured in a way that violates the essence of democracy in many respects. Although our country is reputed to have put parliamentary politics into practice, the Diet [which is the heart of parliamentary politics] is convened for just a brief period every year and we have despotic politics without any Diet for most of the year.57

Even though the Diet convened only infrequently, if the party government had shown respect for debate in the Diet, and the debates had (at a minimum) appeared to be serious in the eyes of opinion leaders, the regime’s legitimacy might have been bolstered. The party government failed to demonstrate the appropriate respect during this period, however, and party politicians likewise failed to engage in serious debates in the Diet,

  The Semi-Democratic Regime, 1929–1932

which betrayed the expectations of the opinion leaders and reduced regime legitimacy. In this regard, the biggest setback for the regime’s legitimacy—to which the party government and the members of the Diet both contributed—­ occurred during the fifty-ninth Diet starting in December 1930, the only regular session of the Diet before the collapse of the semi-democratic regime. The party government harmed the regime’s legitimacy by appointing Foreign Minister Shidehara Kijūrō as acting prime minister after Prime Minister Hamaguchi was shot in October 1930. As Hamaguchi did not quickly recover from his wounds, Shidehara remained acting prime minister for a long time. Even when the Diet opened in December 1930, the Hamaguchi cabinet decided that Shidehara would continue as the acting prime minister. As Shidehara was not a member of the Minseitō, this appointment appeared to many opinion leaders to be a betrayal of the practice of party government rule in which the president of a political party leads the cabinet as prime minister. The Tōkyō Asahi Shimbun, for example, repeatedly criticized ­Shidehara’s acting premiership. Arguing that “the one who has to share ultimate responsibility cannot be someone who does not belong to a political party under the party government,”58 it claimed that the Hamaguchi cabinet was “temporarily” the same as a “transcendentalist cabinet.”59 Sasaki Sōichi, a law professor at the University of Kyoto, expressed concern over the repercussions on the practice of party government and the legitimacy of the semidemocratic regime in which the acting premier was not a party member: There is no doubt that the acting prime minister of a party government should be a party member. Therefore, it is obviously regrettable that a nonparty member has to face the Diet as the acting prime minister. Today there is a particular reason to reiterate what is self-evident and to argue that it is regrettable. [The practice of party government] is finally on the way to consolidation as a constitutional practice. When such a practice is about to be consolidated, it is deplorable that a nonparty member has been put in place as the head of a cabinet and that we have to call it a party government. . . . This might upset the public’s view of the value of party government. . . . I hear that the government and the ruling party ascribe the appointment of the acting prime minister to unavoidable circumstances. . . . [However,] today the Minseitō is a party more powerful than ever, which shares a majority in the Lower House. If this powerful party cannot realize the principles of party government because of external circumstances, the question becomes whether the political parties have enough power to practice party government at all.60

The Semi-Democratic Regime, 1929–1932  139

The problem of an acting premiership became more than a matter of principle. It became a matter of substance relating to the efficacy of the regime. That a nonparty politician was leading the cabinet opened the way for attacks by the opposing political party, the Seiyūkai. Worried that the problem of the acting premiership might complicate the debates over bills submitted by the cabinet, the Hamaguchi cabinet simply delayed introducing important bills.61 Opinion leaders then criticized the cabinet for not respecting the Diet. The Tōkyō Asahi Shimbun persistently denounced the Hamaguchi cabinet, calling its attitude “burying its head in the sand” and criticizing it for being “solely concerned with retaining its position without submitting important bills.”62 The Diet’s reputation also suffered because members of the Diet were often embroiled in disputes, and even physical altercations ending in bloodshed, over trivial issues instead of concentrating on debates on policy. The Ōsaka Mainichi Shimbun lamented, “The floor of the Lower House is not a place of debates and deliberations but literally an arena for battles and quarrels. . . . The current state of the floor of our Lower House is definitely a national disgrace. . . . [It] shakes the foundation of politics based on the constitution and undermines the function of parliamentary politics.”63 Ishibashi Tanzan’s criticisms of the regime’s poor political performance went deeper. In an editorial entitled “The Atmosphere of the World Has Recently Become Extraordinary,” which appeared after the resignation of the Hamaguchi cabinet in April 1931, Ishibashi contended that the regime’s poor performance had almost completely destroyed its legitimacy by this time: Prime Minister Hamaguchi should have decided to resign . . . after the incident. . . . However, by failing to do so . . . he placed our politics in a state in which the Diet virtually does not exist. . . . The Minseitō is a political party which [is known] . . . to respect the Diet. Sincerity was the catch phrase of Prime Minister Hamaguchi. This was why the people welcomed the cabinet he headed as worthy of enforcing official discipline. However, soon after the formation of the cabinet, one of his cabinet ministers was suspected of receiving bribes, discrediting the catch phrase. Hamaguchi also changed the composition of the budget and ignored the Diet’s right to approve it, raising doubts about the catch phrase of respecting the Diet. Furthermore, since [the attempted assassination], for five months including the most important three months during which the Diet was in session he has not looked after national politics at all and is now the subject of strong criticism for negligence and lack of official discipline. What irony. The government became the target of distrust, the Diet was derided, and the leader

  The Semi-Democratic Regime, 1929–1932 of a faction of the opposing party questioned the political responsiveness of the government in a way that was almost tantamount to a resort to direct action. As a result, people of conscience have become deeply concerned that the current state of affairs is no longer normal. The prime minister neglects his duty, we cannot have confidence in the words of the government, the Diet is mocked, and the representatives of the citizens have turned into a band of gangsters. In a word, it is not too much to say that we are almost in a state of anarchy.64

The Combination of Regime Performance in the Socioeconomic and Political Spheres and the Loss of Legitimacy   If its performance had been poor in only the socio­economic sphere or only the political sphere, the regime might not have lost legitimacy among opinion leaders. However, since the it failed in both, a loss of legitimacy became inevitable. The regime’s inability to deal with the economic crisis would not have led, per se, to its loss of legitimacy among opinion leaders if those same leaders had believed that the party government’s failures transpired after it had sincerely tried to incorporate people’s opinions and interests into policy making. Likewise, had there been no actual problems that pointed to the party government and political parties’ failure to respect citizens’ interests and opinions, poor political performance would not have led to the loss of regime legitimacy. In the final stage of the semi-democratic regime’s breakdown, however, legitimacy was lost because opinion leaders associated poor performance in the political sphere with poor performance in the socio­economic sphere. They believed the actions of the party government and the political parties under the semi-democratic regime caused severe problems, most notably the lingering of the economic crisis. The Tōkyō Asahi Shimbun, for example, repeatedly criticized the party government and the political parties for their indifference to society’s critical problems. Denouncing the Hamaguchi cabinet’s reluctance to propose bills to address the socioeconomic crisis, it argued, “[The Hamaguchi cabinet] cannot justify taking [such an attitude] toward the citizens. Even in the face of an unprecedented economic depression, they have been making enormous sacrifices, as they have held some expectations about the future by believing in the commitment of the government.”65 It also accused the second Wakatsuki cabinet of being concerned solely with staying in power, an attitude that “gives rise to anxiety in society and in industry.”66 Furthermore it denounced the sixty-first Diet, the last Diet under the semi-

The Semi-Democratic Regime, 1929–1932  141

democratic regime, for its members’ indifference to the interests of ordinary people and their fighting over trivialities, which contributed to the expanding social unrest.67 After seeing the regime collapse with the May Fifteenth Incident, M ­ inobe Tatsukichi, who had defended party government, acknowledged that the regime’s legitimacy had indeed disappeared: In the past several years, public sentiment has been restless and increasingly ominous. One reason is the extreme economic distress among the masses, particularly in the rural communities. The other reason is the lack of confidence in party politics. On the one hand, due to the world economic depression resulting from the world war, even in Japan trade has shrunk, business has become stagnant, prices have fallen, the number of unemployed has grown, and we cannot predict when economic recovery will ease the distress of those below the middle class both in the cities and in rural communities. On the other hand, the evils of party politics have become more and more conspicuous, politics is conducted not on behalf of ordinary citizens but mainly for the benefit of the political parties, public institutions have been used for private gain, public money has been wasted . . . power has been abused, responsibilities have not been met, false statements have been made, and so on. Such sins have been committed almost as a matter of course. Because of such [practices], together with worries about their economic lives, ordinary people have come to harbor untold grievances against party politics.68

It is difficult to pin down precisely at what point the regime’s legitimacy disappeared, but it seems that many opinion leaders had lost their faith in the regime—or had at least perceived that legitimacy was lost—by the time the Inukai cabinet was toppled in the May Fifteenth Incident. In October 1931, Ishibashi Tanzan had already expressed deep concern about the state of politics. In a shocking editorial titled “Illegal Trends [in Politics] Becoming Even Graver” he lamented that the situation had degenerated into “a kind of politics by violence” and argued that given such a state of affairs “it is no wonder that . . . quite a few people who praise Italian fascism have emerged.”69 In 1932, various newspapers, although they never endorsed rejection of party government and the Diet system, affirmed that many people had outright rejected party government and parliamentary politics. Moreover, though they continued to oppose the threat of force or violence against party government and parliamentary politics, the papers often admitted that, given the current state of politics, it was no wonder

  The Semi-Democratic Regime, 1929–1932

that such rejections emerged. For example, the Tōkyō Asahi Shimbun wrote, “politics by the existing political parties cannot continue to be as it is today. It is necessary to reconstruct parliamentary politics” and “[the state of politics] will change the current trend of defying the Diet into a trend of rejecting the Diet.”70 Several days before the May Fifteenth Incident, newspapers openly acknowledged that the legitimacy of the regime was in crisis. The Ōsaka Asahi Shimbun, for example, warned: Threatened by the voices criticizing the existing political parties, politics nowadays has lost all vitality. . . . What is now emerging, in a bid to take advantage of such circumstances, are voices and movements that reject parliamentary politics and the existing political parties. Our people are now facing an important moment in which they have to decide what attitude they should take toward such movements.71

The Tōkyō Asahi Shimbun also recognized the changes in the political environment, arguing “what we must now think about is how the state of affairs has changed within the past year. When the fifty-ninth Diet was convened, [views about the Diet] were ones of contempt and distrust. However, today they have developed into hatred.”72 In this environment where the legitimacy of the semi-democratic regime had disintegrated among opinion leaders, the Yomiuri Shimbun no longer expressed support for party government: The political parties’ loss of credibility should take most of the blame [for the current situation]. They have not properly managed parliamentary politics. As a result, they have got their just desserts. . . . Those who should reflect most on the end of the practice of party government are party politicians. . . . The place for engaging in debates has turned into a place of hatred and its debates cannot be trusted because of the laziness and irresponsibility of party politicians. . . . We call for the formation of a government that can bring alienated people back . . . and make as its priority bringing benefits to the people. We will not support any government that has no capacity to deal with the current state of affairs and does not deserve the trust of the people, no matter who forms the cabinet.73

The Tōkyō Asahi Shimbun wrote that the rise of nonparty government was inevitable: From the perspective of constitutional politics, it is our greatest regret that the existence of a dominant party was ignored because of the unfavorable reputation

The Semi-Democratic Regime, 1929–1932  143 of parties themselves. However, because the dominant party is in such a plight that it lost its confidence in one night and cried for support for a nationally united government [a nonparty government], we might say that this outcome is rather a matter of course.74

Faced with the state of the existing political parties, the Ōsaka Asahi Shimbun also endorsed the Saitō government: We should not hold genrō Saionji, who is in charge of recommending [the next prime minister to the emperor], responsible for recommending a person other than the president of one of the political parties to form a cabinet, thus deviating from the practice [of party government] and putting aside his own principles. [He made this recommendation] simply because he understood that the accumulated crimes of the existing political parties made it impossible for them to have the confidence of either the elite or the masses. . . . In this emergency, we also believe that the rise of an emergency cabinet cannot be avoided.75

Legitimacy among Political Leaders  Many political leaders during this time also lost belief in the regime’s legitimacy and clearly perceived that its reputation was sinking among the public. The diary of Harada Kumao, one of Saionji’s secretaries, reveals that a nonparty government continued to be viewed as a serious alternative within political circles in this period, just as it was during the preceding regime. Various maneuvers took place toward the end of the second Wakatsuki cabinet to design the succeeding cabinet, including a cabinet led by party politicians, but other leaders who did not belong to political parties, such as Hiranuma Kiichirō, the vice chairman of the Privy Council, Saitō Makoto, an admiral, and Ugaki Kazushige, a general, were also considered as possible candidates for prime minister.76 After the Inukai cabinet was in office, Saionji referred to Ugaki Kazushige and Konoe Fumimaro, the vice chairman of the House of Peers, as possible next prime ministers, though he also mentioned Inoue Junnosuke, a leader of the Minseitō, as another contender.77 As all this shows, party government had become merely one of many forms of government contemplated in political circles. In his examination of the process by which the Saitō Makoto cabinet was founded after the May Fifteenth Incident, Yamamoto Shirō reveals that there was almost no support for party government within political circles except among members of the Seiyūkai, who expected that Suzuki Kisaburō, the succeeding president, would be appointed the next prime minister.78 Before recommending the prime minister to the emperor, Saionji met with politi-

  The Semi-Democratic Regime, 1929–1932

cally important figures to ask their opinions about the next cabinet. In Yamamoto’s description, there is no evidence that some of them supported a party government to replace the Inukai cabinet.79 Kido Kōichi, the chief secretary of the Chamberlain of the Privy Seal, also felt that someone not affiliated with the political parties, such as Saitō Makoto, should be prime minister.80 Aside from the members of the Seiyūkai, Konoe Fumimaro was perhaps the only person who supported a party government to replace the Inukai cabinet. However, he saw it as one of many options, including a military cabinet.81 Legitimacy among the Public  For the years 1929 to 1932, as with the preceding period, it is difficult to assess the level of regime legitimacy at the mass level. Turnout rates in general elections continued to be high: 83.3 percent in 1930 and 81.7 percent in 1932.82 In addition, despite unrelenting criticism of the political parties, two major parties continued to win most of the votes and seats. In both elections, they gained more than 90 percent of the total votes and won more than 95 percent of all seats in the Lower House.83 In prefectural elections voting picked up in 1931, reaching 79.8 percent. Looking at these aggregate data, the elections do not reveal an erosion of regime legitimacy at the mass level. Other evidence, however, suggests that ordinary people felt strongly that Japanese society as a whole had reached a stalemate and that party government could not come up with an effective solution. The main reason for this feeling was the economic depression. As the depression took hold, the number of unemployed workers rose and incomes fell. The ensuing hardships are reflected in articles about such things as the growing number of university students who were unable to pay their tuition, elementary school students who could not bring lunch to school, and peasants suffering from starvation.84 Inomata Tsunao described the suffering of the peasants: “In the year of the famine, it was not uncommon for daughters to be sold for 15 yen. Daughters were sold to pay off debts, to pay rents, and to pay for rice. . . . The daughters were pleased to be sold. . . . They had no prospects even if they had stayed in the villages.”85 According to Shimomura Chiaki, “In Midoumura Village in Iwate Prefecture, boys lacked shirts and socks, even when it snowed. The girls wore only red-colored drawers that covered just their legs. As for food, people ate only steamed acorns and starchy roots to ease their hunger.”86 In the face of hardships such as these and the absence of effective solutions, the grievances of ordinary people mounted during this period, which is consistent with the view that regime legitimacy was declining. As pointed

The Semi-Democratic Regime, 1929–1932  145

out earlier, there was a rapid increase in the number of labor and tenant farmer disputes (see Table 5.1). Furthermore, Japan became plagued by assassinations and other hostile acts against the regime. The first of these, by an ultranationalist, came in November 1930 when Prime Minister Hamaguchi Osachi was shot and wounded. This was followed in early 1932 by the so-called League of Blood (Ketsumeidan) Incident. In February and March of that year, members of an ultranationalist group, the Ketsumeidan, led by Inoue Nissho, assassinated Inoue Junnosuke, the former finance minister under the Wakatsuki cabinet, and Dan Takuma, the head of the Mitsui conglomerate. Inoue Nissho’s group planned to assassinate other prominent figures in political and industrial circles, including Saionji Kinmochi and Wakatsuki Reijirō. When word of this plot got out, however, members of the group were arrested before it could be carried out. The conspirators claimed they had carried out the assassinations because they were outraged by the political situation and believed it necessary to remove people in power by force in order to reform Japan.87 Certainly most people did not engage in violence. Yet it seems that they ceased to worry about what type of regime would govern Japan. This is not to say that they actively supported an alternative to the semi-democratic regime or demonstrated overt hostility to the regime. Regime legitimacy was probably lost at the mass level in the sense that ordinary people, preoccupied with their personal hardships, were no longer concerned about the fate of the semi-democratic regime. Thus, even in the face of the military’s threat against the regime, ordinary citizens were not inclined to defend a system so ineffective in its handling of socioeconomic and political problems. Taken together, the views expressed by the mass media, opinion leaders, and political leaders reveal that many people felt alienated from politics. This was the result of the party government and political parties seeming to seek power above all else, and their utter indifference to the economic and social crisis that gripped Japan. Some articles concretely and directly reflect the spread of this alienation. The Tōkyō Nichinichi Shimbun, for instance, reported the reactions of those in different occupations to the fifty-ninth Diet: A puzzled teacher [said], “I do not even have the courage to take up the recent repulsiveness of the Diet. It no longer leaves an impression on us. Hearing of disputes day after day, I am more and more inclined to regard them as merely disputes within a special class. Until recently, young teachers were often infuri-

  The Semi-Democratic Regime, 1929–1932 ated, but these days they don’t pay attention, they’re just fed up.” A certain policeman in the city said to us, “I really feel that the world has reached an deadlock.” . . . Mr. Aoki Seiichi, a day laborer, described the ugliness of the Diet to us as follows: “I pay attention to the Diet every day. However, it does not interest us at all as they debate issues that have no relevance to our lives.”88

The following passage from an article in Ōsaka Asahi Shimbun highlights the sense of alienation that many workers felt and their desire for relief from the economic crisis: In face of political anarchy, the workers debate one another in dining rooms and offices about the next cabinet, talking about [the May Fifteenth Incident], sometimes as liberals, sometimes as fascists, and sometimes as sympathizers of the left. However, few of them regard the next cabinet as something that will directly affect their lives, regardless of what type it turns out to be. Most of what they say comes either from rage or hopeless cynicism. Yet their words indicate both conscious and unconscious sympathy for anything that will break the current situation because they are discouraged, having experienced all dimensions of the continuing economic depression. For example, those who feel gloomy and worried about the prevailing atmosphere of fascism will not argue that they will be saved by the existing political parties and liberalism.89

Many party politicians were aware of the loss of legitimacy among the public. They ascribed this to the regime’s performance in the political sphere. Before his death in 1931, Hamaguchi Osachi, the former prime minister, gave voice to citizens’ concern over the level of legitimacy in terms of the way the Diet was managed: The Diet, which must be a place to express opinions, may turn into a place of violence as the majority party abuses its majority by silencing the speech of the minority party and the minority abandons speech, disrupting order in the chamber with violence and obstructing debate, which in the extreme leads to physical clashes in the chamber. Were this to happen it would mark the destruction of constitutional politics and a violation of parliamentary politics; this is a matter of grave significance. . . . Alas, given such unruliness in the Diet, the citizens have become disgusted with and then cynical about parliamentary politics. Will they not then come to detest and then to reject parliamentary politics? Today, the citizens have come to the second stage. They are becoming cynical of parliamentary politics although they have not yet begun to demonstrate their hatred nor rejection. However, if we do not make some adjustments today, we cannot know in what direction the feelings of the citizens might go.90

The Semi-Democratic Regime, 1929–1932  147

Ogawa Heikichi, the minister of railways under the Tanaka cabinet, admitted more frankly that the legitimacy of the regime had been lost because of its poor performance. He acknowledged that “the mudslinging among the political parties” had made the public regard them as “a group of criminals” and that “distrust of the political parties even led [the public] to be suspicious of the Diet system itself.”91 Indeed, he ascribed the May Fifteenth Incident and other acts of terror during this period to mudslinging among the political parties. Inukai Takeshi also was aware that the regime’s legitimacy was in critical condition. A week before the May Fifteenth Incident, on May 8, 1932, in a speech at a Seiyūkai convention Prime Minister Inukai stated that rejection of the Diet was prevalent in some segments of society.92 Historians are the last group of individuals whose views of this period should be considered. They seem to endorse the view that what the public cared most about was not the fate of the regime itself, but instead whether they could overcome their own economic hardships. Awaya Kentarō writes: The depression hit the middle class ruthlessly. Until then, [much of the middle class] considered themselves the “backbone of society” and were a strong base for the parties. However, they started to lose confidence in the parties and asked for an overturn of the status quo, protection by the state, and restrictions on large companies.93

And Banno Junji adds: For the citizens who demanded economic recovery, relief from unemployment, and a way out from the economic hardships in rural areas, it was not important whether or not a party government would be formed after the resignation of a Seiyūkai cabinet with the tragic death of Prime Minister Inukai in the May Fifteenth Incident.94

In the final analysis, regime legitimacy seems to have been lost as ordinary people, who cared more about their own economic struggles, grew indifferent to the semi-democratic regime’s fate. Although they did not positively endorse an alternative regime, they no longer believed the semidemocratic regime was the only regime that could govern Japan. A Legitimacy Vacuum  Regime legitimacy was thus lost at both the elite and the mass levels. As we have seen, by the time the Tanaka cabinet fell, it was

  The Semi-Democratic Regime, 1929–1932

likely that the regime was sustaining its legitimacy only by default (see Chapter 5). That is, the belief of both the elite and the masses in the legitimacy of the regime—even given their awareness of its shortcomings—reflected their lack of support for an alternative regime dominated by a nonparty government. By 1932, however, the regime no longer could sustain its legitimacy even by default. This is not to say that there was strong support for an alternative regime that could be observed, for instance, in the rise of Nazism in Germany or fascism in Italy, where there was passionate support for a new regime, at least among some parts of the population. During this period in Japan, no positive support existed for any kind of regime. Many political leaders no longer supported the semi-democratic regime, nor did they support the regime dominated by the military which finally emerged in 1936 after the February Twenty-Sixth Incident. Although these leaders endorsed a nonparty government because of their lack of confidence in the party version, the military’s influence over the Saitō cabinet was not yet prominent. This is best demonstrated by an entry in Kido Kōichi’s diary: [In this situation] it goes without saying that the political parties and the military should cooperate. However, the military’s rejection of the political parties has taken on an emotional tone, which makes it difficult for them to cooperate. Therefore, we should make them both withdraw [from politics]. One effective solution [to the current situation] is to introduce a third, impartial, political figure onto the political scene and let him handle the situation.95

Opinion leaders and the masses, as we have noted, did not actively support the alternative regime either. This is probably why the mass media repeatedly remarked that the rise of the Saitō cabinet could not be “avoided” but did not positively endorse it.96 the semi-loyalty of party politicians

In addition to eroding regime legitimacy, the behavior of important political actors played a significant role in causing the semi-democratic regime to break down. During this crucial period, some party politicians did not hesitate to “encourage, tolerate, cover up, treat leniently, excuse, or justify” various actions taken by the military that overtly challenged the party government and curtailed its ability to constrain the military.97 While such behavior facilitated the military’s challenge of the party government over the London Naval Treaty and the Manchurian Incident, it also played a

The Semi-Democratic Regime, 1929–1932  149

crucial role in the final phase of the regime’s collapse after the May Fifteenth Incident. Here it is necessary to note that there were two important factors that had driven some party politicians to act semi-loyally to the regime. First was the influence from political institutions under the Meiji Constitution. As I have described in Chapter 3, there was no rule on the selection of prime ministers nor on the resignation of cabinets. Since there was the possibility that the fall of a Minseitō cabinet might lead to a government under a Seiyūkai prime minister, the Seiyūkai did all it could to bring down the government, sometimes even seducing the military to interfere in politics. Second was the close affinity of some party politicians with the military. The Seiyūkai was well connected with the military as is symbolized by the fact that Prime Minister Tanaka Giichi had been a general in the army. In addition, some Seiyūkai politicians shared the military’s orientation in foreign policy toward China. The London Naval Treaty  After the London Naval Treaty was signed, serious conflict arose between the party government and the military. The General Staff of the navy raised the issue of violating the supreme command of the emperor in order to obstruct its ratification. Here, opportunistic behavior on the part of party politicians encouraged the military to challenge the party government and thereby played an important role in intensifying the conflict. The Seiyūkai brought up the supreme command issue in order to attack the Hamaguchi cabinet and to engineer its demise, which would have increased the prospects for a Seiyūkai cabinet. The attempt by the Seiyūkai reflected the defect in the semi-democratic regime that no rule had been established on government succession. It expected that a possible showdown between the party government and the military might drive the Hamaguchi cabinet to resign and pave the way for another Seiyūkai cabinet. The way the government ultimately decided to sign the treaty gave rise to the dispute between the party government and the military, which later claimed that the process had violated the supreme command of the emperor. In March 1931, the Japanese delegation in London reached a final compromise with other delegations on the terms of the treaty and asked the government whether it would accept the compromise. The General Staff of the navy was ambiguous.98 Although it expressed its dissatisfaction with

  The Semi-Democratic Regime, 1929–1932

the treaty’s terms, it did not explicitly oppose the compromise when the government finally decided to accept it. On April 1, 1930, shortly before the cabinet met, a meeting was held between the prime minister and representatives of the navy, at which the final decision on whether to accept the compromise was to be made. In the meeting, Admiral Okada, representing the navy, agreed to accept the terms of the treaty, but Katō Hiroharu, chief of the General Staff of the navy, said only, “[with the naval armaments set by the treaty] I cannot take responsibility as the chief of the General Staff from the perspective of military operations.”99 And in a statement to the press on April 2, after the government had finally decided to accept the compromise and sign the treaty, the General Staff of the navy did not oppose the signing itself. In the statement, the General Staff announced that the navy was committed to coping with the new state of affairs without resorting to rash acts. It stated also that although it could not agree with the level of armaments set by the compromise, with its words and deeds it do its best not to undermine Japan’s security.100 Certainly, the General Staff of the navy was not content with the terms of the compromise, yet it did not dare convey its dissatisfaction to the government. At this time it intended to follow the government’s decision. On March 26, a joint summit meeting of the Ministry of Navy and the General Staff of the navy adopted the following policy regarding the treaty:101 Even if the policy of the Navy [on the treaty] does not get accepted by the government, as a matter of course the institutions of the Navy will not go beyond their [proper] administrative affairs and military affairs. Furthermore, it goes without saying that the institutions of the Navy will do their duty within the range set up by government policy [toward the treaty], according to their respective jurisdictions.102

The chief and vice chief of the General Staff of the navy participated in the meeting and did not oppose the policy mentioned above.103 Thus it seems that at this point the General Staff of the navy did not explicitly oppose the signing of the treaty nor had it even reflected on the violation of the emperor’s supreme command as a reason to reject the treaty.104 Why, then, did the violation of the supreme command suddenly emerge as an issue, significantly exacerbating the clash between the military and the party government? Kobayashi Tatsuo and Itō Takashi, who have extensively researched the policy process of the signing and ratifying of the London Naval Treaty, suggest that it was the Seiyūkai that raised the issue of the

The Semi-Democratic Regime, 1929–1932  151

violation of the supreme command of the emperor as a means to force the Hamaguchi cabinet to resign.105 Indeed, it is likely that the Seiyūkai tried to manipulate the navy and the Privy Council to force the cabinet’s resignation. The Seiyūkai expected that the issue of the violation of the supreme command of the emperor would provoke the navy to oppose the government while it would lead the Privy Council to reject ratification of the treaty. On one hand, the Seiyūkai made a series of public attacks on the government, both inside and outside the Diet, alleging that it violated the supreme command of the emperor in signing the treaty. On April 2, Mori Kaku, secretary-general of the Seiyūkai, criticized the government decision to sign the treaty: We should not include national defense in ordinary administrative affairs [of the government] in the first place. Everyone recognizes that direct responsibility for national security . . . resides [with the general staffs of the army and of the navy], and ministers are not directly responsible for national security. The ministers who are not directly responsible [for national security] have made an important decision on national security despite vigorous opposing opinions from the General Staff of the Navy, which is directly responsible [for national security].106

The Seiyūkai went on to castigate the government in the fifty-fifth session of the Diet beginning on April 23, 1931, for allegedly violating the supreme command of the emperor, arguing that the right to determine the level of armaments was primarily within the jurisdiction of the general staffs.107 On the other hand, the Seiyūkai also appealed directly to the navy during this period, anticipating that pressure from the navy would lead to the collapse of the cabinet.108 For example, the biography of Mori Kaku, secretary-general of the Seiyūkai, indicates that he worked persuasively on the General Staff of the navy to topple the government.109 That there was intimate contact between Mori and the General Staff is supported by the fact that he was aware beforehand that Katō Hiroharu, the navy chief of staff, would submit his resignation to the emperor on June 10 in protest of the signing of the treaty.110 In addition, several members of the Seiyūkai, such as Suzuki Kisaburō and Yamamoto Teijirō, visited Admiral Okada Keisuke (who strongly supported the treaty and helped the government by persuading the opposition in the navy to support the treaty) to encourage him to oppose the treaty.111 Their expectations of the navy were clearly demonstrated by Suzuki’s statement to Admiral Okada: “Itō [Miyoji] [a member of the Privy Council] said, ‘The last time the Privy Council was the assassin.

