Civic Engagement in Postwar Japan: The Revival of a Defeated Society

Civic Engagement in Postwar Japan The Revival of a Defeated Society

Despite reduced incomes, diminished opportunities for education, and the psychological trauma of defeat, Japan experienced a rapid rise in civic engagement in the immediate aftermath of World War II. Why? Civic Engagement in Postwar Japan answers this question with a new general theory of the growth in civic engagement in postwar democracies. It argues that wartime mobilization unintentionally instills civic skills in the citizenry, thus laying the groundwork for a postwar civic engagement boom. Meanwhile, legacies of prewar associational activities shape the costs of association-building and information-gathering, thus affecting the actual extent of the postwar boom. Combining original data collection, rigorous statistical methods, and in-depth historical case analyses, this book illuminates one of the keys to making postwar democracies work. Rieko Kage is Associate Professor of Political Science at the University of Tokyo, Japan. She earned her Ph.D. from Harvard University in 2005. She has been a Fulbright Scholar, Advanced Research Fellow at the Program on U.S.–Japan Relations at Harvard, and Toyota Visiting Scholar at the University of Michigan. Her writings have appeared in Comparative Political Studies, Political Psychology, and other publications.

Civic Engagement in Postwar Japan The Revival of a Defeated Society

RIEKO KAGE University of Tokyo

cambridge university press Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo, Delhi, Dubai, Tokyo, Mexico City Cambridge University Press 32 Avenue of the Americas, New York, ny 10013-2473, usa www.cambridge.org Information on this title: www.cambridge.org/9780521192576 © Rieko Kage 2011 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2011 Printed in the United States of America A catalog record for this publication is available from the British Library. Library of Congress Cataloging in Publication data Kage, Rieko, 1973– Civic engagement in postwar Japan : the revival of a defeated society / Rieko Kage. p. cm. Includes bibliographical references and index. isbn 978-0-521-19257-6 1. Civil society – Japan – History – 20th century. 2. Postwar reconstruction – Japan – Citizen participation. 3. Political participation – Japan – History – 20th century. 4. Japan – Politics and government – 1945–1989. i. Title. jq1681.k33 2011 300.952–dc22 2010030585 isbn 978-0-521-19257-6 Hardback Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party Internet Web sites referred to in this publication and does not guarantee that any content on such Web sites is, or will remain, accurate or appropriate.

Contents

List of Tables List of Figures Preface Acknowledgments 1 2 3 4 5 6 7 8 9

Introduction Civic Engagement: The Dependent Variable War and Civic Engagement: A Theoretical Framework Quantitative Analysis: The Rise of Civic Engagement across Forty-Six Japanese Prefectures The Long-Term Effects of Wartime Mobilization: Cross-National Analysis Repression and Revival of the YMCA Japan Wartime Endorsement and Postwar Repression of a Traditional Art Civil Society and Reconstruction in Postwar Japan Conclusions

References Index

page vii ix xi xiii 1 19 44 64 78 95 123 140 161 171 193

v

Tables

1.1 Military and Civilian Casualties for Selected Countries, World War II page 11 1.2 Private Homes Destroyed and Damaged in Selected Countries, World War II 11 1.3 Industrial Production before and after World War II, 1937–1955 13 1.4 Inflation before and after World War II, 1937–1955 14 2.1 YMCA Membership in Selected Japanese Cities, 1922–1950 25 2.2 Top Five and Bottom Five Prefectures, Membership in Youth, Women’s, and Christian Organizations 38 2.3 Pre- to Postwar Growth in Memberships in Youth and Christian Organizations 41 3.1 Predicting Changes in Civic Engagement between Prewar and Postwar Periods 45 4.1 Regression Analysis, Explaining the Growth in Youth Group Membership, 1931–1958, Using Participation Rates as of 1931 and 1922 70 4.2 Regression Analysis, Explaining the Growth in Youth Group Membership, 1922–1958, Using Participation Rates as of 1922 71 4.3 Regression Analysis, Explaining the Growth in Religious Membership, 1936–1955, Using Participation Rates as of 1936 72 4.4 Regression Analysis, Explaining the Growth of Youth and Christian Memberships Using Wartime Damage Figures 76 5.1 Mean Number of Associational Memberships: Pre–World War II and World War II Cohorts 83 5.2 Number of Associational Memberships at the Same Age Level: Pre–World War II and World War II Cohorts 85 5.3 Index of Wartime Mobilization 87 5.4 Intensity of the Great Depression 88 5.5 The Determinants of Associational Membership 90 6.1 The Growth of YMCA Membership, Postwar Period (vis-à-vis Prewar Period) 100 vii

viii

8.1 Descriptive Statistics for Damage and Recovery, by Prefecture: Jobs, Hospitals, Schools, and Library Books 8.2 Correlations across Measures 8.3 The Determinants of Reconstruction in Japan: Jobs, Hospitals, Schools, and Library Books

Tables

150 153 156

Figures

2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.10 2.11 2.12 2.13 5.1

5.2 6.1

Membership in the YMCA Japan, 1925–1956 page 24 Membership in the YWCA Japan, 1927–1960 26 Membership in the Boy Scouts Japan, 1939–1960 28 Membership in the Girl Scouts Japan, 1929–1960 28 New Members in the Nihon Sangakukai [Japan Alpine Club], 1925–1960 29 New Members in the Kodokan, 1925–1960 30 Membership in the Tomo no Kai, 1930–1960 31 Membership in the Rotary Club Japan, 1920–1960 32 Membership in the Japan Consumer Cooperatives’ Union, 1925–1960 33 Membership in the Nada Kobe Consumer Cooperatives’ Union, 1921–1956 33 Membership in the Japan Seafarers’ Relief Association, 1925–1957 34 Membership in Christian Churches, 1925–1960 35 Membership in Labor Unions, 1925–1960 36 Percentage Difference in Number of Associational Memberships between 1911–1920 and 1921–1930 Cohorts, by Level of Mobilization 87 Predicted Number of Associational Memberships by Generational Cohort, Based on Cross-National Analysis 92 Differences in Membership per Population, Kobe and Sapporo YMCAs, 1916–1956 101

ix

Preface

Hiroshima, 1949.1 Four years earlier, the atomic bomb had completely destroyed the city. There were still many visible signs of damage. In the midst of this devastation, Hiroshima’s first professional baseball team came into being. Professional baseball had been founded in Japan in 1934, but Hiroshima had never had a professional team. Yet baseball at the high school level had been enormously popular in Hiroshima before the war, and the local Hiroshima Commercial High School had won three titles at the national high school baseball tournament (Koshien) during the prewar period. As early as November 1945, a team of local baseball players began playing with the U.S. Occupation forces. Thus, in April 1949, when Matsutaro Shoriki, owner of the Yomiuri Giants, announced a plan to re-organize Japanese professional baseball into two leagues and to vastly expand the number of teams, local business leaders in Hiroshima sprang into action and sent in an application. They saw baseball as a way to revitalize the war-torn city. The new Hiroshima Carp, as the team would be called, was accepted into the Central League, and in 1950 the team played its first full season. Crowds flocked to the stadium to watch the games. The problem was funding. Most of the professional teams that were in existence at the time were owned by big corporations that were willing to run a deficit on the team in order to generate publicity for their main line of business. The Yomiuri Giants and the Hanshin Tigers typified this business model. The Hiroshima Carp was the only professional team in Japan not to have a major corporate owner. Initially, the team sought to raise capital from local firms, governments, and wealthy individuals, but it had little success. Local governments in particular were reluctant to provide assistance due to tight budgets at the time. Even before the first season began, the Carp ran into difficulty paying its players, despite the relatively small payroll. There was little travel funding, so once the season started, players often stayed with friends or family when they were on the road; and when this was not possible, players slept in the back rooms of the host stadium. 1

This section draws on Matsunaga (2001), Matsubara (2009), and Abe (2009).

xi

xii

Preface

The Carp finished its first season at forty-one wins and ninety-six losses, fiftynine games behind the first place team and in last place among the eight teams that were then members of the Central League. But rather than being discouraged, the Carp’s weakness only strengthened the resolve of the fans to support their team. Remarkably, the Carp came in fourth in the league in game attendance during its first few seasons, just behind the well-established teams of the Yomiuri Giants, the Hanshin Tigers, and the Chunichi Dragons. This was particularly astounding because many of the teams in the league had much bigger stadiums in more accessible locations. Hiroshima Sogo Stadium, the Carp’s home stadium, had survived the atomic bombing, but this was because the stadium was far from the center of the city. It was also small. The Korakuen and the Koshien Stadiums, home to the Yomiuri Giants and the Hanshin Tigers, respectively, both seated more than 40,000 spectators. In contrast, the Hiroshima Sogo Stadium only accommodated around 10,000 people. There were always more fans than seats, so those who did not have seats simply stood or sat on fences; the lucky ones were allowed to sit at the edge of the outfield to watch the game. The Carp’s financial situation did not improve in the following season, and the team began a serious search for potential buyers. The Taiyo Fishery Company, which had just launched its own new professional team, the Whales, in 1950, offered to buy the Carp. But just as the Carp was about to sign the agreement with Taiyo, Shuichi Ishimoto, the manager of the Carp, came up with the idea of establishing a fan club to raise money. Ishimoto had served as manager of the Hiroshima Commercial High School baseball team before the war and had gained experience there in fundraising. As news broke that day that the Carp was about to be taken over, local residents flooded to the city hall, the local newspaper company, and the local chamber of commerce, petitioning officials not to sell the team. Encouraged by this outpouring of local support, the board of directors of the Carp voted not to sell the team and accepted Ishikawa’s idea of trying to raise funds through a new fan club. The response was enormous. Within just days of its founding, 3,000 members had signed up for the fan club; four months later, more than 13,000 members had joined and more than 2.7 million yen – the equivalent of more than $800,000 today – had been raised. The team was saved. This book is about revival. It is about the revival of societies in the wake of damage beyond imagination. The story of the Hiroshima Carp is a microcosm of the story of Japanese society in the wake of World War II. Even atomic bombs could not erase the memories of recruiting, joining, and participating. Despite having endured one of the most destructive events in human history, there were many, like the Carp’s manager Ishimoto, who remembered how to create opportunities for people to come together; and there were many, like the Carp’s avid fans, who, after losing nearly everything, took those opportunities to come together and gave what little they could. But this story is not particular to Hiroshima, or even Japan. All over the world, before and after Hiroshima, societies have, time and time again, found the strength to rise from the ashes. This book explains how and why.

Acknowledgments

This book grew out of a Ph.D. dissertation that was advised by a “dream team” thesis committee that was chaired by Susan Pharr and included Margarita Estevez-Abe, Torben Iversen, and Sidney Verba. I could not have asked for a more conscientious team of advisers. They generously dispensed words of wisdom on my numerous half-baked draft thesis proposals, and, more importantly, they always showed confidence in me. As the seed of an idea crystallized into a full-fledged project and ultimately into this book, numerous individuals provided thoughtful suggestions for improvement. In the United States, these included Daniel P. Aldrich, Marie Anchordoguy, Adam Berinsky, James Caporaso, Mario Carranza, Christina L. Davis, Michael Desch, Shinju Fujihira, Andrew Gordon, Mary Alice Haddad, Peter Hall, Yoshiko Herrera, Christopher Hood, Yusaku Horiuchi, the late Samuel P. Huntington, Nahomi Ichino, Koji Kagotani, Peter Katzenstein, Elizabeth Kier, Ron Krebs, Margaret Levi, Paul Manuel, Kenneth McElwain, Joel Migdal, Robert Pekkanen, David Reingold, Kim Reimann, Bo Rothstein, S. Wojciech Sokolowski, Richard Vallely, Dennis Yasutomo, and Jong-Sung You. In Japan, these included Ronni Alexander, Ruth Grubel, Nobuyuki Hashimoto, Yu Ishida, Yuko Kasuya, Wataru Kitamura, Jun Matsunami, Koji Nakakita, Hiroshi Okayama, Kengo Soga, Kensuke Takayasu, Masahiko Tatebayashi, Yasuo Tsuji, Yutaka Tsujinaka, Shigeki Uno, Gabriele Vogt, and Naoto Yamauchi. I also thank the two anonymous reviewers for their thorough reading of the entire manuscript and for their insightful comments. As a project that transcends the academic boundaries of comparative politics, international relations, history, and sociology, this book has benefited tremendously from the thriving interdisciplinary intellectual climate at my current affiliation, the Department of Advanced Social and International Studies at the University of Tokyo. My comparative politics colleagues Naoki Takahashi and Yu Uchiyama in particular have been generous with their support; Atsushi Ishida and Yoshiko Kojo have also kindly looked out for both myself and my husband and have eased my transition to the big city. Much of the work on the book was completed while I was with my previous employer, the Graduate xiii

xiv

Acknowledgments

School of Law at Kobe University. I am grateful to Mitsutoshi Ito, Ikuo Kume, Yutaka Onishi, Yutaka Shinada, and other colleagues there who warmly supported my first steps in the world of academia. Thanks also to Yuriko Takahashi and Atsushi Tago for their friendship and stimulating conversations over many good meals, the insights from which no doubt have worked their way into this book. This project also relied heavily on archaeological methods. I am grateful to the numerous archives and archivists in Japan, the United States, and Europe for helping me to locate hundreds of dusty old statistical reports. Those who have helped me in this project are too numerous to list here, but I particularly wish to thank the staff members at the archives of the Boy Scouts Federation of Japan; the Centre for Prospective Studies and Documentation, World Scouts Bureau, Geneva, Switzerland; the Consumer Cooperative Kobe; Girl Scouts of Japan; the Japan Choral Association; the Japan Consumer Cooperatives Union; the Japan Shogi Association; the Kanagawa Judo Federation; Kodokan; the Nihon Kiin; the Shuyodan; Tomo no Kai; the U.S. National Archives at College Park; the Japan YMCA; the Hiroshima YMCA; the Hokkaido YMCA; the Kobe YMCA; the Kyoto YMCA; the Omi HachimanYMCA; the Japan YWCA; and the Tokyo YWCA. Ronni Alexander kindly opened the door to my crucial first contacts with officials of YMCA Japan. Thanks also to the wonderfully efficient library staffs at the University of Tokyo and Kobe University, who cheerfully processed what must have felt like an unending stream of interlibrary loan requests, often on subjects that probably did not seem terribly academic (including chorus groups, sailing clubs, and the Hiroshima Carp). It has been a great pleasure to work with Lew Bateman, Anne Lovering Rounds, Elise Oranges, and the highly professional team at Cambridge University Press. Nobuo Tanaka at the University of Tokyo provided superb research assistance in the final stages of preparing the manuscript. An earlier version of Chapter 8 previously appeared in Comparative Political Studies; I thank Sage Publications for permission to use substantial portions of the article in this book. I thank the Harvard–Yenching Institute and the Fulbright Program for financial support during my initial years of graduate study in the Department of Government, Faculty of Arts and Sciences, at Harvard University. The Reischauer Institute for Japanese Studies at Harvard University made possible my initial fieldwork on this project in Japan between 2001 and 2002, and the Research Seed Grant from the Center for American Politics Studies at Harvard funded my archival research of U.S. occupation documents at the U.S. National Archives in the fall of 2002. Subsequent research for this book was supported by the Rokkodai Koenkai at Kobe University and several Grantsin-Aid for Scientific Research, Ministry of Education, Culture, Sports, Science, and Technology–Japan. The project also benefited from a one-year fellowship (2005–2006) from the Ministry of Education, Culture, Sports, Science, and Technology–Japan, which allowed me to spend a productive year at the Program on U.S. –Japan Relations at Harvard.