  The Semi-Democratic Regime, 1929–1932

This time it is the navy.’ But, alas, it may not work that way!”112 Suzuki meant to say that he was expecting that the navy would contribute to driving the Hamaguchi cabinet to resign, just as the Privy Council had forced the last Minseitō cabinet to resign in 1927 by refusing to approve the government’s emergency decree. The Seiyūkai also encouraged the Privy Council not to approve the ratification of the treaty.113 Again, it did so expecting that the rejection of the treaty would lead to the cabinet’s resignation. Around this time, Mori Kaku told Harada Kumao: “The cabinet will fall because of this problem. . . . In the Privy Council the government has finally reached a dead end. . . . Within four or five days, the government will fall.”114 The Seiyūkai even held a special party convention on September 16, 1930, at which it asserted that the treaty should be abandoned, expecting that would bring down the government.115 The Seiyūkai’s expectations were completely betrayed, however, as the Hamaguchi cabinet overcame obstructions from the military and the Privy Council and ratified the treaty. But, by bringing up the issue of the violation of the supreme command of the emperor, the Seiyūkai contributed greatly to weakening the party government’s position against the military. After signing the treaty, the General Staff of the navy persuaded the navy minister to accept that it also had jurisdiction over naval armaments. In addition, the heated controversy over the violation of the supreme command increased the military’s antipathy toward the party government, as demonstrated in the charter that the Sakurakai adopted to state its objectives at its alleged first meeting in October 1930.116 The charter reeked with hostility to party government. It is obvious that poison of politicians which has been recently directed at the navy will soon be targeted at a reduction of the army. Therefore, we, the middleranking officers of the army, have to unite together and work hard every day. We have to avoid a recurrence of the navy reduction fiasco as a matter of course and have to have the spirit of cleaning the belly of the corrupted political leaders with patriotic passion.117

Although the Seiyūkai had a stake in preserving the semi-democratic regime in the long run, its pursuit of short-term interests—the formation of a Seiyūkai cabinet—actually promoted the military movement against the regime. The Manchurian Incident  Before the Manchurian Incident, some politicians implicitly endorsed the objectives the military sought to achieve through

The Semi-Democratic Regime, 1929–1932  153

the incident. After the incident, the party government, as we have seen, took a lenient position on the initial military actions that had overtly defied it. The behavior of certain political actors severely curtailed the party government’s ability to restrain the military. Behind the military’s continuing defiance of the party government in the Manchurian Incident was its abiding commitment to using force in order to pry Manchuria off from the rest of China and place it under the influence of Japan. Once the military had opened fire, moreover, it could prevent the party government from disciplining it by justifying its operations as falling within the supreme command of the emperor. Only once, at the end of November, did the government succeed in restraining the expansionism of the Kwantung Army, by persuading Army Minister Minami and Chief of the General Staff Kanaya to order the Kwantung Army to reduce its military actions.118 However, the Kwantung Army and officers in other units immediately criticized these governmental restraints as a violation of the supreme command, and the Kwantung Army again began to expand the front line in Manchuria.119 The weakness of politicians also played an important role in allowing the military to ignore the party government. Before the Manchurian Incident, the Minseitō cabinets’ policy was to rely on diplomacy to maintain Japanese interests in the region. However, some politicians implicitly supported the policy that the military considered necessary to protect Japanese interests in Manchuria. Although the Minseitō government was committed to diplomacy (even in the face of Chinese attempts to recover its sovereignty), its policy objectives were identical to those pursued by the military. For example, at a convention of the Minseitō, Prime Minister Wakatsuki proclaimed: “Our country holds vital interests in Manchuria which are related to the survival of our people. We have to stand firm to defend the survival of our country no matter how many sacrifices we have to make.”120 Turning to the role played by the Seiyūkai, the other important political party, the Seiyūkai aligned itself closely with the policy pursued by the military during this period, as is evidenced by the Tanaka cabinet’s policy toward China.121 Adopting an expansionary mission in China, Tanaka dispatched Japanese troops a number of times to Shandong and was eager to expand Japanese interests in Manchuria. In doing so, some politicians of the Seiyūkai collaborated with army officers and radicalized them.122

  The Semi-Democratic Regime, 1929–1932

After the Seiyūkai had lost power to the Minseitō, its politicians not only criticized the Minseitō cabinets’ diplomacy, but also implied that using force was necessary to secure Japanese interests in Manchuria. In a speech in the Diet in January 1931, Matsuoka Yōsuke, a Seiyūkai representative, declared: The issue of Manchuria is directly related to the survival of our country. Manchuria is “the lifeline of Japan” from the perspective of our defense as well as our economy. . . . One and a half years have passed since the present government was formed. What has it done so far in Manchuria? The foreign minister’s policy of “to do absolutely nothing and wait and see” has been thoroughly implemented in this area.123

Mori Kaku was more explicit in rejecting diplomacy as a solution and advocating the use of force: Everything in Manchuria has a close and inalienable causal relationship with the Japanese right of survival. In the past and the future, regardless of the country, the efforts a country makes for the right of survival are absolute. Nothing, even external pressure, environment, restraints from treaties, can affect such efforts. . . . We have to clearly realize this. Weak diplomacy . . . undermines the Japanese right of existence and is very dangerous. . . . In conclusion, as I reported to the Seiyūkai the other day, I must argue that we have no other means but mobilization of national force in order to address the deteriorating [situation] in Manchuria and China under the Chinese anti-Japan policy. Of course, I personally have an idea about what is meant exactly by the mobilization of national power. Yet, as the time has not yet reached the point where I can announce and implement it, I have no choice but to leave it to the interpretation of the reader.124

At a meeting of the Seiyūkai leaders, he also told them that he could not think of anything other than “the mobilization of national force” to protect Japanese interests in Manchuria and that “the situation in Manchuria is about to reach a state of war.”125 Mori even directly encouraged the military. In a speech to the army in August 1931, Army Minister Minami Jirō publicly expressed his “regret” that “the situation in Manchuria is becoming extremely unfavorable for the empire.”126 Although the government and the Minseitō criticized Minami for interfering in politics, Mori said to Minami, “Your speech is far from being radical or deviating from your responsibilities. . . . I fully agree with your understanding of the ‘Manchurian problem.’ I hope you will do your best.”127

The Semi-Democratic Regime, 1929–1932  155

Even after the Manchurian Incident had begun, the government had several chances to recover control. The indulgent attitude of party politicians toward the military, however, made it difficult for the party government to control the military. First, relying on evidence that the Ministry of Foreign Affairs had gathered, Foreign Minister Shidehara Kijūrō judged that the Kwantung Army opening fire in Manchuria was the result of a conspiracy.128 However, though he used this evidence to convince Army Minister Minami to agree to the cabinet policy not to expand military operations, he did not rely on this information to restrain the military later on when it continued to ignore the party government’s policies. That the Kwantung Army triggered the Manchurian Incident through a conspiracy was of pivotal importance. This meant that the Kwantung Army itself violated the supreme command of the emperor, as it moved its troops without orders from the army’s central command. If the cabinet had publicly announced that the Kwantung Army opened fire through a conspiracy, the military would have lost its grounds for justifying the actions it had taken in Manchuria. But the government did not do so. The second opportunity to rein in the army came when the Chōsen Army crossed the border between Korea and China on September 21. This was politically important because the Chōsen Army had failed to secure approval from the cabinet and from the emperor to deploy troops and to finance military operations abroad.129 At the time, it was believed that sending the Chōsen Army to Manchuria was indispensable to achieve the K ­ wantung Army’s objective of occupying all of Manchuria. Had the second Wakatsuki cabinet refused to endorse the action of the Chōsen Army and declined to fund its military operations, the military would have had trouble carrying out its objectives. At the very least, the second Wakatsuki cabinet could have used this violation and its post-hoc endorsement to make a deal with the military that might have enabled it to recover control over the military. Prime Minister Wakatsuki, however, did not dare to make this an issue and the cabinet missed this chance. The cabinet endorsed the crossing without criticizing the military for violating the rules, which actually facilitated the military’s adventures in Manchuria. Behind party politicians’ reluctance around this period to restrain the army was a fear of a coup d’état. Given the defiance of the Kwantung Army and the army’s unwillingness to restrain it, the second Wakatsuki cabinet asked Saionji and jushin such as former prime ministers to meet with the emperor to rein

  The Semi-Democratic Regime, 1929–1932

in the Kwantung Army.130 On October 14, Saionji decided to go to Tokyo by October 21, implying that he would support the cabinet in this goal.131 During this period, although Lieutenant Colonel Hashimoto’s conspiracy to stage another coup d’état—the October Incident—was leaked and the conspirators were arrested, the planned coup d’état scared off the jushin and Saionji from constraining the army, and they decided not to bring the issue before the emperor.132 Having lost the ability to constrain the Kwantung Army, the cabinet in November reversed its policy against expanding military operations in Manchuria and decided there would be no quick withdraw from the occupied territories.133 Thus, the attempt to constrain the Kwantung Army through a meeting of the jushin had failed. Even after the October Incident conspiracy had become known within political circles, some party politicians attempted to constrain the military by forming a grand coalition government.134 Some leaders of the two major parties, such as Adachi Kenzō, a leading figure in the Minseitō and the minister of interior in the Wakatsuki cabinet, and Kuhara Fusanosuke, the secretary-general of the Seiyūkai, discussed possible cooperation between the two major parties. Early in December they agreed to establish a coalition cabinet. ­Although Prime Minister Wakatsuki initially supported this idea, some cabinet ministers were opposed.135 As a result the attempt was abandoned. A major reason for the ministers’ opposition was that it was difficult to reconcile the differences between the economic policies of the two political parties.136 The May Fifteenth Incident  After the May Fifteenth Incident, the lack of unity among political actors in defense of the semi-democratic regime played an important role in its final breakdown. The party politicians failed to form a united front against the military’s threat to refuse to accept the continuation of party government. Although Nagai Ryutarō of the Minseitō approached the Seiyūkai to form a joint movement in order to preserve party government, the Seiyūkai was reluctant and this initiative petered out.137 Rather, by this time prominent party politicians were ready to accept a nonparty government. Mori Kaku, the powerful Seiyūkai leader, no longer supported party government, and aborted the Seiyūkai politicians’ attempt to hold a meeting in which they planned to urge the preservation of party government. Going even further, and working on Army Minister Araki Sadao, Mori tried to have Hiranuma Kiichirō, the vice chairman of the Privy Council, selected as the next prime minister.138

The Semi-Democratic Regime, 1929–1932  157

At the same time, even former prime minister Wakatsuki Reijirō, the president of the Minseitō at the time, felt that Admiral Yamamoto Gombei, a former prime minister and a leader of the Satsuma clique, should form the next cabinet. When Saionji sought his opinion on the next cabinet, he responded: As I am a party member, I am a supporter of the party government system. However, when the military has lost its discipline and triggered [the May Fifteenth Incident] it is not necessarily appropriate to assert a party government. It would be proper to recommend someone who has a strong will and can get support from the military.139

Takahashi Korekiyo, the former prime minister and former president of the Seiyūkai, also conveyed his reluctance to Saionji to see the continuation of the Seiyūkai cabinet.140 There is no way to tell if having all party politicians join together to protect the practice of party government would have been effective. Never­ theless, it is likely that the lack of unity among party politicians affected Saionji’s decision not to nominate Suzuki Kisaburō, who became president of the Seiyūkai after Inukai’s assassination. As Saionji told Harada, “Wakatsuki’s story made the most sense.”141

Conclusion

A semi-democratic regime is one that arises after the long-term democratization of a nondemocratic (e.g., despotic or authoritarian) regime. It is a distinct political regime and needs to be studied in its own right. Such a regime can head in either of two directions: it can develop into a fully democratic regime, or it can collapse into an authoritarian regime. Japan, as we have seen, followed the latter course.

The Birth and Death of the Semi-Democratic Regime As a nondemocratic regime begins to democratize, free competition among politicians, regularly held elections, guarantees of various political rights, and control over political offices gradually emerge. Under a semi-­democratic regime, there is competition for political offices and elections are held regularly to fill those offices, though they are neither fully free nor fair. Even when competitive elections are regularly held, not all effective political offices are accountable to the electorate through elections and the size of the population with the right to vote is limited. Two political forces coexist within a semi-democratic regime: those subject to electoral control (democratic forces) and those that remain independent from electoral control (nondemocratic forces). The semi-democratic regime breaks down when the balance of power between democratic and nondemocratic forces drastically changes in favor of the latter. To explain this change, I have developed an analytical framework that focuses on three factors: political institutions, regime legitimacy, and the semi-loyalty of 159

  Conclusion

­ olitical actors. The first, political institutions, determines the legal political p power held by democratic and nondemocratic forces, and sets parameters for their political behavior. Regime legitimacy influences the political power that democratic and nondemocratic forces can project. When the level of regime legitimacy is high, democratic forces can contain nondemocratic ones and maintain the semi-democratic regime. However, when nondemocratic forces challenge the semi-democratic regime (often resorting to the prerogatives provided by political institutions) and when the level of regime legitimacy is already in decline, democratic forces are constrained in their ability to contain nondemocratic pressures. Finally, the semi-loyalty displayed by some democratic forces diminishes the power of democratic forces as a whole even as it strengthens nondemocratic forces. When regime legitimacy is completely lost and some of the democratic forces lose their loyalty—that is, become semi-loyal—to the regime, democratic forces can no longer sustain the semi-democratic regime in the face of nondemocratic challenges. When this occurs, the semi-democratic regime breaks down. In prewar Japan, the gradual democratization after the Meiji Restoration in 1867 led to the rise of a semi-democratic regime in 1918, when the Hara cabinet was formed as virtually Japan’s first party government. Confronted with a series of threats from the military, this regime broke down in 1932, and was replaced by a military authoritarian regime in 1936. Using our analytical framework, we see that at the outset of the regime the party government held sufficient power to constrain the military. Gradually, however, the balance of power tilted toward the military, and finally, by 1932 had shifted so drastically that the semi-democratic regime collapsed. Under the Meiji Constitution, Japan’s party government was institutionally weak: the cabinet lacked internal cohesiveness and many other political institutions could curtail its power. The military, on the other hand, was institutionally strong; it remained autonomous from the government because of various constitutional prerogatives related to the supreme command of the emperor. Initially, the party government could constrain the military because it enjoyed a high level of legitimacy. However, the regime’s legitimacy declined at both the elite and the mass levels because the regime performed poorly in the socioeconomic and political spheres. Party government was unable to address the lingering economic crisis effectively, while the behavior of the party government and the political parties—political scandals, outbreaks

Conclusion  161

of physical violence in the Lower House, and a lack of respect for the government—increased doubts about the party government’s accountability to the public. Concomitantly, distrust of party government and concern for Japan’s national security politicized the military and emboldened it to challenge the semi-democratic regime. The semi-loyalty of party politicians, which encouraged, tolerated, and justified the military’s challenge, severely curtailed the party government’s power and invested even more power in the military. When the military tried to topple the semi-democratic regime in the aftermath of the May Fifteenth Incident in 1932—by which time the regime’s legitimacy had disappeared completely—some party politicians acted semi-loyally to the regime. They refused to unite with the party government to defend the semi-democratic regime and, accordingly, it broke down with the inauguration of the Saitō cabinet.

Implications for the Study of Democracy The arguments presented in this book have important implications for the study of democracy. In particular, the book considers the significance of the semi-democratic regime as a concept and examines the parallels between what happened in Japan and findings from other studies on the breakdown of democratic regimes. the semi-democratic regime as an analytical concept

In recent years there has been significant progress in the study of hybrid regimes, reversing a tendency in the study of democratization to draw a dichotomy between democratic and authoritarian regimes. The case of Japan shows that there can be hybrid regimes, neither completely democratic nor completely authoritarian, but instead semi-democratic. The transition from an authoritarian regime to a democratic one may be long and drawn out, in which case a country may experience a regime that lies somewhere between democracy and authoritarianism. The transition may also be aborted, as happened in Japan, where the semi-democratic regime failed to mature into a democracy. Although the literature on political regimes has paid increasing attention to hybrid regimes such as competitive authoritarianism and electoral authoritarianism, which, for example, can be found in Mexico since 1980s and postwar Malaysia, in crucial respects the hybrid regime in Japan from 1918 to 1932

  Conclusion

was different. Other examples comparable to Japan are Great Britain in the nineteenth century, postwar Brazil, and Thailand in 1980s. The literature on hybrid regimes has often focused on, for example, the state of political competition. Factors such as this, however, cannot fully explain the nature and fate of the semi-democratic regime in Japan. Instead, one needs to pay particular attention to the degree of control over political offices by the electorate. The case of the semi-democratic regime in prewar Japan also sheds light on models of democratization. Prewar Japan had already achieved a significant degree of democratization in the two areas that Dahl used to model democratization: political competition and political participation. The biggest stumbling block for democratization in prewar Japan consisted of political institutions that did not fall under the electorate’s control—the military, the House of Peers, the Privy Council, and the genrō. Ultimately, the challenges and obstructions posed by these institutions—especially the military—led the semi-democratic regime to break down. Put another way, had these nonelectoral institutions not retained their power to challenge the party government, the semi-democratic regime might have avoided breakdown. This suggests that eliminating the prerogatives of political institutions not accountable to the electorate is a key aspect of democratization. Therefore, in addition to Dahl’s identification of political competition and political participation, the enlargement of electoral control over politically significant offices (or the elimination of the prerogatives of institutions that do not fall under electoral control) can be considered a third dimension of democratization. Today, most countries in the world have more or less opened their doors to political participation through regularly held elections. Thus, a modified form of Dahl’s model—one that incorporates political competition and the expansion of electoral control—can serve as a useful model for analyzing contemporary cases of democratization. semi-loyalty and civil-military relations

The breakdown of the semi-democratic regime in Japan highlights once again the importance of civil-military relations in the development of various political regimes. Changes in civil-military relations are of vital importance in the rise of authoritarian regimes, in the transition to democratic regimes, in democratic consolidation, and in the breakdown of democratic regimes. The semi-democratic regime of Japan broke down because the civil-­ military relationship gradually shifted toward the military to the point that

Conclusion  163

party politicians and political parties could no longer prevent political interference by the military. A major cause of the military’s political meddling was the semi-loyalty of some party politicians. Their failure to whole-heartedly support the regime had an important effect in tilting the balance of power between the party government and the military in favor of the latter. The first turning point in the relationship between party government and the military was the assassination of Chang Tso-lin in 1928. The semi-loyalty of party politicians prevented the party government from disciplining the military after this incident. Prime Minister Tanaka felt he could not disclose the facts behind the incident nor court martial the conspirators. This was because he met with opposition not only from the army but also from within his own party, the Seiyūkai. How the Seiyūkai reacted to the Hamaguchi cabinet’s signing of the London Naval Treaty also had a deep impact on the relationship between the party government and the military. The Seiyūkai raised the sensitive issue of supreme command in debates in the Diet and criticized the Hamaguchi cabinet for violating the supreme command of the military by signing the treaty despite the opposition of the General Staff of the navy. This had two repercussions. First, it weakened the power of the party government vis-à-vis the military. For a political party, this was nothing less than ­suicidal. Until then it was widely admitted that, under the Meiji Constitution, the military had autonomy in military operations. Yet, it was not clear which institution—the government or the military—had control over the organizational affairs of the military. The Seiyūkai endorsed the claim of the military that it had autonomy also in organizational affairs. Second, by drawing attention to the violation of the supreme command of the military, the Seiyūkai radicalized the military. The semi-loyalty of party politicians exhibited in a series of events from the Manchurian Incident to the May Fifteenth Incident further deprived party government of the power to confront the military. Some party politicians openly supported the military’s objectives in the Manchurian Incident. Furthermore, because of the opposition of some party politicians the two major political parties failed to initiate a united front to continue the practice of party government after the May Fifteenth Incident. The question that arises is why some party politicians acted semi-loyally. One possible answer is that many of them identified themselves as the ruling elite and had a close affinity with other political actors including the

  Conclusion

military. For example, Marius Jansen claims that party politicians “became part of the palace system,” implying that they, the military, and the bureaucracy belonged to the same class in that they supported the emperor system and the Japanese empire.1 Such a claim suggests that it was natural for some party politicians to have collaborated with the military. This might have been the case until the emergence of party government. Party politicians, in particular those who belonged to the Seiyūkai, worked closely with the hanbatsu oligarchies, often participating in the cabinets lead by oligarchs. They built up close relationships not only with the oligarchies but also with the bureaucracy and the military. Even after the emergence of party government, the Seiyūkai had a close relationship with the military as symbolized by the fact Tanaka Giichi, a general, became its president in 1925 and the prime minister in 1927. Yet, it is too much to treat all party politicians and the military as a single class with a shared identity. The relationship between political parties and the military was much more complicated. If they were identical, why did they confront each other over the signing of the London Naval Treaty? The truth is that there were divisions among the political parties as well as in the military on several policy issues. For example, although there was consensus both in the political parties and the military on the importance of securing Japanese interests in Manchuria, opinions diverged on the means to secure those interests. Andrew Gordon subscribes to this view when he argues that “parties joined the ruling elite in part because their leaders were extremely practical politicians who had come to see bureaucrats and military men more as allies than as opponents. In social terms, the party leaders were little different from elites in the bureaucracy and military.”2 Yet, he admits “through the 1920s, an important strategic division also emerged among parties and within the bureaucracy and the military.”3 Two institutional reasons were far more decisive in seducing some party politicians to act semi-loyally. The first was the two party system. During the era of the semi-democratic regime, politics evolved under the system of two major parties (initially, the Seiyūkai and the Kenseikai and then the Seiyūkai and the Minseitō) most of the time.4 The two political parties fiercely competed for power. A loss for one party was a gain for the other. Thus, the parties condemned each other, the one trying to spotlight its differences from the other. Sometimes, for the sake of condemning the opposing party, politicians resorted to radically irresponsible arguments, which

Conclusion  165

at times had the effect of endorsing the military’s claims. The wholesale endorsement of the Manchurian Incident by the Seiyūkai is an example.5 Second, the “semi-democraticness” of the regime itself lured some politicians into semi-loyalty. As described in Chapter 3, there was no firmly established rule on the nomination of prime ministers as well as changes of government.6 Having the majority in the Diet was not a prerequisite for political parties to come to power and have one of their own selected as prime minister. General elections were not the ultimate event that led to a change in government. Rather, the selection of prime ministers was at the discretion of the genrō. In practice, the transfer of power from the ruling party to the opposition was usually the outcome of a political deadlock. Thus, the opposition did everything it could to undercut the government. Sometimes it even resorted to inviting the military to condemn the government, expecting a showdown between the government and the military would bring about the fall of the government. This temptation increased political interference by the military. This confirms the proposition on civil-military relations in comparative politics that the military intervenes when politicians go “to knock on the door of the barracks.”7 This was exactly what happened in the case of the London Naval Treaty. legitimacy

Legitimacy, a factor to which much importance has been attached in the literature on the breakdown of democratic regimes, is also analytically important in examining the fall of semi-democratic regimes.8 Indeed, a major cause for the collapse of the semi-democratic regime in Japan was its obvious loss of legitimacy. It is instructive here to make a brief comparison of the Japanese case with the cases of other semi-democratic regimes. Taking the example of Brazil between 1945 and 1964 and Thailand between 1978 and 1997, we see that the semi-democratic regime in Brazil broke down while in Thailand it survived the crisis between 1991 and 1992 and succeeded in reequilibration. The Brazilian semi-democratic regime collapsed in 1964 as a result of the military’s intervention in politics.9 From the end of March to the beginning of April of that year, the Brazilian armed forces staged a coup d’état and seized power from President João Goulart, who subsequently fled the country. Following that, the coup leader, General Castero Branco, was elected

  Conclusion

president by the Congress. What followed was a military authoritarian regime that lasted until 1985. Based on an examination of views expressed by politicians and opinion leaders, Alfred Stepan argues that the Brazilian regime broke down because the legitimacy of the “constitutional regime” had declined to the point that civil-military relations were fundamentally changed.10 He argues that the regime was in the state of a crisis and had lost legitimacy because of deteriorating economic conditions and political disorder.11 In the early 1960s Brazil was hit by high inflation, sluggish economic growth, general strikes, and so on. At the same time, political disorder reigned as evidenced by the presidential election of 1960, the confused succession of the presidency from Jânio Quadros to Goulart, a change in the political system through the introduction of a parliamentary system, and a plebiscite on increasing the power of the president.12 Although the semi-democratic regime in both Japan as well as Brazil broke down, the semi-democratic regime in Thailand survived a crisis between 1991 and 1992 and made a transition to a democratic regime in late 1990s. The Thai semi-democratic regime entered a state of crisis in February 1991 when the military led by General Sunthorn Kongsompong staged a coup d’état.13 He set up a so-called National Peacekeeping Council (NPKC) with commanders of the three branches of the military, including General Suchinda Kraprayoon. The military suspended the constitution, closed the National Assembly, and arrested the prime minister, Chatichai Choonhavan. The pretext for the coup was that the Chatichai cabinet and politicians in general were corrupt.14 The real reason, however, was that the military felt its interests were being undermined under the Chatichai cabinet.15 In the semidemocratic regime of Thailand, as discussed in the Introduction, the military enjoyed significant prerogatives. It is true, however, that these slowly diminished as time passed. A new constitution was adopted in December 1991 and the NPKC chose Anand Punyarachun, a diplomat and a businessman, as prime minister in March. The newly installed Anand cabinet held a general election one year later, in March 1992. In November 1991, when a new constitution was being debated, Suchinda declared he would not accept the office of prime minister.16 In April 1992, after the election, five political parties called on Suchinda to accept the office of prime minister. In spite of his earlier commitment, Suchinda relented and was made prime minister.

Conclusion  167

Suchinda’s acceptance led to the so-called Black May Incident. Shortly after Suchinda had become prime minister, large protests demanding his resignation broke out. Determined to stay in power, Suchinda tried to suppress demonstrators by mobilizing the armed forces and the police. As the confrontation between the government and the protestors escalated, the military and the police fired on demonstrators and reportedly killed several. On May 20 the king intervened and called on the prime minister and an opposition leader to end the confrontation. Suchinda was forced to resign and was succeeded by Anand, the former prime minister. After Black May the constitution of 1991 was amended to make it more democratic. For example, the prime minister now had to be nominated from members of the House of Representatives. Another election was held in September in which the Democrats won a plurality and their leader, Chuan Leekpai, became the new prime minister. The influence of the military in politics was gradually reduced. In short, reequilibration followed the crisis and the semi-democratic regime was sustained. In Black May, the military was contained because of popular protests. The public viewed Suchinda’s accession to the post of prime minister as the military’s attempt to consolidate its political power. Although more extensive studies are necessary, a factor that possibly led the semi-democratic regime of Thailand to a different outcome from the ones in Japan and Brazil is that the people considered the regime legitimate. Massive demonstrations are consistent with a situation in which the public believes in the regime’s legitimacy. Criticisms by mass media and intellectuals of Suchinda’s becoming prime minister as well as the military’s intervention in politics is one piece of evidence suggesting people’s belief in the regime’s legitimacy. For example, the Bangkok Post editorialized that Suchinda’s cabinet lacked legitimacy.17 economic crisis

If the case study of the breakdown of the semi-democratic regime in prewar Japan confirms the importance of legitimacy, the next question is what factors affect the level of regime legitimacy. In Chapter 1, I have argued that a regime’s poor performance is likely to affect its legitimacy. Regarding regime performance, the Japanese case confirms the findings from the study of the breakdown of democratic regimes that economic crisis alone does end regime legitimacy. A close scrutiny of cases of democratic breakdown suggests that political factors must combine with economic crisis to trigger the breakdown of democratic regimes.

  Conclusion

In examining the causes of the collapse of the Weimar Republic, M. Rainer Lepsius acknowledges the influence of the Great Depression on the republic’s fate, but argues that other political factors were necessary “preconditions for the breakdown of democracy.”18 He attaches particular importance to two such preconditions—the disintegration of the party system and presidential rule without parliamentary participation. The former, which paralyzed the Reichstag, and the latter, which was characterized by a reliance on presidential decrees that overtly defied the Reichstag, undermined the credibility of democracy and paved the way for the rise of dictatorship.19 Walter B. Simon draws similar conclusions. Simon, who analyzed the collapse of the First Austrian Republic during the interwar period, does not attribute a deterministic role to the Great Depression, arguing instead that “the economic situation affected the political [sphere] but . . . did not determine it.”20 He suggests that it was the polarization of three major political groups—socialists, pro-clerical conservatives, and German nationalists— that caused the fall of democracy. This polarization paralyzed the Parliament and “created a power vacuum that invited authoritarian intervention.”21 Similarly, in the case of Argentina during the interwar period, Peter H. Smith rejects the view that the Great Depression was the primary cause of the fall of democracy, and argues instead that it was triggered by a combination of the economic crisis with other, political factors.22 Smith contends that Argentina’s democratic regime broke down because the elite conservatives and the military came to consider it illegitimate. This was not solely because of their concern about the economic crisis,23 but also because elite conservatives lost control over politics and the military was outraged in the face of President Yrigoyen’s practice of patronizing officers and his arbitrary exercise of presidential power as demonstrated in intervention through decrees in provincial administration.24 The case of prewar Japan confirms that economic crisis alone does not cause semi-democratic regimes to collapse. What undermined the regime’s legitimacy was the failure of party government to effectively deal with the economic crisis in combination with the fact that its political behavior and that of the political parties began to appear unaccountable to the public. Political scandals, turmoil on the floor of the Diet, backroom political bargaining, and lack of respect for Diet debates all contributed to the regime’s worsening reputation. In other words, legitimacy eroded because poor regime performance in the socioeconomic sphere—the lingering of the eco-

Conclusion  169

nomic crisis—was associated with poor regime performance in the political sphere. It seems unlikely that legitimacy would have eroded if the party government and the political parties had acted in a manner that would have made them appear accountable to the public. Despite the similarities of the Japanese case with other cases of democratic breakdown it is necessary to bear in mind that regime legitimacy is undermined differently in semi-democratic and democratic regimes. In a democratic regime, regime legitimacy is undermined by criticisms from forces that support nondemocratic, in most cases authoritarian, regimes. In a semi-democratic regime, criticism from both democratic and nondemocratic forces contributes to undermining legitimacy. Notably, critics who demand more democratization do not necessarily get it. On the contrary, their actions may actually weaken the democratic forces that sustain the regime, undermine its legitimacy, and, ultimately, contribute to its breakdown. In prewar Japan, although such intellectuals as Yoshino Sakuzō and Ishibashi Tanzan and newspaper editorials criticized the semi-democratic regime for certain nondemocratic practices, they never called for it to be replaced by a military authoritarian regime. Even so, their criticisms undermined the regime’s legitimacy and deprived the party government of the power to contain the military. This paradox arises just as easily in the breakdown of democratic regimes. As this book shows, a number of factors affect regime legitimacy, including nondemocratic practices. When regime legitimacy is lost in a democratic country, it is probably because of nondemocratic elements that remain within the regime or because of nondemocratic governance. This possibility should be further examined by studying the breakdown of democratic regimes in interwar Europe. the importance of time and legitimacy

Past studies of democratization have often emphasized time as an important factor in perpetuating democratic rule. For example, Robert Dahl argues that “the breakdown of democracy is rare” in countries where democratic institutions have survived for nearly twenty years.25 This case study reminds us that time is also a significant factor in the stability of semi-democratic regimes. Why are long years of democratic rule conducive to the perpetuation of democratic regimes? A major reason probably lies in the nature of the relationship between time and legitimacy. The longer the rule under a demo-

  Conclusion

cratic regime the stronger will be the belief of political elites and masses in the legitimacy of the regime. After all, people expect politics to solve social and economic problems. Thus, “both politicians and citizens learn from the successful resolution of some issue to place their faith in the new rules and to apply them to new issues.”26 If democratic regimes can demonstrate good performance, that is, if governments under newly established democratic regimes can find solutions to problems that exist at the time, then such a regime is likely to survive. And if governments can continue to find solutions, then the political elites and masses will come to accept rule under the democratic regime and believe in its legitimacy. Hence, the democratic regime will be perpetuated. A similar process can proceed under a semi-democratic regime. If governments can continue to find solutions to problems facing countries in which a semi-democratic regime has been newly established, the political elites’ and masses’ belief in the legitimacy of the regime will be strengthened. Long years of rule under a semi-democratic regime stabilize and solidify the regime. On that basis, a semi-democratic regime can move to the next stage of democratization by strengthening and expanding democratic institutions while expanding political participation. Unfortunately, the semi-democratic regime in Japan was established under unfavorable conditions. Before it was given enough time—for example, fifteen to twenty years of rule without turmoil in the domestic economy as well as in foreign relations, sufficient to develop regime legitimacy—it was confronted with economic crisis, such as the turmoil in the banking system in 1927 and the Great Depression, as well as upheaval in relations with China. If Japan could have avoided these crises, it might have been possible to gradually develop a custom of appointing the leader of the majority in the Diet as prime minister, thus institutionalizing a parliamentary cabinet system just as Great Britain had done. The semi-democratic regime in Japan, however, simply ran out of time. the probabilistic nature of regime breakdown

Studies have emphasized that the breakdown of democratic regimes is probabilistic. That is, the democratic regimes that breakdown were not destined to do so and “in retrospect, it is possible to identify points at which opportunities existed for alternative courses of action that might have reduced the probability of the fall of the regime.”27

Conclusion  171

The breakdown of semi-democratic regimes is also probabilistic. This was certainly true in the case of Japan in 1932. One could build a counter­factual argument that there were historical moments that might have altered the regime’s course. Opportunities to change direction presented themselves in almost every party government from 1918 to 1932. With the initial support it enjoyed from a wide segment of the population, the Hara cabinet (1918–1921) probably could have introduced universal male suffrage. If this had been done, it might have enabled the cabinets that appeared later—in particular the Katō Takaaki cabinet, which was appointed on the basis of strong electoral support in 1924—to concentrate on curtailing the power of such political institutions as the House of Peers and the Privy Council. Such a step would not only have enhanced the party government’s ability to formulate and implement policies to address society’s problems, but would also have increased its power vis-à-vis the military. Similar chances to curtail the power of these institutions also appeared under the first Wakatsuki cabinet. Namely, the first Wakatsuki cabinet had a chance in 1927 to nullify the Privy Council’s power to affect government policy. When the Privy Council rejected the government’s emergency decree to deal with the financial crisis, the cabinet could have chosen to confront the Privy Council by asking the emperor to decide between the Privy Council and the party government. Given the party government’s indirect mandate from the electorate, it was likely that the emergency decree would have been approved, setting an important precedent that the Privy Council could not obstruct government policy. This missed opportunity was particularly important because it indirectly contributed to the politicization of the military over the signing and ratification of the London Naval Treaty in 1930. The evidence suggests that the Seiyūkai raised the issue of the violation of the supreme command of the emperor in order to attack the Hamaguchi cabinet, expecting that the Privy Council might refuse to approve the ratification of the treaty, thereby triggering the government’s collapse. Had the first Wakatsuki cabinet boldly locked horns with the Privy Council, the Seiyūkai might not have begun to stir up opposition and the regime might have avoided the politicization of the military. The Tanaka cabinet, too, could have countered the politicization of the military. When Chang Tso-lin was assassinated in 1928, it had a chance to show that when the military meddled in politics it would be penalized.

  Conclusion

Saionji and even the emperor favored punishment of the conspirators who had assassinated Chang Tso-lin. Had party politicians maintained party unity and punished the conspirators, the Tanaka cabinet could have established a precedent. However, the party politicians’ failure to do so set precisely the opposite precedent, namely, that the military could get away with arrogant defiance. In 1931, the second Wakatsuki cabinet had a chance to arrest the regime’s further decline either by forestalling the erosion of its legitimacy or constraining the military. It might have prevented the former—even in the face of the continuing economic crisis—by drastically altering the Hamaguchi cabinet’s economic policies (i.e., abandoning the gold standard and a tight fiscal policy). This alone might have helped the economy to recover. The second Wakatsuki cabinet also had at least two chances to contain the military shortly after the Manchurian Incident. It could have revealed to the public that the military triggered the incident through a conspiracy or it could have refused to approve the crossing of the border by the Chōsen Army. While such drastic actions might still have politicized the military, the party government could have recovered its control over the military. Moreover, the Minseitō and the Seiyūkai could have formed a coalition government to contain the military. A coalition government, which would have encompassed more than 90 percent of the members of the Lower House, directly elected from the people, would effectively have represented the whole nation. There were attempts to form a coalition government between the Minseitō and the Seiyūkai, but they came to naught.28 Even at the very last moment, just before the breakdown of the semi-­ democratic regime in 1932, the party government might have been able to avert its fall. After the May Fifteenth Incident, the Seiyūkai and the Minseitō could have cooperated to preserve the practice of party government. They could have turned to the public for support. As I have argued, legitimacy had by this time been lost at both the elite and mass levels. Yet by revealing the series of coercive threats as well as the military’s defiance, and by committing to change their behavior, the political parties might have turned public opinion against the military’s intervention in politics and preserved the practice of party government. Such an attempt was not made, however, and the regime broke down with the emergence of the Saitō cabinet after the May Fifteenth Incident.

Conclusion  173 was it destiny?

To my arguments on the probabilistic nature of the breakdown of Japan’s semi-democratic regime, one might counter that the breakdown was inevitable. Such a deterministic conclusion could be based upon two different structural approaches. In the first, the rise and endurance of a democratic regime is related to the level of economic development. In the second, its rise and endurance is related to a particular political culture. One might argue that democratization was destined to fail because Japan had not yet reached the level of socioeconomic development that enables a country to develop into a full democracy. Indeed, the level of economic development and the likelihood of the emergence and survival of democracy are strongly correlated: “the more well-to-do a nation, the greater the chances that it will sustain democracy.”29 The opposite side of this thematic coin is that the less economically developed a country is, the less likely it will be to develop and sustain democracy. This argument, first proposed by Seymour Martin ­Lipset, is echoed in Larry Diamond’s observation that in virtually every piece of research “[there is] a consistent and strong positive relationship between the level of economic development and democracy.”30 Robert Dahl and Samuel Huntington pinpoint the thresholds above which democracy is likely to develop and be sustained, and below which it is unlikely to emerge and endure. Dahl locates the upper threshold as between $700 and $800 GNP per capita in 1957 U.S. dollars and the lower threshold as between $100 and $200.31 Likewise, Huntington locates the upper threshold at $3,000 GNP per capita and the lower threshold at $1,000 in 1976 U.S. dollars, calling the range between $1,000 and $3,000 a “transition zone.”32 The argument that would ascribe the failure of democratization in prewar Japan to economic development assumes that Japan’s economy at the time was little developed. To examine the validity of this argument, it is first necessary to locate the extent of economic development of prewar Japan on the threshold for democratization. One way to do so is to convert the GNP per capita at the time to present values. I have converted the GNP per capita of Japan in 1918, when the semi-democratic regime emerged; in 1936, when the military authoritarian regime emerged; and in 1940, a date shortly before the war, denominated in 1976 U.S. dollars.33 The GNP per capita in 1918, 1936, and 1940 would be respectively $771, $993, and $1,114. These figures demonstrate that Japan’s level of economic development was below the critical threshold for most of the prewar period. However, this does not invalidate the conten-

  Conclusion

tion that the breakdown of semi-democratic regimes is probabilistic, nor that further democratization was possible. Indeed, though it may be rarer than in countries with higher a GNP per capita, democratization, as present-day data demonstrate, is possible even in countries with a low GNP per capita. According to Huntington, of forty-one countries with a GNP per capita in 1976 of between $200 and $1,000, three were already democratic in 1974 and eleven experienced democratization between 1974 and 1989.34 One might argue that Japan’s failure to democratize was inevitable because the dominant political culture was inimical to its success. Lucian Pye argues that the political culture of Asia, which favors paternalistic power and communal unity, makes it less likely that democracy will develop there.35 In the case of prewar Japan, Robert Scalapino believes that Japan’s political culture at the time, which emphasized the family system and the importance of the group and discouraged respect for individualism, was a major factor in preventing democratization.36 There is a methodological problem in operationalizing political culture as an explanatory variable. It is hard to demonstrate a clear causal relationship between the fall of a regime and a particular form of political culture. The evidence I introduced in the last three chapters of this book runs counter to the simplistic argument that prewar Japanese political culture was utterly averse to democracy. Certainly, some political leaders remained hostile to the semi-democratic regime, believing that it was built on competition among political parties in pursuit of factional interests at the expense of the public good. However, many opinion leaders at the time repeatedly called for greater democracy in the Japanese political system, including better accountability of political parties to the electorate, more debates on policy matters in the Diet, and a reduction in the power of such institutions as the Privy Council and the House of Peers. This suggests that the country’s political culture favored—or at least was not hostile to—democracy among opinion leaders and the masses. Indeed, one major reason the semi-­ democratic regime lost legitimacy was that the party government and the political parties began to appear undemocratic, in that they paid scant attention to the opinions and interests of the public and demonstrated little respect for open debates in the Diet. It is therefore difficult to refute the probabilistic nature of the semi-democratic regime’s breakdown by ascribing it to the level of economic development and political culture.

Notes

Introduction 1.  To list some recent studies on hybrid regimes: Terry Lynn Karl, “The Hybrid Regimes of Central America,” Journal of Democracy 6, no. 3 (1995): 72–86; William Case, “Can the ‘Halfway House’ Stand? Semidemocracy and Elite Theory in Three Southeast Asian Countries,” Comparative Politics 28, no. 4 (1996): 437–464; Larry Diamond, “Thinking about Hybrid Regimes,” Journal of Democracy 13, no. 2 (2002): 21–35; Andreas Schedler, “The Menu of Manipulation,” Journal of Democracy 13, no. 2 (2002): 36–50; Steven Levitsky and Lucan A. Way, “The Rise of Competitive Authoritarianism,” Journal of Democracy 13, no. 2 (2002): 51–65; Nicolas van de Walle, “Africa’s Range of Regimes,” Journal of Democracy 13, no. 2 (2002): 66–80; Wolfgang Merkel, “Embedded and Defective Democracies,” Democratization 11, no. 5. (2004): 33–58; Andreas Schedler, ed., Electoral Authoritarianism (Boulder, CO: Lynne Rienner, 2006); Mikael Wigell, “Mapping ‘Hybrid Regimes’: Regime Types and Concepts in Comparative Politics,” Democratization 15, no. 2 (2008): 230–250; Paul Brooker, Non-Democratic Regimes, 2nd ed. (London: Palgrave Macmillan, 2009); Steven Levitsky and Lucan A. Way, Competitive Authoritarianism (New York: Cambridge University Press, 2010). 2.  Studies on competitive authoritarianism and electoral authoritarianism include: Diamond, “Thinking about Hybrid Regimes”; Schedler, “The Menu of Manipulation”; Levitsky and Way, “The Rise of Competitive Authoritarianism”; Schedler, ed., Electoral Authoritarianism; Levitsky and Way, Competitive Authoritarianism. 3.  Levitsky and Way and Schedler are aware of this. Levitsky and Way, Competitive Authoritarianism, 14; Schedler, “Introduction” in idem, ed., Electoral Authoritarianism, 4–5. 175

176  Notes to Introduction 4.  Levitsky and Way, Competitive Authoritarianism, 3–4. A vast literature exists on democratization. See, for example, Guillermo O’Donnell and Philippe Schmitter, Transitions from Authoritarian Regimes: Tentative Conclusions about Uncertain Democracies (Baltimore: Johns Hopkins University Press, 1985); Samuel Huntington, The Third Wave (Norman: University of Oklahoma Press, 1991); Juan Linz and Alfred Stepan, Problems of Democratic Transition and Consolidation: Southern Europe, South America, and Post-Communist Europe (Baltimore: Johns Hopkins University Press, 1996); Stephen Haggard and Robert Kaufman, The Political Economy of Democratic Transitions (Princeton, NJ: Princeton University Press, 1995); Larry Diamond, Developing Democracy (Baltimore: Johns Hopkins University Press, 1999); Richard Gunther, P. Nikiforos Diamandouros, and Hans-Jürgen Puhle, The Politics of Democratic Consolidation: Southern Europe in Comparative Perspective (Baltimore: Johns Hopkins University Press, 1995). 5.  O'Donnell and Schmitter, Transitions from Authoritarian Rule. 6.  Thomas Carothers, “The End of the Transition Paradigm” Journal of Democracy 13, no. 1 (2002): 5–21. 7.  Juan Linz explicitly makes this assumption. See Juan Linz “Transitions to Democracy,” Washington Quarterly 13, no. 3 (1990): 143–144. 8. Huntington, The Third Wave, 16–17. 9.  Linz, “Transitions to Democracy,” 143. 10. Diamond, Developing Democracy, 15–23, 34–42, 42–50. 11.  Diamond, “Thinking about Hybrid Regimes,” 25–26. 12.  Levitsky and Way, Competitive Authoritarianism, 4–5; Schedler, “The Logic of Electoral Authoritarianism,” in idem, ed., Electoral Authoritarianism, 3. 13.  Levitsky and Way, Competitive Authoritarianism, 6–10, 21. 14.  Robert Dahl, Polyarchy: Participation and Opposition (New Haven, CT: Yale University Press, 1971), 7–8. For a review on how scholars have treated the question of hybrid regimes, see Diamond, “Thinking about Hybrid Regimes,” 24–25. 15.  See Note 1 above. 16.  Larry Diamond articulates nuanced differences between electoral authoritarianism and competitive authoritarianism in Diamond, “Thinking about Hybrid Regimes.” 17.  Robert Dahl uses the term “near-polyarchy.” See Dahl, Polyarchy, 8. 18.  Tetsuo Najita, Hara Kei in the Politics of Compromise, 1905–1915 (Cambridge, MA: Harvard University Press, 1967). 19.  Peter Duus, Party Rivalry and Political Change in Taishō Japan (Cambridge, MA: Harvard University Press, 1968). 20.  Murai Ryōta, Seitō Naikakusei no Seiritsu (Tokyo: Yūhikaku, 2005). 21.  Kitaoka Shin’ichi, Seitō kara Gunbu e 1924–41, vol. 5 of Nihon no Kindai (Tokyo: Chūō Kōron Shinsha, 1999).

Notes to Introduction  177 22.  Gordon Berger, Parties Out of Power in Japan, 1931–1941 (Princeton NJ: Princeton University Press, 1977). 23.  In addition to these scholars, there are others, such as James B. Crowley and John Toland, who have studied Japanese foreign policy from late 1920s and 1930s. They refer to developments in domestic politics; the focus of their research, however, is on Japanese foreign policy and they do not make a causal analysis of the events that led to the collapse of the semi-democratic regime. Also, the period they discuss is after the late 1920s. See James B. Crowley, Japan’s Quest for Autonomy (Princeton, NJ: Princeton University Press, 1966); and John Toland, The Rising Sun (New York: Random House, 1970). 24.  Banno Junji, Kindai Nihon Seijishi (Tokyo: Iwanami Shoten, 2006); Banno Junji, Nihon Kensei Shi (Tokyo: Tōkyō Daigaku Shuppankai, 2008). 25.  Andrew Gordon, Labor and Imperial Democracy in Prewar Japan (Berkeley: University of California Press, 1991). 26.  Sheldon Garon, The State and Labor in Modern Japan (Berkeley: University of California Press, 1987). 27.  Gregory Kasza, The State and Mass Media in Japan, 1918–1945 (Berkeley: University of California Press, 1988). 28.  Gregory Kasza, The Conscription Society: Administered Mass Organization (New Haven, CT: Yale University Press, 1995). 29.  Richard J. Samuels, Machiavelli’s Children: Leaders and Their Legacies in Italy and Japan (Ithaca, NY: Cornell University Press, 2003). 30.  Maruyama Masao, “The Ideology and Dynamics of Japanese Fascism,” in idem, Thought and Behaviour in Modern Japanese Politics, ed. Ivan Morris (London: Oxford University Press, 1963), 25–83; Maruyama “Thought and Behaviour Patterns of Japan’s Wartime Leaders,” in Thought and Behaviour in Modern Japanese Politics, 84–134. 31.  Robert Scalapino, Democracy and the Party Movement in Prewar Japan (Berkeley: University of California Press, 1953). 32. Ibid., 393–394. 33.  On the objectives of social science and theory, see Adam Przeworski and Henry Teune, The Logic of Comparative Social Inquiry (New York: Wiley-Interscience, 1970), 1–23; and Kenneth Waltz, Theory of International Politics (New York: McGraw Hill, 1979), 1–17. 34.  Przeworski and Teune, Logic of Comparative Social Inquiry, 20–23. 35. Ibid., 20–21. 36.  Schedler, “Logic of Electoral Authoritarianism,” 3. 37.  Levitsky and Way, Competitive Authoritarianism, 5. 38.  Larry Diamond describes the differences. Diamond, “Thinking about Hybrid Regimes.”

178  Notes to Introduction 39.  Larry Diamond, Juan J. Linz, and Seymour Martin Lipset, “Introduction,” in Politics in Developing Countries, ed. Diamond, Linz, and Lipset (Boulder, CO: Lynne Rienner, 1995), 7–8. 40.  Levitsky and Way, Competitive Authoritarianism, 14. 41. Dahl, Polyarchy, 3. 42. Ibid., 1. 43.  Philippe C. Schmitter and Terry Lynn Karl, “What Democracy Is . . . and Is Not,” Journal of Democracy 2, no. 3 (Summer 1991): 81; Diamond, Developing Democracy, 9. 44. Diamond, Developing Democracy, 8–13. 45.  William Case, Politics in Southeast Asia: Democracy or Less (Richmond, UK: Curson, 2002), 6–7. 46. Brooker, Non-Democratic Regimes, 23. 47.  Samuel Fitch, The Armed Forces and Democracy in Latin America (Baltimore: Johns Hopkins University Press, 1998), 38–39. See also Adam Przeworski, “Democracy as a Contingent Outcome of Conflicts,” in Constitutionalism and Democracy, ed. Jon Elster and Rune Slagstad (Cambridge: Cambridge University Press, 1988), 60–61. 48.  J. Samuel Valenzuela, “Consolidation in Post-Transitional Settings,” in Issues in Democratic Consolidation, ed. Scott Mainwaring, Guillermo O’Donnell, and J. Samuel Valenzuela (Notre Dame, IN: University of Notre Dame Press, 1992), 64–65. 49.  Concise Oxford Dictionary of Current English. 50.  Valenzuela, “Consolidation in Post-Transitional Settings,” 62–63. 51. Dahl, Polyarchy, 7. 52.  Juan Linz, “An Authoritarian Regime: Spain,” in Mass Politics, ed. Erik Allardt and Stein Rokkan (New York: Free Press, 1970), 255. 53. Ibid., 251–283. See also Juan Linz, “Totalitarian and Authoritarian Regimes,” in Handbook of Political Science, vol. 3, ed. Fred I. Greenstein and Nelson W. Polsby (Reading, MA: Addison-Wesley, 1975), 264–274. 54.  Linz, “Totalitarian and Authoritarian Regimes,” 267. 55. Ibid., 268. 56.  R. K. Webb, Modern England: From the Eighteenth Century to the Present, 2nd ed. (New York: Harper Collins, 1980), 421–423; Martin Pugh, The Making of Modern British Politics, 1867–1939 (Oxford: Basil Blackwell, 1982), 9–13; H. J. Hanham, Elections and Party Management: Politics in the Time of Disraeli and Gladstone (London: Longmans, Green, 1959), 56–67. 57. Hanham, Elections and Party Management, 63. 58. Pugh, Making of Modern British Politics, 13. 59. Webb, Modern England, 347. 60.  Chris Cook and John Stevenson, The Longman Handbook of Modern British History, 1714–1995, 3rd ed. (London: Longman, 1996), 89–90.

Notes to Introduction  179 61.  Thomas Skidmore, Brazil: Five Centuries of Change (Oxford: Oxford University Press, 2010), 7–8. 62.  Austin F. MacDonald, Latin American Politics and Government (New York: Thomas Y. Cromwell, 1954), 170–172; Thomas Skidmore, Politics in Brazil (Oxford: Oxford University Press, 1967), 75. 63.  Phyllis Peterson, “Brazil: Institutionalized Confusion,” in Political Systems of Latin America, ed. Martin C. Needler (Princeton, NJ: Van Nostrand, 1964), 478–479. 64. Fitch, Armed Forces and Democracy in Latin America, 17. 65.  For political developments during this period, see Chai-Anan Samudavanija, “Thailand: A Stable Semidemocracy,” in Diamond, Linz, and Lipset, eds., Politics in Developing Countries, 323–367. 66.  Surin Maisrikrod and Duncan McCargo, “Electoral Politics: Commercialization and Exclusion,” in Political Change in Thailand, ed. Kevin Hewison (London: Routledge, 1997), 132–148; William A. Callahan and Duncan McCargo, “Vote-Buying in Thailand’s Northeast,” Asian Survey 36, no. 4 (1996): 376–392. 67.  Callahan and McCargo, “Vote-Buying in Thailand’s Northeast,” 382–382. 68.  Maisrikrod and McCargo, “Electoral Politics,” 140. 69.  Paul M. Handley, The King Never Smiles: A Biography of Thailand’s Bhumibol Adulyadej (New Haven, CT: Yale University Press, 2005), 277–278. 70.  Chai-Anan, “Thailand: A Stable Semidemocracy,” 339. 71.  Kevin Hewison, “The Monarchy and Democratization,” in idem, ed., Political Change in Thailand, 72–73. 72.  Larry A, Nikisch, “Thailand in 1981: The Prem Government Feels the Heat,” Asian Survey 22, no. 2 (1982): 194. 73.  Suchitra Punyaratabandhu-Bhakdi, “Thailand in 1983: Democracy Thai Style,” Asian Survey 24, no. 2 (1984): 187. 74. Handley, King Never Smiles, 284. 75.  Clark D. Neher, “Thailand in 1986: Prem, Parliament, and Political Pragmatism,” Asian Survey 27, no. 2 (1987): 227; Clark D. Neher, “Thailand in 1987: Semi-Successful Semi-Democracy,” Asian Survey 28, no. 2 (1988): 195–197. 76.  Pasuk Phongpaichit and Chris Baker, Thailand: Economy and Politics, 2nd ed. (Oxford: Oxford University Press), 417. 77. Ibid., 417. 78.  James Ockey, “Thailand: The Struggle to Redefine Civil-Military Relations,” in Coercion and Governance, ed. Muthiah Alagappa (Stanford, CA: Stanford University Press, 2001), 199; Surachart Bamrungsuk, “Changing Patterns of Civil-Military Relations and Thailand’s Regional Outlook,” in Civil Military Relations, ed. David Mares (Boulder, CO: Westview, 1998), 193.

180  Notes to Chapter 1

Chapter 1 1.  Steven Levitsky and Lucan A. Way, Competitive Authoritarianism: Hybrid Regimes after the Cold War (Cambridge: Cambridge University Press, 2010). 2.  Seymour Martin Lipset, Political Man, exp. ed. (Baltimore: Johns Hopkins University Press, 1981), 31. 3. Ibid., 33. 4.  Robert Dahl, Polyarchy: Participation and Opposition (New Haven, CT: Yale University Press, 1971), 65. 5. Ibid., 68–71. 6.  Samuel Huntington, The Third Wave (Norman: University of Oklahoma Press, 1991), 63. 7. Ibid., 62–63. 8.  Larry Diamond, “Economic Development and Democracy Reconsidered,” American Behavioral Scientist 35 (March–June 1992): 468. 9.  Gabriel Almond and Sidney Verba, The Civic Culture: Political Attitudes and Democracy in Five Nations (Newbury Park, CA: Sage, 1989), 12. 10. Ibid., 6. 11. Ibid., 30. 12.  Larry Diamond, “Introduction: Political Culture and Democracy,” in Political Culture and Democracy in Developing Countries, ed. Larry Diamond (Boulder, CO: Lynne Rienner, 1993), 10. 13.  Lucian Pye, Asian Power and Politics: The Cultural Dimensions of Authority (Cambridge, MA: Belknap Press, 1985), 342. 14.  For South Korea, see Larry Diamond and Byung-Kook Kim, eds., Consolidating Democracy in South Korea (Boulder, CO: Lynne Rienner, 2000); for Taiwan, see Yun-Han Chu, Crafting Democratization in Taiwan (Taipei: Institute for National Policy Research, 1992). 15.  Samuel Huntington, Political Order in Changing Societies (New Haven, CT: Yale University Press, 1968). 16. Ibid., 55. 17. Ibid., 78–81. 18. Ibid., 80. 19.  In oligarchical praetorianism, landowners and the clergy are major political forces based in rural areas and ruling classes play multiple roles such as political, economic, religious and so on. See ibid., 199. Radical praetorianism is characterized by expansion of political participation and instability in cities. Major political players are students and the military. See ibid., 208–210. In mass praetorianism, political participation has significantly expanded and the military intervenes in politics to protect the interests of the middle class against the urban lower class.

Notes to Chapter 1  181 See ibid., 219–222. For the role of the military in respective praetorianisms, see also ibid., 201, 213, 221–222. 20.  In recent years, political scientists have paid greater attention to the impact of political institutions on political processes and political behavior. There is a vast literature on the role of political institutions in politics. For leading examples, see James E. Alt and Kenneth Shepsle, eds., Perspectives on Positive Political Economy (Cambridge: Cambridge University Press, 1990); and Sven Steinmo, Kathleen Thelen, and Frank Longstreth, eds., Structuring Politics: Historical Institutionalism in Comparative Analysis (Cambridge: Cambridge University Press, 1992). 21.  Juan Linz, Crisis, Breakdown, and Reequilibration: The Breakdown of Demo­ cratic Regimes (Baltimore: Johns Hopkins University Press, 1978); Juan Linz and Alfred Stepan, eds., Europe, The Breakdown of Democratic Regimes (Baltimore: Johns Hopkins University Press, 1978); and Juan Linz and Alfred Stepan, eds., Latin America, The Breakdown of Democratic Regimes (Baltimore: Johns Hopkins University Press, 1978). 22.  See Linz, Crisis, Breakdown, and Reequilibration. 23.  Juan Linz and Alfred Stepan, “Editors’ Preface and Acknowledgments,” in Linz, Crisis, Breakdown, and Reequilibration, xi. 24. Linz, Crisis, Breakdown, and Reequilibration, 32. 25.  Seymour Martin Lipset, “The Social Requisites of Democracy Revisited,” American Sociological Review 59, no. 2 (February 1994): 8. 26.  Max Weber, Economy and Society, ed. Guenther Roth and Claus Wittich (New York: Bedminster, 1968), 213. 27. Ibid., 213–216. 28.  Juan Linz, “Legitimacy of Democracy and the Socioeconomic System,” in Comparing Pluralist Democracies, ed. Mattei Dogan (Boulder, CO: Westview, 1988), 65. 29. Linz, Crisis, Breakdown, and Reequilibration, 16. He uses the same definition in “Legitimacy of Democracy and the Socioeconomic System” and calls it “a minimalist definition.” See Linz, “Legitimacy of Democracy and the Socioeconomic System,” 65. 30. Lipset, Political Man, 64. 31.  Richard M. Merelman, “Learning and Legitimacy,” American Political Science Review 60, no. 3 (1966): 548. 32.  On the role of legitimacy in the breakdown of democracies, see Linz, ­Crisis, Breakdown, and Reequilibration. A vast literature has dealt with its role in the consolidation of democracy. See, for example, Larry Diamond, Juan Linz, and Seymour Martin Lipset, eds., Politics in Developing Countries (Boulder, CO: Lynne Rienner, 1995); and Peter McDonough, Samuel H. Barnes, and Antonio López Pina, “The Growth of Democratic Legitimacy in Spain,” American Political Science Review 80, no. 3 (1986): 735–760.

182  Notes to Chapter 1 33.  Peter H. Merkl, “Comparing Legitimacy and Values,” in Dogan, ed., Comparing Pluralist Democracies, 21. 34.  Leonardo Morlino and José R. Montero, “Legitimacy and Democracy in Southern Europe,” in The Politics of Democratic Consolidation: Southern Europe in Comparative Perspective, ed. Richard Gunther, P. Nikiforos Diamandouros, and Hans-Jürgen Puhle (Baltimore: Johns Hopkins University Press, 1995), 232. 35. Lipset, Political Man, 64. 36.  McDonough, Barnes, and Lopez, “The Growth of Democratic Legitimacy in Spain”; and Morlino and Montero, “Legitimacy and Democracy in Southern Europe.” 37. Dahl, Polyarchy, 131. 38. Linz, Crisis, Breakdown, and Reequilibration, 19. 39.  Larry Diamond, Juan J. Linz, and Seymour Martin Lipset, “Introduction,” in Diamond, Linz, and Lipset, eds., Politics in Developing Countries, 9 (emphasis added). 40. Dahl, Polyarchy, 131. 41.  Diamond, Linz, and Lipset, “Introduction,” 10. 42. Ibid. 43.  Linz, “Legitimacy of Democracy and the Socioeconomic System,” 91. 44.  Morlino and Montero, “Legitimacy and Democracy in Southern Europe,” 245. Diamond, Linz, and Lipset also argue that the firm rejection of an authoritarian regime prevented economic problems from triggering political crises in Spain. See Diamond, Linz, and Lipset, Politics in Developing Countries, 11. 45.  Juan Linz and Alfred Stepan, Problems of Democratic Transition and Consolidation: South Europe, South America, and Post-Communist Europe (Baltimore: Johns Hopkins University Press, 1996), 80–81. 46. Linz, Crisis, Breakdown, and Reequilibration, 20, 22. 47.  Larry Diamond argues that regime performance should be assessed not only in terms of economic growth and social reform, but also along political dimensions. See Diamond, “Introduction: Political Culture and Democracy,” 13. 48. Lipset, Political Man, 64–66. See also Diamond, Linz, and Lipset, “Introduction,” 11. 49.  Larry Diamond, Developing Democracy (Baltimore: Johns Hopkins University Press, 1999), 80. 50.  Morlino and Montero, “Legitimacy and Democracy in Southern Europe,” 233. See also Ulrich Widmaier, “Tendencies toward an Erosion of Legitimacy,” in Dogan, ed., Comparing Pluralist Democracies, 144. M. Stephen Weatherford criticizes the current level of research on legitimacy, claiming that the available method for measuring legitimacy is determining how theory is elaborated instead of theory determining the way to do research and measure legitimacy. See M. Stephen

Notes to Chapters 1 and 2  183 Weatherford, “Measuring Legitimacy,” American Political Science Review 86, no. 1 (1992): 149–166. 51.  McDonough, Barnes, and López Pina, “The Growth of Democratic Legitimacy in Spain,” 740; Merkl, “Comparing Legitimacy and Values,” 23–24. 52.  For examples of research measuring legitimacy through opinion polls, see Morlino and Montero, “Legitimacy and Democracy in Southern Europe”; and McDonough, Barnes, and López Pina, “The Growth of Democratic Legitimacy in Spain.” 53. Linz, Crisis, Breakdown, and Reequilibration, 32–34. 54. Ibid., 33. 55. Ibid., 32. 56. Ibid., 33. 57.  Imre Lakatos, “Falsification and the Methodology of Scientific Research Programmes,” in Criticism and the Growth of Knowledge, ed. Imre Lakatos and Alan Musgrave (Cambridge: Cambridge University Press, 1970): 91–196.

Chapter 2 1.  For a detailed description of the nature of political competition during this period, see Banno Junji, Taishō Seihen (Kyoto: Mineruba Shobō, 1994). 2.  Itō Yukio, “Nichiro Sensō e no Seiji Katei,” in Nihon Kindai Kokka no Keisei to Hatten, ed. Yamamoto Shirō (Tokyo: Yoshikawa Kōbunkan, 1996), 252–253; Masuda Tomoko, “Nichiro Kyoshō to Nichibei Domei,” in Meiji Kenpō Taisei no Seiritsu, vol. 1, ed. Ōkubo Toshiaki, vol. 14 of Nihon Rekishi Taikei, ed. Inoue Mitsusada et al. (Tokyo: Yamakawa Shuppansha, 1996), 323–328. 3.  Competition between the Seiyūkai and Katsura is a point Banno stresses throughout his analysis of this period in his Taishō Seihen. See Banno, Taishō Seihen, 96–97. 4.  Tokinoya Masaru, ed., Nihon Kindaishi Jiten (Tokyo: Tōyō Keizai Shimpōsha, 1958), 766; Umemura Mataji et al., Rōdōryoku, vol. 2 of Chōki Keizai Tōkei, ed. Ōkawa Kazushi, Shinohara Miyohei, and Umemura Mataji (Tokyo: Tōyō Keizai Shimpōsha, 1988), 166–169. 5.  See Note 4 above. 6.  Miyazaki Ryūji uses the terms “representative forces” and “non-representative forces” to make a distinction between institutions that were under electoral control and those that were not. See Miyazaki Ryūji, “Senzen Nihon no Seiji Hatten to Rengō Seiji,” in Rengō Seiji, vol. 1, ed. Shinohara Hajime (Tokyo: Iwanami Shoten, 1984), 198–200. 7.  For example, Juan Linz, “Opposition In and Under an Authoritarian Regime: The Case of Spain,” in Regimes and Oppositions, ed. Robert Dahl (New Haven, CT:

184  Notes to Chapter 2 Yale University Press, 1973), 171; Richard Gunther, Public Policy in a No-Party State (Berkeley: University of California Press, 1980), 32. 8. Gunther, Public Policy in a No-Party State, 39. 9.  Ibid., citing Ley Constitutiva de las Cortes, art. 13. 10.  Juan Linz, “From Falange to Movimiento-Organización: The Spanish Single Party and the Franco Regime, 1936–1968,” in Authoritarian Politics in Modern Society, ed. Samuel P. Huntington and Clement H. Moore (New York: Basic Books, 1970), 169. 11. Gunther, Public Policy in a No-Party State, 39. 12. Ibid., 32. 13.  Richard Gunther, Spain after Franco: The Making of a Competitive Party System (Berkeley: University of California Press, 1986), 24. 14. Gunther, Public Policy in a No-Party State, 40. 15.  Banno Junji, Meiji Kenpō Taisei no Kakuritsu (Tokyo: Tōkyō Daigaku Shuppankai, 1971), 147–151. 16. Ibid., 21–24; Rekishigaku Kenkyūkai, ed., Kindai, vol. 4 of Nihonshi Shiryō (Tokyo: Iwanami Shoten, 1997), 278–279. 17.  Masumi Junnosuke, Nihon Seitōshi Ron, vol. 5 (Tokyo: Tōkyō Daigaku Shuppankai, 1979), 299. 18. Ibid., 235. 19.  On the policy options they advocated, see Awaya Kentarō, Shōwa no Seitō, vol. 6 of Shōwa no Rekishi (Tokyo: Shōgakukkan, 1988), 105–109. 20.  Peter Duus, Party Rivalry and Political Change in Taishō Japan (Cambridge, MA: Harvard University Press, 1968), 154–161. 21.  Kawato Sadafumi, Nihon no Seitō Seiji, 1890–1937 (Tokyo: Tōkyō Daigaku Shuppankai, 1992), 267–271. 22.  On autonomous institutions under the Meiji Constitution, see Mitani Taichirō, Nihon Seitō Seiji no Tenkai, exp. ed. (Tokyo: Tōkyō Daigaku Shuppankai, 1995), 4–6. 23.  Centre for East Asian Cultural Studies, The Meiji Japan through Contemporary Sources, vol. 1. (Tokyo: Centre for East Asia Cultural Studies, 1969), 101. 24.  Masuda Tomoko, “Seitō Naikaku to Sūmitsuin,” in Seitō Naikaku no Seiritsu to Hōkai, ed. Kindai Nihon Kenkyūkai (Tokyo: Yamakawa Shuppansha, 1984), 143. 25.  On the prerogatives enjoyed by the military, see Awaya, Shōwa no Seitō, 91–94. See also, Ōe Shinobu, Tennō no Guntai, vol. 3 of Shōwa no Rekishi (Tokyo: Shōgakukkan, 1982), 139–144. 26.  Centre for East Asian Cultural Studies, ed., The Meiji Japan through Contemporary Sources, vol. 1, 96. 27.  Ibid., vol. 3 (Tokyo: Centre for East Asia Cultural Studies, 1972), 215. 28.  Naikaku Tōkeikyoku, ed., Nihon Teikoku Tōkei Nenkan, vol. 48 (Tokyo:

Notes to Chapter 2  185 Naikaku Tōkeikyoku, 1929), 390. The percentage of people with the franchise is calculated from data in Umemura et al., Rōdōryoku, 166–171. 29.  On why it is problematic to label this regime fascist, see Itō Takashi, “Fashizumu Ronsō sonogo,” in Kindai Nihon Kenkyū no Kentō to Kadai, ed. Kindai Nihon Kenkyūkai (Tokyo: Yamakawa Shuppansha, 1988), 310–323; Itō Takashi, “Shōwa Seijishi Kenkyū e no Ichi Shikaku,” Shisō, no. 624 (1976): 949–962; Peter Duus and Daniel I. Okimoto, “Fascism and the History of Pre-War Japan: The Failure of a Concept,” Journal of Asian Studies 39, no. 1 (1979): 65–76. 30. Awaya, Shōwa no Seitō, 303–304. 31.  Inumaru Giichi, “Han Fashizumu Undō to sono Kaitai,” in Iwanami Kōza: Nihon Rekishi Kindai, vol. 20, ed. Asao Naohiro et al. (Tokyo: Iwanami Shoten, 1976), 172–173, 182–184. 32.  Miyamoto Moritarō, “Hirota Naikaku: Jun-Senji Taisei no Kakuritsu,” in Nihon Naikaku Shiroku, vol. 3, ed. Hayashi Shigeru and Tsuji Kiyoaki (Tokyo: Daiichi Hōki Shuppan, 1981), 415–416. 33. Awaya, Shōwa no Seitō, 358–366. 34. Ibid., 315–319. 35. Ibid., 316, 318. 36.  Imperial Rule Assistance Association is the translation used by Gordon Berger in his Parties Out of Power in Japan, 1931–1941 (Princeton, NJ: Princeton University Press, 1977). On the formation of the Imperial Rule Assistance Association, see Kisaka Jun’ichirō, “Taisei Yokusankai no Seiritsu,” in Asao et al., eds, Iwanami Kōza, vol. 20, 269–314. 37.  Kisaka Jun’ichirō, “Taisei Yokusankai no Seiritsu,” 297. 38. Awaya, Shōwa no Seitō, 371–373. 39.  Ibid. See also Miyake, “Dainiji Konoe Naikaku,” 241–245. 40.  As the last genrō, Saionji, became too old to continue recommending the next prime ministers to the emperor, the few leading aristocrats who had occupied such offices as prime minister and chairman of the Privy Council collectively took over this function. These leading aristocrats were called jushin. 41.  On how the efforts by the army failed, see Mikuriya Takashi, “Kokusaku Tōgō Kikan Secchi Mondai no Shiteki Tenkai,” in Shōwaki no Gunbu, ed. Kindai Nihon Kenkyūkai (Tokyo: Yamakawa Shuppansha, 1979), 122–172. 42.  Hata Ikuhiko, Gun Fashizumu Undōshi (Tokyo: Kawaide Shobō, 1962), 173. 43.  Katō Yoko, “Ni Ni Roku Jiken to Gunbu no Seijiteki Taito,” in Kakushin to Sensō no Jidai, ed. Ōkubo Toshiaki, vol. 17 of Nihon Rekishi Taikei, ed. Inoue Mitsusada et al. (Tokyo: Yamakawa Shuppan, 1997), 109. 44.  Iokibe Makoto, “Rikugun ni yoru Seiji Shihai,” in Tairiku Shinkō to Senji Taisei, vol. 2 of Shōwashi no Gunbu to Seiji, ed. Miyake Masaki et al. (Tokyo: Daiichi Hōki Shuppan, 1982), 34–35.