Acknowledgments

xv

I began my graduate study at Kyoto University, and I thank Hideo Otake for sparking my interest in political science as well as for being a role model of … irreverence. The legendary Yoko Otake also has staunchly supported my career. Michio Muramatsu and Masahiro Hosoya also provided stimulating thoughts during my early graduate years, supported my progress, and have continued to inspire my work. My parents have been a constant source of support to me, showing forbearance during the many years I spent working on this project and regularly plying me with yakiniku. This book would not have been possible without the help of my husband, Jacques Hymans, who is also a co-conspirator in our twoperson family business of reading, writing, and teaching political science. The business may still be struggling, but the partnership is a joy. Rieko Kage Tokyo, Japan

Civic Engagement in Postwar Japan The Revival of a Defeated Society

1 Introduction

Overview War destroys. It claims lives, it damages economic and physical infrastructure, and it reduces opportunities for educational advancement. If a country is defeated, the citizenry inevitably is left traumatized as well. These are hardly conditions that would seem conducive to a growth and enrichment of civic engagement. This book, however, finds considerable evidence that civic engagement rose dramatically in Japan after World War II and some initial evidence that it also rose in many other war-affected countries during the same period. The rise of civic engagement in Japan also notably predated the country’s postwar economic takeoff. How could this be? This book seeks to explain the causes of the postwar growth in civic engagement. A standard explanation points to U.S. occupation policies, which democratized the Japanese regime and introduced freedom of association. But this explanation is inadequate; although the occupation regime removed barriers to civic engagement, it hardly forced Japanese citizens to participate in organizations. The civic space opened up by the occupation had to be filled by the citizens themselves. The study proceeds in three steps. First, it documents the postwar surge in Japan’s voluntary participation. Second, it takes a step back and offers a general theory of the growth in civic engagement in the wake of wars. Third, it tests the theory using both quantitative analyses of data from Japan and twelve other developed countries, as well as qualitative analyses of specific local associations in Japan. The theory focuses on two independent variables: wartime mobilization and legacies of prewar associational activities. First, wartime mobilization lays the groundwork for postwar civic engagement by instilling civic skills in the citizens, which they can transfer to new contexts if the postwar political regime permits it. Second, path-dependency effects from the prewar structures of participation influence the availability of actual opportunities for civic engagement by shaping the costs of association-building and information-gathering. 1

Civic Engagement in Postwar Japan

2

Thus, the postwar growth in civic engagement should be fastest in extensively mobilized regions where prewar levels of associational activities had been high, and slower otherwise. This study also brings a new perspective to the recent debate over whether and why there has been a rise in the number of individuals who “bowl alone” in many societies. On the one hand, scholars primarily studying the U.S. case, such as Robert Putnam, argue that the gradual passing of the “long civic generation” that came of age around the time of World War II has produced a recent decline in civic engagement (e.g., Putnam 1995a, 1995b, 2000). On the other hand, scholars primarily studying certain Western European countries such as Belgium, Denmark, and the Netherlands have argued that more recent generations are no less willing to engage than their predecessors were (Hooghe 2003; de Hart and Dekker 1999; Torpe 2003).1 The present study advances a framework that simultaneously accounts for both the existence of the “long civic generation” in some countries, like the United States, that were significantly mobilized in World War II, and the lack of such a phenomenon in others, such as Belgium, Denmark, and the Netherlands, which were much less extensively mobilized. The theory offered in the book speaks to contemporary policy issues as well. Wars aiming at “regime change” have recently acquired a bad name, but the fact remains that war has served as a major catalyst for democratization over the past decades. According to Nancy Bermeo, more than half of all democracies that were founded after 1945 and are still in existence today emerged during or in the immediate aftermath of war (Bermeo 2003: 159; see also Bermeo 2007). So when does war promote democratization and when does it not? Examining war’s effects on civic engagement can help us to begin to answer this question. Many studies have shown that a vibrant civic life is crucial not only for “making democracy work” but also for “making democracy” in the first place (Almond and Verba 1963; Inglehart 1988; Putnam 1993; Diamond 1999). By studying the effects of war on civil society, this book thus illuminates some key conditions for successful democratic transition.

The Effects of War Writers and novelists have long recognized the vast and far-reaching consequences of war on societies and have sought to capture their devastating effects in their many manifestations. It is not a coincidence that the twentieth century, the bloodiest century in human history, also gave rise to such literary explorations on the effects of war as The Sun Also Rises and Tender is the Night. Many of the social science disciplines have also long understood the importance of the question of how wars may impact societies. There exists, for instance, a long-standing debate in the field of sociology as to whether 1

Hall (1999) also fi nds no decline in association memberships in Britain over the last few decades, although he does fi nd evidence of some decline in levels of social trust.

Introduction

3

conflict with outside powers increases social cohesion within. In his classic study, Suicide, Emile Durkheim (1951) argued that war lowers the level of social anomie and thereby enhances social cohesion; others have claimed that wars may also fragment societies, depending on how the costs of waging war, whether financial or otherwise, are distributed across different segments of society (Stein 1980; Stohl 1980). Sociologists and historians have also found that periods immediately following wars are often characterized by a frenzy of activity, including higher rates of marriage, divorce, childbearing, violent crimes, and minority activism, regardless of whether the society was victorious or vanquished (Archer and Gartner 1984; Brandes 1950; Ember and Ember 1994; Jacobson 1959; Rodgers and Thornton 1985; South 1985; Wynn 1971). More recently, sociologists, psychologists, and psychiatrists have spawned a large body of work on the relationship between wartime experiences, posttraumatic stress disorder, and postwar social problems (e.g., Kulka et al. 1990; Norris 1992; Bracken et al. 1995; Parker 1996). The literature on the effects of war on the state has also flourished. Political scientists and sociologists have built a large and thriving body of work examining the impact of war on state-building (Tilly 1975; Skocpol 1979; Centeno 1997; Ertman 1997; Herbst 2000) and the rise of the welfare state (Kaufman 1983; Skocpol 1992), as well as on levels of state extraction and expenditures (Rasler and Thompson 1985; Levi 1988; Fujihira 2003). A number of political scientists have also explored the effects of war on public opinion, especially on presidential approval ratings (Mueller 1970; Kernell 1978; Jentleson 1992; Lian and Oneal 1993; Baker and Oneal 2001), and, more recently, public support for war itself (Goldsmith et al. 2005; Berinsky 2007). But political scientists have largely neglected the question of how wars may impact the long-term well-being of societies. Some political scientists have begun to investigate the long-term consequences of war on public health (e.g., Ghobarah et al. 2003). Theda Skocpol and her associates have conducted pathbreaking work on the effects of war on civic engagement (Skocpol 1999, 2000; Crowley and Skocpol 2001; Skocpol et al. 2002). This important research will be discussed extensively in the next section. But how the processes and outcomes of war affect the various facets of social life is a major issue that deserves much more focused attention from political scientists because, after all, as Arthur Stein notes, war is a public policy (Stein 1980: 1). A better understanding of the effects of this public policy should yield important insights for both political scientists and policymakers in thinking about how war-torn societies may be rebuilt and, indeed, whether to go to war in the first place. By examining the impact of war on civic engagement, then, this study seeks to fill an important gap in political science research.

Victory, Defeat, and Civic Engagement Civic engagement is a crucial component of social capital, which in turn produces an assortment of beneficial social outcomes, from higher rates of economic

4

Civic Engagement in Postwar Japan

growth (Fukuyama 1995) to improved political performance (Putnam 1993) and even better health and longer life expectancy (Putnam 2000; Rose 2000). These are undoubtedly important concerns for countries in times of peace, but they should be particularly pressing issues for societies seeking to recover from war and its various forms of damages. Precisely for this reason, international efforts at peace-building and reconstruction have devoted a substantial amount of resources to the strengthening of indigenous civil societies in recent years (Belloni 2001; Azimi et al. 2003). Some existing studies do offer insights into the issue of how war may affect the long-term levels of participation of the affected cohorts. Theda Skocpol and her associates in the Harvard University Civic Engagement Project (Skocpol 1999, 2000; Crowley and Skocpol 2001; Skocpol et al. 2002) draw on previously unavailable data on participation in membership associations in the United States to argue that major wars have historically served as the single most important catalyst for discontinuous jumps in levels of civic engagement in the United States (Skocpol 1999; Crowley and Skocpol 2001; Skocpol et al. 2002). To explain this phenomenon, the Skocpol group invokes what can be termed the “victory/defeat hypothesis.” This hypothesis suggests that defeat should produce a decrease in participation, whereas victory should lead to no fall, and perhaps even an increase. Scholars have offered at least three reasons for this “victory/defeat hypothesis.” First, wartime defeat is often accompanied by extensive physical destruction, which leads to economic downturn and, by implication, a loss of incomes. Second, war, especially when it becomes prolonged, is likely to interrupt formal educational opportunities. As numerous studies have shown, income and education strongly affect an individual’s level of participation (Verba et al. 1995; Brehm and Rahn 1997; Hall 1999; Oliver 2000). Of course, victorious countries may also see declines in incomes and diminished educational opportunities, but their recovery is likely to be faster and the extent of their losses much less severe compared to defeated countries. Victorious countries may also be able to compensate for the loss of formal educational opportunities through an equivalent of the U.S. G.I. Bill (Mettler 2005). But such a luxury is usually not an option for defeated societies, at least not in the short term. In defeated societies, then, education interrupted is likely to become education terminated, with severe long-term consequences for voluntary participation. Third, the effects of defeat may also be psychological. Here, Skocpol and her associates have offered the most systematic formulation to date. The reasoning behind this argument is one of social learning. Skocpol et al. begin from the premise that major wars, particularly those of the modern era, have generally caused the extensive mobilization of civilians into the war effort. Modern wars have typically been fought via joint action by state and society – that is, through what Peter Evans (1997) has termed “state–society synergy.” If a country is victorious in war, the argument goes, citizens find that their participation on the “home front,” or what might be termed “involuntary civic participation,” yielded real and tangible results, and, they are in turn likely to

Introduction

5

feel encouraged to continue to participate voluntarily after the war. Defeat in war, in contrast, is likely to lead citizens to believe that their participation had been futile and therefore to withdraw from public life altogether once the war is over. “After victory in war,” Skocpol et al. contend, “former combatants may have renewed energy and will to cooperate with their allies and friends. But groups that mobilize and then suffer defeat may well dissolve and fragment, as participants downplay their unsuccessfully realized identity” (Skocpol et al. 2002: 140).2

Empirical Evidence How well are these claims borne out by the empirical evidence? To what extent and how does war affect postwar levels of civic engagement? The fact is that, thus far, we have very little information. To date, only Skocpol’s Civic Engagement Project has collected any systematic empirical data on this question, and we thus only have reliable evidence from the United States. While data from the United States are valuable in their own right, generalizing on the relationship between war and civic engagement on the basis of the American case alone may be problematic, for several reasons. To begin with, it is just one case. Second, the United States lacks a clear-cut case of defeat in a major war. The Skocpol group invokes the U.S. South in the wake of the Civil War as a case of defeat, but the repercussions of defeat in civil wars may be quite different from those in international wars. In particular, civil wars may in general be much more disruptive to the social fabric of the defeated side than international wars. Third, the United States may also constitute an outlier within the class of victorious countries in World Wars I and II. Mainland United States has hardly ever faced invasion by a foreign power, which sets it apart from most other countries. In short, at the very least, the U.S. case needs to be compared against other country-cases. This book presents data from a broader range of cases, both inter- and intranational. The data challenge the victory/defeat hypothesis. First, national-level data from Japan show that after its complete defeat, levels of civic engagement actually rose compared to prewar levels. Second, regional-level data from Japan reveal substantial variation in the growth of civic engagement across its forty-six prefectures. And third, the study provides data on the long-term impact of war on civic engagement in thirteen countries that suggest that the

2

The Skocpol group also points to the manner in which war is conducted as a second variable to explain the rise or the fall of civic engagement following major wars (Skocpol et al. 2002: esp. 139–141). Specifically, they argue that the state–society synergy that evolves over the course of conducting war often presents an important opportunity for citizens to learn to cooperate with one another. However, in arguing that participation dwindled in the South following the Civil War and increased in the United States following World Wars I and II and also in the North after the Civil War, much of the explanation seems to fall on the victory/defeat variable.