186  Notes to Chapter 2 45.  Takeyama Morio, “Yonai Naikaku: Ishoku no Dai Yonji Senji Naikaku,” in Nihon Naikaku Shiroku, vol. 4, ed. Hayashi Shigeru and Tsuji Kiyoaki (Tokyo: Daiichi Hōki Shuppan, 1981), 194–97. 46.  Itō Takashi, “Konoe Fumimaro no Tōjō,” in Ōkubo, ed., Kakushin to Sensō no Jidai, 123. 47.  Hara Akira, “Shigenkyoku to Naikaku Chōsakyoku,” in Ōkubo, ed., Kakushin to Sensō no Jidai, 134–135. 48.  Itō Takashi, “Konoe Naikaku no Seiritsu to Matsuoka no Tojō,” in Ōkubo, ed., Kakushin to Sensō no Jidai, 196–199; Miyake Masaki, “Dainiji Konoe Naikaku: Gensō to Zasetsu,” in Hayashi and Tsuji, eds., Nihon Naikaku Shiroku, vol. 4, 213–217. 49.  Throughout his Parties Out of Power in Japan, Gordon Berger emphasizes that political parties and the Diet retained political power. See Berger, Parties Out of Power in Japan, 294. 50. Awaya, Shōwa no Seitō, 297–298. 51. Ibid., 298. 52. Ibid., 307. 53. Ibid. 54. Ibid., 305. 55.  Ōkurashō, ed., Saikei, vol. 3. of Shōwa Zaisei Shi (Tokyo: Tōyō Keizai Shimpōsha, 1955), 314. 56.  Hara Akira, “Senji Tōsei Keizai no Tenkai,” in Ōkubo, ed., Kakushin to Sensō no Jidai, 140–141; Awaya, Shōwa no Seitō, 307–308. 57. Berger, Parties Out of Power in Japan, 157. 58.  Hara, “Senji Tōsei Keizai,” 140–141; Awaya, Shōwa no Seitō, 308. 59. Berger, Parties Out of Power, 294. 60.  Masumi Junnosuke, Nihon Seijishi, vol. 2 (Tokyo: Tōkyō Daigaku Shuppankai, 1988), 140. 61.  Muroyama Yoshimasa, “Jōi Tōgō to Taishō Demokurashii,” in Meiji Kenpō Taisei no Tenkai, vol. 2, ed. Ōkubo Toshiaki, vol. 15 of Nihon Rekishi Taikei, ed. Inoue Mitsusada et al. (Tokyo: Yamakawa Shuppan, 1996), 70. 62.  R. K. Webb, Modern England, 2nd ed. (New York: Harper Collins, 1980), 49. 63. Ibid. 64.  Hugh Ross Williamson, “The Evolution of Parties and the Party System: The Seventeenth Century,” in Political Parties and the Party System in Britain, ed. Sydney D. Bailey (Westport, CT: Hyperion, 1979), 23. 65. Webb, Modern England, 217. 66.  Eric Evans, The Forging of the Modern State: Early Industrial Britain, 1783– 1870, 2nd ed. (London: Longman, 1996), 257–258. 67. Ibid., 261.

Notes to Chapters 2 and 3  187 68.  John A. Hawgood, “The Evolution of Parties and the Party System: The Nineteenth Century,” in Bailey, ed., Political Parties and the Party System in Britain, 27, 32; Alan R. Ball, British Political Parties (London: Macmillan, 1981), 13–14. 69.  Hawgood, “Evolution of Parties and the Party System,” 32; Alan Beattie, English Party Politics, vol. 1 (London: Weidenfeld and Nicolson, 1970), 139. 70.  Hawgood, “Evolution of Parties and the Party System,” 34; Beattie, English Party Politics, 140–145. 71.  Samuel Huntington, The Solider and the State (Cambridge, MA: Harvard University Press, 1957), 187. 72. Webb, Modern England, 347. 73. Ibid., 216. 74. Ibid., 465. 75. Ibid. 76.  Chris Cook and John Stevenson, The Longman Handbook of Modern British History, 1714–1995, 3rd ed. (London: Longman, 1996), 89–90. 77. Webb, Modern England, 421–423; Martin Pugh, The Making of Modern British Politics, 1867–1939 (Oxford: Basil Blackwell, 1982), 9–13; H. J. Hanham, Elections and Party Management (London: Longmans, Green 1959), 56–67. 78.  Robert Dahl, Polyarchy: Participation and Opposition (New Haven, CT: Yale University Press, 1971), 4. 79.  Masumi Junnosuke, Nihon Seitōshi Ron, vol. 5 (Tokyo: Tōkyō Daigaku Shuppankai, 1979), 235. 80.  Shinobu Seizaburō, Taishō Demokurashii Shi, vol. 3 (Tokyo: Nihon Hyōron Shinsha, 1959), 809–811. 81.  Itō Takashi, “Chukan Naikaku to Seitō Seiji,” chap. 3 in Daiichiji Sekai Taisen to Seitō Naikaku, ed. Ōkubo Toshiaki, vol. 16 of Nihon Rekishi Taikei, ed. Inoue Mitsusada et al. (Tokyo: Yamakawa Shuppan, 1996), 260–262. 82.  Igarashi Takeshi points this out and suggests that this was a crucial factor in postwar Japanese democratization. Igarashi Takeshi, interview by author, Tokyo, Japan, June 1997. 83.  Felipe Agüero, “Democratic Consolidation and the Military in Southern Europe and South America,” in The Politics of Democratic Consolidation, ed. Richard Gunther, P. Nikiforos Diamandouros, and Hans-Jürgen Puhle (Baltimore: Johns Hopkins University Press, 1995), 146. 84. Ibid.

Chapter 3 1.  Awaya Kentarō, Shōwa no Seitō, vol. 6 of Shōwa no Rekishi (Tokyo: Shōga­ kukkan, 1988), 51. 2. Ibid.

188  Notes to Chapter 3 3.  On the collapse of the second Wakatsuki cabinet, see Miyamoto Moritarō, “Dainiji Wakatsuki Naikaku: Manshū Jihen to Minseitō Naikaku no Kunō,” in Nihon Naikaku Shiroku, vol. 3, ed. Hayashi Shigeru and Tsuji Kiyoaki (Tokyo: Daiichi Hōki Shuppan, 1981), 243–247. 4.  On the fragmentation and lack of coordination of political power under the Meiji Constitution, see Mitani Taichirō, Nihon Seitō Seiji no Tenkai, exp. ed. (Tokyo: Tōkyō Daigaku Shuppankai, 1995), 4–6. See also Miyazaki Ryūji, “Senzen Nihon no Seiji Hatten to Rengō Seiji,” in Rengō Seiji, vol. 1, ed. Shinohara Hajime (Tokyo: Iwanami Shoten, 1984), 197–199; and Awaya, Shōwa no Seitō, 49–52. 5.  Mitani and Miyazaki both emphasize the role of political parties in coordinating various political institutions to increase the power of political parties. See Mitani, Nihon Seitō Seiji no Tenkai, 7–8; and Miyazaki, “Senzen Nihon no Seiji Hatten to Rengō Seiji,” 201, 240. 6.  Satogami Ryōhei, “Taishō Demokurashii to Kizokuin,” in Taishōki no Seiji to Shakai, ed. Inoue Kiyoshi (Tokyo: Iwanami Shoten, 1969), 278. See also Miyazaki, “Senzen Nihon no Seiji Hatten to Rengō Seiji,” 240–241. 7.  The Labor Accident Assistance Law was designed to provide financial compensation for workplace injury or death. 8.  Tōkyō Asahi Shimbun, March 26, 1929. 9. Article 1, Regulations on the Privy Council (translated by the author). 10.  Masuda Tomoko, “Seitō Naikaku to Sūmitsuin,” Seitō Naikaku no Seiritsu to Hōkai, ed. Kindai Nihon Kenkyūkai (Tokyo: Yamakawa Shuppansha, 1984), 143. 11. Ibid., 145. 12.  Tōkyō Asahi Shimbun, March 1, 1929; Ōsaka Asahi Shimbun, March 2, 1929; Tōkyō Asahi Shimbun, March 13, 1929. 13. Awaya, Shōwa no Seitō, 84–85. 14. Ibid., 49. 15.  For discussion of the increase of power held by the prime minister within the government, see John P. Mackintosh, The British Cabinet, 3rd ed. (London: Stevens & Sons, 1977), 308–319. See also Patrick Gordon Walker, The Cabinet: Political Authority in Britain (New York: Basic Books, 1970), 80–85. 16. Mackintosh, British Cabinet, 312; Walker, Cabinet, 82–83. 17.  R. K. Webb, Modern England, 2nd ed. (New York: Harper Collins, 1980), 465; Samuel Huntington, The Solider and the State (Cambridge, MA: Harvard University Press, 1957), 187. 18. Awaya, Shōwa no Seitō, 92; Kobayashi Tatsuo, ed., Suiusō Nikki (Tokyo: Hara Shobō, 1966), 826–827. 19.  Centre for East Asian Cultural Studies, ed., The Meiji Japan through Contemporary Sources, vol. 1 (Tokyo: Centre for the East Asia Cultural Studies, 1969), 96. 20.  Kobe Ryōichi, Gyakusetsu no Guntai, vol. 9. of Nihon no Kindai (Tokyo:

Notes to Chapter 3  189 Chūō Kōron Shinsha, 1998), 73–77; Fujita Tsukugo, Meiji Gunsei (Tokyo: Shinzansha, 1992), 82–85. 21.  Kobayashi, ed., Suiusō Nikki, 877. 22. Article 6 of “On the Offices of Cabinet Ministers” stipulated: “the Minister of War [the army minister] must report to the Prime Minister even those matters which [are] secrets of the Army and are to be laid directly before the Emperor by the Chief of the General Staff Office.” Article 7 of the “Cabinet-Official Organization” stipulated: “Matters which pertain to military secrecy or military orders and are presented to the Emperor, except for those items which the Emperor has seen fit to entrust to the Cabinet, are to be reported by the Minister of War [the army minister] and the Minister of the Navy to the Cabinet Prime Minister.” This did not stipulate who held the right to report to and secure approval from the emperor; it was considered that at least the chiefs of the General Staff of the army and navy were permitted to do so. Article 7 of “the CabinetOfficial Organization” was also interpreted as giving the army and navy ministers the same right as their chiefs vis-à-vis the emperor. See Ōe Shinobu, Tennō no Guntai, vol. 3 of Shōwa no Rekishi (Tokyo: Shōgakukkan, 1982), 166. For translations of “On the Offices of Cabinet Ministers” and the “Cabinet-Official Organization,” see the Centre for East Asian Cultural Studies, ed., Meiji Japan through Contemporary Sources, vol. 3. (Tokyo: Centre for East Asian Cultural Studies, 1972), 210–211, 215. 23.  Centre for East Asian Cultural Studies, The Meiji Japan through Contemporary Sources, vol. 1, 96. 24. Fujita, Meiji Gunsei, 195; and Kobayashi, ed., Suiusō Nikki, 826–827. 25.  Nagai Kazu argues that the army and navy ministers acquired the right by practice before the introduction of the “Cabinet-Official Organization” law in 1889. He believes that Article 7 of the “Cabinet-Official Organization” did not stipulate which institutions held the right to report directly to the emperor, removing the term “chief of the General Staff” from Article 6 of On the “Offices of Cabinet Ministers” to give ex post facto legal recognition to the practice. See Nagai Kazu, Kindai Nihon no Gunbu to Seiji (Kyoto: Shibunkaku, 1993), 306–335. 26. Ibid., 375. 27. Ibid. 28. Ibid., 331. 29. Awaya, Shōwa no Seitō, 92. 30.  Matsushita Yoshio, Meiji Gunseishi Ron, vol. 2 (Tokyo: Kokusho Kankō Kai, 1978), 298–300. 31. Ibid. 32. Nagai, Kindai Nihon no Gunbu to Seiji, 256.

190  Notes to Chapter 4

Chapter 4 1.  Kobe demonstrates how party government was superior to the military during this period. See Kobe Ryōichi, Gyakusetu no Gundai, vol. 9 of Nihon no Kindai (Tokyo: Chūō Kōron Shinsha, 1999), 178–184. 2.  On the Hara cabinet’s attempts in the first and third arenas, see Amemiya Shōichi, Kindai Nihon no Sensō Shidō (Tokyo: Yoshihara Kōbunkan, 1997), 130– 144, 144–148, respectively. For its attempt in the second arena, see Asano Kazuo, Taishō Demokurashii to Rikugun (Ōta: Kantō Gakuen Daigaku, 1993), 207–236. 3.  In August 1918, Japan sent troops along with the United States to Siberia under the pretext of saving Czech troops under attack by the Bolsheviks. After dispatching its troops, Japan changed the objectives of its military operations in Siberia to preempting the Bolshevik threat to Korea and Manchuria. See Ikei Masaru, Nihon Gaikōshi Gaisetsu, rev. ed. (Tokyo: Keiō Tsūshin, 1983), 126–131. 4. Amemiya, Kindai Nihon no Sensō Shidō, 131–144. See also Tanaka Giichi Denki Kankōkai, Tanaka Giichi Den, vol. 2 (Tokyo: Tanaka Giichi Denki Kankōkai, 1960), 189–203. 5.  Tanaka Giichi Denki Kankōkai, Tanaka Giichi Den, vol. 2, 194. 6. Amemiya, Kindai Nihon no Sensō Shidō, 142; Tanaka Giichi Denki Kankōkai, Tanaka Giichi Den, vol. 2, 194. 7.  Hara Takashi, Hara Takashi Nikki, vol. 5, ed. Hara Keiichirō (Tokyo: Fukumura Shuppan, 1965), 456–459. 8. Ibid., 457. 9. Amemiya, Kindai Nihon no Sensō Shidō, 145; Kobe, Gyakusetu no Gundai, 183. 10. Amemiya, Kindai Nihon no Sensō Shidō, 145. 11. Ibid., 145. 12.  Kisaka Jun’ichirō, “Gunbu to Demokurashii,” in Heiwa to Sensō no Kenkyū, vol. 2, ed. Nihon Kokusai Seiji Gakkai (Tokyo: Yūhikaku, 1969), 18–19. 13.  Shinobu Seizaburō, Taishō Demokurashii Shi, vol. 3 (Tokyo: Nihon Hyōron Shinsha, 1959), 809–811. 14.  Itō Takashi, “Gunshuku to Teikoku Kokubo Hoshin no Kaitei,” in Daiichiji Sekai Taisen to Seitō Naikaku, ed. Ōkubo Toshiaki, vol. 16 of Nihon Rekishi Taikei, ed. Inoue Mitsusada et al. (Tokyo: Yamakawa Shuppansha, 1996), 227. 15.  Kōketsu Atsushi, Sōryokusen Taisei Kenkyū (Tokyo: San’ichi Shobō, 1981), 82–83. 16.  Takahashi Hidenao, “Rikugun Gunshuku no Zaisei to Seiji,” in Kanryōsei no Keisei to Tenkai, ed. Kindai Nihon Kenkyūkai (Tokyo: Yamakawa Shuppansha, 1986), 157. 17.  Kisaka, “Gunbu to Demokurashii,” 33. 18.  Takahashi, “Rikugun Gunshuku no Zaisei to Seiji,” 175.

Notes to Chapter 4  191 19.  Juan Linz, Crisis, Breakdown, and Reequilibration: The Breakdown of Democratic Regimes (Baltimore: Johns Hopkins University Press, 1978), 17. 20.  Felipe Agüero, “Democratic Consolidation and the Military in Southern Europe and South America,” in The Politics of Democratic Consolidation: Southern Europe in Comparative Perspective, ed. Richard Gunther, P. Nikiforos Diamandouros, and Hans-Jürgen Puhle (Baltimore: John Hopkins University Press, 1995), 151. 21.  Chūō Kōron 33, no. 11 (October 1918): 70. 22.  Tōkyō Asahi Shimbun, September 26, 1918. 23.  Yoshino Sakuzō, “Hara Naikaku ni taisuru Yobo,” Chūō Kōron 33, no. 11 (October 1918): 82. 24.  Furukawa Gakuto, “Hara Shushō ni teisuru Sho,” Chūō Kōron 33, no. 12 (November 1918): 49. 25.  Yamakawa Hitoshi, “Gunpuku no Seiji yori Furokku Kōto no Seiji e,” in Yamakawa Hitoshi Zenshū, vol. 2 (Tokyo: Keisō Shobō, 1966), 109. First published in Shin Nihon, November 1918. 26.  Matsumoto Gōkichi, Taishō Demokurashii Ki no Seiji: Matsumoto Gōkichi Nikki, ed. Oka Yoshitake and Hayashi Shigeru (Tokyo: Iwanami Shoten, 1959), 35. 27.  Yoshino Sakuzō was one of the most prominent advocates of minponshugi. According to him, it meant that “the activities of the sovereign should politically be aimed at the people.” He distinguished minponshugi from minshushugi, the term the Japanese today use for democracy, because minshushugi meant that “the sovereignty of the state resided in the people.” This latter concept goes against the polity of prewar Japan, in which the state’s sovereignty resided in the emperor. Yoshino argued that, in minponshugi, the government should formulate policies according to the people’s will. He considered universal suffrage and a party government system the two institutional pillars of minponshugi. See Yoshino Sakuzō, “Kensei no Hongi o toite sono Yūshū no bi o Nasu no Michi o ronzu,” in Yoshino Sakuzō, Yoshino Sakuzō Senshū, vol. 2 (Tokyo: Iwanami Shoten, 1996), 3–98. First published in Chūō Kōron, January 1916. See also Matsuzawa Hiroshi, “Yoshino Sakuzō to Seiji Kaikaku,” in Yoshino, Yoshino Sakuzō Senshū, vol. 2, 313. 28.  Maeda Renzan, Hara Takashi Den (Tokyo: Takayama Shoin, 1943), 365–366. 29.  Suzuki Bunji, “Rōdōsha no Tachiba yori Hara Shin Naikaku ni Nozomu,” in Yūaikai Kikanshi: Rōdō Oyobi Sangyō, vol. 8, ed. Hōsei Daigaku Ōhara Shakai Mondai Kenkyūjo and Sōdōmei Gojūnenshi Kankō Iinkai (Tokyo: Hōsei Daigaku Shuppankyoku, 1977), 238–239. First published in Rōdō oyobi Sangyō 7, no. 11 (1918): 2–3. 30.  Takahashi Hidenao, “Hara Takashi Naikaku ka no Gikai,” in Nihon Gikai Shiroku, vol. 2, ed. Uchida Kenzō, Kinbara Samon, and Furuya Tetsuo (Tokyo: Daiichi Hōki Shuppan, 1990), 254; Oka Yoshitake, Tenkanki no Taishō, vol. 3 of Oka Yoshitake Senshū (Tokyo: Iwanami Shoten, 1992), 151–153. See also Toriumi ­Yasushi,

192  Notes to Chapter 4 “Hara Naikaku: ‘Jun Seitō Naikaku no Meian,’” in Nihon Naikaku Shiroku, vol. 2, ed. Hayashi Shigeru and Tsuji Kiyoaki (Tokyo: Daiichi Hōki Shuppan, 1981), 332–334. 31.  Ishibashi Tanzan, “Seitō Seiji no Kakuritsu to Futsū Senkyo no Yobo,” in Ishibashi Tanzan Zenshū, vol. 3, ed. Ishibashi Tanzan Zenshū Hensan Iinkai (Tokyo: Tōyō Keizai Shimpōsha, 1971), 3–7. First published in Tōyō Keizai Shimpō, January 1919. 32.  For example, see Tōkyō Asahi Shimbun, July 30, 1920. 33.  For example, see ibid., May 16, 1920. 34.  Yoshino, “Hara Naikaku ni taisuru Yobo,” 84; Furukawa, “Hara Shushō ni teisuru Sho,” 50–51; Matsuo Takayoshi, Futsū Senkyo Seido Seiritushi no Kenkyū (Tokyo: Iwanami Shoten, 1989), 129. 35. Matsuo, Futsū Senkyo Seido Seiritushi no Kenkyū, 130–131. 36.  Banno Junji, “Seitō Seiji no Kakuritsu,” in Kindai, vol. 3, ed. Ishii Kanji et al.—vol. 9 of Kōza Nihon Rekishi, ed. Rekishigaku Kenkyūkai and Nihonshi Kenkyūkai (Tokyo: Tōkyō Daigaku Shuppankai, 1985), 97–101. 37. Ibid., 98, referring to “Kaisan Yokujitsu ni okeru Hara Sō sai no Enzetsu” [Prime Minister Hara Takashi’s speech the Day after the Dissolution], in Seiyū, no. 239 (March 15, 1920), 30. 38. Ibid., 101. 39. Oka, Tenkanki no Taishō, 139–141. 40. Ibid., 152–153, introducing articles from Tōkyō Asahi Shimbun, November 5, 1921; Tōkyō Nichinichi Shimbun, November 7, 1921; and Tōkyō Nichinichi Shimbun, November 14, 1921. See also Tōkyō Asahi Shimbun, July 30, 1920; and Takahashi, “Rikugun Gunshuku no Zaisei to Seiji,” 237–238. 41. Maeda, Hara Takashi Den, 365–366. 42. Linz, Crisis, Breakdown, and Reequilibration, 16. 43.  Yoshino Sakuzō, “Shakai Kaizō no Daiichi Kaidan toshite no Futsū Senkyo,” in Yoshino Sakuzō Senshū, vol. 3 (Tokyo: Iwanami Shoten, 1995), 290. First published in Warera, March 1920. 44.  The Seiyūkai split over whether to support the Kiyoura cabinet. Those who supported the cabinet formed the Seiyū Hontō. 45.  Ōsaka Asahi Shimbun, January 5, 1924. 46.  Tōkyō Nichinichi Shimbun, January 5, 1924. 47.  Jiji Shimpō, January 9, 1924. 48.  Masumi Junnosuke, Nihon Seitō Shi Ron, vol. 5 (Tokyo: Tōkyō Daigaku Shuppankai, 1979), 72–79. 49.  Kawato Sadafumi, Nihon no Seitō Seiji, 1890–1937 (Tokyo: Tōkyō Daigaku Shuppankai, 1992), 217. 50.  Yoshino Sakuzō, “Shin Naikaku ni taisuru kitai,” in Yoshino Sakuzō Senshū, vol. 4 (Tokyo: Iwanami Shoten, 1996), 84. First published in Chūō Kōron, July 1924.

Notes to Chapters 4 and 5  193 51.  Tōkyō Asahi Shimbun, May 11, 1925. 52.  For example, see Yomiuri Shimbun, May 15, 1925. See also Ōsaka Asahi Shimbun, June 10, 1925. 53.  Yomiuri Shimbun, June 8, 1925. 54.  Ōsaka Asahi Shimbun, January 10, 1924. 55.  Minobe Tatsukichi, “Seigi ni motozuku Seiji,” Kaizō 6, no. 7 (1924): 112. 56.  Tōkyō Asahi Shimbun, June 10, 1925. 57. Ibid. 58.  Yamakawa Hitoshi, “Goken Naikaku,” Kaizō 6, no. 7 (1924): 114. 59.  Irie Kannichi, the chief secretary of the Chamberlain of the Privy Seal at the time and a close associate of Saionji, is one of those who made this speculation. See Utsumi Tōru, “Seihen Omoidegusa,” in Taishō Shoki Yamagata Aritomo Danwa Hikki: Seihen Omoidegusa, ed. Itō Takashi (Tokyo: Yamakawa Shuppansha, 1981), 158–159. Utsumi Tōru is believed to be the pen name used by of Irie Kannichi, although Seihen Omoidegusa was never published. See also Kawahara Hiroshi, “Kiyoura Naikaku: Saigo no Chōzen Naikaku,” in Nihon Naikaku Shiroku, vol. 3, ed. Hayashi Shigeru and Tsuji Kiyoaki (Tokyo: Daiichi Hōki Shuppan, 1981), 5. 60.  Peter Duus, Party Rivalry and Political Change in Taishō Japan (Cambridge, MA: Harvard University Press, 1968), 169. 61.  The exact date of this entry is not specified. See Ugaki Kazushige, Ugaki Kazushige Nikki, vol. 1 (Tokyo: Misuzu Shobō, 1968), 457. 62. Ibid., 478. 63.  Letter to Marshal Uehara Yūsaku from Hoshino Shōzaburō, May 26, 1924, in Uehara Yūsaku Kankei Bunsho, ed. Uehara Yūsaku Kankei Bunsho Kenkyūkai (Tokyo: Tōkyō Daigaku Shuppankai, 1976), 422. 64.  Letter to Marshal Uehara Yūsaku from Hoshino, June 12, 1924, in Uehara Yūsaku Kankei Bunsho Kenkyūkai, ed., Uehara Yūsaku Kankei Bunsho, 423. 65. Ibid. 66.  Kisaka, “Gunbu to Demokurashii,” 36. 67.  Banno, “Seitō Seiji no Kakuritsu,” 98, quoting ““Kaisan Yokujitsu ni okeru Hara Sō sai no Enzetsu” in Seiyū, no. 239 (March 15, 1920), 27. 68. Ibid., 106. 69. Ibid., 104.

Chapter 5 1.  On the assassination of Chang Tso-lin, see Seki Hiroharu, “Manshū Jihen Zenshi,” in Manshū Jihen Zennya, vol. 1 of Taiheiyō Sensō e no Michi, ed. Nihon Kokusai Seiji Gakkai (Tokyo: Asahi Shimbunsha, 1963), 299–309; Rekishigaku Kenkyūkai, ed., Manshū Jihen, vol. 1 of Taiheiyō Sensōshi (Tokyo: Aoki Shoten, 1971), 143–153.

194  Notes to Chapter 5 2.  Seki, “Manshū Jihen Zenshi,” 305–306; Rekishigaku Kenkyūkai, ed., Manshū Jihen, 147–150. 3.  Kōmoto Daisaku, “Watashi ga Chō Sakurin o koroshita,” in “Bungei Shunjū” ni miru Shōwashi, vol. 1, ed. Bungei Shunjū (Tokyo: Bungei Shunjū, 1988), 48–49. First published in Bungei Shunjū, December 1954. Kōmoto found that Chief in Commander Muraoka was going to send Major Takeshita Yoshiharu as a messenger to the Japanese troops stationed in Beijing and Tianjin. Uncovering the plan, Kōmoto told Major Takeshita that he would carry out the assassination and asked Takeshita to let him know when Chang was leaving Beijing. 4.  Harada Kumao, Saionji Kō to Seikyoku, vol. 1 (Tokyo: Iwanami Shoten, 1950), 3–4; Matsumoto Gōkichi, Taishō Demokurashii Ki no Seiji: Matsumoto Gōkichi Nikki, ed. Oka Yoshitake and Hayashi Shigeru (Tokyo: Iwanami Shoten, 1959), 605–607; Oka Yoshitake et al., eds., Ogawa Heikichi Kankei Bunsho, vol. 1 (Tokyo: Misuzu Shobō, 1973), 628–629. 5.  Awaya Kentarō, Tōkyō Saibanron (Tokyo: Ōtsuki Shoten, 1989), 232–233; Oka et al., eds., Ogawa Heikichi Kankei Bunsho, vol. 1, 629. 6.  Oka et al., eds., Ogawa Heikichi Kankei Bunsho, vol. 1, 260; Matsumoto, Taishō Demokurashii Ki no Seiji, 612. 7.  Seki, “Manshū Jihen Zenshi,” 322; Oka et al., eds., Ogawa Heikichi Kankei Bunsho, vol. 1, 630; Funaki Shigeru, Okamura Yasuji (Tokyo: Kawaide Shobō Shinsha, 1984), 199–204. 8.  Oka et al., eds., Ogawa Heikichi Kankei Bunsho, vol. 1, 630. 9. Matsumoto, Taishō Demokurashii Ki no Seiji, 567. 10.  Tōkyō Asahi Shimbun, July 3, 1929. 11.  Kitaoka Shin’ichi, Seitō kara Gunbu e, vol. 5 of Nihon no Kindai (Tokyo: Chūō Kōron Shinsha, 1999), 48–51; Awaya Kentarō, Shōwa no Seitō, vol. 6 of Shōwa no Rekishi (Tokyo: Shōgakukkan, 1988), 22–25. 12.  For a detailed description of the Matsushima Scandal, see Ōshima Mitsuko, “Matsushima Yūkaku Iten Mondai,” in Nihon Seiji Saiban Shiroku: Shōwa Zen, ed. Wagatsuma Sakae et al. (Tokyo: Daiichi Hōki Shuppan, 1970), 97–122. 13. Kitaoka, Seitō kara Gunbu e, 50; Awaya, Shōwa no Seitō, 24. See also Matsumoto Seichō, Shōwashi Hakkutsu, vol. 1 (Tokyo: Bungei Shunjū Shinsha, 1965), 5–70. 14.  For a detailed description of the Boku Retsu Scandal, see Kyo Seikai, “Boku Retsu Jiken,” in Nihon Seiji Saiban Shiroku: Taishō, ed. Wagatsuma Sakae et al. (Tokyo: Daiichi Hōki Shuppan, 1969), 379–411. See also Matsumoto, Shōwashi Hakkutsu, vol. 1, 149–151. 15.  Banno Junji, “Seitō Seiji no Kakuritsu,” in Kindai, vol. 3, ed. Ishii Kanji et al.—­vol. 9 of Kōza Nihon Rekishi, ed. Rekishigaku Kenkyūkai and Nihonshi Kenkyūkai (Tokyo: Tōkyō Daigaku Shuppankai, 1985), 112–115; Takahashi Susumu and Miyazaki Ryūji, “Seitō Seiji no Teichaku to Hōkai,” in Nihon Kindaishi ni

Notes to Chapter 5  195 okeru Tenkanki no Kenkyū, ed. Banno Junji and Miyaji Masato (Tokyo: Yamakawa Shuppan, 1985), 236. 16.  Ishibashi Tanzan, “Konki Gikai no Koseki,” in Ishibashi Tanzan Zenshū, vol. 5, ed. Ishibashi Tanzan Zenshū Hensan Iinkai (Tokyo: Tōyō Keizai Shimpōsha, 1971), 77. 17.  Ōsaka Asahi Shimbun, March 6, 1926. 18.  Tōkyō Nichinichi Shimbun, March 11, 1926. 19.  Takahashi and Miyazaki, “Seitō Seiji no Teichaku to Hōkai,” 235. 20.  Tōkyō Asahi Shimbun, March 24, 1927. 21. Ibid. 22.  The Taishō emperor died at the end of 1926 and the Shōwa emperor came to the throne, marking the beginning of the Shōwa era. 23. Awaya, Shōwa no Seitō, 30–31; Kitaoka, Seitō kara Gunbu e, 51; Itō Takashi, “Taishō Tennō Hogyo,” in Daiichiji Sekai Taisen to Seitō Naikaku, ed. Ōkubo Toshiaki, vol. 16 of Nihon Rekishi Taikei, ed. Inoue et al. (Tokyo: Yamakawa Shuppansha, 1996), 279; Wakatsuki Reijirō, Kofūan Kaikoroku (Tokyo: Yomiuri Shimbunsha, 1950), 321–322; Matsumoto, Taishō Demokurashii Ki no Seiji, 551. 24.  Tōkyō Asahi Shimbun, January 21, 1927. 25.  Ibid., January 22, 1927. 26.  Ibid., January 26, 1927. 27.  Ōsaka Asahi Shimbun, January 23, 1927. 28.  Ōsaka Mainichi Shimbun, January 23, 1927. 29.  Ishibashi Tanzan, “Shugiin o Kaisan subeshi,” in Ishibashi Tanzan Zenshū, vol. 5, 102. 30.  Tōkyō Asahi Shimbun, April 20, 1927. 31.  When the fifty-fourth Diet was convened at the end of 1927, the Seiyūkai only had 190 members versus the Minseitō’s 219. See Itō Takashi, “Tanaka Naikaku to Tōhō Kaigi,” in Ōkubo, ed., Daiichiji Sekai Taisen to Seitō Naikaku, 290–291. 32.  The total number of seats in the Lower House was 466. 33.  Ōsaka Asahi Shimbun, February 24, 1928. 34. Ibid. 35. Ibid. 36.  Ōsaka Mainichi Shimbun, February 25, 1928. 37.  Ibid., February 26, 1928. 38.  Yoshii Ken’ichi, “Tai Chūgoku Seisaku no Tenkan to Gikai,” in Nihon Gikai Shiroku, vol. 3., ed. Uchida Kenzō, Kinbara Samon, and Furuya Tetsuo (­Tokyo: Daiichi Hōki, 1991), 40–42. 39.  Itō, “Tanaka Naikaku to Tōhō Kaigi,” 292. 40.  Tōkyō Asahi Shimbun, February 12, 1929. 41.  Yoshino Sakuzō, “Gendai Seikyoku no Tenbo,” in Yoshino Sakuzō Senshū,

196  Notes to Chapter 5 vol. 4 (Tokyo: Iwanami Shoten, 1996), 251–252. First published in Chūō Kōron, February 1929. 42.  Ōsaka Mainichi Shimbun, September 4, 1928. 43.  Yoshii, “Tai Chūgoku Seisaku no Tenkan to Gikai,” 37–38. 44.  Ōsaka Mainichi Shimbun, May 31, 1928. 45.  Ōsaka Asahi Shimbun, June 18, 1928. See also ibid., June 22, 1928. 46.  For a detailed description of the crisis of the financial system in 1927 and its origins, see Nakamura Masanori, Shōwa no Kyōkō, vol. 2 of Shōwa no Rekishi (Tokyo: Shōgakukkan, 1988), 34–88; Kitaoka, Seitō kara Gunbu e, 52–54, 68–69; Itō, “Tanaka Naikaku to Tōhō Kaigi,” 286–288. 47.  For the details of the bills, see Nakamura, Shōwa no Kyōkō, 52–53. 48.  Ōsaka Mainichi Shimbun, May 9, 1927. 49.  Tōkyō Asahi Shimbun, April 19, 1927. 50.  Ibid., April 17, 1927. 51.  Ōsaka Mainichi Shimbun, May 4, 1927. 52.  Nihon Tōkei Kenkyūjo, ed., Nihon Keizai Tōkeishū: Meiji, Taishō, Shōwa (Tokyo: Nihon Hyōron Shinsha, 1958), 296. 53.  Takeda Haruhito, “Rōshi Kankei,” in Daiichiji Sekai Taisenki, vol. 1 of Nihon Teikoku Shugishi, ed. Ōishi Kaichirō (Tokyo: Tōkyō Daigaku Shuppankai, 1985), 288–289; Futamura Kazuo, “Rōdōsha Kaikyū no Jōtai to Rōdō Undō,” in Iwanami Kōza: Nihon Rekishi, vol. 18, ed. Asao Naohiro et al. (Tokyo: Iwanami Shoten, 1975), 118–119. 54.  Tōkyō Asahi Shimbun, March 17, 1927. 55.  Ibid., February 10, 1926 (emphasis added). 56. Matsumoto, Taishō Demokurashii Ki no Seiji, 529–530. 57. Ibid., 531. 58. Ibid., 544, 561–562. 59.  Itō Takashi, Shōwa Shoki Seijishi Kenkyū (Tokyo: Tōkyō Daigaku Shuppankai, 1969), 380–381. Itō suspects that the group led by Hiranuma Kiichirō saw Hiranuma himself as a candidate for prime minister. 60. Ibid., 380. 61.  This became a reality as Ogawa Heikichi, the minister of the railways of the Tanaka cabinet, was arrested under the Hamaguchi cabinet. 62.  Hayashi Yūichi, “Kaikyū no Seiritsu to Chiiki Shakai,” in Gendai Shakai e no Tenkei, vol. 3. of Nihon Gendaishi: Kōzō to Hendō, ed. Banno Junji et al. (­Tokyo: Iwanami Shoten, 1993), 43–44. 63. Ibid. 64.  For example, the Japan Labor Union Alliance called for the adoption of the Labor Union Law, which would have guaranteed the right to unite, the right to bargain collectively, and the right to strike, at its seventeenth convention in 1928.