Civic Engagement in Postwar Japan

6

overall war-induced growth in levels of participation was not specific to the case of Japan. The striking empirical puzzle with which this study grapples – the rise of Japanese civic engagement in the immediate aftermath of World War II – has long been “known” in scholarly folk wisdom.3 But this study is the first to offer systematic evidence that demonstrates the extent of the postwar Japanese civic engagement boom across a diverse array of membership associations, including youth groups, recreational organizations, women’s associations, social service organizations, and religious groups. The data were assembled through extensive fieldwork that drew on archival records of the Allied occupation of Japan; prewar Japanese government documents; the archives of various Japanese voluntary associations themselves; and available secondary resources.

Overview of the Argument If it is indeed the case that civic engagement may rise in the wake of wars, the next task is to offer a new theory to explain the relationship between war, on the one hand, and civic engagement, on the other. Why, contrary to the expectations of existing perspectives, should civic engagement rise, even in cases of devastating and total defeat, as in post–World War II Japan? And what explains the variations in this rise across different countries and subnational regions? The study advances a two-step framework that pays greater attention to the effects of citizens’ accumulated experiences. Specifically, it focuses on two independent variables: 1) wartime mobilization, and 2) legacies of prewar voluntary activities. First, wartime mobilization drives the overall rise in levels of civic engagement in the wake of wars by instilling important civic skills into the citizenry. Second, path-dependency effects from the prewar structures of participation account for the variations in the extent of this growth. Wartime Mobilization The experience of war, whether it ends in victory or defeat, is more than just a story of lost incomes, educational opportunities, or psychological devastation. Charles Tilly (1975: 42) noted, “War made the state and state made war,” and as numerous scholars have argued, wars generally strengthen state capacities (Skocpol 1992; Rasler and Thompson 1985). But wars can also strengthen society as well. As new wartime needs, both military and civilian, multiply, the state becomes increasingly incapable of providing all of the needed services on its own. The resources necessary to wage war must be mobilized and extracted from society, and this actually deepens the state’s dependence on society (Tilly 2004: 65). As Stein and Russett (1980: 402) suggested, “Many of the domestic effects of war are indeed a function of the extent of wartime mobilization.” 3

Drawing on extensive historical data, Tsujinaka (2002), for instance, has documented the explosion of the number of associations in Japan in the wake of World War II.

Introduction

7

At the micro-level, this means that mobilized citizens, on both the war front and the home front, come to assume increasingly large public responsibilities. They come into contact with individuals, officials, and groups that they would not have encountered otherwise, and in so doing they may acquire important communication skills, political and social awareness, organizational savvy, or, in short, what scholars have termed “civic skills” (Verba et al. 1995). War, then, may provide vast opportunities for social learning and create a large pool of citizens who are both willing and able to engage in collective endeavors. Freshly equipped with the skills for participation, these citizens should be well prepared to continue to participate voluntarily in collective activities of their choice once the war is over, whether that war ended in victory or defeat. Indeed, the rise of social movements in the United States such as the civil rights movement or the women’s movement during the 1950s and 1960s has often been attributed to experiences of mobilization during World War II (McAdam 1982; Freeman 1972). Former draftees have also been found to participate at higher rates than non-draftees (Jennings and Markus 1976). These phenomena have generally been viewed to support the victory/defeat hypothesis, but only because the United States happens to have been a victorious country. In fact, this study argues that the same phenomena can be observed in defeated countries as well. Path-Dependency Effects Wartime mobilization, then, should generally produce growth in civic engagement in the wake of wars. Mobilization, however, does not fully account for the extent of participation growth. The accumulation of civic skills per se does not guarantee that citizens will continue to voluntarily exercise those skills once war – or mobilization – is over. The extent to which these citizens actually participate in voluntary associations after wars depends crucially on the postwar social context, specifically the availability of opportunities. The availability of such opportunities, in turn, is shaped by the legacies of prewar associational activities. Organizational legacies affect the availability of opportunities for participation once the war is over, in two ways. First, where voluntary associations had flourished before the war, the organizational apparatus of those associations is likely to have at least partly survived, even if the associations themselves had ceased to operate during the war. Citizens seeking groups in which to participate may easily join those revived organizations. In contrast, where levels of associational activity had been low before the war, fewer associations are likely to have survived. With a smaller heritage of associations, citizens seeking to participate must often organize their own associations. As Paul Pierson (2004: 33) argues, however, the start-up costs for associations are considerable, and these barriers to entry are likely to impede the participation of many civicminded citizens.

8

Civic Engagement in Postwar Japan

Second, even if preexisting organizations have been revived or new ones have been founded, information on them must also be available to the citizens in order for participation to occur. This kind of information is likely to be available in greater abundance in societies where levels of participation had been high prior to the war. In areas or regions that had exhibited high levels of voluntary activities before the war, citizens seeking opportunities to participate in the wake of war are likely to know larger numbers of people who used to be members of voluntary associations before the war, who may provide them with information on the rebirth of associations or even with personal contacts at the organization. This should significantly facilitate postwar participation. In areas that had exhibited lower levels of participation, in contrast, the task of information-gathering should pose a much greater challenge. Thus, while agreeing with Putnam’s (1993) view that path-dependency effects exert a significant impact over cross-regional patterns in participatory behavior, this study suggests that those effects may operate in a slightly different fashion from those envisaged by Putnam. In Making Democracy Work, Putnam argued that path-dependency effects should shape the participatory behavior of citizens even over a period of centuries because social norms of reciprocity and the threat of sanctions ensured that, once in place, cooperation would persist. The insight of the present study is that citizens’ behavior may be more malleable than Putnam suggests, at least for absolute levels of participation. Social learning via mobilization may enable citizens to overcome the weight of the past. At the same time, the mediating effects of prewar legacies may exacerbate the relative differences in levels of participation across different societies. Between State and Society In emphasizing the effects of wartime mobilization on the postwar growth of civic engagement, this study also highlights the role of the state in shaping civil society. This emphasis on the effect of state policies over civil society follows recent studies by Levy (1999), Skocpol (2000), and Schwartz and Pharr (2003), which have argued against Putnam’s (1993, 1995b, 2000) primarily societalbased account for the patterns and changes in civil society. The state and its policies may crucially influence the subsequent configuration of civil society, sometimes inadvertently. However, this book also points to the effects of pathdependency, that the impact of the state is constrained by the social context in which it operates. That is, the state may influence society, but within limits that are imposed by the preexisting configuration of society. In this way, the book seeks to strike a more nuanced balance between the society-centric and statecentric perspectives. This view that the state may strengthen and empower civil society in significant ways (and vice versa) also challenges the conventional view that the relationship between state and society is inherently zero-sum. Scholars such as Peter Drucker and Everett Carl Ladd have argued that an expansion in the

Introduction

9

role of the state invariably crowds out the voluntary sector (Drucker 1993: 9; Ladd 1999). Even Alexis de Tocqueville, to some degree, subscribed to this view: “The more [government] puts itself in place of associations, the more particular persons, losing the idea of associating with each other, will need it to come to their aid; these are causes and effects that generate each other without rest” (Tocqueville 2000: 491). From this perspective, the ideal way to promote voluntary activities would be for the state, to the extent possible, to retreat from the private sphere. The findings from this study, however, suggest that state capacities and societal capacities may expand in tandem. Indeed, the insight that the state may empower societies by equipping them with crucial civic skills echoes some of the recent studies on human capital in the field of political economy (e.g., Iversen and Stephens 2008). Not only may the relationship between state and society be “synergistic,” in the sense that “active government and mobilized communities may enhance each other” (Evans 1997: 178); this study suggests that the effects of this synergy may persist long after mobilization is over. It should be stressed here that while this study points to the crucial effects of mobilization on subsequent levels of civic engagement, it by no means intends to endorse either war or military mobilization at a normative level. The empirical question of how mobilization affects citizens’ levels of participation is entirely separate from the normative issue of whether such mobilization is in fact desirable. To the extent that the hypothesis advanced in this study is valid, that mobilization yields positive benefits for postwar participation, it does not necessarily mean that societies should mobilize, and much less that they should mobilize for war. The normative question is also an important one, but one that the author of the present book is not qualified to address. It is an issue better left to political theorists and philosophers. The book will return to this point in the concluding chapter.

Empirical Testing and Case Selection The theoretical framework presented in this study will be tested in three steps. First, the study assembles data and conducts quantitative analyses of the combined effects of path-dependency and wartime mobilization on the postwar rise in civic engagement in forty-six Japanese prefectures. Second, it process-traces the postwar growth in participation through in-depth comparative case studies of the Japanese YMCA and Japanese judo in two cities respectively. The case studies complement the quantitative analyses in important ways because they are well suited for illuminating the precise pathways through which civic engagement rose in the wake of the war. Third, to test the broader applicability of the framework presented in this project beyond simply the Japanese context, the study conducts quantitative tests of the long-run impact of World War II on levels of civic engagement in thirteen industrialized countries. For purposes of assessing the impact of war on civic engagement, Japan in the wake of World War II represents a particularly important case.

10

Civic Engagement in Postwar Japan

Specifically, Japan immediately after World War II exhibited a confluence of factors that would lead one to expect a particularly good fit with the victory/ defeat hypothesis: high levels of destruction, unconditional surrender, and relatively slow economic recovery. The extensive damage that Japan incurred should have led to a severe weakening of the social fabric, and the unconditional surrender of the Japanese regime should have led to a complete loss of faith in the nation’s collective potential. These factors, in turn, should have led to a marked decline in civic activities. Thus, if the victory/defeat perspective should apply anywhere, it should apply to post–World War II Japan. World War II was the bloodiest war in human history, claiming an estimated 61 million military and civilian deaths around the world (Sivard 1986), roughly double the number of casualties from World War I. Moreover, World War II was especially notable in the extent of civilian damage; according to one estimate, 67 percent of all casualties in World War II were civilians, compared to only 5 percent in World War I (Goralski 1981). And Japan’s damage from World War II was among the most extensive of belligerent countries. Casualty figures from World War II continue to be contested to this day, but Table 1.1 provides a sense of the relative levels of casualties across the major belligerent countries. As shown in Table 1.1, Japan lost almost 1.8 million soldiers and more than 600,000 civilians in the war (Ellis 1993: 253), or the equivalent of almost 4 percent of the population. Close to 300,000 more were wounded or missing (Ellis 1993: 253). Among the major powers, Japan’s level of human casualties is third only to the Soviet Union and Germany (Ellis 1993: 253). Japan also incurred extensive physical damage. As shown in Table 1.2, almost 2.3 million private homes, or 65 percent of all residences, were destroyed, by far the most among the belligerent countries, except perhaps the Soviet Union, for which data are not available. The exceedingly high level of property damage probably owes to the fact that most residential housing in Japan during this period was built from wood, thereby making it much more vulnerable to fire raids than in European countries, where housing is generally built from less flammable material. More than eight times as many private homes were destroyed in Japan as in Germany, and almost ten times as many as in France (Goralski 1981: 427). An additional 1.3 million private homes were damaged. Loss of property amounted to 65.3 billion yen (calculated on the basis of prices as of August 15, 1945; Akimoto 1974). This high level of physical destruction also caused enormous social disruptions. Roughly 9 million people in Japan are said to have been homeless at the end of the war (Dower 1999: 47–48). According to estimates by the Japanese government, Tokyo Prefecture’s population fell by more than half between 1944 and 1945, from 7.2 million to 3.5 million people (Somusho, n.d.[1]). Since the number of deaths from fire raids has been estimated to be roughly around 100,000, most of the loss of population in Tokyo has been attributed to people fleeing to the countryside to avoid the fire raids, especially after March 1945, when the most extensive attacks on Tokyo occurred, than

Introduction

11

Table 1.1. Military and Civilian Casualties for Selected Countries, World War II Country

Armed Forces Casualties (Killed and Missing)

Civilian Casualties

Total Casualties

Population (in thousands, 1940)

Total Casualties per Population (%)

Austria Belgium Finland France Germany Italy Japan Netherlands UK USA USSR

29,395 9,561 89,900 205,707 3,250,000 226,900 1,740,000 13,700 271,311 405,400 11,000,000

145,000 75,000 3,400 173,260 2,300,000 60,000 668,000 236,300 60,600 0 6,700,000

174,395 84,561 93,300 378,967 5,550,000 286,900 2,408,000 250,000 331,911 405,400 17,700,000

6,705 8,346 3,698 41,000 69,835 44,341 62,967 8,879 48,226 132,637 192,379

2.60 1.01 2.52 0.92 7.95 0.65 3.82 2.82 0.69 0.31 9.20

Source: Casualty figures for Austria, Belgium, France, Japan (civilians only), the Netherlands, and the UK (armed forces) are from Goralski (1981); the rest of the figures are from Ellis (1993). Population figures as of 1940 are from Maddison (2003).

Table 1.2. Private Homes Destroyed and Damaged in Selected Countries, World War II Country France Germany Japan Netherlands UK

Private Homes Destroyed 255,500 270,000 2,251,928 82,530 456,000

Private Homes Damaged 1,051,000 1,300,000 1,270,000 427,620 4,703,000

Source: Goralski (1981: 425–427).

to deaths from the fire raids themselves (Taeuber 1958: 340). In particular, because the Japanese were aware of their vulnerability to fire raids, children were evacuated from the cities into the countryside as the war wore on; an estimated 450,000 students around the country were thus evacuated away from the urban centers, where most of the American bombing was concentrated, into rural areas (Monbusho 1972). Tokyo’s population figures would not return to prewar levels until 1953 (Somusho, n.d.[1]).

12

Civic Engagement in Postwar Japan

Economic damages were no less catastrophic. The real gross national product (GNP) fell by more than 40 percent between 1935 and 1946 and the GNP per capita by 35 percent (Hara 1998: 227). Between 1940 and 1947, the number of jobs in Japan fell from 3,737,713 to 2,707,132, or by about 28.1 percent (Somucho 1987). Indeed, over the same period, Tokyo lost 60 percent of its jobs, while Osaka, the next largest urban center, lost 43 percent (Somucho 1988b). Education was also cut short. This is important for this study because scholars have consistently found educational attainment to be the best predictor of civic engagement (Oliver 1984, 2000; Schlozman et al. 1994; Smith 1994; Putnam 1995b; Brady et al. 1999; Hall 1999; Wilson 2000). In Japan, the education of an entire generational cohort was cut short as students were mobilized into the war and home fronts. Beginning in 1942, for instance, college students, excluding those in the natural sciences, engineering, and medicine, were no longer exempted from the draft. Students who were too young to be eligible for the draft were increasingly mobilized into the home front toward the end of the war, which meant that in practice academic classes were no longer functional even when students could in theory stay in school. After 1944, secondary school students were removed from the schools and placed to work in war-related factories. It is estimated that more than 4 million students had thus been mobilized by July 1945 (Gakuto Doin n.d.). According to one account, most children over the age of ten were working in the fields or in war plants on a nearly or completely full-time basis by the end of the war (Havens 1978: 178). While in the postwar United States the G.I. Bill more than made up for the loss of educational opportunities during wartime (Elder and Meguro 1987: 451; Mettler 2002), in Japan no comparable counterpart to the G.I. Bill was passed, which meant that education once interrupted was typically never completed even once the war was over.4 Not only were economic and social conditions unfavorable for the expansion of civic engagement during the war; the immediate postwar conditions were hardly conducive to the expansion of participation. There is a myth that Japan’s economy “bounced back” from the destruction of war stronger than ever. In fact, the initial postwar years were an exceedingly trying time for the country, with recovery slow even compared to other countries emerging from World War II. Postwar poverty also discouraged many individuals who had been removed from school during wartime from returning to school after the war. Industrial production fell more dramatically and also recovered much more slowly than in many other countries, as shown in Table 1.3.

4

A recent study by Ichino and Winter-Ebmer (2004) compares the loss of human capital suffered by school-age children in Austria and Germany during World War II to their Swiss and Swedish counterparts. They argue that the mobilization of teachers, the destruction of school buildings, and the forced evacuation of children from the major cities of Austria and Germany led not only to lost educational opportunities compared to Switzerland and Sweden but also to a loss of income that persisted even forty years after the war.

Introduction

13

Table 1.3. Industrial Production before and after World War II, 1937–1955 Year 1937 1940 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955

Belgium France Germany Italy 100

100

100

100

31 74 86 93 94 95 110 107 107 116 124

39 73 87 102 112 113 128 133 129 141 155

34 40 60 89 109 131 140 152 160 196

95 99 105 126 143 145 159 173 188

Japan Netherlands 100 114 61 34 40 52 65 79 104 115 140 151 164

100 104 32 75 95 114 127 144 146 146 160 178 189

UK

United States

100

100 111 180 150 165 170 156 177 196 203 218 203 227

90 98 109 116 124 127 123 131 140 148

Calculated on the basis of 1937 = 100. Source: United Nations Secretariat, Statistical Office (various years).

In 1950, five years after the end of the war, industrial production in Japan was still only at 79 percent of 1937 levels, the lowest level among the countries listed in Table 1.3. Indeed, as the table shows, industrial production had returned to prewar levels in most countries by 1949, but not Japan. Even in 1952, Japanese industrial production was still the lowest among the former belligerents, with the exception of Belgium. Japan’s industrial production did not pick up momentum until 1953. Part of the reason for the slow economic recovery in Japan lay in inflation. Inflation, which had been creeping up steadily throughout the war,5 accelerated dramatically in the immediate postwar years. As Table 1.4 indicates, inflation hit Japan significantly harder than in most of the other participants in World War II. Price levels in Japan as of 1955 stood at 27,591 percent of 1937 levels, roughly five times higher than in Italy and ten times higher than in France. Not only did these staggering levels of inflation impose severe hardships on the citizenry; they also impeded long-term economic investments that would accelerate recovery. The slow economic recovery had serious public health consequences as well. Citizens faced difficulties in filling basic needs such as food and clothing. Food shortages became a major problem soon after the war, as consumption

5

Shinju Fujihira (2003) argues that the Japanese government made a deliberate decision to fi nance the war through infl ation rather than through taxation.

Civic Engagement in Postwar Japan

14

Table 1.4. Infl ation before and after World War II, 1937–1955 Year

Belgium

1937 1940 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955

100

330 355 389 384 467 442 413 409 409

France Germany 100 139 375 648 989 1,712 1,917 2,070 2,645 2,777 2,645 2,592 2,502

100 92.5 48 62.3 52 51 51 62 63 61 60 62

Italy 100 130 2,203 3,085 5,518 5,821 5,528 5,239 5,996 5,657 5,657 5,714 5,714

Japan Netherlands

UK

United States

100 130 278 1,290 3,837 10,190 16,580 19,667 27,309 28,154 28,154 27,872 27,591

100 126 155 161 176 202 212 242 295 304 304 307 319

100 91 123 140 176 191 180 189 210 204 202 202 204

100 120 167 232 250 260 271 304 372 362 342 345 349

Calculated on the basis of 1937 = 100. Source: United Nations Secretariat, Statistical Office (various years).

of grains plummeted to 75 percent of 1934 levels in 1946. Even in 1952, grain consumption stood only at 82 percent of 1934 levels (Havens 1978). In 1946, consumption of textiles was only 16 percent of 1934 levels and still stood at only 22 percent of 1934 levels in 1949 (Nakamura 1989: 22). This left a mark on the development of an entire generation of youth. For instance, the average twelve-year-old boy in Japan in 1941 was 143.1 centimeters tall, or roughly four feet eight inches. In 1948, the average twelveyear-old boy was only 135 centimeters tall, or roughly four feet five inches (Somusho 2008, n.d.[3]). In addition to the economic, social, and physical damage, the immediate postwar period inflicted psychological traumas on the Japanese population. Not only did Japan submit to unconditional surrender; it faced foreign occupation and sweeping reforms of its domestic institutions, or “nation-building,” which proceeded on an unprecedented scale except in post–World War II West Germany. The reforms ranged from constitutional revision and land reform, to the breakup of Japan’s corporate conglomerates (zaibatsu) such as Mitsubishi and Mitsui. And the reforms did not stop at institutional change; the Allied occupation also sought to re-make the Japanese mind and culture through educational reforms as well, emphasizing the Western values of free deliberation and compromise (Kage 2003a). The occupation era is often portrayed as an “emancipating” period of liberation (Dower 1999: 26), but it is difficult to imagine that this complete rejection not only of Japan’s pre–World War II institutions, both political and economic but also of social practices

Introduction

15

and culture, was anything but traumatizing for the citizenry (Dower 1999, esp. ch. 3).6 Thus, extensive physical, economic, and social damage, both during and after the war, and the postwar occupation reforms created severe social disruptions that should have been particularly unfavorable for the rise of civic voluntarism in Japan in the immediate postwar period. But, as the next chapter shows, civic engagement rose briskly in postwar Japan. If the victory/defeat hypothesis fails in this easy case, then a new approach to understanding the course of civic engagement after wars is necessary. An examination of the Japanese case should therefore shed light on the conditions under which civic engagement may expand in war-torn nations more generally. The choice to focus primarily, though not exclusively, on Japan between 1945 and 1955 rests not only on theoretical considerations, but also on the contemporary policy relevance of the Japanese case as well. As noted earlier, a vibrant civil society constitutes an important precondition for successful democratic transition. And in the runup to the Iraq War, the Bush administration invoked Japan’s democratization, as well as that of Germany, on repeated occasions as the ideal-typical case on which the Iraqi occupation would be modeled (e.g., Sanger and Schmitt 2002).7 The analogy with Iraq was a poor one, but the contemporary prominence of the Japan “model” suggests that a study of Japan’s civil society in the wake of World War II may help to illuminate the conditions under which war-torn countries may democratize even today.

Scope of Study This study seeks to explain the impact of war on postwar levels of civic engagement. Following much of the existing literature, this study defines civic engagement as participation in membership associations (e.g., Skocpol 1999; Schofer and Foucarde-Gourinchas 2001; Andolina et al. 2003). Membership associations are those “groups in which people do things together as fellow members” (Skocpol 1999: 34, fn.19). Non-membership civic associations, such as non-profit research institutes, were excluded because they do not provide as useful a measure of the extent to which individual citizens are willing to engage in collective endeavors as membership associations do. For the same reason, associations with only corporate, rather than individual, members, most typically industrial and/or business associations, were also excluded. Among associations with individual members, professional associations were also excluded

6

7

This, of course, is in addition to the traumas incurred from the experiences of war. Given the high levels of civilian damage in Japan, the psychological wounds from destruction and the loss of family and friends should have been considerable. In addition, soldiers who served in the war should have experienced mental and emotional health issues as well. For a recent study of war trauma among Japanese World War II veterans, see Noda (1998). See Dower (2005) for a rebuttal of this view.

16

Civic Engagement in Postwar Japan

if, as with the American Bar Association, membership is not voluntary but a condition for practicing one’s trade. For specialists of Asia, another note is in order here about the scope of civic engagement being analyzed in this book. As in many other Asian societies, there exists in Japan a variety of “administered” mass organizations: “mass civilian organization[s] created and managed by a political regime to implement public policy” (Kasza 1995: 7). As Haddad (2006: 1223) notes, these groups have “close, ongoing relationships with municipalities and/or specific government ministries and participate in the formation and the implementation of policy.” The most prominent of these associations in Japan include neighborhood organizations, the Red Cross Volunteers, Seinendan (youth organizations), and Chiiki Fujinkai (local women’s organizations). While these groups clearly enroll some of the largest memberships in Japan, encompassing almost every household particularly in the rural areas, it is not the goal of this study to explain levels of membership in these associations. This is not to say that these groups may not serve an important purpose in public life; to the contrary, their role is often very much a public one. However, because membership in these associations is typically not voluntary, it is exceedingly difficult to measure the genuine levels of civic engagement or changes therein that they bring about. This book therefore focuses on more clearly voluntary associations. This book focuses on Japan during the immediate decade following defeat in World War II, or 1945–1955. This is due to two considerations. First, the initial decade after defeat should be the most affected by defeat. Second, and in a related point, Japan’s economic growth accelerated after 1955, which means that factors other than the war experience are likely have to come into play.8 This is not to suggest that the impact of defeat dissipated after 1955. But after 1955, the rise in civic engagement may have been affected also by growing incomes and educational attainments, as well as by the increase in civic engagement prior to 1955. In order to distinguish analytically the effects of war from the effects of economic growth, this book will be primarily interested in the initial decade following 1945. It should also be noted that this book is primarily concerned with the trajectory of civic engagement in postwar democracies. The reason for this is straightforward. More than other forms of government, the success of a democracy relies on voluntary participation on the part of its citizens (Putnam 1993; Tocqueville 2000). Indeed, Robert Dahl, in his classic study, Polyarchy, lists the “freedom of association” first among his twenty requirements for democracy (Dahl 1971: 3). At the same time, this study includes countries that had been either democratic or authoritarian prior to war. If we are interested in democratization, we 8

Some economists would argue that Japan’s economic growth in fact took off after 1952, at the time of the Korean War (Kosai 1986). However, as the previous section shows, Japan’s economic recovery as compared with levels of production in the prewar period still trailed those of its European counterparts as of 1955.

Introduction

17

cannot only study democracies. In addition, the distinction between democracies and non-democracies is generally not a difference in kind but in degree. While scholars of democratization typically view the freedom of association as one of the distinguishing characteristics of democracies (Dahl 1971: 2–3; O’Donnell and Schmitter 1986: 8; Diamond 1996: 23–24), it is not unusual for authoritarian regimes to provide a measure of freedom to some, if not all, types of groups. Guilds flourished in pre-democratic England; Putnam speaks of the flourishing of communes in twelfth-century northern Italy (Putnam 1993: ch. 5). Even Communist Poland allowed religious groups and the Solidarity movement to operate and to expand. As Tsai (2007) points out, “solidary groups” may serve as an important conduit for local public goods provision, even in non-democratic regimes. And if one of the motivations for this study is to explore the preconditions for successful democratization, it is crucial to explore the path-dependent sources of voluntarism in these former authoritarian countries as well. It may well be that the framework advanced in this volume can also be applied to explain the trajectory of civic engagement in non-democracies in the wake of wars, in particular for segments of civil society that are permitted to operate under relatively liberal conditions. For the time being, however, this study seeks to set more modest goals for itself.