Notes to Chapters 5 and 6  197 See Ōkōchi Kazuo and Matsuo Hiroshi, Nihon Rōdō Kumiai Monogatari: Shōwa (Tokyo: Chikuma Shobō, 1965), 134. To give another example, the All ­Japan Farmers Union pressed for the adoption of the Tenancy Law to protect tenant farmers at its founding convention in 1927. See Nōmin Kumiai Gojūshunen Kinensai Jikkō Iinkai, ed., Nōmin Kumiai Gojūnenshi (Tokyo: Ochanomizu Shobō, 1972), 100. People such as Suzuki Bunji and Takahashi Kamekichi, leaders of the labor and tenant farmers movements, demanded the legalization of labor unions and ­legal recognition of tenancy rights. See Suzuki Bunji, “Kinrō Musansha no Seikatsu ­Antei o,” Kaizō 10, no. 3 (1928): 102–104; Takahashi Kamekichi, “Seisansha Hon’i no Kosakuho Seitei no tameni,” Kaizō 10, no. 3. (1928): 96–97. 65.  For example, see Ōsaka Mainichi Shimbun, April 19, 1926; and Tōkyō Nichinichi Shimbun, June 30, 1929. 66. Matsumoto, Taishō Demokurashii Ki no Seiji, 510. 67.  Ugaki Kazushige, Ugaki Kazushige Nikki, vol. 1, ed. Tsunoda Jun (Tokyo: Misuzu Shobō, 1968), 538. 68. Ibid., 528. 69. Ibid., 643. 70.  Hata Ikuhiko, Shōwashi no Nazo o ou, vol. 1 (Tokyo: Bungei Shunjūsha, 1993), 35. 71. Matsumoto, Taishō Demokurashii Ki no Seiji, 612. 72.  Juan Linz, Crisis, Breakdown, and Reequilibration: The Breakdown of Democratic Regimes (Baltimore: Johns Hopkins University Press, 1978), 32. 73. Matsumoto, Taishō Demokurashii Ki no Seiji, 628–629. 74.  Seki, “Manshū Jihen Zenshi,” 321.

Chapter 6 1.  For example, Nakamura Masanori writes, “[The signing of the treaty] was the last victory of party politics in foreign policy at the beginning of the Shōwa era.” Nakamura Masanori, Shōwa no Kyōkō, vol. 2 of Shōwa no Rekishi (Tokyo: Shōgakukkan, 1988), 201. See also Itō Yukio, Taishō Demokurashii to Seitō Seiji (Tokyo: Yamakawa Shuppansha, 1987), 253. 2.  Kobayashi Tatsuo, “Kaigun Gunshuku Jōyaku,” in Manshū Jihen Zennya, vol. 1 of Taiheiyō Sensō e no Michi, ed. Nihon Kokusai Seiji Gakkai (Tokyo: Asahi Shimbunsha, 1963). 3.  Shimada Toshihiko, “Manshū Jihen no Tenkai,” in Manshū Jihen, vol. 2 of Taiheiyō Sensō e no Michi, ed. Nihon Kokusai Seiji Gakkai (Tokyo: Asahi Shimbunsha, 1962), 12–13; Katakura Tadashi, “Manshū Jihen Kimitsu Seiryaku Nisshi sono Ichi,” in Manshū Jihen, ed. Kobayashi Tatsuo and Shimada Toshihiko, Gendaishi Shiryō, vol. 7 (Tokyo: Misuzu Shobō, 1964), 184–185.

198  Notes to Chapter 6 4.  Ogata Sadako, Manshū Jihen to Seisaku no Keisei Katei (Tokyo: Hara Shobō, 1966), 106–107; Shimada, “Manshū Jihen no Tenkai,” 44. 5.  Morishima Gorō and Yanai Hisao, eds., Manshū Jihen, vol. 18 of Nihon Gaikōshi (Tokyo: Kashima Kenkyūjo Shuppankai, 1973), 144–146. 6.  Shimada, “Manshū Jihen no Tenkai,” 86. After the Kwantung Army occupied major cities in southern Manchuria, Chang Hsüeh-liang (the son of Chang Tso-lin) established control over the remainder of Manchuria. See Rekishigaku Kenkyūkai, ed., Manshū Jihen, vol. 1 of Taiheiyō Sensōshi (Tokyo: Aoki Shoten, 1971), 279. 7.  Shimada, “Manshū Jihen no Tenkai,” 16. 8.  Eguchi Keiichi, Jūgonen Sensō no Kaimaku, vol. 4 of Shōwa no Rekishi (­Tokyo: Shōgakukkan, 1982), 76. 9. Ogata, Manshū Jihen to Seisaku no Keisei Katei, 209, 234; Sakai Tetsuya, Taishō Demokurashii Taisei no Hōkai: Naisei to Gaikō (Tokyo: Tōkyō Daigaku Shuppankai, 1992), 19–20; Katō Yōko, “Seitō Naikaku no Shuen,” in Kakushin to Sensō no Jidai, ed. Ōkubo Toshiaki, vol. 17 of Nihon Rekishi Taikei, ed. Inoue ­Mitsusada et al. (Tokyo: Yamakawa Shuppansha, 1997), 23. 10.  Shimada, “Manshū Jihen,” 180. 11.  Hata Ikuhiko, Gun Fashizumu Undōshi (Tokyo: Kawaide Shobō, 1962), 21. 12.  Since this plan was never carried out and no official investigation was undertaken, its exact content and the degree of involvement of the army’s central command still remain to be clarified. See Itō Takashi, “Sangatsu Jiken,” in Daiichiji Sekai Taisen to Seitō Naikaku, ed. Ōkubo Toshiaki, vol. 16 of Nihon Rekishi Taikei, ed. Inoue Mitsusada et al. (Tokyo: Yamakawa Shuppansha, 1996), 329–330, note 2 in particular. 13.  The degree of Ugaki’s commitment remains unclear. On Ugaki’s involvement in this affair, see ibid., 330; and Itō Takashi, “Dainiji Wakatsuki Naikaku,” in Ōkubo, ed., Daiichiji Sekai Taisen to Seitō Naikaku, 332, note 10. 14.  For a description of this attempted coup d’état, see Katō Yōko, “Sangatsu Jiken,” 11–14. 15. Ibid., 13–14. 16.  Katō Yōko, “Go Ichi Go Jiken,” in Ōkubo, ed., Kakushin to Sensō no Jidai, 35. 17.  Hara Akira, “Keiki Junkan,” in Sekai Daikyōkōki, vol. 2 of Nihon Teikoku Shugishi, ed. Ōishi Kaichirō (Tokyo: Tōkyō Daigaku Shuppankai, 1987), 391; Hara Akira, “Kin Kaikin to Inoue Zaisei,” in Ōkubo, ed., Daiichiji Sekai Taisen to Seitō Naikaku, 315. 18.  Hara, “Kin Kaikin to Inoue Zaisei,” 315. 19.  Hara, “Keiki Junkan,” 393. 20. Ibid., 396; Hara Akira, “Kyubō no Nōson,” in Ōkubo, ed., Daiichiji Sekai Taisen to Seitō Naikaku, 334. 21.  Hara, “Kyubō no Nōson,” 334.

Notes to Chapter 6  199 22.  Nakamura Masanori, “Daikyōkō to Nōson Mondai,” in Iwanami Kōza: ­Nihon Rekishi, vol. 19, ed. Asao Naohiro et al. (Tokyo: Iwanami Shoten, 1976), 139. 23.  Hara, “Kyubō no Nōson,” 335. 24.  Hara, “Keiki Junkan,” 400. 25. Ibid., 401. 26. Nakamura, Shōwa no Kyōkō, 295. 27.  Nagaoka Shinkichi, ed., Kindai Nihon no Keizai: Tōkei to Gaisetu (Kyoto: Mineruba Shobō, 1988), 235. 28. Itō, Taishō Demokurashii to Seitō Seiji, 252–253. 29.  Various works have addressed the problems with the economic policies of the Hamaguchi and Wakatsuki cabinets. Among them I am most indebted to Itō Yukio’s analysis of the ineffectiveness of the regime in dealing with the economic crisis. See Itō, Taishō Demokurashii to Seitō Seiji, 248–255. On this issue, see also Nakamura, Shōwa no Kyōkō, 271–290; and Chō Yukio, “Shōwa Kyōkō 2,” in Shōwa Kyōkō: Sono Rekishiteki Igi to Zentaizō, ed. Sumiya Mikio (Tokyo: Yūhikaku, 1974), 234–245. 30.  For example see Itō, Taishō Demokurashii to Seitō Seiji, 248. 31.  For example, Ishibashi Tanzan strongly opposed the return to the gold standard at par determined before the First World War and proposed a return at devalued par. Among various articles in which he expressed this view, see Ishibashi Tanzan, “Inoue Zōsho no Kyuheika Kaikinron o hyōsu,” in Ishibashi Tanzan Zenshū, vol. 7 (Tokyo: Tōyō Keizai Shimpōsha, 1971), 72–104. First published in Tōyō Keizai Shimpō, September 21, 28; October 5, 12, 19, 26; November 2, 1929. 32.  Ōsaka Mainichi Shimbun, November 22, 1929. 33.  Tōkyō Asahi Shimbun, January 11, 1930. 34.  Hara, “Kin Kaikin to Inoue Zaisei,” in Ōkubo, ed., Daiichiji Sekai Taisen to Seitō Naikaku, 315–316; Itō, Taishō Demokurashii to Seitō Seiji, 250–251. 35.  Ōkurashō, ed., Saikei, vol. 3 of Shōwa Zaiseishi (Tokyo: Tōyō Keizai Shimpōsha, 1955), appendix, 5. 36. Itō, Taishō Demokurashii to Seitō Seiji, 251–253. 37.  Inoue Junnosuke, “Sekai Fukeiki to waga Kokumin no Kakugo,” Minsei 4, no. 9 (September 1930), quoted in ibid., 252. 38.  Ishibashi Tanzan, “Nokoka suru Seihenrai no Yōsō,” in Ishibashi Tanzan Zenshū, vol. 7, 426. First published in Tōyō Keizai Shimpō, August 2, 1930. 39.  Tōkyō Asahi Shimbun, April 14, 1931. 40.  Ibid., April 17, 1931. 41.  During this time key policy makers in the government and the Bank of Japan had become aware that it was impossible to sustain the gold standard. For their views, see Nakamura Takafusa, Shōwashi, 1926–1945 (Tokyo: Tōyō Keizai Shimpōsha, 1993), 142. 42.  Tōkyō Asahi Shimbun, November 4, 6, 1931; Nakamura, Shōwa no Kyōkō, 289.

200  Notes to Chapter 6 43.  Ishibashi Tanzan, “Shidō Kaikyu no otoshiireru Zetsudai no Kiken Shiso,” in Ishibashi Tanzan Zenshū, vol. 8, 14–18. First published in Tōyō Keizai Shimpō, May 2, 1931. 44.  Tōkyō Asahi Shimbun, December 3, 1931. 45.  Jiji Shimpō, November 6, 1931, quoted in Itō, Taishō Demokurashii to Seitō Seiji, 254–255. 46.  For a detailed description of these scandals, see Ōshima Tarō, “Kunsho Tetsudō Gigoku Jiken: Seitō Seiji ni okeru Oshoku no Rotei,” in Nihon Seiji Saiban Shiroku: Shōwa Zen, ed. Wagatsuma Sakae et al. (Tokyo: Daiichi Hōki Shuppan, 1970), 305–358. 47.  Ōsaka Mainichi Shimbun, November 22, 1929. 48.  Tōkyō Nichinichi Shimbun, November 28, 1929. 49.  Kawahara Hiroshi, “Hamaguchi Naikaku: Seitō Seiji no Eikō to Higeki,” in Nihon Naikaku Shiroku, vol. 3, ed. Hayashi Shigeru and Tsuji Kiyoaki (Tokyo: Daiichi Hōki Shuppan, 1981), 192. 50.  Awaya Kentarō, Shōwa no Seitō, vol. 6 of Shōwa no Rekishi (Tokyo: Shōgakukkan, 1988), 162. 51.  Yoshino Sakuzō, “Senkyo to Kane to Seitō,” in Yoshino Sakuzō Senshū, vol. 4 (Tokyo: Iwanami Shoten, 1996), 353–355. First published in Chūō Kōron, June 1932. 52. Ibid., 355. 53.  In addition, in postwar Japan, extra sessions of the Diet are nearly always convened for a substantial period in the second half of the year. 54.  Sessions of the Diet convened after a general election were called special sessions. 55.  Furuya Tetsuo, “Kin Kaikin, Rondon Jōyaku, Manshū Jihen,” in Nihon ­Gikai Shiroku, vol. 3, ed. Uchida Kenzō, Kinbara Samon, and Furuya Tetsuo (­Tokyo: Daiichi Hōki, 1990), 81–82. 56. Ibid. 57.  Ishibashi Tanzan, “Shushō Sōnan no Kōnin: Kairyō o yosuru Gikaiseido,” in Ishibashi Tanzan Zenshū, vol. 7, 408–409. First published in Tōyō Keizai Shimpō, November 22, 1930. 58.  Tōkyō Asahi Shimbun, January 3, 1931. 59.  Ibid., January 22, 1931. 60.  Ōsaka Asahi Shimbun, January 19, 1931. 61.  Furuya, “Kin Kaikin, Rondon Jōyaku, Manshū Jihen,” 90. 62.  Tōkyō Asahi Shimbun, March 21, 1932; February 5, 1931. 63.  Ōsaka Mainichi Shimbun, January 29, 1931. 64.  Ishibashi Tanzan, “Kinrai no Sesō Tadagoto narazu,” in Ishibashi Tanzan Zenshū, vol. 8, 7–9. First published in Tōyō Keizai Shimpō, April 18, 1931. 65.  Tōkyō Asahi Shimbun, January 8, 1931.

Notes to Chapter 6  201 66.  Ibid., October 8, 1931. 67.  Ibid., March 25, 1932. 68.  Minobe Tatsukichi, “Kyokoku Icchi Naikaku no Seiritsu,” Chūō Kōron 47, no. 7 (July 1932): 30–31. 69.  Ishibashi Tanzan, “Higōhō Keikō iyoiyo Shinokuka sento su,” in Ishibashi Tanzan Zenshū, vol. 8, 36. First published in Tōyō Keizai Shimpō, October 31, 1931. 70.  Tōkyō Asahi Shimbun, February 10, 1932; March 3, 1932. 71.  Ōsaka Asahi Shimbun, May 9, 1932. 72.  Tōkyō Asahi Shimbun, May 14, 1932. 73.  Yomiuri Shimbun, May 18, 1932. 74.  Tōkyō Asahi Shimbun, May 23, 1932. 75.  Ōsaka Asahi Shimbun, May 23, 1932. 76.  Harada Kumao, Saionji Kō to Seikyoku, vol. 2 (Tokyo: Iwanami Shoten, 1950), 129–130. 77. Ibid., 182, 261. 78.  Yamamoto Shirō, “Saitō Naikaku no Seiritsu o megutte,” Shirin 59, no. 4 (1976): 42–78. 79. Ibid., 62–63. Nor is any evidence present in Harada’s diary. See Harada, Saionji Kō to Seikyoku, vol. 2, 288–293. 80.  Yamamoto, “Saitō Naikaku no Seiritsu o megutte,” 52–53. 81. Ibid., 53. 82.  Tokinoya Masaru, ed., Nihon Kindaishi Jiten (Tokyo: Tōyō Keizai Shimpōsha, 1958), 766. 83. Ibid., 768–769. 84.  Miyako Shimbun, July 13, 1931; Ōsaka Asahi Shimbun, June 21, 1930. On the suffering in the countryside, see Nakamura, Shōwa no Kyōkō, 306–317. 85.  Inomata Tsunao, Kyubō no Nōson (Tokyo: Kaizōsha, 1934), 126–127. 86.  Shimomura Chiaki, “Kiga Chitai o aruku: Tōhoku Nōson Sanjō Hōkokusho,” quoted in Nakamura, Shōwa no Kyōkō, 308. First published in Chūō Kōron, February 1932. 87.  For example, see Sagōya Yoshiaki, “Naze Hamaguchi Shushō o koroshita ka,” Shōwa Kyōkō, vol. 2 of Mokugekisha ga kataru Shōwashi, ed. Inose Naoki and Yamazaki Hiroshi (Tokyo: Shin Jimbutsu Ōraisha, 1989), 209–220. In this article Sagōya describes his motives for attempting to assassinate Hamaguchi Osachi. See also Matsumoto Seichō, Shōwashi Hakkutsu, vol. 4 (Tokyo: Bungei Shunjū, 1966), 232, on Inoue Nisshō’s motives for forming such conspiracies. 88.  Tōkyō Nichinichi Shimbun, February 7, 1931. 89.  Ōsaka Asahi Shimbun, May 20, 1932, quoted in Yamamoto, “Saitō Naikaku no Seiritsu o megutte,” 67. 90.  Hamaguchi Osachi, Nikki Zuikanroku (Tokyo: Misuzu Shobō, 1991), 558–559.

202  Notes to Chapter 6 91.  Oka Yoshitake et al., eds., Ogawa Heikichi Kankei Bunsho, vol. 2 (Tokyo: Misuzu Shobō, 1973), 339–340. 92.  Tōkyō Asahi Shimbun, May 9, 1932. 93. Awaya, Shōwa no Seitō, 252–253. 94.  Banno Junji, “Seitō Seiji no Hōkai,” in Nihon Kindaishi ni okeru Tenkanki no Kenkyū, ed. Banno Junji and Miyaji Masato (Tokyo: Yamakawa Shuppansha, 1985), 367. 95.  Kido Kōichi, Kido Kōichi Nikki, vol. 1, ed. Kido Nikki Kenkyūkai (Tokyo: Tōkyō Daigaku Shuppankai, 1966), 168. 96.  For example, see Ōsaka Asahi Shimbun, May 23, 1932. 97.  Juan Linz, Crisis, Breakdown, and Reequilibration: The Breakdown of Democratic Regimes (Baltimore: Johns Hopkins University Press, 1978), 32. 98.  For a detailed description of how the General Staff of the navy acted shortly before the final decision on whether or not to accept the treaty, see Kobayashi, “Kaigun Gunshuku Jōyaku,” 84–91. See also Itō Takashi, Shōwa Shoki Seijishi Kenkyū (Tokyo: Tōkyō Daigaku Shuppankai, 1969), 159–163. 99.  Okada Keisuke, “Okada Keisuke Nikki,” in Manshū Jihen, ed. Kobayashi and Shimada, 8. 100.  Kobayashi, “Kaigun Gunshuku Jōyaku,” 88. 101. Itō, Shōwa Shoki Seijishi Kenkyū, 162; Kobayashi, “Kaigun Gunshuku Jōyaku,” 81. 102.  Inaba Masao et al., eds., Shiryōhen, Supplement to Taiheiyō Sensō e no Michi (Tokyo: Asahi Shimbunsha, 1963), 26. 103. Itō, Shōwa Shoki Seijishi Kenkyū, 162; Kobayashi, “Kaigun Gunshuku Jōyaku,” 81–82. 104. Itō, Shōwa Shoki Seijishi Kenkyū, 162; Kobayashi, “Kaigun Gunshuku Jōyaku,” 82. 105. Itō, Shōwa Shoki Seijishi Kenkyū, 246–247; Kobayashi, “Kaigun Gunshuku Jōyaku,” 110. I am indebted to the work of these scholars for their introduction of most of the evidence of the semi-loyalty of the Seiyūkai regarding the London Naval Treaty. 106.  Kobayashi, “Kaigun Gunshuku Jōyaku,” 109. 107.  Furuya, “Kin Kaikin, Rondon Jōyaku, Manshū Jihen,” 76–78; Kobayashi, “Kaigun Gunshuku Jōyaku,” 113–114. 108. Itō, Shōwa Shoki Seijishi Kenkyū, 246–247; Kobayashi, “Kaigun Gunshuku Jōyaku,” 110. 109.  Yamaura Kan’ichi, ed., Mori Kaku (Tokyo: Mori Kaku Denki Kankōkai, 1940), 672. 110. Harada, Saionji Kō to Seikyoku, vol. 1, 83. 111.  Okada, “Okada Keisuke Nikki,” 13–19.

Notes to Chapter 6  203 112. Ibid., 19. 113. Itō, Shōwa Shoki Seijishi Kenkyū, 246–247. 114. Harada, Saionji Kō to Seikyoku, vol. 1, 177–178. 115. Nakamura, Shōwashi, 1926–1945, 198. 116. Hata, Gun Fashizumu Undōshi, 21–22. The first meeting might have been convened at the end of September. 117.  Kobayashi, “Kaigun Gunshuku Jōyaku,” 151. 118.  On the government’s failed attempts to restrain the military, see Banno Junji, “Gaikōkan no Gokai to Manshū Jihen no Kakudai,” Shakai Kagaku Kenkyū 35, no. 5 (1984). See also Henry Stimson, The Far Eastern Crisis (New York: Harper and Brothers, 1936), 78–79. 119.  It was thought to be appropriate that the General Staff of the army should decide on its own whether to send orders to the troops. However, it was considered a violation of the supreme command of the emperor when the General Staff of the army sent orders at the request of the government. 120.  Eguchi Keiichi, Jūgonen Sensō no Kaimaku, vol. 4, 47. 121.  On the foreign policy of the Tanaka cabinet, see Baba Nobuya, Manshū Jihen e no Michi (Tokyo: Chūō Kōronsha, 1979), 185–217. 122.  Amemiya Shōichi, “Tanaka Naikaku: Kensei Jodō Taisei Shoki ni okeru Seiji Handō,” in Hayashi and Tsuji, eds., Nihon Naikaku Shiroku, vol. 3, 168. 123. Eguchi, Jūgonen Sensō no Kaimaku, 20–21. 124.  Mori Kaku, “Kyuhakuseru Manmo Taisaku,” Keizai Ōrai (October, 1931), quoted in Yamaura, ed., Mori Kaku, 709–711. 125.  Seki Hiroharu, “Manshū Jihen Zenshi,” in Nihon Kokusai Seiji Gakkai, ed., Manshū Jihen Zennya, 411. 126.  Speech on August 4, 1931, quoted in Mitearai Tatsuo, Minami Jirō (Tokyo: Minami Jirō Denki Kankōkai, 1957), 218, 219. 127. Ibid., 224. 128.  Shimada, “Manshū Jihen no Tenkai,” 12. 129. Ibid., 6–7. 130.  Jushin, men who had the informal power to give advice to the emperor, gradually increased their political role as the number of genrō decreased due to death. Jushin included such people as former prime ministers and the chairman of the Privy Council. On the attempt by the Wakatsuki cabinet to convince the jushin of the need for this meeting, see Banno Junji, “‘Kensei Jōdō’ to ‘Kyōryoku Naikaku,’” in Seitō Naikaku no Seiritsu to Hōkai, ed. Kindai Nihon Kenkyūkai (Tokyo: Yamakawa Shuppansha, 1984), 188–189. 131. Ibid., 188; Harada, Saionji Kō to Seikyoku, vol. 2, 90–93. 132.  See Banno, “‘Kensei Jōdō’ to ‘Kyōryoku Naikaku,’” 188; Harada, Saionji Kō to Seikyoku, vol. 2, 90–93; Rekishigaku Kenkyūkai, ed., Manshū Jihen, 300–301.

204  Notes to Chapter 6 and Conclusion 133. Eguchi, Jūgonen Sensō no Kaimaku, 76. 134.  On the attempt to establish a grand coalition cabinet after the Manchurian Incident had begun, I have relied on Inoue Toshikazu, Seiyūkai to Minseitō (Tokyo: Chūō Kōron Shinsha, 2012), 126–135. See also, Banno, “‘Kensei Jōdō’ to ‘Kyōryoku Naikaku,’” 188–200. He argues that there were different attempts to form a coalition including one aimed at seeking the army’s cooperation. 135. Inoue, Seiyūkai to Minseitō, 131–133. 136. Ibid., 133. 137.  Yamamoto, “Saitō Naikaku no Seiritsu o megutte,” 47–48; Kitaoka Shinichi, Seitō kara Gunbu e 1924–41, vol. 5 of Nihon no Kindai (Tokyo: Chūō Kōron Shinsha, 1999), 163–164. 138.  Yamamoto, “Saitō Naikaku no Seiritsu o megutte,” 61; Yamaura, ed., Mori Kaku, 815–818. 139.  Wakatsuki Reijirō, Kofūan Kaikoroku (Tokyo: Yomiuri Shimbunsha, 1950), 390. 140.  Yamamoto, “Saitō Naikaku no Seiritsu o megutte,” 62; Yamaura, ed., Mori Kaku, 816. 141. Harada, Saionji Kō to Seikyoku, vol. 2, 293.

Conclusion 1.  Marius Jansen, The Making of Modern Japan (Cambridge, MA: Harvard University Press, 2000), 504. 2.  Andrew Gordon, A Modern History of Japan: From Tokugawa Times to the Present (New York: Oxford University Press, 2009), 169. 3. Ibid., 170. 4.  Note that three parties—the Seiyūkai, Kenseikai, and Seiyū Hontō—coexisted between 1924 and 1927. 5.  Inoue Toshikazu, Seiyūkai to Minseitō (Tokyo: Chūō Kōron Shinsha, 2012), 126–133. See also Banno Junji, “‘Kensei Jōdō’ to ‘Kyōryoku Naikaku,’” in Seitō Naikaku no Seiritsu to Hōkai, ed. Kindai Nihon Kenkyūkai (Tokyo: Yamakawa Shuppan, 1984), 133–134. 6.  Awaya Kentarō, Shōwa no Seitō, vol. 6 of Shōwa no Rekishi (Tokyo: Shōgakukkan, 1988), 49. 7.  Guillermo A. O’Donnell, Modernization and Bureaucratic-Authoritarianism (Berkeley: University of California Press, 1973), 152. See also Alfred Stepan, Rethinking Military Politics: Brazil and the Southern Cone (Princeton, NJ: Prince—ton University Press, 1988), 128. 8.  On the importance of legitimacy, see Juan Linz, Crisis, Breakdown, and Reequilibration: The Breakdown of Democratic Regimes (Baltimore: Johns Hopkins University Press, 1978), 16–23.

Notes to Conclusion  205 9.  On the political history of Brazil in the early 1960s and the collapse of the semi-democratic regime, see Thomas E. Skidmore, Politics in Brazil, 1930–1964: An Experiment in Democracy (New York: Oxford University Press, 1967); Leslie Bethell, “Politics in Brazil under the Liberal Republic, 1945–1985,” in Cambridge History of Latin America. Vol. 9: Brazil since 1930, ed. Leslie Bethell (Cambridge: Cambridge University Press, 2008), 87–164; Alfred Stepan, The Military in Politics: Changing Patterns in Brazil (Princeton, NJ: Princeton University Press, 1971); Alfred Stepan, “Political Leadership and Regime Breakdown: Brazil,” in The Breakdown of Democratic Regimes: Latin America, ed. Juan J. Linz and Alfred Stepan (Baltimore: Johns Hopkins University Press, 1978), 110–137. 10. Stepan, Military in Politics, 149–152. 11. Ibid., 138–151. 12. Ibid., 151. 13.  On the 1991–1992 crisis, see Pasuk Phongpaichit and Chris Baker, Thailand: Economy and Politics, 2nd ed. (New York: Oxford University Press, 2002), 366–384; Suchit Bunbongkarn, “Thailand in 1991: Coping with Military Guardianship,” Asian Survey 32, no. 2 (1992): 131–139; Daniel King, “The Thai Parliamentary Elections of 1992: Return to Democracy in an Atypical Year,” Asian Survey 32, no. 12 (1992): 218–223; Suchit Bunbongkarn, “Thailand in 1992: In Search of a Democratic Order,” Asian Survey 33, no. 2 (1993): 218–223. 14.  Bunbongkarn, “Thailand in 1991,” 132; Phongpaichit and Baker, Thailand: Economy and Politics, 373. 15.  Phongpaichit and Baker, Thailand: Economy and Politics, 368–372. 16.  Paul. H. Handley, The King Never Smiles (New Haven, CT: Yale University Press, 2006), 341–342; King, “Thai Parliamentary Elections of 1992,” 1113. 17.  Bangkok Post, April 13, 1992. 18.  M. Rainer Lepsius, “From Fragmented Party Democracy to Government by Emergency Decree and National Socialist Takeover: Germany,” in The Breakdown of Democratic Regimes: Europe, ed. Juan Linz and Alfred Stepan (Baltimore: Johns Hopkins University Press, 1978), 46. 19. Ibid., 45–49. 20.  Walter B. Simon, “Democracy in the Shadow of Imposed Sovereignty: The First Republic of Austria,” in Linz and Stepan, eds., The Breakdown of Democratic Regimes: Europe, 95. 21. Ibid., 112–114. 22.  Peter H. Smith, “The Breakdown of Democracy in Argentina, 1916–1930,” in Linz and Stepan, eds., The Breakdown of Democratic Regimes: Latin America, 3–27. 23. Ibid., 17–18. 24. Ibid., 17; David Rock, Politics in Argentina, 1890–1930 (Cambridge: Cambridge University Press, 1975), 247.

206  Notes to Conclusion 25.  Robert Dahl, Democracy and Its Critics (New Haven, CT: Yale University Press, 1989), 315. 26.  Dankwart A. Rustow, “Transitions to Democracy: Toward a Dynamic Model,” Comparative Politics 2, no. 3 (1970): 360. 27.  Linz, Crisis, Breakdown, and Reequilibration, 81. 28.  Banno, “‘Kensei Jōdō’ to ‘Kyōryoku Naikaku,’” 183–203. 29.  Seymour Martin Lipset, Political Man, exp. ed. (Baltimore: Johns Hopkins University Press, 1981), 31. 30.  Larry Diamond, “Economic Development and Democracy Reconsidered,” American Behavioral Scientist 35 (March–June 1992): 450–499. 31.  Robert Dahl, Polyarchy: Participation and Opposition (New Haven, CT: Yale University Press, 1971), 67–68. 32.  Samuel Huntington, The Third Wave (Norman: University of Oklahoma Press, 1991), 63. 33.  Ōkawa Kazushi, Takamatsu Nobukiyo, and Yamamoto Yuzō, Kokumin Shotoku, vol. 1 of Chōki Keizai Tōkei, ed. Ōkawa Kazushi, Shinohara Miyohei, and Umemura Mataji (Tokyo: Tōyō Keizai Shimpōsha, 1974); Ōkawa Kazushi et al., Bukka, vol. 8 of Chōki Keizai Tōkei, ed. Ōkawa Kazushi, Shinohara Miyohei, and Umemura Mataji (Tokyo: Tōyō Keizai Shimpōsha, 1966); Nihon Ginkō Chōsa Tōkeikyoku, ed., Economic Statistics Annual (1984) (Tokyo: Nihon Ginkō Chōsa Tōkeikyoku, 1985); Ōsato Katsuma, ed., Meiji Ikō Shuyō Keizai Tōkei (Tokyo: Nihon Ginkō, 1966). 34. Huntington, Third Wave 62. 35.  Lucian Pye, Asian Power and Politics: The Cultural Dimension of Authority (Cambridge: Belknap Press of Harvard University Press, 1985), 340–341. 36.  Robert Scalapino, Democracy and the Party Movement in Prewar Japan (Berkeley: University of California Press, 1953), 130–133, 394–395.

Selected Bibliography

Listed here are only the major writings that have been of use in the making of this book.