Plan of the Book This book will proceed as follows. The next chapter presents new data that show that civic engagement grew at an impressive rate in Japan between 1945 and 1955. The data show that civic engagement in Japan grew across a wide range of groups, from youth/recreational, women’s, and social service, to religious organizations, but that there also was substantial cross-regional variation in the extent of this rise. Chapter 3 more fully lays out the theoretical framework to explain this increase, focusing on the effects of a) wartime mobilization and b) legaciesof prewar participation. First, mobilization prepares the ground for an overall growth in levels of civic engagement by instilling civic skills into a citizenry. Second, the extent to which citizens make use of those newly acquired skills in the postwar period depends crucially on the preexisting legacies of participation, which shape the availability of opportunities for citizen participation by shaping the costs of association-building and information- gathering once the war is over. Chapters 4 through 7 test the argument presented in Chapter 3, in four stages. First, Chapter 4 shows through quantitative tests on data presented in Chapter 2 that a combination of wartime mobilization and levels of prewar voluntarism accounts for the rates of membership growth in the postwar period in Japan. Chapter 5 tests the broader applicability of the argument beyond the Japanese context by exploring the long-term impact of World War II on rates of participation in thirteen industrialized countries. The analysis shows that

18

Civic Engagement in Postwar Japan

the rise in participation among individuals in the war-affected cohorts of moremobilized countries was much greater than in countries that mobilized less. Chapters 6 and 7 then present case studies of two associations, the Japanese YMCA and the umbrella organization for Japanese judo (Kodokan), in two cities, respectively. These case studies further illuminate the complex relationship between wartime mobilization and preexisting prewar legacies in the acceleration of membership growth in the postwar period. Chapter 8 addresses the “so what?” question. If civic engagement grew rapidly in the wake of World War II in Japan, what might have been some of the macro-level consequences of this growth? The chapter presents results from preliminary analyses that suggest that higher levels of civic engagement led to more a rapid postwar reconstruction of Japan’s key social infrastructure. Chapter 9 concludes by summarizing the findings and by offering possible avenues for further research.

2 Civic Engagement The Dependent Variable

Overview As noted in Chapter 1, little empirical data outside of the United States have thus far been compiled on the changes in the trajectory of civic engagement in the wake of wars. Chapter 1 further explained why collecting data on the case of mid-twentieth-century Japan would be especially useful as a test of the standard victory/defeat hypothesis. What in fact happened to levels of civic engagement in Japan in the wake of defeat? How do the postwar levels of voluntary participation compare with the prewar period? This chapter presents the results of an original mining of Japanese voluntary association and government archives. Much of the data presented in this chapter have never been systematically compiled, either in the Japanese or English language. Drawing on these data, the chapter advances five main claims: 1) In absolute terms, levels of membership grew to much higher levels in the postwar period compared to the prewar period. The available data generally exhibit a sharp rise in memberships during the first postwar decade, even before Japan’s economic “takeoff” began around the early to mid-1950s. 2) Levels of postwar membership growth were at least as rapid for associations that operated under fairly liberal conditions during the prewar and wartime periods as those that had been more repressed. 3) In the immediate postwar era, membership in groups with more “indigenous” Japanese origins appears to have risen at rates that are comparable with groups with more “Western” origins. 4) There is some evidence that associations that were suppressed by the U.S. occupation exhibited less rapid increases in memberships compared to those that were not; but in either case, by 1955, membership appears to have grown in absolute terms. 5) There is considerable intra-Japan variation in the extent to which membership in voluntary associations grew in the wake of World War II. 19

20

Civic Engagement in Postwar Japan

The chapter is structured as follows. The next section outlines the procedures that were used to assess the changes in levels of civic engagement from prewar to postwar Japan. The third section lays out the figures for youth/recreational groups, women’s groups, social service groups, and religious organizations at the national level. The fourth section presents figures on the cross-prefectural variation in the growth of civic engagement. Finally, the last section summarizes the findings.

Data Selection Which Groups? No comprehensive data source on associational memberships is available for the period under study. Therefore, this study sought to collect data from as broad a range of voluntary membership associations as was possible. Toward this end, I identified four types of membership associations for which data would be collected: youth/recreational groups, women’s organizations, social service groups, and religious organizations. Other types of groups could have been included, such as environmental groups, human rights groups, and labor unions. However, it was only after the 1960s that environmental and human rights issues became a major focus of civic groups around the world (Iriye 2002), and this pattern is also evident in Japan (e.g., Nihon Bunka Chuo Renmei 1938, 1939, 1943; Mainichi Shinbunsha various years). Labor unions are also not ideal for the purposes of this study because in postwar Japan labor union membership was often not entirely voluntary, due to the emergence in many industries of the union shop. However, since the surge in the labor movement is often noted, membership figures have been compiled and will be presented in this chapter for comparison with other types of groups. Nevertheless, they will be excluded from the quantitative analyses that will be presented in Chapter 4. Membership in religious organizations requires a separate discussion. Membership in Shinto and Buddhism in Japan are more often matters of routine than of deliberate choice for many Japanese citizens; that is, most individuals become affiliated early in life with the same religions as their parents and ancestors and remain so throughout the course of their lives without making a real commitment to them. Of course, many people do make a real commitment, but raw membership figures cannot provide a sense of this. The figures simply show near-universal membership of the Japanese population (Monbusho various years). Therefore, these data do not yield much insight for the puzzle at hand. By contrast, membership in Christianity, as well as the “new” religions, does reflect more conscious decisionmaking, especially in the mid-twentieth century, when hardly any Japanese citizen had Christian ancestors. Careful examination of the official figures for the “new” religions revealed that membership figures tended to fluctuate wildly, casting some doubt on

Civic Engagement: The Dependent Variable

21

their reliability. This study thus presents figures only on Christian church membership. Data Sources Conceivably, there are two methods for investigating the levels of voluntary association memberships. The first would be to start with a full list of Japanese civic organizations that were in existence in Japan at the beginning of World War II and to trace their membership figures into the postwar period. Alternatively, one might also search for public opinion surveys that track citizens’ responses regarding their affiliations with various membership organizations before, during, and after World War II. For religious groups, the present study used the first of these approaches. The Japanese government has conducted surveys on all religious groups in existence in Japan sporadically during the prewar period and for much of the postwar period. For the prewar period, this information is available from several sources, including the National Diet Library (1955); the Dai Nihon Teikoku Monbusho Nenpo [Yearbook of the Imperial Ministry of Education], published annually by the Ministry of Education; and the Shukyo Nenkan of 1939 (Murata 1939). For the postwar period, data since 1952 have been publicly available through government statistics published by the Ministry of Education, but information for the late 1940s had not been available. In this research, I found these data in the records of the U.S. Occupation of Japan at the U.S. National Archives in College Park, Maryland. For labor unions, prewar data are available through the Labor Almanac [Rodo Nenkan], which was compiled by the Ohara Institute for Social Research. Postwar data are available from the Ministry of Labor. Data collection for youth/recreational groups, women’s associations, and social service organizations presented greater challenges. In contrast to religious activities, government survey figures for membership in youth groups are only available for the years 1922, 1924, 1931, 1936, and 1956. For women’s groups, results of national-level government surveys are available for most years of the postwar period beginning in 1949, but no corresponding data are available for the prewar period. For social service organizations, aggregate national-level data simply do not exist at all. While numerous public opinion surveys that have been conducted in the postwar period have asked questions on associational memberships, to the best of my knowledge such questions were not asked in any prewar surveys. An alternative method for tracking membership figures needed to be devised. Following Theda Skocpol’s methodology (Skocpol 1999, 2000, 2002), this study chose to focus on the larger groups in the above-mentioned categories. This method is based on the assumption that tracking membership figures for the largest organizations should also shed light on the patterns of participation among the broader citizenry. Admittedly, it is possible that membership patterns for large associations may not fully reflect broader societal patterns

22

Civic Engagement in Postwar Japan

in participation, so the results shown here will be a rough approximation. Nevertheless, as a full list of voluntary associations from the immediate postwar period is not available, examining simply the largest groups is a reasonable second-best method to ensure that the study captures broader society-wide trends, rather than trends that are idiosyncratic to any particular organization. In order to illuminate trends that transcend various subsectors of civil society rather than those that are particular to specific segments of it, this study draws on four different subfields within civil society, and on as many groups as possible. Two sources of information list the major associations for all four types of groups from the prewar period. First, the Nihon Bunka Dantai Nenkan [Yearbook of Cultural Groups in Japan], which was issued in 1938, 1939, and 1943, is the largest surviving sourcebook for associations from the prewar and wartime periods. The 1943 version, for instance, lists 998 groups that were in existence as of November 1942 (Nihon Bunka Chuo Renmei 1943). While these yearbooks do not appear to have selected groups in a rigorously scientific manner, they do seem to cover most of the major organizations that were active at the time. This is confirmed by a second major source of information on associations from the prewar period, the Mainichi Nenkan [Mainichi Almanac], which lists the major organizations from the prewar into the postwar periods in varying depths every year beginning in 1920. These two sources were then supplemented by sources that focus more narrowly on specific segments of civil society. For women’s groups, for instance, the Fujin Nenkan [Women’s Almanac], first published in 1920, then annually between 1936 and 1940, and also between 1948 and 1950, provides a wealth of information on the major women’s organizations that were in existence during the period under study. For youth/recreational groups, information from the Nihon Bunka Dantai Nenkan and the Mainichi Nenkan was supplemented by surveys of youth organizations compiled by the Ministry of Education in 1939, 1940, and 1942. For social service groups, the Nihon Shakai Jigyo Nenkan [Japanese Almanac of Social Services], published annually between 1920 and 1925, 1933 and 1943, and 1947 by the Chuo Shakai Jigyo Kyokai [National Association for Social Services], also served as a valuable resource for major groups that were active around the prewar and immediate postwar periods. These sources helped to establish the range of major associations that were in existence in Japan during the prewar and the immediate postwar periods. The groups were then further narrowed down by data availability. To gather the individual group-level data, I drew on the published histories of the groups, visited the groups’ offices and archives, and conducted research at the U.S. National Archives for the records of the U.S. occupation. Unfortunately, the number of groups for which data are available and/or have survived is not exceedingly large. For many of the groups, information on memberships was scattered across a number of different sources that ranged from Japanese government documents, to the groups’ annual reports and internal documents, to documents of the U.S. occupation. Moreover, different

Civic Engagement: The Dependent Variable

23

sources at times provided contradictory figures, perhaps owing partly to different counting schemes.1 When the membership figures reported by the group itself contradicted figures reported by other sources, the former were accorded priority. Many groups also had missing data for many years. When this was the case, I tried, to the extent possible, to consult the published histories of the groups for clues as to whether memberships were likely to be growing or declining during the missing years. Piecing together the participation data thus presented something of an enormous jigsaw puzzle, but the following represents a first attempt at presenting the available information.

The Rise in Civic Engagement in Postwar Japan, 1945–1955 The information that I have uncovered on youth/recreational, women’s, social service, and religious organizations largely suggests that participation in civic activities rose dramatically in Japan following World War II. This section first presents participation data for youth/recreational organizations, followed by those for women’s, social service, and religious organizations. As noted in Chapter 1, this project focuses primarily on levels of participation up to 1955. This allows for the separation of the effects of war from the effects of Japan’s high-speed growth, which began in the early to mid-1950s. However, to provide a sense of the trajectory of civic engagement during the postwar period, the chapter presents data from the prewar period until 1960 wherever the data are available. Youth/Recreational Groups The study compiled data for four of the largest youth organizations: the YMCA Japan, the YWCA Japan, the Boy Scouts of Japan, and the Girl Scouts of Japan. Prewar data for the YWCA Japan, the Boy Scouts Japan, and the Girl Scouts Japan are much more limited than for the YMCA Japan. In addition, the study presents data on two recreational groups open to participants of all ages: the Japan Alpine Club and the Kodokan judo association. As more “indigenous” activities, these groups, and especially the judo data, provide a useful contrast against the four youth groups, all of which have more “Western” roots. The YMCA Japan, founded in 1903, is one of the oldest youth organizations in Japan.2 Annual reports of the YMCA Japan for many of the prewar years report membership figures, allowing us to gain a fairly accurate picture of the trajectory of membership figures, as seen in Figure 2.1.

1

2

For instance, for the YMCA and the YWCA, some of the figures appear to include members of university outposts of the YMCA, while others do not. As will be seen in Chapter 6, some local-level chapters of the YMCA were founded before the YMCA Japan.

Civic Engagement in Postwar Japan

24

Number of Members

40000 30000 20000 10000 0 1925

1930

1935

1940

1945

1950

1955

Year

Figure 2.1. Membership in the YMCA Japan, 1925–1956. Source: Annual Reports, YMCA Japan (various years).

Figure 2.1 reveals that membership in the YMCA grew dramatically in the immediate years after the war. Prewar membership peaked in 1922, at 19,500 members (YMCA Japan various years). While data are missing for much of the late 1930s and the early 1940s, a reading of the histories of the YMCA Japan during this period suggests that the group not only encountered repression by the authorities during this period for on account of its Western origins but also increasingly lost members to conscription as the war drew on, which brought the group’s activities to a virtual halt by the early 1940s (Nara 1958; YMCA Japan 2003). Thus, it is reasonably safe to conclude that membership was certainly not growing, indeed more likely to be falling, during the wartime years between the late 1930s and the early 1940s. In 1950, the first year for which postwar data are available at the national level, membership stood at 20,841, already surpassing prewar peak levels; thereafter, it grew rapidly to 35,582 members in 1953 before declining slightly to about 32,130 in 1956.3 Surviving data for some of the YMCA’s major cities also confirm this general upward trend in memberships during the immediate postwar period. Indeed, postwar membership in the YMCA Japan rose steadily and soon outnumbered prewar levels in every major city except for Sendai. Note also that this rapid growth in memberships takes place before 1950, well before Japan’s economic growth takes off. This is shown in Table 2.1. Table 2.1 not only confirms the rapid rise in YMCA memberships during the immediate postwar period; it also helps to fill in some of the gaps in the national-level membership figures during the late 1930s and early 1940s. In particular, as expected, memberships do appear to have fallen substantially between the late 1930s and the mid-1940s. The YMCAs in Nagoya and Kobe, for instance, lost a sizable portion of their memberships between 1936 and 3

Membership in the YMCA continued to grow in the late 1950s and into the 1960s.

25

Table 2.1. YMCA Membership in Selected Japanese Cities, 1922–1950 City Sendai Tokyo Yokohama Nagoya Kyoto Osaka Kobe

1922

1923

1925

1936

1,746

1937 2,247 1,107

283 899

563

837

541 502

1,736 677

1940

1943

1945

1948

1950

326 2,093 650 511 392 1,600 583

257 1,755 313 405 1,363 632 306

528 200 350 215 120

2,093 600 997 795 1,166 1,333

284 3,973 1,593 2,909 963 2,199 2,395

Source: Compiled from Nara (1958); Yoshimura (1964); Osaka YMCA (1982); Kobe YMCA Hyakunenshi Hensanshitsu (1987); Owaku (1987).

Civic Engagement in Postwar Japan

26

Number of Members

6000 5000 4000 3000 2000 1000 0 1925

1935

1945

1955

Year

Figure 2.2. Membership in the YWCA Japan, 1927–1960. Source: Data for 1927, 1948, and 1957–1960 are from YWCA Japan, internal documents; data for 1947 are from Nihon YWCA (2005a: 85); data for 1950 and 1951 are from Monbusho Shakai Kyoiku-kyoku (1954a); data for 1952 are from Monbusho Shakai Kyoiku-kyoku (1953b); data for 1956 are from Monbusho Shakai Kyoikukyoku (1956).

1940. Tokyo, Yokohama, and Osaka similarly lost members between 1937 and 1940. Not surprisingly, the YMCAs in all cities saw a drastic decline in memberships between 1940 and 1945. The Kyoto YMCA’s membership, in particular, fell almost 85 percent between 1943 and 1945, from 1,363 to 215 members; the Osaka YMCA also lost more than 80 percent of its members during the same period, from 632 members in 1943 to 120 members in 1945. These figures make the postwar rise of the YMCA all the more remarkable. Surviving prewar data for the YWCA Japan, the Boy Scouts Japan, and the Girl Scouts Japan are much more limited compared to the YMCA, and therefore the figures must be interpreted with some caution. These are shown in Figures 2.2 through 2.4. For the YWCA Japan, founded in 1905, total membership in 1927, only a few years after the height of the era of liberalism in Japan known as the period of “Taisho democracy,” stood at 2,364. By contrast, in 1948, only the second year for which postwar data are available, YWCA membership had already reached 3,783 despite all of the hardships of the time. The YWCA continued to grow steadily thereafter. Admittedly, the evidence as to whether the YWCA Japan saw a postwar surge in memberships equivalent to the one that occurred with the YMCA Japan is somewhat inconclusive. If we assume that prewar membership levels for the YWCA Japan grew at similar rates to the YMCA, then, at its peak, membership figures for the YWCA Japan in the prewar period may have been somewhat higher than the 1927 figure of 2,364. However, a reading of the history of the era does not suggest rapid membership growth after this year, and especially after the early 1930s; for instance, while a wave of new local YWCA chapters

Civic Engagement: The Dependent Variable

27

had been established across Japan during the late 1910s into the 1920s, the last new local chapter of the YWCA Japan in the prewar era appears to have been founded in 1931 in Nagoya. This suggests that if growth in YWCA membership occurred at all after 1927, it is unlikely to have been dramatic, particularly after 1931 (Nihon YWCA 1987: 101–137; 2005a: chs. 2–4). But membership growth after the war was certainly quite strong. And the growth in membership occurred in tandem with the founding of more local branches. Eight new branches sprung up between 1945 and 1950, mostly in areas where the YWCA had not had a presence in the prewar era: Shonan (1946), Fukuoka (1948), Kure (1948), Sendai (1948), Sapporo (1949), Numazu (1949), Hakodate (1949), and Seien (Hamamatsu, 1950) (Nihon YWCA 2005b: 76–124). Three more were founded in the next three years. The story for the Boy Scouts also confirms the pattern of a postwar surge. Historical accounts of the Boy Scouts Japan note that the Boy Scouts increasingly encountered suspicion from authorities after the late 1920s for its Western origins. Tanaka (1999) argues that the Boy Scouts launched an extensive recruitment drive during the 1930s to shield itself from this suspicion and successfully expanded its membership. As shown in Figure 2.3, prewar data, which are only available from 1939, show that membership at that time stood at 3,100. It is difficult to imagine that Boy Scout memberships rose to new heights during the 1940–1945 period. While it is conceivable that the 1939 figures may not represent peak prewar membership for the Boy Scouts, they are likely to have been close to it. In the postwar period, membership figures quickly surpassed the prewar figure; in 1949, the first year for which data are available, membership already stood at 14,410. And given the considerable difference in levels of membership between 1939 and 1949, it seems reasonably safe to conclude that the Boy Scouts in Japan became a much bigger organization in the wake of the war compared to the prewar period.4 If we assume, as with the YWCA Japan, that membership trends for the Boy Scouts closely paralleled those of the YMCA, this story of membership reaching unprecedented new heights in the postwar period becomes even more plausible. Surviving data for the Girl Scouts Japan also appear to suggest a sharp rise in memberships between the prewar and postwar periods. This is shown in Figure 2.4. 4

Some surviving sources suggest that the membership levels for the Boy Scouts Japan during the prewar period far exceeded that of the postwar period: For instance, the Shonendan Nihon Renmei (various years) lists membership levels as of 1930, 1932, 1934, and 1936 as 71,920, 88,327, 82,979, and 102,510, respectively. However, as Tanaka (1999) notes, the prewar Boy Scout membership included both Boy Scouts and Sea Scouts members. In the postwar era, the two organizations parted ways and became separate associations, especially because SCAP was skeptical of the Sea Scouts for having counted many members of the Japanese Navy among its leaders and did not allow for the revival of the Sea Scouts until 1951. The Boy Scouts were allowed to resume activities much sooner, in late 1946 (Tanaka 1995, 1999). This lack of continuity in the membership figures makes it particularly difficult to trace the Boy Scouts’ membership figures between the prewar and postwar periods.

Civic Engagement in Postwar Japan

28

Number of Members

75000

50000

25000

0 1935

1940

1945

1950

1955

1960

Year

Figure 2.3. Membership in the Boy Scouts Japan, 1939–1960. Source: Data for 1939 from e-mail communication, Malaï Séraj, Centre for Prospective Studies and Documentation, World Scouts Bureau, Geneva, Switzerland, October 23, 2002; postwar data are from Zaidan Hojin Boy Scout Nihon Renmei (1978).

Number of Members

125000 10000 7500 5000 2500 0 1925

1935

1945

1955

Year

Figure 2.4. Membership in the Girl Scouts Japan, 1929–1960. Source: Prewar data are from Yaguchi (2008: 153–161), except for 1942 data, which are from Girl Scout Nihon Renmei (1971: 30); data for 1948 are from Girl Scout Nihon Renmei (2000: 40); data for 1949–1960 are from Girl Scout Nihon Renmei (2000).

The peak membership figure for the Girl Scouts, at least on the basis of surviving figures, stood only at 316, in 1931. In 1948, the first year for which data have survived, membership had already reached 3,581, more than a tenfold increase compared to the 1931 figures. As with the YWCA and the Boy Scouts, there is a relatively long period for which no data are available, so the findings must be interpreted with some caution. But Yaguchi (2008: 166), who conducted an extensive study of the Girl Scouts in prewar Japan, notes that Girl Scout activities stagnated during the 1930s and after 1934 the group’s

Civic Engagement: The Dependent Variable

29

Number of New Members

800 600 400 200 0 1925

1935

1945

1955

Year

Figure 2.5. New Members in the Nihon Sangakukai [Japan Alpine Club], 1925–1960. Source: Nihon Sangakukai (2007: 136–137).

newsletter was suspended. And in 1942, the Girl Scouts Japan was formally disbanded (Yaguchi 2008: 168). Moreover, in general the 1930s are not known to have been an era of explosive growth in voluntary associational activities, and certainly not for organizations with Western roots, as with the Girl Scouts. Thus, while there are no data to disprove that a rapid growth in Girl Scout memberships occurred during the 1930s, in reality this seems quite unlikely. While more information would certainly be desirable, it seems fairly reasonable to conclude that that the dramatic rise in Girl Scout memberships that occurred in the wake of World War II did not simply represent a return to prewar levels but instead signified new heights for the association. Data are also available for the Japan Alpine Club. Founded in 1905, the Alpine Club is one of the oldest, as well as the largest, recreational associations in Japan. Figure 2.5 presents the available membership figures. Note that the numbers shown are for the number of new members every year rather than total numbers of members. In a somewhat different pattern from the YMCA, the number of new membership figures for the Alpine Club actually grew steadily through the whole prewar period and even the early 1940s before dropping off at the tail end of the war. This may be due to the fact that as a less “Western” organization, the Alpine Club faced less repression during wartime than the YMCA. But in any case it is clear that after the war, membership grew much faster than previously. The average number of new members per year for 1925–1943 was 69.8 members, while for the period 1947–1955 it rose to 230.0. Note that the Japan Alpine Club reports that chaos in the club offices during the early postwar years resulted in many of the membership applications that arrived in the mail in those years not being processed properly in the year that they were received (Nihon Sangakukai 2007: 152–153). This means that at least some of the new members who are shown in Figure 2.5 to have joined in

Civic Engagement in Postwar Japan

30

Number of New Members

35000 30000 25000 20000 15000 10000 5000 0 1925

1935

1945

1955

Year

Figure 2.6. New Members in the Kodokan, 1925–1960. Source: Internal documents courtesy of Kodokan.

1948, the peak year for this period, may have actually enrolled in 1947 or even 1946. Nevertheless, this does not change the overall pattern of surging growth in the immediate aftermath of war. Figure 2.6 presents data for the Kodokan, which is the national umbrella association for judo in Japan. For purposes of this study, the Kodokan provides a particularly interesting case, as, in contrast to many of the youth groups shown earlier, such as the YWCA or the Scouts, the U.S. occupation viewed judo as promoting militaristic values and actively sought to discourage its growth. The Supreme Commander, Allied Powers (hereafter SCAP), went so far as to ban judo from Japanese schools between 1945 and 1950, either as part of the school curriculum or as an extracurricular activity.5 Note that, as with the Japan Alpine Club, the figures are for new enrollments rather than total membership figures. Figure 2.6 shows that the fastest growth in judo membership occurred both before 1945, before the occupation began, and following the end of the occupation, in 1952. As with the Alpine Club, the fact that judo did not face repression during the war is likely to have facilitated the continued membership growth during the late 1930s and into the early 1940s. The lower rate of membership growth between 1945 and 1950, on the other hand, should owe at least in part to occupation policies that discouraged participation in judo. Nevertheless, it is notable that, despite SCAP’s efforts to discourage judo, membership still continued to grow, and the growth accelerated even while the ban on school judo was still in place, during the late 1940s. Then when the ban was lifted, the rate of growth was even faster than it had been when the Japanese state was actively encouraging judo participation during the 1930s into the 5

Similarly, the occupation also discouraged participation in kendo as well as kyudo (traditional Japanese archery). As with judo, both were banned from Japanese schools during the occupation.

Civic Engagement: The Dependent Variable

31

Number of Members

16000

12000

8000

4000

0 1930

1935

1940

1945

1950

1955

1960

Year

Figure 2.7. Membership in the Tomo no Kai, 1930–1960. Source: Zenkoku Tomo no Kai Chuobu (2000).

early 1940s. Thus, the judo data highlight the fact that while U.S. occupation policies did have some bearing on the rate of membership growth, growth itself was resilient and continued despite efforts to discourage it. Again, data from the Alpine Club and Kodokan are particularly valuable because, in contrast to groups such as the YMCA or the Scouts, the Alpine Club and especially Kodokan are more indigenously rooted and are not Japanese outposts of organizations that originated in the West. Thus, the data for the two groups suggest that the rise in civic engagement was not a function of Japan “embracing its defeat” by the United States, but rather reflected a new willingness on the part of ordinary citizens to join groups of all kinds. This point will be explored in further detail in Chapter 3. Women’s Groups As with the Alpine Club and Kodokan, but in contrast to many of the major youth associations, women’s organizations in Japan generally tend to have indigenous roots. Again, this allows us to compare membership trends in the more “Western” organizations against the more “indigenous” associations. The Tomo no Kai [Friends], founded in 1930 by Motoko Hani, presents a prime example of the more domestically rooted women’s organizations. Membership figures for the Tomo no Kai are presented in Figure 2.7. As with judo membership, membership in the Tomo no Kai appears to have continued to grow during the early years of World War II, at least between 1940 and 1942. Again, this growth may owe at least in part to the fact that the indigenous Tomo no Kai did not face stringent repression during wartime. But even so, as with judo, membership in the Tomo no Kai also grew rapidly in the immediate postwar period, quickly recovering and then surpassing earlier peak levels. Whereas the wartime peak membership was just 9,278 in 1942, that level was matched by 1951, and by 1955 it had soared to just over 11,000.

Civic Engagement in Postwar Japan

32

Social Service Organizations This section presents membership figures for four different social service organizations: the Rotary Club of Japan, the Japan Consumer Cooperatives’ Union (Nihon Seikatsu Kyodo Kumiai Rengokai), Kobe Consumer Cooperatives’ Union (Nada/Kobe Seikyo), and the Japan Seafarers’ Relief Association (Nihon Kaiin Ekisaikai). These groups provide useful variation not only for the types of activities in which they engage, but also for their different origins. The Rotary Club has largely Western roots; as for consumer cooperatives, the idea of a consumer cooperative was originally conceived in the West, but the organization itself was founded and evolved quite indigenously (Seikatsu Kyodo Kumai Coop Kobe 1991). The Seafarers’ Relief Association is also relatively indigenous in origin and indeed was one of the first voluntary organizations recognized by the state after the Meiji Restoration (Nihon Kaiin Ekisaikai 1981). As seen in the previous sections, the general trend with social service organizations also seems to be that the Western and the more indigenous groups do not in fact differ markedly from each other in postwar membership growth patterns. Membership in the Rotary Club grew slowly during the prewar period and reached a little more than 2,000 in 1940, the last year for which figures are available for the prewar period (Rotary Club of Japan n.d.). During the postwar period, membership quickly surpassed the prewar peak in 1950 and continued to grow rapidly thereafter, reaching more than 4,000 just two years later, in 1952, and surpassing 6,000 in another two years (Rotary Club of Japan n.d.). This is shown in Figure 2.8. Membership figures for the Japan Consumer Cooperatives’ Union are shown in Figure 2.9. Figures for the Japanese Consumer Cooperatives’ Union from the prewar period are available only for sporadic years, during which

Number of Members

20000

15000

10000

5000

0 1925

1935

1945

1955

Year

Figure 2.8. Membership in the Rotary Club Japan, 1920–1960. Source: Rotary Japan (n.d.).