Works in English Agüero, Felipe. “Democratic Consolidation and the Military in Southern Europe and South America.” In The Politics of Democratic Consolidation, edited by Richard Gunther, P. Nikiforos Diamandouros, and Hans-Jürgen Puhle, 124–165. Baltimore: Johns Hopkins University Press, 1995. ———. Soldiers, Civilians, and Democracy: Post-Franco Spain in Comparative Perspective. Baltimore: Johns Hopkins University Press, 1995. Alagappa, Muthiah, ed. Coercion and Governance: The Declining Political Role of the Military in Asia. Stanford, CA: Stanford University Press, 2001. Almond, Gabriel A., and Sidney Verba. The Civic Culture: Political Attitudes and Democracy in Five Nations. Princeton, NJ: Princeton University Press, 1963; Newbury Park, CA: Sage, 1989. Alt, James E., and Kenneth Shepsle, eds. Perspectives on Positive Political Economy. Cambridge: Cambridge University Press, 1990. Bailey, Norman A., ed. Latin America: Politics, Economics, and Hemispheric Security. New York: Praeger, 1965. Bailey, Sydney D. Political Parties and the Party System in Britain, reprint ed. Westport, CT: Hyperion Press, 1979. Baker, Chris, and Pasuk Phongpaichit. A History of Thailand. Melbourne: Cambridge University Press, 2005. Ball, Alan R. British Political Parties. London: Macmillan, 1981. Bamrungsuk, Surachart. “Changing Patterns of Civil-Military Relations and 207

208  Selected Bibliography Thailand’s Regional Outlook.” In Civil-Military Relations, edited by David R. Mares, 187–205. Boulder, CO: Westview, 1998. Banno Junji. Democracy in Prewar Japan: Concepts of Government, 1871–1937. Trans. Andrew Fraser. London: Routledge, 2001. ———. The Establishment of the Japanese Constitutional System. Trans. J. A. A. Stockwin. London: Routledge, 1992. Beasley, W. G. Japanese Imperialism, 1894–1945. Oxford: Clarendon, 1987. Beattie, Alan, ed. English Party Politics, vol. 1. London: Weidenfeld and Nicolson, 1970. Beeson, Mark, ed. Contemporary Southeast Asia, 2nd ed. New York: Palgrave Macmillan, 2009. Bentley, Michael. Politics without Democracy 1815–1914. London: Fontana Press, 1984. Berger, Gordon. Parties Out of Power in Japan, 1931–1941. Princeton, NJ: Princeton University Press, 1977. ———. “Politics and Mobilization in Japan, 1931–1945.” In The Cambridge History of Japan. Volume 6: The Twentieth Century, edited by Peter Duus, 97–153. Cambridge: Cambridge University Press, 1988. Bethell, Leslie. “Politics in Brazil under the Liberal Republic, 1945–1985.” In The Cambridge History of Latin America. Volume 9: Brazil since 1930, edited by Leslie Bethell, 87–164. Cambridge: Cambridge University Press, 2008. ———, ed. The Cambridge History of Latin America. Volume 8: 1930 to the Present. Cambridge: Cambridge University Press, 1991. ———, ed. The Cambridge History of Latin America. Volume 9: Brazil since 1930. Cambridge: Cambridge University Press, 2008. Bowornwathana, Bidhya. “Thailand in 1999: A Royal Jubilee, Economic Recovery, and Political Reform.” Asian Survey 40, no. 1 (January–February 2000): 87–97. Brooker, Paul. Non-Democratic Regimes, 2nd ed. London: Palgrave Macmillan, 2009. Bunbongkarn, Suchit. “Thailand in 1991: Coping with Military Guardianship.” Asian Survey 32, no. 2 (1992): 131–139. ———. “Thailand in 1992: In Search of a Democratic Order.” Asian Survey 33, no. 2 (February 1993): 218–223. ———. “Thailand’s Successful Reforms.” Journal of Democracy 10, no. 4 (October 1999): 54–68. Bunce, Valerie, Michael McFaul, and Kathryn Stoner-Weiss. Democracy and Authoritarianism in the Postcommunist World. Cambridge: Cambridge University Press, 2010. Burton, Michael, Richard Gunther, and John Higley. “Introduction: Elite Transformations and Democratic Regimes.” In Elites and Democratic Consolidation in

Selected Bibliography  209 Latin America and Southern Europe, edited by John Higley and Richard Gunther, 1–37. Cambridge: Cambridge University Press, 1992. Callahan, William A., and Duncan McCargo. “Vote-Buying in Thailand’s Northeast.” Asian Survey 36, no. 4 (1996): 376–392. Carothers, Thomas. “The End of the Transition Paradigm.” Journal of Democracy 13, no. 1 (January 2002): 5–21. ———. “The ‘Sequencing’ Fallacy.” Journal of Democracy 18, no. 1 (January 2007): 12–27. Case, William. “Can the ‘Halfway House’ Stand? Semidemocracy and Elite Theory in Three Southeast Asian Countries.” Comparative Politics 28, no. 4 (July 1996): 437–464. ———. Politics in Southeast Asia: Democracy or Less. London: Routledge, 2002. ———. “Thai Democracy, 2001: Out of Equilibrium.” Asian Survey 41, no. 3 (May–June 2001): 525–547. ———, ed. Contemporary Authoritarianism in Southeast Asia: Structures, Institutions and Agency. London: Routledge, 2010. Centre for East Asian Cultural Studies. The Meiji Japan through Contemporary Sources. 3 vols. Tokyo: Centre for East Asian Cultural Studies, 1969–1972. Chai-Anan, Samudavanija. “Thailand: A Stable Semidemocracy.” In Politics in Developing Countries: Comparing Experiences with Democracy, 2nd ed., edited by Larry Diamond, Juan J. Linz, and Seymour Martin Lipset, 323–367. Boulder, CO: Lynne Rienner, 1995. Christensen, Scott R. “Thailand in 1989: Consensus at Bay.” Asian Survey 30, no. 2 (February 1990): 178–186. ———. “Thailand in 1990: Political Tangles.” Asian Survey 31, no. 2 (February 1991): 196–204. Chu, Yun-Han. Crafting Democratization in Taiwan. Taipei: Institute for National Policy Research, 1992. Collier, David, and Steven Levitsky. “Democracy with Adjectives: Conceptual Innovation in Comparative Research.” World Politics 49 (April 1997): 430–451. Connors, Michael K. Democracy and National Identity in Thailand. Copenhagen: Nordic Institute of Asian Studies, 2007. ———. “Liberalism, Authoritarianism and the Politics of Decisionism in Thailand.” In Contemporary Authoritarianism in Southeast Asia: Structures, Institutions and Agency, edited by William Case, 85–103. London: Routledge, 2010. ———. “When the Walls Come Crumbling Down: The Monarchy and Thai-Style Democracy.” Journal of Contemporary Asia 41, no. 4 (November 2011): 657–673. Cook, Chris, and John Stevenson. The Longman Handbook of Modern British History, 1714–1995, 3rd ed. London: Longman, 1996.

210  Selected Bibliography Cox, Gary W. The Efficient Secret: The Cabinet and the Development of Political Parties in Victorian England. Cambridge: Cambridge University Press, 1987. Crowley, James B. Japan’s Quest for Autonomy. Princeton, NJ: Princeton University Press, 1966. Dahl, Robert. Democracy and Its Critics. New Haven, CT: Yale University Press, 1989. ———. Polyarchy: Participation and Opposition. New Haven, CT: Yale University Press, 1971. Diamond, Larry. Developing Democracy. Baltimore: Johns Hopkins University Press, 1999. ———. “Economic Development and Democracy Reconsidered.” American ­Behavioral Scientist 35, no. 4–5 (March–June 1992): 450–499. ———. “Is the Third Wave Over?” Journal of Democracy 7, no. 3 (July 1996): 20–37. ———. The Spirit of Democracy. New York: Times Books, 2008. ———. “Thinking about Hybrid Regimes.” Journal of Democracy 13, no. 2 (April 2002): 21–35. ———, ed. Political Culture and Democracy in Developing Countries. Boulder, CO: Lynne Rienner, 1993. Diamond, Larry, and Byung-Kook Kim, eds. Consolidating Democracy in South Korea. Boulder. CO: Lynne Rienner, 2000. Diamond, Larry, and Juan J. Linz, eds. “Introduction: What Makes for Democracy?” In Politics in Developing Countries: Comparing Experiences with Democracy, 2nd ed., edited by Larry Diamond, Juan J. Linz, and Seymour Martin Lipset, 1–66. Boulder, CO: Lynne Rienner, 1995. ———. Politics in Developing Countries: Comparing Experiences with Democracy, 2nd ed. Boulder, CO: Lynne Rienner, 1995. Diamond, Larry, and Marc F. Platter, eds. Civil-Military Relations and Democracy. Baltimore: Johns Hopkins University Press, 1996. ———. Economic Reform and Democracy. Baltimore: Johns Hopkins University Press, 1995. ———. The Global Resurgence of Democracy, 2nd ed. Baltimore: Johns Hopkins University Press, 1996. Diamond, Larry, Marc F. Platter, and Philip J. Costopoulos, eds. Debates on Democratization. Baltimore: Johns Hopkins University Press, 2010. Diamond, Larry, Marc F. Platter, Yun-han Chu, and Hung-mao Tien, eds. Consolidating the Third Wave Democracies: Themes and Perspectives. Baltimore: Johns Hopkins University Press, 1997. Dogan, Mattei, ed. Comparing Pluralist Democracies. Boulder, CO: Westview, 1988.

Selected Bibliography  211 Dulles, John W. F. Unrest in Brazil Political-Military Crisis, 1955–1964. Austin: University of Texas Press, 2012. Duus, Peter. Party Rivalry and Political Change in Japan. Cambridge, MA: Harvard University Press, 1968. ———. The Rise of Modern Japan. Boston: Houghton Mifflin, 1976. ———, ed. The Cambridge History of Japan. Volume 6: The Twentieth Century. Cambridge: Cambridge University Press, 1988. Duus, Peter, and Daniel I. Okimoto. “Fascism and the History of Pre-War Japan: The Failure of a Concept.” Journal of Asian Studies 39, no. 1 (November 1979): 65–76. Elster, John, and Rune Slagstad, eds. Constitutionalism and Democracy. Cambridge: Cambridge University Press, 1988. Evans, Eric. The Forging of the Modern State: Early Industrial Britain, 1783–1870, 2nd. ed. London: Longman, 1996. Finer, S. E. The Man on Horseback: The Role of the Military in Politics, 2nd enl. ed. Harmondsworth, England: Penguin Books, 1976. Fitch, John Samuel. Armed Forces and Democracy in Latin America. Baltimore: Johns Hopkins University Press, 1998. ———. “Armies and Politics in Latin America: 1975–1985.” In Armies and Politics in Latin America, rev. ed., edited by Abraham F. Lowenthal and J. Samuel Fitch, 26–55. New York: Holmes and Meier, 1986. Fletcher, W. Miles. The Search for a New Order: Intellectuals and Fascism in Prewar Japan. Chapel Hill: University of North Carolina Press, 1982. Garon, Sheldon. The State and Labor in Modern Japan. Berkeley: University of California Press, 1987. Gordon, Andrew. Labor and Imperial Democracy in Prewar Japan. Berkeley: University of California Press, 1991. ———. A Modern History of Japan: From Tokugawa Times to the Present. New York: Oxford University Press, 2009. Gunther, Richard. Public Policy in a No-Party State. Berkeley: University of California Press, 1980. ———. Spain after Franco: The Making of a Competitive Party System. Berkeley: University of California Press, 1986. Gunther, Richard, P. Nikiforos Diamandouros, and Hans-Jürgen Puhle. The Politics of Democratic Consolidation: Southern Europe in Comparative Perspective. Baltimore: Johns Hopkins University Press, 1995. Haggard, Stephen, and Robert R. Kaufman. The Political Economy of Democratic Transitions. Princeton, NJ: Princeton University Press, 1995. Handley, Paul M. The King Never Smiles. New Haven, CT: Yale University Press, 2006.

212  Selected Bibliography Hane, Mikiso. Modern Japan. Boulder, CO: Westview, 1986. Hanham, H. J. Elections and Party Management: Politics in the Time of Disraeli and Gladstone. London: Longmans Green, 1959. Hawgood, John A. “The Evolution of Parties and the Party System: The Nineteenth Century.” In Political Parties and the Party System in Britain, edited by Sydney D. Bailey, 26–34. Westport, CT: Hyperion, 1979. Hewison, Kevin. “The Monarchy and Democratization.” In Political Change in Thailand, edited by Kevin Hewison, 58–74. London: Routledge, 1997. ———, ed. Political Change in Thailand: Democracy and Participation. London: Routledge, 1997. Higley, John, and Richard Gunther, eds. Elites and Democratic Consolidation in Latin America and Southern Europe. Cambridge: Cambridge University Press, 1992. Hill, Brian. The Early Parties and Politics in Britain, 1688–1832. New York: St. Martin’s, 1996. Himmelfarb, Gertrude. “The Politics of Democracy: The English Reform Act of 1867.” Journal of British Studies 6, no. 1 (November 1966): 97–138. Holmes, Geoffrey. The Making of a Great Power: Late Stuart and Early Georgian Britain, 1660–1772. London: Longman, 1993. Holmes, Geoffrey, and Daniel Szechi. The Age of Oligarchy: Pre-industrial Britain, 1722–1783. London: Longman, 1993. Horowitz, Donald L. Coup Theories and Officers’ Motives: Sri Lanka in Comparative Perspective. Princeton, NJ: Princeton University Press, 1980. Huntington, Samuel P. Political Order in Changing Societies. New Haven, CT: Yale University Press, 1968. ———. “Reforming Civil-Military Relations.” Journal of Democracy 6, no. 4 (October 1995): 9–17. ———. The Soldier and the State: The Theory and Politics of Civil-Military Relations. Cambridge, MA: Belknap Press of Harvard University Press, 1957. ———. The Third Wave. Norman: University of Oklahoma Press, 1991. Hyman, Gerald. “Tilting at Straw Men.” Journal of Democracy 13, no. 3 (July 2002): 26–38. Ivarsson, Soren, and Lotte Isager. Saying the Unsayable. Copenhagen: Nordic Institute of Asian Studies, 2010. Janowitz, Morris. The Military in the Political Development of New Nations. Chicago: University of Chicago Press, 1964. Jansen, Marius B. The Making of Modern Japan. Cambridge, MA: Belknap Press of Harvard University Press, 2002. Johnson, John, ed. The Role of the Military in Underdeveloped Countries. Princeton, NJ: Princeton University Press, 1962.

Selected Bibliography  213 Juree, Vichi-Vadakan. “Thailand in 1984: Year of Administering Rumors.” Asian Survey 25, no. 2 (February 1985): 232–240. ———. “Thailand in 1985: Year of Facing up to Facts.” Asian Survey 26, no. 2 (February 1986): 174–185. Karl, Terry Lynn. “The Hybrid Regimes of Central America.” Journal of Democracy 6, no. 3 (July 1995): 72–86. ———. “Imposing Consent? Electoralism vs. Democratization in El Salvador.” In Elections and Democratization in Latin America, 1980–1985, edited by Paul W. Drake and Eduardo Silva, 9–36. San Diego: Center for Iberian and Latin American Studies, Center for U.S.-Mexican Studies, and Institute of the Americas, University of California, San Diego, 1986. Kasza, Gregory. The Conscription Society: Administered Mass Organization. New Haven, CT: Yale University Press, 1995. ———. The State and Mass Media in Japan, 1918–1945. Berkeley: University of California Press, 1988. King, Daniel E. “The Thai Parliamentary Elections of 1992.” Asian Survey 32, no. 12 (December 1992): 1109–1123. ———. “Thailand in 1995: Open Society, Dynamic Economy, Troubled Politics.” Asian Survey 36, no. 2 (February 1996): 135–141. ———. “Thailand in 1996: Economic Slowdown Clouds Year.” Asian Survey 37, no. 2 (February 1997): 160–166. Kuhonta, Erik M. “The Paradox of Thailand’s 1997 ‘People’s Constitution’: Be Careful What You Wish For.” Asian Survey 48, no. 3 (May–June 2008): 373–392. Lakatos, Imre. “Falsification and the Methodology of Scientific Research Programmes.” In Criticism and the Growth of Knowledge, edited by Imre Lakatos and Alan Musgrave, 91–196. Cambridge: Cambridge University Press, 1970. Large, Stephen S. “Oligarchy, Democracy, and Fascism.” In A Companion to Japanese History, edited by William M. Tsutsui, 156–171. Oxford: Blackwell, 2007. ———. Organized Workers and Socialist Politics in Interwar Japan. Cambridge: Cambridge University Press, 1981. Lepsius, M. Rainer. “From Fragmented Party Democracy to Government by Emergency Degree and National Socialist Takeover: Germany.” In The Breakdown of Democratic Regimes: Europe, edited by Juan J. Linz and Alfred Stepan, 34–79. Baltimore: Johns Hopkins University Press, 1978. Levitsky, Steven, and Lucan A. Way. Competitive Authoritarianism: Hybrid Regimes after the Cold War. New York: Cambridge University Press, 2010. ———. “The Rise of Competitive Authoritarianism.” Journal of Democracy 13, no. 2 (April 2002): 51–65. Linz, Juan J. “An Authoritarian Regime: Spain.” In Mass Politics, edited by Erik Allardt and Stein Rokkan, 251–283. New York: Free Press, 1970. First published

214  Selected Bibliography in Cleavages, Ideologies and Party Systems: Contributions to Comparative Political Sociology, edited by Erik Allardt and Yrjö Littunen, 291–341. Helsinki: Academic Bookstore, 1964. ———. The Breakdown of Democratic Regimes: Crisis, Breakdown, and Reequilibration. Baltimore: Johns Hopkins University Press, 1978. ———. “From Falange to Movimiento-Organización: The Spanish Single Party and the Franco Regime, 1936–1968.” In Authoritarian Politics in Modern Society, edited by Samuel P. Huntington and Clement H. Moore, 128–201. New York: Basic Books, 1970. ———. “Legitimacy of Democracy and the Socioeconomic System.” In Comparing Pluralist Democracies, edited by Mattei Dogan, 65–113. Boulder, CO: Westview, 1988. ———. “Opposition In and Under an Authoritarian Regime: The Case of Spain.” In Regimes and Oppositions, edited by Robert Dahl, 171–259. New Haven, CT: Yale University Press, 1973. ———. “Totalitarianism and Authoritarian Regimes.” In Handbook of Political Science, vol. 3, edited by Fred Greenstein and Nelson Polsby, 175–411. Reading, PA: Addison Wesley, 1975. ———. “Transitions to Democracy.” Washington Quarterly 13, no. 3 (Summer 1990): 143–161. Linz, Juan J., and Alfred Stepan, eds. The Breakdown of Democratic Regimes: Europe. Baltimore: Johns Hopkins University Press, 1978. ———, eds. The Breakdown of Democratic Regimes: Latin America. Baltimore: Johns Hopkins University Press, 1978. ———. Problems of Democratic Transition and Consolidation: Southern Europe, South America, and Post-Communist Europe. Baltimore: Johns Hopkins University Press, 1996. Lipset, Seymour Martin. Political Man: The Social Bases of Politics, exp. ed. Baltimore: Johns Hopkins University Press, 1981. ———. “The Social Requisites of Democracy Revisited.” American Sociological Review 59, no. 2 (February 1994): 1–22. Lowenthal, Abraham F., and J. Samuel Fitch., eds. Armies and Politics in Latin America, rev. ed. New York: Holmes & Meier, 1986. Luckham, A. R. “A Comparative Typology of Civil-Military Relations.” Government and Opposition 6, no. 1 (Winter 1971): 5–35. MacDonald, Austin F. Latin American Politics and Government, 2nd ed. New York: Thomas Y. Crowell, 1954. MacDorman, Ted L. “The 1991 Constitution of Thailand.” Pacific Rim Law & Policy Journal 3, no. 2 (February 1995): 258–298. Mackintosh, John P. The British Cabinet, 3rd. ed. London: Stevens & Sons, 1977.

Selected Bibliography  215 Maier, Joseph, and Richard W. Weatherhead, eds. Politics of Change in Latin America. New York: Praeger, 1964. Mainwaring, Scott, Daniel Brinks, and Anibal Perez-Linan. “Classifying Political Regimes in Latin America, 1945–1999.” Studies in Comparative International Development 36, no. 1 (Spring 2001): 37–65. Maisrikrod, Surin, and Duncan McCargo. “Electoral Politics: Commercialization and Exclusion.” In Political Change in Thailand, edited by Kevin Hewison, 132–148. London: Routledge, 1997. Maravall, Jose Maria, and Adam Przeworski, eds. Democracy and the Rule of Law. Cambridge: Cambridge University Press, 2003. Maruyama Masao. “The Ideology and Dynamics of Japanese Fascism.” In idem, Thought and Behaviour in Modern Japanese Politics, edited by Ivan Morris, 25–83. London: Oxford University Press, 1963. ———. Thought and Behaviour in Modern Japanese Politics. Edited by Ivan Morris. London: Oxford University Press, 1963. ———. “Thought and Behaviour Patterns of Japan’s Wartime Leaders.” In idem, Thought and Behaviour in Modern Japanese Politics, edited by Ivan Morris, 84–134. London: Oxford University Press, 1963. McDonough, Peter, Samuel H. Barnes, and Antonio López Pina. “The Growth of Democratic Legitimacy in Spain.” American Political Science Review 80, no. 3 (September 1986): 735–760. Meade, Teresa A. A Brief History of Brazil. New York: Checkmark Books, 2010. Merelman, Richard M. “Learning and Legitimacy.” American Political Science Review 60, no. 3 (September 1966): 548–561. Merkel, Wolfgang, “Embedded and Defective Democracies.” Democratization 11, no. 5. (2004): 33–58. Merkl, Peter H. “Comparing Legitimacy and Values.” In Comparing Pluralist Democracies, edited by Mattei Dogan, 19–63. Boulder, CO: Westview, 1988. Mitani, Taichiro. “The Establishment of Party Cabinets.” In The Cambridge History of Japan. Volume 6: The Twentieth Century, edited by Peter Duus, 55–96. Cambridge: Cambridge University Press, 1988. Moore, Barrington, Jr. Social Origins of Dictatorship and Democracy. Boston: Beacon Press, 1966. Morlino, Leonardo, and José R. Montero. “Legitimacy and Democracy in Southern Europe.” In The Politics of Democratic Consolidation: Southern Europe in Comparative Perspective, edited by Richard Gunther, P. Nikiforos Diamandouros, and Hans-Jürgen Puhle, 231–260. Baltimore: Johns Hopkins University Press, 1995. Muftuler-Bac, Meltem, and E. Fuat Keyman. “The Era of Dominant-Party Politics.” Journal of Democracy 23, no. 1 (January 2012): 85–99.

216  Selected Bibliography Muller, Edward N., and Thomas O. Jukam. “On the Meaning of Political Support.” American Political Science Review 71, no. 4 (December 1977): 1561–1595. Murray, David. “The 1995 National Elections in Thailand: A Step Backward for Democracy?” Asian Survey 36, no. 4 (April 1996): 361–375. Najita, Tetsuo. Hara Kei in the Politics of Compromise, 1905–1915. Cambridge, MA: Harvard University Press, 1967. Najita, Tetsuo, and J. Victor Koschmann, eds. Conflict in Modern Japanese Politics. Princeton, NJ: Princeton University Press, 1987. Needler, Martin. “Military Motivations in the Seizure of Power.” Latin American Research Review 10, no. 3 (Fall 1975): 63–79. ———. Political Systems of Latin America. Princeton, NJ: D. Van Nostrand, 1964. Neher, Clark D. “Political Succession in Thailand.” Asian Survey 32, no. 7 (July 1992): 585–605. ———. “Thailand in 1986: Prem, Parliament, and Political Pragmatism.” Asian Survey 27, no. 2 (February 1987): 219–230. ———. “Thailand in 1987: Semi-Successful Semi-Democracy.” Asian Survey 28, no. 2 (February 1988): 192–201. Nikisch, Larry A. “Thailand in 1980: Confrontation with Vietnam and the Fall of Kriangsak.” Asian Survey 21, no. 2 (February 1981): 223–231. ———. “Thailand in 1981: The Prem Government Feels the Heat.” Asian Survey 22, no. 2 (February 1982): 191–199. Nodia, Ghia. “The Democratic Path.” Journal of Democracy 13, no. 3 (July 2002): 13–19. Ockey, James. “Thailand: The Struggle to Redefine Civil-Military Relations.” In Coercion and Governance: The Declining Political Role of the Military in Asia, edited by Muthiah Alagappa, 187–208. Stanford, CA: Stanford University Press, 2001. O’Donnell, Guillermo A. “Delegative Democracy.” Journal of Democracy 5, no. 1 (January 1994): 55–69. ———. “Illusions about Consolidation.” Journal of Democracy 7, no. 2 (April 1996): 34–51. ———. Modernization and Bureaucratic-Authoritarianism: Studies in South American Politics. Berkeley: Institute of International Studies, University of California, 1973. ———. “Modernization and Military Coups: Theory, Comparisons, and the Argentine Case.” In Armies and Politics in Latin America, rev. ed., edited by Abraham F. Lowenthal and J. Samuel Fitch, 96–133. New York: Holmes and Meier, 1986. O’Donnell, Guillermo, and Philippe C. Schmitter. Transitions from Authoritarian Rule: Tentative Conclusions about Uncertain Democracies. Baltimore: Johns Hopkins University Press, 1986.

Selected Bibliography  217 Ogata, Sadako. Defiance in Manchuria. Berkeley: University of California Press, 1964. O’Gorman, Frank. The Emergence of the British Two-Party System. London: Edward Arnold, 1982. O’Kane, Rosemary H. T. “Probabilistic Approach to the Causes of Coups d’Etat.” British Journal of Political Science 11, no. 3 (July 1981): 287–308. Overholt, William H. “Thailand’s Financial and Political Systems: Crisis and ­Rejuvenation.” Asian Survey 39, no. 6 (November–December 1999): 1009–1035. Paribatra, Sukhumbhand. “State and Society in Thailand: How Fragile the ­Democracy?” Asian Survey 33, no. 9 (September 1993): 879–893. Payne, Stanley. A History of Fascism, 1914–45. Madison: University of Wisconsin Press, 1995. Pearce, Malcolm, and Geoffrey Stewart. British Political History, 1867–1995: ­Democracy and Decline, 2nd ed. London: Routledge, 1996. Peterson, Phyllis. “Brazil: Institutionalized Confusion.” In Political Systems of Latin America, edited by Martin C. Needler, 463–510. Princeton, NJ: Van Nostrand, 1964. Peukert, Detlev J. K. The Weimar Republic. New York: Hill and Wang, 1987. Phongapaichit, Pasuk, and Chris Baker. Thailand: Economy and Politics. New York: Oxford University Press, 2002. Pion-Berlin, David. “The Armed Forces and Politics: Gains and Snares in Recent Scholarship.” Latin American Research Review 30, no. 1 (Winter 1995): 147–162. Potash, Robert A. The Army & Politics in Argentina: 1928–1945. Stanford, CA: Stanford University Press, 1969. Powell, G. Bingham, Jr. Contemporary Democracies: Participation, Stability, and Violence. Cambridge, MA: Harvard University Press, 1982. Przeworski, Adam. “Democracy as a Contingent Outcome of Conflicts.” In Constitutionalism and Democracy, edited by Jon Elster and Rune Slagstad, 59–80. Cambridge: Cambridge University Press, 1988. Przeworski, Adam, Michael Alvarez, José Antonio Cheibub, and Fernando Limongi. Democracy and Development: Political Institutions and Well-Being in the World, 1950–1990. Cambridge: Cambridge University Press, 2000. Pugh, Martin. The Making of Modern British Politics 1867–1939. Oxford: Basil Blackwell, 1982. Punyaratabandhu, Suchitra. “Thailand in 1997: Financial Crisis and Constitutional Reform.” Asian Survey 38, no. 2 (February 1998): 161–167. ———. “Thailand in 1998: A False Sense of Recovery.” Asian Survey 39, no. 1 (January–February 1999): 80–88. Punyaratabandhu-Bhakdi, Suchitra. “Thailand in 1982: General Arthit Takes Center Stage.” Asian Survey 23, no. 2 (February 1983): 172–177.

218  Selected Bibliography ———. “Thailand in 1983: Democracy, Thai Style.” Asian Survey 24, no. 2 (February 1984): 187–194. Putnam, Robert D. Making Democracy Work: Civic Traditions in Modern Italy. Princeton, NJ: Princeton University Press, 1993. ———. “Toward Explaining Military Intervention in Latin America.” World Politics 20, no. 1 (October 1967): 83–110. Pye, Lucian. Asian Power and Politics: The Cultural Dimensions of Authority. Cambridge, MA: Harvard University Press, 1985. Ramsay, Ansil. “Thailand 1978: Kriangsak—The Thai Who Binds.” Asian Survey 19, no. 2 (February 1979): 104–114. ———. “Thailand 1979: A Government in Trouble.” Asian Survey 20, no. 2 (February 1980): 112–122. Ramseyer, J. Mark, and Frances M. Rosenbluth. The Politics of Oligarchy: Institutional Choice in Imperial Japan. Cambridge: Cambridge University Press, 1995. Reischauer, Edwin O. Japan: The Story of a Nation. New York: Knopf, 1970. Remmer, Karen L. “The Political Impact of Economic Crisis in Latin America in the 1980s.” American Political Science Review 85, no. 3 (September 1991): 777–800. Robbins, Keith. The Eclipse of a Great Power: Modern Britain, 1870–1992, 2nd ed. London: Longman, 1994. Rock, David. Politics in Argentina, 1890–1930: The Rise and Fall of Radicalism. Cambridge: Cambridge University Press, 1975. Rosi, Ian Sheffield. “On Pragmatism and Politics in Thailand: Major General Chamlong Srimuang.” Asian Survey 35, no. 11 (November 1995): 1042–1054. Rustow, Dankwart A. “Transitions to Democracy.” Comparative Politics 2, no. 3 (April 1970): 337–364. Samuels, Richard J. Machiavelli’s Children: Leaders and Their Legacies in Italy and Japan. Ithaca, NY: Cornell University Press, 2003. Scalapino, Robert A. Democracy and the Party Movement in Prewar Japan. Berkeley: University of California Press, 1962. Schedler, Andreas, ed. Electoral Authoritarianism: The Dynamics of Unfree Competition. Boulder, CO: Lynne Rienner, 2006. ———. “The Menu of Manipulation.” Journal of Democracy 13, no. 2 (April 2002): 36–50. Schmitter, Philippe C., and Terry Lynn Karl. “What Democracy Is . . . and Is Not.” Journal of Democracy 12, no. 3 (Summer 1991): 75–88. Schneider, Ronald M. “Order and Progress”: A Political History of Brazil. Boulder, CO: Westview, 1991. Shugart, Matthew Soberg, and John M. Carey. Presidents and Assemblies: Constitutional Design and Electoral Dynamics. Cambridge: Cambridge University Press, 1992.

Selected Bibliography  219 Siaroff, Alan. Comparing Political Regimes. Toronto: University of Toronto Press, 2009. Silberman, Bernard. “The Bureaucratic State in Japan: The Problem of Authority and Legitimacy.” In Conflict in Modern Japanese History: The Neglected Tradition, edited by Tetsuo Najita and J. Victor Koschmann, 226–257. Princeton, NJ: Princeton University Press, 1982. Silberman, Bernard S., and H. D. Harootunian, eds. Japan in Crisis: Essays on Taishō Democracy. Princeton, NJ: Princeton University Press, 1974. Simon, Walter B. “Democracy in the Shadow of Imposed Sovereignty: The First Republic of Austria.” In The Breakdown of Democratic Regimes: Europe, edited by Juan J. Linz and Alfred Stepan, 80–121. Baltimore: Johns Hopkins University Press, 1978. Skidmore, Thomas E. Brazil: Five Centuries of Change. New York: Oxford University Press, 2010. ———. Politics in Brazil, 1930–1964: An Experiment in Democracy. New York: Oxford University Press, 1967. Sklar, Richard. “Developmental Democracy.” Comparative Studies in Society and History 29, no. 4 (October 1987): 686–714. Smith, Peter H. Argentina and the Failure of Democracy: Conflict among Political Elites, 1904–1955. Madison: University of Wisconsin Press, 1974. ———. “The Breakdown of Democracy in Argentina, 1916–1930.” In The Breakdown of Democratic Regimes: Latin America, edited by Juan J. Linz and Alfred Stepan, 3–27. Baltimore: Johns Hopkins University Press, 1978. Snitwongse, Kusuma. “Thailand in 1993: Politics of Survival.” Asian Survey 34, no. 2 (February 1994): 147–152. ———. “Thailand in 1994: The Trials of Transition.” Asian Survey 35, no. 2 (February 1995): 194–200. Steinmo, Sven, Kathleen Thelen, and Frank Longstreth, eds. Structuring Politics: Historical Institutionalism in Comparative Analysis. Cambridge: Cambridge University Press, 1992. Stepan, Alfred. The Military in Politics: Changing Patterns in Brazil. Princeton, NJ: Princeton University Press, 1971. ———. “Political Leadership and Regime Breakdown: Brazil.” In The Breakdown of Democratic Regimes: Latin America, edited by Juan J. Linz and Alfred Stepan, 110–137. Baltimore: Johns Hopkins University Press, 1978. ———. Rethinking Military Politics: Brazil and the Southern Cone. Princeton, NJ: Princeton University Press, 1988. Stimson, Henry. The Far Eastern Crisis. New York: Harper and Brothers, 1936. Surachai, Sirikrai. “General Prem Survives on a Conservative Line.” Asian Survey 22, no. 11 (November 1982): 1093–1104.