Number of Members (in Thousands)

Civic Engagement: The Dependent Variable

33

4000

3000

2000

1000

0 1925

1935

1945

1955

Year

Figure 2.9. Membership in the Japan Consumer Cooperatives’ Union, 1925–1960. Source: Nisseikyo Soritsu 50-Shunen Kinen Rekishi Hensan Iinkai (2001).

Number of Members

40000 30000 20000 10000 0 1925

1935

1945

1955

Year

Figure 2.10. Membership in the Nada Kobe Consumer Cooperatives’ Union, 1921–1956. Source: Nada Kobe Seikatsu Kyodo Kumiai (1971).

membership crept up to around 400,000 by 1940. During World War II, the consumer cooperatives were repressed for their allegedly communistic leanings, which brought their activities to a halt in many places. However, the cooperatives bounced back quickly after the war. Figure 2.9 shows a dramatic rise in membership in the postwar period; in 1949, the first year for which postwar data are available, the membership figure, at 670,000, already vastly surpassed the prior membership peak. By 1955, total membership had reached 2.77 million. A glimpse of the local-level changes in membership can be seen in the figures that are available from Kobe, one of the largest local cooperative societies in Japan. Membership figures for the Nada Kobe Consumer Cooperatives are reported in Figure 2.10.

Civic Engagement in Postwar Japan

Number of Members (in 1,000)

34 300 250 200 150 100 50 0 1925

1935

1945

1955

Year

Figure 2.11. Membership in the Japan Seafarer’s Relief Association, 1925–1957. Source: Nihon Kaiin Ekisaikai (1981).

As with the national-level figures for the consumer cooperatives, the Nada Kobe Consumer Cooperatives also show a rapid growth in memberships in the immediate postwar period. Unlike membership in the national consumer cooperatives’ union, which grew quite steadily over the course of the 1950s, membership in the Nada Kobe Consumer Cooperatives at first grew sharply between 1946 and 1948 to almost 45,000 members but then leveled off at around 33,000. Despite the stasis of the 1950s, membership levels remained at much higher levels compared to the prewar period. Membership in the Japan Seafarers’ Relief Association (JSRA) is shown in Figure 2.11. As with the Tomo no Kai and other more indigenously rooted associations that have been presented thus far in this chapter, the growth in the membership in the JSRA began before, not after, the end of war. A possible explanation for this may be that the JSRA, as one of the oldest voluntary organizations given non-profit status by the Meiji government in the late nineteenth century, had a much closer relationship with the wartime regime compared to its more “Western” counterparts. For this reason, it may have faced fewer constraints on its activities during the wartime period.6 But even having grown during the war, the JSRA continued to grow after the war, albeit at a slower rate after 1948. By 1955, the membership of 319,360 was 63 percent higher than the 1940 figure of 195,607. Membership in Christian Churches Membership in Christian churches also grew following World War II. As noted earlier in the chapter, this section focuses primarily on membership in Christian 6

A similar pattern is exhibited by many British groups, which begin to show a rise in memberships during, rather than after, World War II. See Kage (2006).

Civic Engagement: The Dependent Variable

35

Number of Members (in 1,000)

600 500 400 300 200 100 0 1925

1935

1945

1955

Year

Figure 2.12. Membership in Christian Churches, 1925–1960. Source: Data for 1925, 1930, and 1935 are from National Diet Library (1955); data for 1934 and 1939 are from Murata (1939); data for 1944–1947 are from “Statistics on Religions in Japan (as of December 31, 1947)” (n.d.); data for 1948–49 are from “Statistics on Religions in Japan (as of December 31, 1949)” (n.d.); data for 1952 are from Monbusho (1953); data for 1953–1960 are from Monbusho, Shukyo Nenkan (Annual Report on Religions in Japan), various years.

churches, since memberships in Buddhism and Shintoism are generally considered to be less “voluntary” in Japan than membership in Christianity. Once pieced together, the data reveal that Christian memberships grew rapidly in the immediate postwar period. The prewar peak figure for Christian membership was 315,517 in 1939, but the first available postwar figures, from 1946, at 332,907, already surpassed this prewar peak, and memberships continued to grow thereafter. Between 1946 and 1955, memberships grew at a faster rate than any time in the prewar period, roughly doubling from a little more than 300,000 members to almost 600,000. Christianity was strongly favored by Douglas MacArthur, so this may be one reason for its growth during the occupation. But this was not the entire story. The fact that Christian membership continued to rise even after MacArthur left also suggests that the postwar rise in Christianity cannot be attributed simply to American power. Figure 2.12 presents the available figures. Labor Unions As noted earlier, membership in labor unions was not necessarily voluntary in the postwar period. Nevertheless, since the surge in the labor movement in the immediate postwar period is well noted, the present study presents data on union membership for purposes of comparison with other types of associations. The figures are shown in Figure 2.13.

Civic Engagement in Postwar Japan

Number of Members (in 100,000)

36 70 60 50 40 30 20 10 0 1925

1935

1945

1955

Year

Figure 2.13. Membership in Labor Unions, 1925–1960. Source: Prewar data are from Ohara Shakai Mondai Kenkyusho (various years); postwar data are from Kosei Rodo Sho (n.d.). Note that there is a slight change in the counting method between 1952 and 1953. The percentage of union members denotes the proportion of unionized workers among the overall work force.

Figure 2.13 shows that, as with other organizations, the rise in labor union membership is also impressive. According to surviving data, roughly 7 percent of all industrial workers had been unionized in 1936, compared to 45.3 percent in 1947 (Ohara Shakai Mondai Kenkyusho 1937). The percentage of unionized workers fell to 35.6 percent in 1955, but this still dwarfed the prewar levels of membership. Note that measurement methods are different between 1947 and 1952 and after 1953. While the change in measurement methods makes it difficult to draw simple conclusions from the data, membership in labor unions exhibits a consistent increase for much of the early postwar period. This rise in union membership is particularly revealing because it shows that the postwar surge in civic engagement transcended class lines. Unusually for the data on other groups, note that union membership did fall significantly between 1949 and 1951. This was likely due to a change in U.S. occupation policy that increasingly soured toward left-wing movements during this period (Otake 1996, ch.1; Kume 1998) rather than a lack of enthusiasm on the part of workers toward participation itself. Once the occupation ended in 1952, membership figures began to recover again. National-Level Trends in Memberships: Summing Up The foregoing presented just a series of glimpses of Japanese civic engagement over the mid-twentieth century from the prewar to postwar periods. The lack of systematic national-level data for the prewar and wartime periods is disappointing. Nevertheless, we must do with what we have, and the research performed here draws on the alternative methodology devised by the Skocpol group.

Civic Engagement: The Dependent Variable

37

The data that are available seem to indicate the following: a) Overall, levels of civic engagement were generally much higher in postwar Japan compared to prewar Japan; b) postwar growth rates in civic engagement do not merely represent a return to prewar levels but appear to far outpace them; and c) this growth occurred both in the more “Western” as well as more “indigenous” groups, despite the repression of the former during the war and some of the latter during the occupation. The only caveat to this last point is that some groups that grew strongly during the war did not grow as strongly after the war, although all certainly did continue to grow in absolute terms. The surge in postwar participation should not be seen as a simple consequence of “democratization,” which is too blunt a concept for the shifting regulatory environment faced by the groups between the prewar and postwar periods. Some organizations, such as the Rotary Club, were repressed before the war but not after it. Others, like judo, were encouraged before and during the war but repressed after it. Some, such as labor unions, saw stark changes in policy during the course of the occupation; SCAP initially encouraged membership in unions but then abruptly reversed course. These policies clearly affected membership levels, and the timing of membership growth, but they could not explain the basic phenomenon of postwar growth across different types of groups. After all, people may not organize even if they are allowed to do so. Indeed, the fact that membership grew not only among elite organizations such as the Rotary Club or the Tomo no Kai but also among grassroots groups such as youth groups, consumer cooperatives, judo, and labor unions also suggests that this surge in participation was not confined to specific segments of society that had previously been repressed, but was strong regardless of the evolution of the relationship of different social segments to the state.

Cross-Prefectural Variation in Civic Engagement, 1945–1955 This section breaks down the national-level membership figures into the prefectural levels. Available sources suggest that although overall levels of participation may have grown rapidly in the immediate postwar period, the variation in the rates of growth across the forty-six Japanese prefectures, as well as in the resulting differences in prefectural levels of participation, was considerable. In the paragraphs that follow, I simply report the prefectures with the highest and lowest rates of participation to provide a flavor for the range of the variation. These prefectural-level data will also be used in the quantitative analysis presented in Chapter 4. Cross-Prefectural Variation in Membership Levels circa 1955 This section first presents cross-sectional data on the variations in membership levels across different prefectures during the initial decade after the war.

38

Table 2.2. Top Five and Bottom Five Prefectures, Membership in Youth, Women’s, and Christian Organizations Youth Groups (1958) Top Five Prefectures

Bottom Five Prefectures

Mean (46 Prefectures) Standard Error

Niigata Toyama Tokushima Aomori Iwate Kyoto Chiba Aichi Kanagawa Tokyo

Membership per 1,000 Population 37.96 37.71 37.66 37.15 35.37 15.52 14.95 14.93 8.22 5.90 25.08

Christian Organizations (1955) Tokyo Hokkaido Osaka Fukuoka Aichi Shiga Yamanashi Nara Fukui Tottori

7.75

Sources: Monbusho Shakai Kyoiku-kyoku (1959a, 1959b); Monbusho (1956).

Membership per 1,000 Population 80.37 47.73 46.18 38.60 37.69 8.54 8.07 7.77 7.54 6.14 19.41 13.69

Women’s Organizations (1955) Nagano Aomori Shiga Akita Shizuoka Kanagawa Osaka Saitama Tochigi Ibaraki

Membership per 1,000 Population 37.95 31.41 29.32 28.30 27.08 3.47 3.03 2.65 1.03 .81 15.84 8.66

Civic Engagement: The Dependent Variable

39

Since prefectural-level data are unfortunately missing for social service and recreational organizations, the section focuses on the remaining three types of organizations: youth, religious, and women’s. Table 2.2 shows substantial cross-prefectural variation in the levels of memberships across different types of organizations. For youth groups, for instance, Niigata Prefecture, the prefecture with the highest membership per 1,000 people, had more than six times as large a membership as in Tokyo, which had the lowest membership per 1,000 people. Across the forty-six prefectures, the mean membership was 25.08 per 1,000 people (standard deviation: 7.75). For youth groups, membership levels in the more rural prefectures (such as Niigata and Toyama) generally appear to have been higher than in the more urban prefectures (such as Tokyo and Kanagawa). For women’s groups, the variation in membership levels across prefectures is even starker. Levels of memberships in women’s groups in Nagano Prefecture, which had the highest membership per 1,000 people, were more than forty times as large as in Ibaraki Prefecture, which had the smallest membership per 1,000 people. The mean membership level per population across the fortysix prefectures was 15.84 per 1,000 people (standard deviation: 8.66). As with youth groups, the more rural prefectures (such as Aomori and Akita) generally appear to exhibit higher levels of membership than the more urbanized prefectures (such as Kanagawa and Osaka), although the pattern is not entirely clear-cut; Shizuoka Prefecture, which had one of the highest levels of memberships, is relatively urbanized, while Tochigi and Ibaraki, which report some of the lowest number of members per population, are relatively rural prefectures. Cross-prefectural variation in the levels of Christian membership was also substantial. Tokyo is the clear outlier here, with a membership per population that was almost twice as large as Hokkaido Prefecture, which had the second highest membership per population. Christian membership per population in Tokyo Prefecture was also more than thirteen times as large as in Tottori Prefecture, which had the smallest membership per population. Generally, the urban prefectures such as Tokyo and Osaka tend to exhibit higher levels of members compared to their more rural counterparts. The mean membership level per 1,000 population was 19.41 (standard deviation: 13.69). It should be noted here that not only was there considerable cross-prefectural variation in levels of civic engagement but also in the types of groups that individuals in different prefectures participate in. Thus, for instance, Tokyo Prefecture had large memberships in Christian groups but had the smallest perpopulation membership in youth groups. Similarly, Shiga Prefecture had one of the largest memberships in women’s groups but one of the smallest memberships per population in Christian organizations. Japan does not present a simple geographical pattern of “high” versus “low” civic engagement as Putnam (1993) found between Northern and Southern Italy.