220  Selected Bibliography Toland, John. The Rising Sun. New York: Random House, 1970. Valenzuela, Arturo. “A Note on the Military and Social Science Theory.” Third World Quarterly 7, no. 1 (January 1985): 132–143. Valenzuela, J. Samuel. “Democratic Consolidation in Post-Transitional Settings: Notion, Process, and Facilitating Conditions.” In Issues in Democratic Consolidation: The New South American Democracies in Comparative Perspective, edited by Scott Mainwaring, Guillermo O’Donnell, and J. Samuel Valenzuela, 57–104. Notre Dame, IN: University of Notre Dame Press, 1992. Van de Walle, Nicolas. “Africa’s Range of Regimes.” Journal of Democracy 13, no. 2 (2002): 66–80. Walker, Patrick Gordon. The Cabinet: Political Authority in Britain. New York: Basic Books, 1970. Ward, Robert E., ed. Political Development in Modern Japan. Princeton, NJ: Princeton University Press, 1968. Weatherford, M. Stephen. “Measuring Political Legitimacy.” American Political Science Review 86, no. 1 (March 1992): 149–166. Webb, R. K. Modern England: From the Eighteenth Century to the Present, 2nd. ed. New York: Harper Collins, 1980. Weber, Max. Economy and Society. Edited by Guenther Roth and Claus Wittich. New York: Bedminster, 1968. Weyland, Kurt. “The Growing Sustainability of Brazil’s Low-Quality Democracy.” In The Third Wave of Democratization in Latin America: Advances and Setbacks, edited by Frances Hagopian and Scott P. Mainwaring, 90–120. Cambridge: Cambridge University Press, 2005. Widmaier, Ulrich. “Tendencies Toward an Erosion of Legitimacy.” In Comparing Pluralist Democracies, edited by Mattei Dogan, 143–167. Boulder, CO: Westview, 1988. Wigell, Mikael. “Mapping ‘Hybrid Regimes’: Regime Types and Concepts in Comparative Politics.” Democratization 15, no. 2 (2008): 230–250. Williams, Mary Wilhelmine, Ruhl J. Bartlett, and Russell E. Miller. The People and Politics of Latin America, 4th ed. Boston: Ginn, 1955. Williamson, Hugh Ross. “The Evolution of Parties and the Party System: The Seventeenth Century.” In Political Parties and the Party System in Britain, edited by Sydney D. Bailey, 13–18. Westport, CT: Hyperion, 1979. Zakaria, Fareed. “The Rise of Illiberal Democracy.” Foreign Affairs 76, no. 6 (November–December 1997): 22–41.

Selected Bibliography  221

Works in Japanese The place of publication, unless otherwise mentioned, is Tokyo. Adachi Mikio. “Kosaku Chotei Hō” [The Tenancy Arbitration Law]. In Nihon Kindaihō Hattatsushi [The History of the Development of Modern Laws in Japan], vol. 7, edited by Ukai Nobushige et al., 37–86. Keisō Shobō, 1959. Amemiya Shōichi. Kindai Nihon no Sensō Shidō [Modern Japan’s Wartime Leadership]. Yoshihara Kōbunkan, 1997. ———. “Tanaka Naikaku: Kensei Jodō Taisei Shoki ni okeru Seiji Handō” [The Tanaka Cabinet: Reactionary Politics in the Early Period of Politics Based on the Constitution]. In Hayashi and Tsuji, eds., Nihon Naikaku Shiroku, vol. 3, 137–176. Andō Yoshio, ed. Ryōtaisenkan no Nihon Shihon Shugi [Japanese Capitalism in the Inter-War Period]. Tōkyō Daigaku Shuppankai, 1979. Asano Kazuo. Taishō Demokurashii to Rikugun [Taishō Democracy and the Army]. Ōta: Kantō Gakuen Daigaku, 1993. Asao Naohiro et al., eds. Iwanami Kōza: Nihon Rekishi [Iwanami Lecture: History of Japan], vols. 15–20. Iwanami Shoten, 1975–1976. Awaya Kentarō. Shōwa no Seitō [Political Parties of the Shōwa Period], vol. 6 of Shōwa no Rekishi [History of the Shōwa Period]. Shōgakukkan, 1988. ———. Tōkyō Saibanron [A Study of the Tokyo Tribunal]. Ōtsuki Shoten, 1989. Baba Nobuya. Manshū Jihen e no Michi [The Road to the Manchurian Incident]. Chūō Kōronsha, 1979. Baba Tsunego. “Gen Naikaku no Jishoku subeki Toki” [The Current Cabinet Should Resign]. Chūō Kōron 43, no. 4 (April 1928): 33–38. Banno Junji. “Gaikōkan no Gokai to Manshū Jihen no Kakudai” [H. Stimpson’s Press Interview and the Escalation of the Manchurian Incident]. Shakai Kagaku Kenkyū 35, no. 5 (February 1984): 45–68. ———. “‘Kensei Jōdō’ to ‘Kyōryoku Naikaku’” [English title: “Political Parties and the Crisis of 1931”]. In Seitō Naikaku no Seiritsu to Hōkai [The Establishment and Collapse of the Party Cabinet System in Prewar Japan], edited by Kindai Nihon Kenkyūkai, 183–203. Yamakawa Shuppan, 1984. ———. Kindai Nihon no Kokka Kosō: 1871–1936 [State Design of Modern Japan: 1871–1936]. Iwanami Shoten, 1997. ———. Kindai Nihon Seijishi [Political History of Modern Japan]. Iwanami Shoten, 2006. ———. Meiji Kenpō Taisei no Kakuritsu [The Establishment of the Meiji Constitutional System]. Tōkyō Daigaku Shuppankai, 1971. ———. Nihon Kensei Shi [Political History of Japan]. Tōkyō Daigaku Shuppankai, 2008.

222  Selected Bibliography ———. “Seitō Naikaku no Hōkai” [The Collapse of Party Government]. In Banno and Miyaji, eds., Nihon Kindaishi ni okeru Tenkanki no Kenkyū, 349–401. ———. “Seitō Seiji no Kakuritsu” [The Establishment of Party Politics]. In Kindai [Modern], vol. 3, edited by Ishii Kanji et al.—vol. 9 of Kōza Nihon Rekishi, edited by Rekishigaku Kenkyūkai and Nihonshi Kenkyūkai, 83–121. ———. Taishō Seihen [The Political Crisis in the Taishō Period]. Kyoto: Mineruba Shobō, 1994. Banno Junji and Miyaji Masato, eds. Nihon Kindaishi ni okeru Tenkanki no Kenkyū [A Study of the Transition Periods in the Modern History of Japan]. Yamakawa Shuppansha, 1985. Chō Yukio. “Shōwa Kyōkō 2” [The Shōwa Great Depression, Part 2]. In Sumiya, ed., Shōwa Kyōkō: Sono Rekishiteki Igi to Zentaizō, 197–245. Eguchi Keiichi. Jūgonen Sensō no Kaimaku [The Opening of the Fifteen Years War], vol. 4 of Shōwa no Rekishi [History of the Shōwa Period]. Shōgakukkan, 1982. Fujita Tsukugo. Meiji Gunsei [Military Institutions in the Meiji Period]. Shinzansha, 1992. Fukayama Kiichirō. “Daiichiji Sekai Taisengo no Wagakuni ni okeru Rōdō ­Kumiaihoan no Tenkai” [Bills on Labor Unions in Japan after the First World War]. In Takahashi, ed., Nihon Kindaika no Kenkyū, vol. 2, 89–116. Funaki Shigeru. Okamura Yasuji [Okamura Yasuji]. Kawaide Shobō Shinsha, 1984. Furuya Tetsuo. “Kinkaikin, Rondon Jōyaku, Manshū Jihen” [The Return to the Gold Standard, the London Naval Treaty, and the Manchurian Incident]. In Uchida, Kinbara, and Furuya, eds., Nihon Gikai Shiroku, vol. 3, 65–125. Futamura Kazuo. “Rōdōsha Kaikyū no Jōtai to Rōdō Undō” [The Condition of the Working Class and the Labor Movement]. In Asao et al., eds., Iwanami Kōza, vol. 18, 93–140. Hamaguchi Osachi. Nikki Zuikanroku [Diary and Memoirs]. Misuzu Shobō, 1991. Hara Akira. “Keiki Junkan” [Economic Cycles]. In Sekai Daikyōkōki [The Worldwide Great Depression], vol. 2 of Ōishi, ed., Nihon Teikoku Shugishi, 367–410. Hara Takashi. Hara Takashi Nikki [Diary of Hara Takashi], vols. 4 and 5. Edited by Hara Keiichirō. Fukumura Shuppan, 1965. Harada Kumao. Saionji Kō to Seikyoku [Prince Saionji and Politics]. 9 vols. Iwanami Shoten, 1950. Hata Ikuhiko. Gun Fashizumu Undōshi [History of Military Fascist Movements]. Kawaide Shobō, 1962. ———. Shōwashi no Nazo o ou [Exploring the Enigma of the Shōwa Period], vol. 1. Bungei Shunjūsha, 1993. Hayashi Shigeru and Tsuji Kiyoaki, eds. Nihon Naikaku Shiroku [History of Japanese Cabinets], vols. 2–4. Daiichi Hōki Shuppan, 1981. Hayashi Yūichi. “Kaikyū no Seiritsu to Chiiki Shakai” [The Establishment of Class

Selected Bibliography  223 and Regional Society]. In Gendai Shakai e no Tenkei [Transformation into a Modern Society], vol. 3 of Nihon Gendaishi: Kōzō to Hendō [History of Modern Japan: Structure and Change], edited by Banno Junji et al., 25–64. Iwanami Shoten, 1993. Hayashi Yūichi and Yasuda Hiroshi. “Shakai Undō no Shosō” [Various Dimensions of Social Movements]. In Rekishigaku Kenkyūkai and Nihonshi Kenkyūkai, eds., Kōza Nihon Rekishi, vol. 9, 171–219. Ikei Masaru. Nihon Gaikōshi Gaisetsu [A Concise Diplomatic History of Japan], rev. ed. Keiō Tsūshin, 1983. Inaba Masao et al., eds. Shiryōhen [Historical Materials], Supplement to Taiheiyō Sensō e no Michi [The Road to the Pacific War]. Asahi Shimbunsha, 1963. Inomata Tsunao. Kyubō no Nōson [Villages in Distress]. Kaizōsha, 1934. Inoue Toshikazu. Seiyūkai to Minseitō [The Seiyūkai and the Minseitō]. Chūō Kōron Shinsha, 2012. Inumaru Giichi. “Han Fashizumu Undō to sono Kaitai” [The Anti-Fascist Movement and Its Dissolution]. In Asao et al., eds., Iwanami Kōza, vol. 20, 171–216. Iokibe Makoto. “Rikugun ni yoru Seiji Shihai” [Political Rule by the Army]. In Tairiku Shinkō to Senji Taisei [Invasions of the Continent and the Wartime System], vol. 2 of Shōwaki no Gunbu to Seiji [The Military and Politics in the Shōwa Period], edited by Miyake Masaki et al., 3–56. Daiichi Hōki Shuppan, 1983. Ishibashi Tanzan. “Higōhō Keikō iyoiyo Shinkokuka sento su” [Politics Are Becoming More and More Illegitimate]. In Ishibashi Tanzan Zenshū, vol. 8, 33–37. ———. “Inoue Zōsho no Kyuheika Kaikinron o hyōsu” [Evaluating the Return to the Gold Standard on the Old Parity by Finance Minister Inoue]. In Ishibashi Tanzan Zenshū, vol. 7, 72–104. ———. “Kinrai no Sesō Tadagoto narazu” [The Atmosphere of the World Has Recently Become Extraordinary]. In Ishibashi Tanzan Zenshū, vol. 8, 7–10. ———. “Konki Gikai no Kōseki” [Accomplishments of the Diet This Session]. In Ishibashi Tanzan Zenshū, vol. 5, 77–79. ———. “Nōkōka suru Seihenrai no Yōsō” [A Great Prospect for Change in the Government]. In Ishibashi Tanzan Zenshū, vol. 7, 424–427. ———. “Seitō Seiji no Kakuritsu to Futsū Senkyo no Yōbō” [The Establishment of Party Politics and the Request for Implementation of Elections Based on Universal Male Suffrage]. In Ishibashi Tanzan Zenshū, vol. 3, 3–7. ———. “Shidō Kaikyu no otoshiireru Zetsudai no Kiken Shisō” [The Dangerous Ideology Which Traps the Leaders]. In Ishibashi Tanzan Zenshū, vol. 8, 14–18. ———. “Shugiin o Kaisan subeshi” [I Demand the Dissolution of the Lower House]. In Ishibashi Tanzan Zenshū, vol. 5, 101–103. ———. “Shushō Sōnan no Kōnin: Kairyō o yosuru Gikaiseido” [The Successor

224  Selected Bibliography after the Attempted Assassination of the Prime Minister: The Need to Change the Parliamentary System]. In Ishibashi Tanzan Zenshū, vol. 7, 407–410. Ishibashi Tanzan Zenshū [Collected Works of Ishibashi Tanzan], vols. 3, 5, 7, 8. Edited by Ishibashi Tanzan Zenshū Hensan Iinkai. Tōyō Keizai Shimpōsha, 1971. Ishigami Ryohei. Seitōshi Ron: Hara Takashi Botsugo [A Study of the History of ­Political Parties after Hara Takashi’s Death]. Chūō Kōronsha, 1960. Itō Takashi. “Fashizumu Ronsō sonogo” [The “Fascism Debate” Revisited]. In Kindai Nihon Kenkyū no Kentō to Kadai [Themes in Modern Japan Studies], edited by Kindai Nihon Kenkyūkai, 310–323. Yamakawa Shuppansha, 1988. ———. “Shōwa Seijishi Kenkyū e no Ichi Shikaku [A Perspective on the Study of the Political History of the Shōwa Period]. Shisō, no. 624 (June 1976): 215–228. ———. Shōwa Shoki Seijishi Kenkyū [Study of the Political History of Early Shōwa Period]. Tōkyō Daigaku Shuppankai, 1969. ———, ed. Taishō Shoki Yamagata Aritomo Danwa Hikki: Seihen Omoidegusa [­Records of a Monologue by Yamagata Aritomo in the early Taishō Period and a Memorandum of Changes in the Government]. Yamakawa Shuppansha, 1981. Itō Yukio. Nichiro Sensō e no Seiji Katei [The Political Process Leading to the Russo-Japanese War]. In Nihon Kindai no Keisei to Hatten [Formation and Development of Modern Japan], edited by Yamamoto Shirō, 227–288. Yoshikawa Kōbunkan, 1996. ———. Taishō Demokurashii to Seitō Seiji [Taishō Democracy and Party Politics]. Yamakawa Shuppansha, 1987. Kawahara Hiroshi. “Hamaguchi Naikaku: Seitō Seiji no Eikō to Higeki” [Hamaguchi Cabinet: The Glory and Tragedy of Party Politics]. In Hayashi and Tsuji, eds., Nihon Naikaku Shiroku, vol. 3, 177–219. ———. “Kiyoura Naikaku: Saigo no Chozen Naikaku” [The Kiyoura Cabinet: The Last Transcendentalist Cabinet]. In Hayashi and Tsuji, eds., Nihon Naikaku Shiroku, vol. 3, 1–23. Kawato Sadafumi. Nihon no Seitō Seiji, 1890–1937 [Party Politics of Japan, 1890– 1937]. Tōkyō Daigaku Shuppankai, 1992. Kido Kōichi. Kido Kōichi Nikki [Diary of Kido Kōichi], vol. 1. Edited by Kido Nikki Kenkyūkai. Tōkyō Daigaku Shuppankai, 1966. Kindai Nihon Kenkyū Kai, ed. Shōwaki no Gunbu [The Military in the Shōwa Period]. Yamakawa Shuppansha, 1979. Kisaka Jun’ichirō. “Gunbu to Demokurashii” [English title: “The Military Clique and Taishō Democracy”]. In Heiwa to Sensō no Kenkyū [A Study of Peace and War], vol. 2, edited by Nihon Kokusai Seiji Gakkai, 1–41. Yūhikaku, 1969. ———. “Taisei Yokusankai no Seiritsu” [The Formation of the Imperial Rule Assistance Association]. In Asao el al., eds., Iwanami Kōza, vol. 20, 269–314.

Selected Bibliography  225 Kitaoka Shin’ichi. “Rikugun Habatsu Tairitsu (1931–1935) no Saikentō” [A Reexamination of Factional Conflict within the Army in Terms of Foreign and National Defense Policy, 1931–1935]. In Kindai Nihon Kenkyukai ed., Shōwaki no Gunbu, 44–95. ———. Seitō kara Gunbu e 1924–41 [From Political Parties to the Military, 1924– 41], vol. 5 of Nihon no Kindai [Modern Japan]. Chūō Kōron Shinsha, 1999. Kobayashi Tatsuo. “Kaigun Gunshuku Jōyaku” [The London Naval Treaty]. In Nihon Kokusai Seiji Gakkai, ed., Manshū Jihen Zennya [The Eve of the Manchurian Incident], vol. 1 of Taiheiyō Sensō e no Michi, 3–160. ———, ed. Suiusō Nikki [Suiusō Diary]. Hara Shobō, 1966. Kobayashi Tatsuo and Shimada Toshihiko, eds. Manshū Jihen [The Manchurian Incident], vol. 7 of Gendaishi Shiryō [Materials for Modern History]. Misuzu Shobō, 1964. Kobe Ryōichi. Gyakusetsu no Gundai [A Paradoxical Army], vol. 9 of Nihon no Kindai [Modern Japan]. Chūō Kōron Shinsha, 1998. Kōketsu Atsushi. Sōryokusen Taisei Kenkyū [Study of the Total War System]. San’ichi Shobō, 1981. Kōmoto Daisaku. “Watashi ga Chō Sakurin o koroshita” [I Killed Chang Tso-lin]. In “Bungei Shunjū” ni miru Shōwashi [History of the Shōwa Period Seen through the Bungei Shunjū], vol. 1, edited by Bungei Shunjū, 44–52. Bungei Shunjū, 1988. Kyo Seikai. “Boku Retsu Jiken” [The Pak Yŏl Scandal]. In Wakatsuma, ed., Nihon Seiji Saiban Shiroku: Taishō, 379–411. Maeda Renzan. Hara Takashi Den [Biography of Hara Takashi]. Takayama Shoin, 1943. Masuda Tomoko. “Seitō Naikaku to Sūmitsuin” [English title: “Party Cabinets and the Privy Council: Disfunctioning of the Meiji Constitutional System”]. In Seitō Naikaku no Seiritsu to Hōkai [The Establishment and Collapse of the Party Cabinet System in Prewar Japan], edited by Kindai Nihon Kenkyūkai, 143–181. Yamakawa Shuppan, 1984. Masumi Junnosuke. Nihon Seijishi [Political History of Japan], vol. 2. Tōkyō Daigaku Shuppankai, 1988. ———. Nihon Seitō Shi Ron [History of Political Parties in Japan], 7 vols. Tōkyō Daigaku Shuppankai, 1965–1980. Matsumoto Gōkichi. Taishō Demokurashii Ki no Seiji: Matsumoto Gōkichi Nikki [Politics under Taishō Democracy: The Diary of Matsumoto Gōkichi]. Edited by Oka Yoshitake and Hayashi Shigeru. Iwanami Shoten, 1959. Matsumoto Seichō. Shōwashi Hakkutsu [Excavation of the History of the Shōwa Period], vols. 1 and 4. Bungei Shunjū Shinsha, 1965 and 1966. Matsuo Takayoshi. Futsū Senkyo Seido Seiritushi no Kenkyū [A Study of the History of the Introduction of Universal Suffrage]. Iwanami Shoten, 1989.

226  Selected Bibliography Matsushita Yoshio. Meiji Gunseishi Ron [A Study of the Military System in the Meiji Period]. Yūhikaku, 1956. Mikuriya Takashi. “Kokusaku Tōgō Kikan Secchi Mondai no shiteki Tenkai” [English title: “The Historical Development of the Problems Surrounding the Establishment of the Basic State Policy Integrating Organ—The Political Dynamics Leading to the Founding of the Cabinet Planning Board”]. In Kindai Nihon Kenkyūkai, ed., Shōwaki no Gunbu, 122–172. Minobe Tatsukichi. “Kyokoku Icchi Naikaku no Seiritsu” [The Establishment of the All Nation Cabinet]. Chūō Kōron 47, no. 7 (July 1932): 30–31. ———. “Seigi ni motozuku Seiji” [Politics Based on Justice]. Kaizō 6, no. 7 (July 1924): 112–113. Mitani Taichirō. Nihon Seitō Seiji no Keisei [The Formation of Party Politics in Japan], exp. ed. Tōkyō Daigaku Shuppankai, 1995. Mitearai Tatsuo. Minami Jirō [Biography of Minami Jirō]. Minami Jirō Denki Kankōkai, 1957. Miwa Ryōichi. “Rōdō Kumiaiho Seitei Mondai no rekishiteki Ichi” [The Labor Union Bills and Their Historical Significance]. In Andō, ed., Ryōtaisenkan no Nihon Shihon Shugi, 237–288. Miyake Masaki. “Dainiji Konoe Naikaku: Gensō to Zasetsu” [The Second Konoe Cabinet: Illusion and Setback]. In Hayashi and Tsuji, eds., Nihon Naikaku Shiroku, vol. 4, 201–276. Miyamoto Moritarō. “Dainiji Wakatsuki Naikaku: Manshū Jihen to Minseitō Naikaku no Kunō” [The Second Wakatsuki Cabinet: Agonies of a Minseitō Cabinet]. In Hayashi and Tsuji, eds., Nihon Naikaku Shiroku, vol. 3, 221–249. ———. “Hirota Naikaku: Jun-Senji Taisei no Kakuritsu” [The Hirota Cabinet: The Establishment of the Semi-Wartime System]. In Hayashi and Tsuji eds., Nihon Naikaku Shiroku, 393–419. Miyazaki Ryuji. “Senzen Nihon no Seiji Hatten to Rengō Seiji” [Political Development in Prewar Japan and Coalition Politics]. In Rengō Seiji [Coalition Politics], vol. 1, edited by Shinohara Hajime, 197–256. Iwanami Shoten, 1984. ———. “Taishō Demokurashii Ki no Nōson to Seitō” [Changes in Rural Communities and Their Effects on Party Politics under Taishō Democracy]. Kokka Gakkai Zasshi 93, nos. 7, 8, 9, 10, 11, 12 (July, September, November, 1980): 445–511, 693–750, 855–923. Morishima Gorō and Yanai Hisao, eds. Manshū Jihen [The Manchurian Incident], vol. 18 of Nihon Gaikōshi [Diplomatic History of Japan]. Kashima Kenkyūjo Shuppankai, 1973. Murai Ryōta. Seitō Naikakusei no Seiritsu [Establishment of Party Government]. Yūhikaku, 2005.

Selected Bibliography  227 Nagai Kazu. Kindai Nihon no Gunbu to Seiji [The Military and Politics in Modern Japan]. Kyoto: Shibunkaku, 1993. Nagaoka Shinkichi, ed. Kindai Nihon no Keizai: Tōkei to Gaisetu [The Economy of Modern Japan: Statistics and a Brief Study]. Kyoto: Mineruba Shobō, 1988. Naikaku Tōkeikyoku, ed. Nihon Teikoku Tōkei Nenkan [Annual Statistics of the Japanese Empire], vol. 48. Naikaku Tōkeikyoku, 1929. Nakamura Masanori. “Daikyōkō to Nōson Mondai” [The Great Depression and Problems in Rural Areas]. In Asao et al., eds., Iwanami Kōza, vol. 19, 135–186. ———. Shōwa no Kyōkō [The Shōwa Depression], vol. 2 of Shōwa no Rekishi [­History of the Shōwa Period]. Shōgakukkan, 1988. Nakamura Takafusa. Nihon Keizai: Sono Seichō to Kozō [The Japanese Economy: Its Development and Structure]. Tōkyō Daigaku Shuppankai, 1978. ———. Shōwashi, 1926–1945 [History of the Shōwa Period, 1926–1945]. Tōyō Keizai Shimpōsha, 1993. Nihon Kindaishiryō Kenkyūkai and Kido Nikki Kenkyūkai, eds. Suzuki Teiichi shi Danwa Sokkiroku [Stenographic Records of Mr. Suzuki Teiichi’s Talk], 2 vols. Nihon Kindaishiryō Kenkyūkai. Nihon Kokusai Seiji Gakkai. Taiheiyō Sensō e no Michi [The Road to the Pacific War], vols. 1 and 2. Asahi Shimbunsha, 1962 and 1963. Nihon Tōkei Kenkyūjo, ed. Nihon Keizai Tōkeishū: Meiji, Taishō, Shōwa [Statistics on the Japanese Economy: Meiji, Taishō, and Shōwa Periods]. Nihon Hyōron Shinsha, 1958. Nishida Miaki. “Nōmin Undō no Hatten to Jinushisei” [The Development of the Farmers Movement and the Landlord System]. In Asao et al., eds., Iwanami Kōza, vol. 18, 141–181. ———. “Nōmin Undō to Nōgyō Seisaku” [The Farmers Movement and Agricultural Policies]. In Sekai Daikyōkōki [The Worldwide Great Depression], vol. 2 of Ōishi, ed., Nihon Teikoku Shugishi, 295–330. Nōmin Kumiai Gojūshunen Kinensai Jikkō Iinkai, ed. Nōmin Kumiai Gojūnenshi [History of the Farmers Union over the Past Fifty Years]. Ochanomizu Shobō, 1972. Ōe Shinobu. Tennō no Guntai [The Emperor’s Troops], vol. 3 of Shōwa no Rekishi [History of the Shōwa Period]. Shōgakukkan, 1982. Ogata Sadako. Manshū Jihen to Seisaku no Keisei Katei [The Process of Policy Formulation in the Manchurian Incident]. Hara Shobō, 1965. Ōishi Kaichirō, ed. Nihon Teikoku Shugishi [History of Japanese Imperialism], vols. 1 and 2. Tōkyō Daigaku Shuppankai, 1985 and 1987. Oka Yoshitake. Tenkanki no Taishō [The Taishō Period as a Period of Transformation], vol. 3 of Oka Yoshitake Senshū [Selected Works of Oka Yoshitake]. Iwanami Shoten, 1992.

228  Selected Bibliography Oka Yoshitake et al., eds. Ogawa Heikichi Kankei Bunsho [Documents Related to Ogawa Heikichi], 2 vols. Misuzu Shobō, 1973. Okada Keisuke. “Okada Keisuke Nikki” [Diary of Okada Keisuke]. In Kobayashi and Shimada, eds., Manshū Jihen, 3–34. Ōkawa Kazushi, Takamatsu Nobukiyo, and Yamamoto Yuzō. Kokumin Shotoku [National Income], vol. 1 of Chōki Keizai Tōkei [Estimates of Long-Term Economic Statistics of Japan since 1868], edited by Ōkawa Kazushi, Shinohara Miyohei, and Umemura Mataji. Tōyō Keizai Shimpōsha, 1974. Ōkawa Kazushi et al. Bukka [Prices], vol. 8 of Chōki Keizai Tōkei [Estimates of Long-Term Economic Statistics of Japan since 1868], edited by Ōkawa Kazushi, Shinohara Miyohei, and Umemura Mataji. Tōyō Keizai Shimpōsha, 1966. Ōkōchi Kazuo and Matsuo Hiroshi. Nihon Rōdō Kumiai Monogatari: Shōwa [The Story of Japanese Labor Unions: The Shōwa Period]. Chikuma Shobō, 1965. ———. Nihon Rōdō Kumiai Monogatari: Taishō [The Story of Japanese Labor Unions: The Taishō Period]. Chikuma Shobō, 1965. Ōkubo Toshiaki, ed. Daiichiji Sekai Taisen to Seitō Naikaku [The First World War and Party Government], vol. 16 of Nihon Rekishi Taikei [Comprehensive History of Japan], edited by Inoue Mitsusada et al. Yamakawa Shuppansha, 1996. ———, ed. Kakushin to Sensō no Jidai [The Era of Reformation and War], vol. 17 of Nihon Rekishi Taikei [Comprehensive History of Japan], edited by Inoue Mitsusada et al. Yamakawa Shuppansha, 1997. ———, ed. Meiji Kenpō Taisei no Tenkai [The Evolution of the Meiji Constitutional System], vols. 1 and 2—vols. 14 and 15 of Nihon Rekishi Taikei [Comprehensive History of Japan], edited by Inoue Mitsusada et al. Yamakawa Shuppansha, 1996. ———, ed. Meiji Kokka no Seiritsu [The Foundation of the Meiji State], vol. 13 of Nihon Rekishi Taikei [Comprehensive History of Japan], edited by Inoue Mitsusada et al. Yamakawa Shuppansha, 1996. ———, ed. Seijishi [Political History], vol. 3 of Taikei Nihonshi Sōsho [Comprehensive History of Japan], edited by Andō Yoshio et al. Tokyo: Yamakawa Shuppansha, 1967. Ōkurashō, ed. Meiji Taishō Zaiseishi [Fiscal History of the Meiji and Taishō Periods], vol. 6. Keizai Ōraisha, 1957. ———, ed. Saikei [The Budget], vol. 3 of Shōwa Zaiseishi [Fiscal History of the Shōwa Period]. Tōyō Keizai Shimpōsha, 1955. Ōsato Katsuma, ed. Meiji Ikō Shuyō Keizai Tōkei [One Hundred Years of Statistics on the Japanese Economy]. Nihon Ginkō, 1966. Ōshima Mitsuko. “Matsushima Yūkaku Iten Mondai” [The Matsushima Red-Light District Relocation Scandal]. In Wagatsuma et al., eds., Nihon Seiji Saiban Shiroku: Shōwa Zen, 97–122. Ōshima Tarō. “Kunsho Tetsudō Gigoku Jiken: Seitō Seiji ni okeru Oshoku no

Selected Bibliography  229 Rotei” [Decorations and Railway Scandals: Exposure of Corruption in Party Politics]. In Wagatsuma et al., eds., Nihon Seiji Saiban Shiroku: Shōwa Zen, 305–358. Rekishigaku Kenkyūkai, ed. Kindai [Modern Times], vol. 4 of Nihonshi Shiryō [Materials for the History of Japan]. Iwanami Shoten, 1997. ———, ed. Manshū Jihen [The Manchurian Incident], vol. 1 of Taiheiyō Sensōshi [History of the Pacific War]. Aoki Shoten, 1971. Rekishigaku Kenkyūkai and Nihonshi Kenkyūkai, eds. Kōza Nihon Rekishi [Lecture: History of Japan], vols. 8–10. Tōkyō Daigaku Shuppankai, 1985. Sagōya Yoshiaki. “Naze Hamaguchi Shushō o koroshita ka” [Why Did I Assassinate Prime Minister Hamaguchi?]. In Shōwa Kyōkō [The Shōwa Depression], vol. 2 of Mokugekisha ga kataru Shōwashi [History of Shōwa Told by Witnesses], edited by Inose Naoki and Yamazaki Hiroshi, 209–220. Shin Jimbutsu Ōraisha, 1989. Sakai Tetsuya. Taishō Demokurashii Taisei no Hōkai: Naisei to Gaikō [The Collapse of the Taishō Democratic Regime]. Tōkyō Daigaku Shuppankai, 1992. Satogami Ryōhei. “Taishō Demokurashii to Kizokuin” [Taishō Democracy and the House of Peers]. In Taishōki no Seiji to Shakai [Politics and Society in the Taishō Period], edited by Inoue Kiyoshi, 251–279. Iwanami Shoten, 1969. Seki Hiroharu. “Manshū Jihen Zenshi” [History before the Manchurian Incident]. In Nihon Kokusai Seiji Gakkai, ed., Manshū Jihen Zennya [The Eve of the Manchurian Incident], vol. 1 of Taiheiyō Sensō e no Michi, 287–440. Shimada Toshihiko. “Manshū Jihen no Tenkai” [The Development of the Manchurian Incident]. In Nihon Kokusai Seiji Gakkai, ed., Manshū Jihen [The Manchurian Incident], vol. 2 of Taiheiyō Sensō e no Michi, 3–188. Shimizu Yōji. “Nōgyō Kyōkō” [The Depression in Agriculture]. In Sekai Daikyōkōki [The Worldwide Great Depression], vol. 2 of Ōishi, ed., Nihon T­eikoku Shugishi, 255–294. Shinobu Seizaburō. Taishō Demokurashii Shi [History of Taishō Democracy], 3 vols. Nihon Hyōron Shinsha, 1959. Shinohara Miyohei. Kōkōgyō [Mining and Manufacturing], vol. 10 of Chōki Keizai Tōkei [Estimates of Long-Term Economic Statistics of Japan since 1868], edited by Ōkawa Kazushi, Shinohara Miyohei, and Umemura Mataji. Tōyō Keizai Shimpōsha, 1973. Sumiya Mikio, ed. Shōwa Kyōkō: Sono Rekishiteki Igi to Zentaizō [The Shōwa Great Depression: Its Significance and Whole Image]. Yūhikaku, 1974. Suzuki Bunji. “Kinrō Musansha no Seikatsu Antei o” [To Stabilize the Lives of Laborers without Property]. Kaizō 10, no. 3 (March 1928): 102–104. ———. “Rōdōsha no Tachiba yori Hara Shin Naikaku ni nozomu” [Request to the Hara Cabinet from Labor]. In Yūaikai Kikanshi: Rōdō oyobi Sangyō [Labor and Industry], vol. 8, edited by Hōsei Daigaku Ōhara Shakai Mondai

230  Selected Bibliography Kenkyūjo and Sōdōmei Gojūnenshi Kankō Iinkai, 238–241. Hōsei Daigaku Shuppankyoku, 1977. Takahashi Hidenao. “Hara Takashi Naikaku ka no Gikai” [The Diet under the Hara Cabinet]. In Uchida, Kinbara, and Furuya, eds., Nihon Gikai Shiroku, vol. 2, 189–260. ———. “Rikugun Gunshuku no Zaisei to Seiji” [Finance and Politics of the Reduction of the Army: Civil-Military Relations in the Period of Party Maturity]. In Kanryōsei no Keisei to Tenkai [The Formation and Development of the Bureaucratic System], edited by Kindai Nihon Kenkyūkai, 141–183. Yamakawa Shuppansha, 1986. Takahashi Kamekichi. “Seisansha Hon’i no Kosakuhō Seitei no tameni” [For the Tenancy Law for the Producers]. Kaizō 10, no. 3 (March 1928). Takahashi Kohachirō, ed. Nihon Kindaika no Kenkyū [Study on the Modernization of Japan], 2 vols. Tōkyō Daigaku Shuppankai, 1971. Takahashi Masae. Shōwa no Gunbatsu [Military Cliques of the Shōwa Period]. Chūō Kōronsha, 1969. Takahashi Susumu and Miyazaki Ryūji. “Seitō Seiji no Teichaku to Hōkai” [The Consolidation and Collapse of Party Politics]. In Banno and Miyaji, eds., Nihon Kindaishi ni okeru Tenkanki no Kenkyū, 225–256. Takeda Haruhito. “Rōshi Kankei” [The Labor and Employer Relationship]. In Daiichiji Sekai Taisenki [During the First World War], vol. 1 of Ōishi, ed., Nihon Teikoku Shugishi, 273–310. Takeyama Morio. “Yonai Naikaku: Ishoku no Dai Yonji Senji Naikaku” [The ­Yonai Cabinet: Unique Fourth War Cabinet]. In Hayashi and Tsuji eds., Nihon Naikaku Shiroku, vol. 4, 133–200. Tanaka Giichi Denki Kankōkai. Tanaka Giichi Den [Biography of Tanaka Giichi], 2 vols. Tanaka Giichi Denki Kankōkai, 1960. Teruoka Shozō. Nihon Nōgyō Mondai no Tenkai [The Evolution of Problems in Japanese Agriculture], 2 vols. Tōkyō Daigaku Shuppankai, 1970. Toriumi Yasushi. “Hara Naikaku: ‘Jun Seitō Naikaku’ no Meian” [The Hara Cabinet: The Fate of a Pure Party Cabinet]. In Hayashi and Tsuji, eds., Nihon Naikaku Shiroku, vol. 2, 285–340. Tsutsui Kiyotada. Shōwaki Nihon no Kozō [Structure of Japan in the Shōwa Period]. Yūhikaku, 1984. Uchida Kenzō, Kinbara Samon, and Furuya Tetsuo, eds. Nihon Gikai Shiroku [History of the Japanese Diet], vols. 2–3. Daiichi Hōki Shuppan, 1992. Uehara Yūsaku Kankei Bunsho Kenkyūkai, ed. Uehara Yūsaku Kankei Bunsho [Documents Related to Uehara Yūsaku]. Tōkyō Daigaku Shuppankai, 1976. Ugaki Kazushige. Ugaki Kazushige Nikki [Diary of Ugaki Kazushige], vols. 1 and 2. Edited by Tsunoda Jun. Misuzu Shobō, 1968, 1970.

Selected Bibliography  231 Ukai Nobushige et al., eds. Nihon Kindaihō Hattatsushi [The History of the Development of Modern Laws in Japan], vol. 7. Keisō Shobō, 1959. Umemura Mataji, et al. Rōdōryoku [Manpower], vol. 2 of Chōki Keizai Tōkei [Estimates of Long-Term Economic Statistics of Japan since 1868], edited by Ōkawa Kazushi, Shinohara Miyohei, and Umemura Mataji. Tōyō Keizai Shimpōsha, 1988. Utsumi Tōru. “Seihen Omoidegusa” [Memories of Political Change]. In Itō ­Takashi, ed., Taishō Shoki Yamagata Aritomo Danwa Hikki: Seihen Omoidegusa, 145–180. Wagatsuma Sakae et al., eds. Nihon Seiji Saiban Shiroku: Shōwa Zen [History of Political Trials in Japan: The First Half of the Shōwa Period]. Daiichi Hōki Shuppan, 1970. ———. Nihon Seiji Saiban Shiroku: Taishō [History of Political Trials in Japan: The Taishō Period]. Daiichi Hōki Shuppan, 1969. Wakatsuki Reijirō. Kofūan Kaikoroku [Memoir of Wakatsuki Reijirō]. Yomiuri Shimbunsha, 1950. Yamakawa Hitoshi. “Goken Naikaku” [The Cabinet to Preserve Constitutional Politics]. Kaizō 6, no. 7 (July 1924): 113–115. ———. “Gunpuku no Seiji yori Furokku Kōto no Seiji e” [From the Politics of Military Uniforms to the Politics of Frock Coats]. In Yamakawa Hitoshi Zenshū, vol. 2, 109–118. ———. Yamakawa Hitoshi Zenshū [Selected Works of Yamakawa Hitoshi], vol. 2. Keisō Shobō, 1966. Yamamoto Shirō. “Saitō Naikaku no Seiritsu o megutte” [Concerning the Formation of the Saitō Cabinet]. Shirin 59, no. 5 (September 1976): 42–78. Yamaura Kan’ichi, ed. Mori Kaku [Mori Kaku]. Mori Kaku Denki Kankōkai, 1940. Yoshii Ken’ichi. “Tai Chūgoku Seisaku no Tenkan to Gikai” [Changes in Policy toward China and the Diet]. In Uchida, Kinbara, and Furuya, eds., Nihon Gikai Shiroku, vol. 3, 1–63. Yoshino Sakuzō. “Gendai Seikyoku no Tenbo” [Prospects for Contemporary Politics]. In Yoshino Sakuzō Senshū, vol. 4, 247–299. ———. “Hara Naikaku ni taisuru Yōbō” [Request to the Hara Cabinet]. Chūō Kōron 33, no. 11 (October 1918): 80–87. ———. (Under the pen name of “Furukawa Gakuto.”) “Hara Shushō ni teisuru Sho” [A Letter for Prime Minister Hara]. Chūō Kōron 33, no. 12 (November 1918): 49–51. ———. “Kensei no Hongi o toite sono Yūshū no Bi o nasu no Michi o ronzu” [On the Essence of Politics Based on the Constitution and the Way to Achieve It]. In Yoshino Sakuzō Senshū, vol. 2, 3–98.

232  Selected Bibliography ———. “Senkyo to Kane to Seitō” [Elections, Money, and Political Parties]. In Yoshino Sakuzō Senshū, vol. 4, 353–360. ———. “Shakai Kaizō no Daiichi Kaidan toshite no Futsū Senkyo” [Elections Based on Universal Male Suffrage as a First Step to Social Reform]. In Yoshino Sakuzō Senshū, vol. 3, 290–292. ———. “Shin Naikaku ni taisuru Kitai” [Hopes for the New Cabinet]. In Yoshino Sakuzō Senshū, vol. 4, 83–85. Yoshino Sakuzō Senshū [Selected Works of Yoshino Sakuzō], vols. 2–4. Iwanami Shoten, 1995 and 1996.

Index

Accountability, in authoritarian regimes, 16–17. See also Electoral control Acting premiership, 138–39 Adachi Kenzō, 78, 156 Agriculture, 130 Agüero, Felipe, 90, 121 Alienation, from politics, 145–46, 148 All Japan Farmers Union, 196n64 Almond, Gabriel, 26 Amaoka Naoki, 134 Anne, Queen, 68 Argentina, 168 Arms reductions, 89, 126 Army, size of, 89 Asia, political culture in, 27–28, 174 Assassinations and attempts: of Japanese political leaders, 129, 132–33, 145; Manchurian Incident, 78, 103–5, 121–23, 126–27, 152–56, 163, 171–72 Authoritarian regimes: characteristics of, 16–17; competitive oligarchy compared to, 49–51. See also Competitive authoritarianism; Electoral authoritarianism; Military authoritarian regimes Awaya Kentarō, 63, 147 Balance of power, in semi-democratic regimes: breakdown attributable to disturbances in, 30–33, 42, 159; legitimacy as factor in, 43; political institutions

and, 42–43; semi-loyalty as factor in, 42, 43; shifts in, 125 Bangkok Post (newspaper), 167 Bank of Japan, 113 Bank of Taiwan, 113–14 Banno Junji, 8, 93, 99–100, 147 Barnes, Samuel, 35 Berger, Gordon, 7–8, 63–64 Black May Incident (Thailand, 1992), 167 Boku Retsu (Pak Yŏl) Scandal, 106 Branco, Castero, 165 Brazil, 18–19, 162, 165–66 Breakdown (collapse) of democratic regimes, 33, 40, 170; factors in, 167, 169; research on, 32 Breakdown (collapse) of semi-democratic regimes: analytical framework for, 11, 42–44, 159–60; conditional approach to, 24–28; crises contributing to, 125–57, 170; descriptive model of, 31–33; deterministic perspective on, 173–74; development as distinct from, 25–26, 28; factors in, 1, 6, 30–42, 85; imbalance of power leading to, 30–33, 42, 159; in Japan, 1, 4, 99–101, 129; legitimacy as factor in, 140, 165–67; political incidents leading to final, 123, 156–57; probabilistic nature of, 170–74; scholarship on, 9–10, 177n23; structural approach to, 28–30, 173 Brooker, Paul, 15

233

234  Index Cabinet: formation of, 72; in Great Britain, 72; succession of, 80–81, 149 Capitalism, 10 Cardwell, Edward, 69 Carothers, Thomas, 4 Case, William, 15 Chamanan, Kriengsa, 20 Chang Tso-lin, 103–5, 121–23, 126, 163, 171–72 China, 54, 61 China Incident, 61, 64 Chōsen Army, 127, 155, 172 Choonhavan, Chatichai, 166 Chūō Kōron (magazine), 90 Civic culture, 26–27 Civic regimes, 29 Civil-military relationship, 120–23, 149–56, 162–65 Civil rights, 2–3 Collapse of semi-democratic regimes. See Breakdown (collapse) of semi-democratic regimes Communism, 100 Communist Party, 55, 60 Competitive authoritarianism, 2, 5, 12, 14, 24 Competitive oligarchical regimes, 16, 45, 47–53, 64–66 Conservative Party (Britain), 69 Coups d’état, 41, 127, 128, 156 Crises and problems, in semi-democratic regimes, 31–33, 113–14, 125–57, 170 Crowley, James B., 177n23 Dahl, Robert, 5, 13, 16, 24–25, 35, 70–71, 73, 162, 169–70, 173 Dan Takuma, 145 Democracy: characteristics of, 13; development of semi-democratic regimes into, 2–3; legitimacy in relation to, 34–35; legitimacy undermined in, 169; minponshugi distinguished from, 191n27; spectrum of, from electoral to liberal, 14; study of, 161–74 Democratic transitions, 4–6 Democratization: conditions for, 24–28, 173–74; in Japan’s competitive oligar-

chical regime, 51–53; models of, 162; patterns of, 67–75 Den Kenjirō, 118 Depression. See Great Depression; Showa Great Depression Diamond, Larry, 5, 12, 14, 25, 27, 35, 39, 173 Diet: disputes involving, 139; establishment of, 47; in Japan’s competitive oligarchical regime, 50–51; in Japan’s military authoritarian regime, 63–64; party government’s handling of, 136–39, 146; and political competition, 70; power of, 67; semi-loyalty of, 122 Disloyalty, 31–33, 41, 122 Disraeli, Benjamin, 72 Dōshikai, 7, 48–49, 53, 65 Duus, Peter, 7 Economic development, and democratization, 24–26, 173–74 Economy. See Socioeconomic sphere Effectiveness, of governments, 38–39 Efficacy, of governments, 38–39 Electoral authoritarianism, 2, 5, 12, 14 Electoral control: as characteristic of democracy, 2–3, 13–14; in democratization process, 73–75; in Great Britain, 70, 72–73; in Japan’s competitive oligarchical regime, 49, 51–52, 65–66; in Japan’s military authoritarian regime, 62–64, 67; in Japan’s semi-democratic regime, 55–58, 65–67, 70, 72–73; and reserved domains, 73 Electoral democracy, 14 Electoral participation: as characteristic of democracy, 2–4, 13–14; in competitive oligarchical regimes, 16; in Japan’s competitive oligarchical regime, 49, 51, 66; in Japan’s semi-democratic regime, 58–59, 66. See also Voter turnout Elites: beliefs of, concerning legitimacy, 35–36; legitimacy of Japan’s semidemocratic regime among, 90–92; in semi-democratic regimes, 35; types of, 35. See also Intellectuals; Mass media; Military

Index  235

Falange, 51 Farmers. See Tenant farmer disputes; Tenant farmers February Twenty-Sixth Incident (1936), 63, 148 Feudal tradition, 9 Financial system, crises in, 57, 79, 110, 113–14 First Austrian Republic, 168 Fitch, Samuel, 15 Franco, Francisco, 50–51 Freedom of association, 2 Freedom of expression, 2

Harada Kumao, 143, 152 Hara Takashi (also called Hara Kei), 8, 45, 54, 56, 58, 65, 87–88, 90–94, 101, 160, 171 Hashimoto Kingorō, 128, 156 Hayashi Yūichi, 118 Hiranuma Kiichirō, 118, 143, 156 Hirota Kōki, 60–62 Hoshino Shōzaburō, 98 House of Commons (Britain), 68–69, 70, 72, 81 House of Lords (Britain), 69, 72, 81 House of Peers: accountability of, 18; after Second World War, 73; in Japan’s competitive oligarchical regime, 51; in Japan’s semi-democratic regime, 56–57; power of, 73, 78–79, 171; reform of, 72, 100 Huntington, Samuel, 25, 28–30, 173–74 Hybrid regimes: defined, 1; development of, into authoritarian regimes, 6; incidence of, 5; studies of, 2, 4–6, 161–62

Garon, Sheldon, 8 General elections, holding of, 80, 108–10 General Staff, 58, 82–83, 88, 104, 126, 149–50, 203n119 Genrō, 56, 80, 165, 185n40 Germany, 63. See also Weimar Republic Gladstone, William, 72 Gold standard, 54, 130–33, 137 Gordon, Andrew, 8, 164 Goulart, João, 19, 165 Governors, of Japanese colonies, 88 Great Britain: as competitive oligarchical regime, 16; democratization in, 67–75; and gold standard, 131, 133; relations with, 54; as semi-democratic regime, 3, 18, 162 Great Depression, 10, 130, 168. See also Showa Great Depression Guomindang, 104

Ideology, 16–17 Industrialization, 9 Inomata Tsunao, 144 Inoue Junnosuke, 132, 143, 145 Inoue Nissho, 145 Intellectuals, legitimacy of Japan’s semidemocratic regime among, 106–17, 130–40 Inukai Takeshi, 147 Inukai Tsuyoshi, 1, 72, 125, 128, 129, 136, 141, 143–44 Irie Kannichi, 193n59 Ishibashi Tanzan, 92–93, 107, 110, 131–34, 137, 139–41, 169 Italy, 63 Itō Hirobumi, 8, 47, 48, 52, 65, 83 Itō Miyoji, 83, 151 Itō Takashi, 118, 150–51 Iwasaki Isao, 106

Hamada Kumimatsu, 60 Hamaguchi Osachi, 8, 54, 72, 106, 123, 125–26, 128, 130–34, 136–40, 145–46, 149, 151–52, 163, 171 Hanbatsu cliques, 46–48, 56, 64–65, 164

Jansen, Marius, 164 Japan: as a democracy, 27; political culture of, 174. See also Prewar Japan Japanese semi-democratic regime (1918– 1932): breakdown of, 1, 4, 99–101, 129;

Emergency decrees, 50, 57, 79–80, 110, 112–14 Emperor: and Manchurian Incident, 121; military’s relationship to, 82–84, 88, 149–51, 189n22, 189n25 (see also ­Supreme command, of the emperor over the military); power of, 60, 78

236  Index characteristics of, 3–4, 17–18, 53–59; competitive oligarchy compared to, 64–66; crises in, 113–14, 125–57, 170; early stages (1918–1926), 87–101; electoral control in, 55–58, 65–66; electoral participation in, 58–59, 66; elites in, 90–92; late stages (1929–1932), 125–57; legitimacy of, 85, 90–99, 105–22, 129–48; mass media in, 106–17; middle stages (1926–1929), 103–23; military authoritarian regime compared to, 66–67; the military in, 56–58, 87–90, 98, 103–5; parliamentary government compared to, 81; pattern of development of, 67–75; political competition in, 53–55, 64–67; the public in, 92 Japan Labor Union Alliance, 196n64 Japan Labor Union Council, 117 Jiji Shimpō (newspaper), 134 Jiyutō, 48 Jushin, 155–56, 185n40, 203n130 Kaishintō, 48 Kakushinkurabu, 59, 96 Kanaya Hanzo, 153 Kasza, Gregory, 8 Kataoka Naoharu, 113 Katō Hiroharu, 150, 151 Katō Takaaki, 57, 72, 89, 96–101, 171 Katō Tomozaburō, 56, 72, 88–89 Katsura Tarō, 48–49, 52–53, 56 Kensei Hontō, 48–49 Kensei Isshinkai, 111 Kenseikai, 7, 48, 53–54, 59, 65, 93, 96, 97, 106–10, 164. See also Minseitō Kenseitō, 48 Kensei Yogokai (Association to Protect Politics Based on the Constitution), 120 Ketsumeidan, 145 Kido Kōichi, 144, 148 Kita Ikki, 128 Kitaoka Shinichi, 7 Kiyoura Keigo, 59, 65, 94–96, 98, 120, 192n44 Kōmoto Daisaku, 103–5, 121–22, 194n3 Kobashi Ichita, 134–35

Kobayashi Tatsuo, 150–51 Koiso Kuniaki, 128 Kokumintō, 48, 93 Kongsompong, Sunthorn, 166 Konoe Fumimaro, 61, 62, 143–44 Korea, 88. See also South Korea Kraprayoon, Suchinda, 20, 166–67 Kubischek de Oliveria, Juscelio, 19 Kuhara Fusanosuke, 156 Kwantung Army, 103–4, 121, 127–28, 153, 155–56 Kwantung Leased Territory, 88 Labor. See Workers Labor Accident Assistance Law, 79 Labor and Farmers Party, 117 Labor disputes, 119, 131, 145 Labor movement, 115–17 Labor Union Law, 79, 119, 196n64 Labour Party (Britain), 69 Lakatos, Imre, 43 Latin America, 75 Law on the Organization of the Court, 79 League of Blood (Ketsumeidan) Incident (1932), 145 Leekpai, Chuan, 167 Legitimacy: alternative possibilities as factor in, 37–39, 94; and balance of power in regimes, 43; breakdown of semidemocratic regime related to, 165–67; and civil-military relationship, 120–21; defining, 34–35; democracy in relation to, 34–35; economic crises affecting, 167–69; elites and, 90–92; factors affecting, 37–39; government performance as factor in, 37–39, 106–17, 130–43, 146–47, 167–69; intellectuals and, 106–17, 130–40; of Japan’s semidemocratic regime, 85, 90–99, 105–22, 129–48; mass media and, 106–17, 120, 130–40; measurement of, 40, 182n50; the military and, 98, 149–56; nature of groups holding beliefs in, 35–36; political developments affecting, 168; political leaders and, 117–18, 120, 143–44; the public and, 92, 118–19, 144–47; in semidemocratic regimes, 31–40; semi-loyalty

Index  237 in relation to, 41; time as factor in, 169–70; undermining of, in democratic vs. semi-democratic regimes, 169 Lepsius, M. Rainer, 168 Levitsky, Steven, 5, 12 Liberal democracy, 14 Liberal Party (Britain), 69 Linz, Juan, 12, 14, 16, 31–35, 37, 38, 40, 41, 50, 90, 94, 121 Lipset, Seymour Martin, 12, 14, 24–25, 33, 34, 35, 173 Lloyd George, David, 69 London Naval Treaty (1930), 8, 84, 123, 125–26, 149–52, 163, 165, 171 Lower House: dissolution of, 47; in Japan’s competitive oligarchical regime, 49–53, 55–56; in Japan’s military authoritarian regime, 63–64; in Japan’s semi-democratic regime, 56–57 Loyalty, 40–41. See also Disloyalty; Semiloyalty Maeda Renzan, 92 Makino Nobuaki, 105, 117 Malaysia, 161 Manchuria, occupation of, 127–28, 153–56, 198n6 Manchurian Incident, 78, 103–5, 121–23, 126–27, 152–56, 163, 165, 171–72 March Incident (1931), 128, 198n12 Maruyama Masso, 8–9 Masses. See Public, the Mass media, and legitimacy of Japan’s selfdemocratic regime, 106–17, 120, 130–40 Mass praetorianism, 180n19 Matsukata Masayoshi, 48, 52, 56, 80, 94–96 Matsumoto Gōkichi, 91–92, 117 Matsuoka Yōsuke, 154 Matsuo Takayoshi, 93 Matsushima Club Scandal, 106 May Fifteenth Incident (1932), 1, 4, 129, 141, 143, 147, 156, 163, 172 McDonough, Peter, 35 Meiji Constitution: breakdown of semidemocratic regime and, 8–9; and the Diet, 47, 52; and the emperor, 60; and

Japan’s competitive oligarchical regime, 46–47; and the Lower House, 55–56; and the military, 57, 82–85, 126, 163; political institutions under, 77–85 Meiji Restoration, 10, 46 Melbourne, Viscount, 56, 68 Mentalities, 16–17 Merelman, Richard, 34 Merkl, Peter, 34 Mexico, 161 Military: after Second World War, 73; in Brazil, 19, 165–66; civil-military relationship, 120–23, 149–56, 162–65; and control over military operations, 82, 163; control over/power of, 57–58, 82–85, 87–90, 101, 121; emperor’s relationship to, 82–84, 88, 149–51, 189n22, 189n25 (see also Supreme command, of the emperor over the military); in Great Britain, 69; in Japan’s semidemocratic regime, 18, 56–58, 87–90, 98, 103–5; and legitimacy of Japan’s self-democratic regime, 98, 149–56; and the Manchurian Incident, 103–5, 121–23, 126–27, 152–56, 163, 165, 171–72; Meiji Constitution and, 57, 82–85, 126, 163; and military organization, 83–84, 163; party government’s relationship to, 103–5, 125–29; party politicians’ affinity with, 149; in praetorian regimes, 29; size of the army, 89; in Thailand, 20–21, 166–67; in tutelary regimes, 15. See also Military authoritarian regimes Military authoritarian regimes, 45, 60–64, 66–67 Military ministers, 84, 89, 189n22, 189n25 Minami Jirō, 154–55 Ministers, authority and accountability of, 77–78, 83 Ministry of Interior, 55 Minobe Tatsukichi, 60, 97, 141 Minoura Katsundo, 106 Minponshugi (principle of the people as the base), 92, 191n27 Minseitō, 7–8, 55, 110–11, 118, 122, 131, 133–35, 138, 149, 153–54, 156, 164, 172. See also Kenseikai

238  Index Mitani Taichirō, 78 Miyazaki Ryūji, 78 Modernization, 28–30 Monarchy, 20 Montero, José, 34, 35, 38 Moore, Barrington, 9–10 Mori Kaku, 151–52, 154, 156 Morlino, Leonardo, 34, 35, 38 Mudslinging, 118, 147 Murai Ryōta, 7 Muraoka Chōtarō, 104 Nagai Kazu, 189n25 Nagai Ryutarō, 156 Nakano Seigo, 106 National Mobilization Law (1938), 63–64, 67 National Peacekeeping Council (Thailand), 166 Ninomiya Harushige, 128 No-confidence votes, 70, 80 Nonelected officials: examples of, 15; political power of, 6; in tutelary regimes, 15 Nonelectoral institutions, 162 Nonparty government: as alternative to party government, 106, 117–18, 120, 143; criticisms of, 120; and the military, 87; political parties in competition with, 65; ruling of Japan by (1922–1924), 59, 95–96, 98 Ōkawa Shūmei, 128 Ōsaka Asahi Shimbun (newspaper), 95, 96, 107–11, 113, 142–43, 146 Ōsaka Mainichi Shimbun (newspaper), 109, 111–14, 132, 135, 139 October Incident (1931), 128, 156 O’Donnell, Guillermo, 4 Ogawa Heikichi, 134, 147 Okada Keisuke, 60, 150, 151 Okuma Shigenobu, 65 Oligarchical praetorianism, 180n19 Oligarchical regimes, 46. See also Competitive oligarchical regimes Pak Yŏl (Boku Retsu), 106 Parliamentary government, 56, 68, 80–84

Parochial political culture, 26 Participant political culture, 26 Party government: in contemporary Japan, 72; criticisms of, 114, 117; the Diet defied by, 136–39; institutional weakness of, 77–81; military’s relationship to, 103–5, 125–29; rejection of, 141–44 Party politicians, semi-loyalty of, 121–22, 148–57, 163–65 Peace Preservation Law (1925), 54–55, 100, 112 Peasants, 10, 130, 144 Peel, Robert, 68, 72 Performance, regime: legitimacy affected by, 37–39, 106–17, 130–43, 146–47, 167–69; in political sphere, 39, 106–13, 134–43, 146–47; in socioeconomic sphere, 39, 113–17, 130–34, 140–43 Pina, Antonio López, 35 Pluralism, 16 Political competition: in authoritarian regimes, 2, 5, 12, 13–14, 17; as characteristic of democracy, 2–3, 13–14; in competitive oligarchical regimes, 16; in democratization process, 75; in Great Britain, 70; in Japan’s competitive oligarchical regime, 47–49, 64–65; in Japan’s military authoritarian regime, 60–62, 66–67; in Japan’s semi-democratic regime, 53–55, 64–67, 70; under Meiji Constitution, 77–85. See also Political parties Political culture: Asian, 27–28, 174; and democratization, 26–28, 173–74; Japanese, 174; types of, 26 Political institutions: balance of power affected by, 42–43; in Huntington’s structural theory, 28–30; in semi-democratic regimes, 31, 33, 42–43 Political leaders, legitimacy of Japan’s semi-democratic regime among, 117–18, 120, 143–44 Political participation: in authoritarian regimes, 16; and regime development, 29. See also Electoral participation Political parties: in authoritarian regimes, 51; formation of, 48; in Great Britain,

Index  239 68–70; in Japan’s competitive oligarchical regime, 47–49, 51, 64–65; in Japan’s military authoritarian regime, 61, 63; in Japan’s semi-democratic regime, 53–56, 64–65, 70; scandals and culpable behavior of, 94, 106–13, 134–36, 146–47; and two-party system, 7, 68–69, 164–65, 204n4. See also Political competition Political sphere: legitimacy affected by activities in, 168; performance in socioeconomic sphere in relation to, 140–43; regime performance in, 39, 106–13, 134–43, 146–47 Praetorian regimes, 29, 180n19 Prewar Japan: comparison of regimes of, 64–67; competitive oligarchy in (1889– 1918), 45, 47–53; military authoritarian regime in (1936–1945), 45, 60–64; political development in, 6–12; pre-1889 regimes, 45–47; regimes of, 45–75; semi-democratic regime in (1918–1932), 53–59 Prime ministers, nominations of, 56, 61, 80, 165, 185n40 Privy Council: accountability of, 18; after Second World War, 73; in Japan’s semi-democratic regime, 56–57; power of, 73, 79–80, 123, 171; and socioeconomic crises, 113–14 Pseudo democracy, 5 Public, the: beliefs of, concerning legitimacy, 35–36; in Japan’s competitive oligarchical regime, 52–53; legitimacy of Japan’s semi-democratic regime among, 92, 118–19, 144–47; minponshugi and, 92, 191n27; in semi-democratic regimes, 35 Punyarachun, Anand, 20, 166–67 Pye, Lucian, 27–28, 174 Quadros, Jânio, 19 Radical praetorianism, 180n19 Reform Act (Britain, 1867), 68 Reform Act (Britain, 1884), 69 Reserved domains, 73 Rice Riots (1918), 53

Russo-Japanese War (1904–1905), 48, 52 Saionji Kinmochi, 1, 48–49, 51–53, 56, 65–66, 80, 95, 98, 104–6, 117–18, 120, 121, 129, 143, 145, 155–57, 172, 185n40 Saitō Makoto, 1, 129, 143–44, 148, 172 Saitō Takao, 60–61 Sakurakai, 128, 152 Samuels, Richard, 8 Sasaki Sōichi, 138 Satogami Ryōhei, 78 Scalapino, Robert, 9–10, 174 Scandals, 94, 106–8, 134–36 Schedler, Andreas, 12 Schmitter, Philippe, 4 Scholarship: on breakdown of semi-democratic regimes, 9–10, 177n23; on hybrid regimes, 4–6; on Japan’s prewar political development, 6–12 Second Pro-Constitution Movement, 96 Seiyūkai, 7–8, 47, 48–49, 51, 53–55, 59, 64– 65, 78, 93–94, 96, 106–11, 118, 122–23, 131, 134–35, 139, 143–44, 149–54, 156–57, 163–65, 171–72, 192n44 Seiyu Hontō, 53, 59, 96, 106–11, 192n44 Semi-democratic regimes: alternatives to, 39; as analytical concept, 161–62; defining, 2, 12–15; development of, into democracies, 2–3; examples of, 17–21; legitimacy of, 35; legitimacy undermined in, 169; life course of, 159–61; other regimes compared to, 14–17; parliamentary government compared to, 81. See also Breakdown of semi-democratic regimes; Japanese semi-democratic regime Semi-loyalty, 40–42, 163–65; balance of power affected by, 42, 43; defined, 41; of the Diet, 122; effects of, 42; legitimacy in relation to, 41; of party politicians, 121–22, 148–57, 163–65; in semidemocratic regimes, 31–33, 41 Shidehara Kijūrō, 138, 155 Shimomura Chiaki, 144 Shimpotō, 48 Shintōkurabu, 111 Showa Great Depression, 130–32, 144

240  Index Siberia, 87–88, 190n3 Simon, Walter B., 168 Sino-Japanese War (1894–1895), 47 Skidmore, Thomas, 18 Smith, Peter H., 168 Social frustration, 28–29 Socioeconomic sphere: legitimacy affected by crises in, 167–69; performance in political sphere in relation to, 140–43; regime performance in, 39, 113–17, 130–34, 140–43. See also Economic development South Korea, 27. See also Korea Soviet Union, 25, 75 Spain, 38, 50–51, 75 Srimuang, Chamlong, 20 Stepan, Alfred, 38, 166 Structural approach to regime characteristics, 28–30, 173 Subject political culture, 26 Succession, government, 80–81, 149 Suehiro Gantarō, 117 Suffrage, extent of, 3–4, 15, 18, 51, 54, 57– 59, 66, 68–69, 73, 92–94, 98–100, 136 Sugiyama Hajime, 128 Supreme command, of the emperor over the military, 57, 82–84, 88, 149–52, 155, 163, 203n119. See also Military: emperor’s relationship to Suzuki Bunji, 92, 197n64 Suzuki Kisaburō, 143, 151–52, 157 Suzuki Shōten, 113 System-oriented political actors, 41 Taisei Yokusan Kai (Imperial Rule Assistance Association), 61 Taiwan, 27, 88 Takahashi Kamekichi, 197n64 Takahashi Korekiyo, 89, 98, 114, 157 Tanaka Giichi, 79, 98, 104–6, 110–12, 114, 116–23, 134, 147–49, 153, 163, 164, 171–72 Tenancy Law, 79, 119, 197n64 Tenant farmer disputes, 118–19, 130–31, 145 Tenant farmers, 130–33 Terauchi Juichi, 60 Terauchi Masatake, 53, 64 Tetsuo Najita, 7

Thailand, 19–21, 162, 165–67 Theory, in the social sciences, 11 Time, legitimacy affected by, 169–70 Tinsulanonda, Prem, 20–21 Tōkyō Asahi Shimbun (newspaper), 91, 93, 96, 97, 106, 114, 117, 132–34, 138–43 Tōkyō Nichinichi Shimbun (newspaper), 95, 108, 135–36, 145–46 Tokonami Takejirō, 111 Tokugawa Shogunate, 45–46 Toland, John, 177n23 Tories, 68 Totalitarian regimes, 16–17 Tutelary regimes, 15 Two-party system, 7, 68–69, 164–65, 204n4 Ugaki Kazushige, 62, 98, 120, 121, 128, 143 United States, 25, 54, 63 Universal suffrage. See Suffrage, extent of Valenzuela, J. Samuel, 15 Vargas, Getúlio, 19 Verba, Sidney, 26 Voter turnout, 118–19, 144. See also Electoral participation Votes of no confidence, 70, 80 Voting rights. See Electoral participation; Suffrage, extent of Wakatsuki Reijirō, 57, 77–80, 105–10, 113–14, 116–20, 123, 125, 127–28, 131–33, 136, 140, 143, 145, 153, 155, 157, 171–72 Washington Conference (1921), 88 Washington Naval Treaty (1922), 89 Watanabe Bank, 113 Way, Lucan A., 5, 12 Weatherford, M. Stephen, 182n50 Weber, Max, 33–35 Weimar Republic, 37, 38, 168 West, political culture in, 27 Whig regimes, 29 Whigs, 68 William IV, King, 68 Workers, 116–17, 119, 132, 144, 146. See also Labor disputes; Labor movement Yamagata Aritōmo, 47, 65, 91–92

Index  241 Yamakawa Hitoshi, 91, 97 Yamamoto Gombei, 49, 53, 72, 84, 95, 157 Yamamoto Shirō, 143–44 Yamamoto Teijirō, 151–52

Yomiuri Shimbun (newspaper), 96, 142 Yonai Mitsumasa, 62 Yoshino Sakuzō, 91, 93, 94, 111–12, 136, 169, 191n27