40

Civic Engagement in Postwar Japan

Cross-Prefectural Variation in Civic Engagement: Growth Rates Compared with the Prewar Period A useful way to measure the prefectural-level growth of civic engagement over time is to compare levels of participation between the prewar and postwar periods. This raises the question of just when the war began. For Japan, World War II already broke out in 1931 in China, although on a limited scale. By 1936, the war had already been in progress for five years. However, since, by most accounts, full-scale mobilization in Japan did not begin until 1940 (Akimoto 1974; Noguchi 1995), comparisons with data up to the end of the 1930s for measuring prewar levels of associational activity are still justifiable. While prefectural-level data are not sufficient for women’s groups, prewar membership figures at the prefectural level are available from 1922 and 1931 for youth groups and from 1936 for Christian groups. The changes in levels of memberships at the prefectural level between the prewar and postwar periods are shown in Table 2.3. The table reveals two trends. First, at the aggregate level, participation in youth and Christian groups grew dramatically between the prewar and postwar periods. This confirms, at the aggregate level, the findings from the individual group-level data shown earlier in this chapter. On average, memberships in youth groups grew 1,407 percent between 1922 and 1958 and 269 percent between 1931 and 1958, while membership in Christian groups grew 149 percent between 1936 and 1955. Indeed, all but six prefectures saw a growth in Christian membership during this period. Second, the figures shown in Table 2.3 also reveal substantial variation in the extent of this growth across different Japanese prefectures. For instance, while youth group memberships in Mie Prefecture grew more than ninetyfold between 1922 and 1958, levels of membership barely changed in Yamaguchi and Wakayama Prefectures. The regional pattern here is not immediately obvious; among the prefectures in which youth membership grew at a rapid rate, Mie, Tottori, and Akita Prefectures are rural, while Osaka Prefecture, as well as parts of Miyagi Prefecture, are urbanized. Among the prefectures in which the growth in membership was much slower, large parts of Hiroshima and Hyogo Prefectures are urbanized, while Yamaguchi and Wakayama are less so. Hiroshima and, to a lesser extent, Hyogo Prefectures were severely damaged in the war, but Nagasaki, another prefecture that experienced heavy destruction, does not appear among the bottom five. Table 2.3 also shows that rates of membership growth in Christianity also varied considerably by prefecture, although to a lesser extent than for youth groups. Christian membership almost tripled in Kagoshima and Mie Prefectures, while it actually decreased in other prefectures such as Kochi, Nagano, and Toyama. As with youth groups, both the prefectures in which Christian memberships grew rapidly and those in which they grew more slowly appear to be a mix of urban and rural prefectures.

41

Table 2.3. Pre- to Postwar Growth in Memberships in Youth and Christian Organizations Youth Groups (1922–1958) Top Five Prefectures

Bottom Five Prefectures

Mean (46 Prefectures) Standard Error

％

Youth Groups (1931–1958)

％

Christian Organizations (1936–1955)

％

Mie

9260.60

Yamagata

1666.95

Kagoshima

300.00

Tottori Akita Miyagi Osaka Hiroshima

5542.60 5311.77 3521.98 2767.99 320.46

Gunma Akita Kochi Miyazaki Kagoshima

1337.23 1232.33 761.58 592.54 51.89

Mie Saitama Akita Gifu Hokkaido

267.35 248.99 224.56 218.07 81.65

Hyogo Yamanashi Wakayama Yamaguchi

229.52 178.66 135.17 101.39 1407.33

Aichi Saga Yamaguchi Kyoto

46.35 46.07 37.80 25.80 269.38

Yamanashi Toyama Nagano Kochi

79.95 71.00 70.41 68.76 148.93

1711.04

343.13

Source: See Table 2.2; also Monbusho Futsu Gakumu-kyoku (1924), Monbusho Shakai Kyoiku-kyoku (1937a), Shukyo Nenkan (1939).

51.85

42

Civic Engagement in Postwar Japan

Conclusion This chapter does not pretend to offer the last word on the broader trends in Japan’s civic engagement between the prewar and the immediate postwar periods. Much information is still lacking, and much more research still needs to be conducted. Nevertheless, the chapter has presented an initial attempt at illuminating the patterns in the way that people participated in the immediate aftermath of war. The chapter has made five main claims. First, it showed that overall membership in Japan’s voluntary associations, whether in youth, women’s, social service, or religious groups, consistently increased during the first postwar decade. Contrary to the expectations of existing hypotheses, there seems to be no systematic evidence that defeat in war led to a decline in civic engagement in Japan. Rather, the case of Japan seems to show that defeat is indeed compatible with a considerable rise in levels of civic engagement. Levels of postwar participation not only grew rapidly, but membership levels far surpassed those of the prewar period. And not only were absolute levels of membership much higher in the postwar than in the prewar period; the rate of membership growth often also accelerated markedly in the postwar compared to the prewar period as well. Across the board, the growth in memberships started well before Japan’s economic “takeoff” began, around the early to mid-1950s. Second, the chapter revealed that levels of postwar participation grew just as rapidly for groups that had operated under relatively liberal policies during the prewar and wartime periods as those that had been repressed at that time. Third, the chapter also showed that membership also grew in groups that the U.S. occupation repressed. There is some evidence, however, that associations that were more favored by the occupation saw even more rapid increases in memberships than those that were less favored. The significance of this finding will be addressed further in Chapter 3. Fourth, in a related point, the data indicate little systematic difference in the growth rates in memberships between groups with “Western” origins, such as the YWCA and the Girl Scouts, and those with more indigenous origins, such as the Japan Alpine Club and Tomo no Kai. Fifth, while the basic national story is one of overall growth, the data presented in this chapter also revealed substantial cross-regional variation in the growth of civic engagement across Japanese prefectures. The variations were sometimes dramatic; for instance, in the case of youth groups, membership grew ninetyfold in Mie Prefecture between 1922 and 1958 while it barely changed in Kochi Prefecture. An explanation of trends in civic engagement must be able to account for these regional variations as well as the overall national picture. In sum, rather than destroying the social fabric, wartime destruction was quickly followed by a rapid rise in civic engagement. In the face of devastating defeat, citizens mustered the energy to reach out to each other with

Civic Engagement: The Dependent Variable

43

renewed vigor. Civic engagement rose sharply in Japan in the wake of World War II despite the confluence of economic stagnation, lost educational opportunities, and psychological damage, all of which are typically considered to adversely affect participation. The victory/defeat hypothesis thus requires reconsideration.

3 War and Civic Engagement A Theoretical Framework

Introduction Chapter 1 pointed out that most of the existing theories of participation would expect a country’s victory or defeat to crucially determine its trajectory of civic engagement in the wake of wars. Chapter 2 presented new data to show that, contrary to these expectations, civic engagement rose at an impressive rate in defeated Japan in the wake of World War II. It also demonstrated the substantial variation in the extent to which civic engagement grew in the wake of war across different Japanese prefectures. Given the confluence of stagnant economic growth, loss of educational opportunities, and psychological devastation that Japanese citizens faced in the immediate postwar period, conditions in Japan should have been particularly adverse to any rise in civic engagement. If the victory/defeat hypothesis cannot explain a case of defeat as complete as Japan’s, this leads us to question the applicability of the hypothesis in any case. Why might levels of civic engagement rise in the wake of war, even in the face of devastating defeat? And what accounts for the variations in the extent to which voluntary participation rises in the wake of war? The present chapter takes a step back from the Japanese case and develops a new theoretical framework to explain the more general relationship between war, on the one hand, and the postwar trajectory of civic engagement, on the other. Specifically, it points to the combination of wartime mobilization and preexisting legacies of prewar civic activities as the key factors that shape the growth of civic engagement in the postwar period. It argues that wartime mobilization sets the stage for an overall rise in civic engagement in the wake of wars, while preexisting legacies from the prewar period account for the actual variations in this rise across different subnational regions. Table 3.1 summarizes the theory’s predicted postwar patterns in civic engagement. The rest of this chapter delineates the precise mechanisms that produce these predictions. 44

War and Civic Engagement: A Theoretical Framework

45

Table 3.1. Predicting Changes in Civic Engagement between Prewar and Postwar Periods Prewar Participation

Low High

Wartime Mobilization Limited

Extensive

Little or no growth in participation Little or no growth in participation

Some growth in participation Highest growth in participation

The chapter is structured as follows. The next section considers three possible explanations for the rise in voluntary civic engagement in postwar Japan: the birth of a democratic regime, specific policy reforms introduced by the occupation forces (Supreme Commander for Allied Powers, hereafter SCAP), and the experience of war disasters. While these indeed offer useful insights for explaining the postwar surge in civic engagement, ultimately they are not fully satisfying. Thus, we turn to theory-building. The third section explains why wartime mobilization and recruitment into war-related tasks should be expected to prepare the groundwork for postwar civic engagement. The fourth section discusses how preexisting structures of participation should be expected to mediate the effects of mobilization on the ultimate behavioral outcomes. The fifth and final section summarizes the argument presented in this chapter.

Possible Explanations: Democratization and Disasters This section explores the plausibility of three possible explanations for the trends in civic engagement in postwar Japan: democratization, occupation reforms, and wartime destruction. Possible Explanation #1: Democratization As noted in Chapter 1, this study is primarily concerned with the trajectory of postwar civic engagement in democracies, whether they had been democratic or authoritarian prior to the war. For countries that had not been democratic prior to war, regime change, that is, democratization itself, may provide an important explanation for the rise in civic engagement after war. Indeed, studies often find a rise in civic engagement when countries democratize in the absence of war. Torcal and Montero, for instance, report a growth in associational activities in post–Franco-era Spain (Torcal and Montero 1999: 169); a recent study by Tsujinaka and Yeom (2002) also finds a surge in levels of voluntary civic activities in the wake of democratization in South Korea in the 1980s.

46

Civic Engagement in Postwar Japan

This study does not deny that democratization often leads to higher levels of civic engagement. As noted in Chapter 1, however, civic engagement in the United States jumped up after World War II, even though the United States was a democracy before the war as well. And as Chapter 5 will show, this also occurred in other cases, such as the UK and Finland. So there would appear to be something about the effects of war, irrespective of regime type, that produce a surge in civic engagement. If postwar civic engagement can rise even in the absence of democratization, then the postwar rise in civic engagement cannot be attributed solely to democratization. Possible Explanation #2: U.S. Occupation Policies A more precise variant of the argument that democratization may have played a key role in affecting the rise of civic engagement in Japan in the wake of World War II suggests that U.S. occupation policies crucially changed Japanese society. As is well known, the United States after World War II sought to remake West Germany and Japan from the ground up, and it devoted vast amounts of resources, both financial and human, to that end. As in the case of the more general democratization variable, this study does not deny that U.S. occupation policies played an important role in shaping the evolution of civic engagement in postwar Japan. But it must be noted that the occupation regime did not consistently promote increased civic engagement. Rather, it encouraged the growth of some associations while severely repressing others (Kage 2003a). Christian groups and Western-based organizations and, for a brief period, labor unions were encouraged, but other organizations that were deemed to espouse and/or promote militarism or nationalistic values, like judo, were discouraged and sometimes even vigorously broken up. As noted in Chapter 2, such policies are likely to have had a significant effect on the types of groups that enjoyed the most membership growth.1 Nevertheless, such policies do not provide the full, or even the most important, explanation for the puzzle at hand, which is an overall surge in civic engagement, and not merely a surge in particular, favored types of groups. Furthermore, a careful examination of the history of the period suggests that while some of SCAP’s policies certainly encouraged the general rise in voluntary activities in Japan, their effects also should not be overestimated. The main pillars of SCAP’s policies toward voluntary associations in Japan were threefold: a) the breakup of groups that were viewed to promote militarist and/or ultranationalist values, b) the dissociation of the Japanese state from private associational activities, especially through the new Social Education Law, and c) the (re-)training of associational leaders in democratic 1

This echoes Robert Pekkanen and Susan Pharr’s argument that state policies have a crucial impact on the types of associations that flourish under a particular political system (Pekkanen 2006; Pharr 2003).

War and Civic Engagement: A Theoretical Framework

47

associational procedures, particularly in the methods of democratic deliberation (Kage 2003a). Clearly, breaking up extremist groups should not encourage participation but would rather dampen it. As for the Social Education Law, which came into effect in June 1949, it stipulated what is known as the “no control–no support” policy, which prohibited national and local governments from channeling funds into voluntary associations except to provide information or to assist in the holding of public meetings (Kage 2003a). In short, the occupation greatly reduced the degree to which associations could count on government support. It is hard to see how such a reduction in government support could cause their memberships to soar thereafter. If anything, such a policy should have further constrained the resources of organizations that were already strapped for cash during the immediate postwar period. True, some associations with Western connections, such as the Girl Scouts, received financial assistance from their sister associations in the United States, but such groups merely comprised the lucky few. Finding this policy restrictive, the Japanese Ministry of Education eventually devised legal measures that would allow for the public funding of private activities, thus reestablishing a semblance of the old system (Tanaka 1988: ch. 4). Finally, it is also difficult to see that SCAP’s training programs for associational leaders had much impact on overall membership figures. The training programs were by no means meant to serve as nonprofit management programs. Little if any emphasis was placed on the more practical dimensions of operating a voluntary association, such as recruitment and financial accounting. Rather, the programs were meant to instill democratic values and methods into the associations. As such, the programs focused heavily on the teaching of democratic deliberation, the place of voluntary associations within a democratic society, and related abstract principles (Trainor 1983: 259; Tanaka 1988). Again, there can be no doubt that the U.S. occupation conferred special social legitimacy on the groups that it promoted, thereby encouraging participation in them. Groups with Anglo-American roots, such as the Girl Scouts and the Rotary Club, gained in social status with the arrival of the occupation forces, and this was helpful for attracting members. But this is a point about the distribution of civic engagement, not its extent. Moreover, there is little evidence that Japanese citizens only participated in groups that SCAP viewed favorably. As shown in Chapter 2, SCAP initially outlawed judo at the school level but membership grew steadily. Another case in point is the neighborhood associations. Although SCAP outlawed the associations in 1947, citizens of all ages continued to participate in them as a matter of course and have, to a large extent, done so to this day.2 This indicates that the Japanese were not simply 2

The breakup of ultranationalist organizations was more successful, as the occupation purged many of the leaders of those associations. For neighborhood associations, however, it was not

48

Civic Engagement in Postwar Japan

accepting American policies, but that they had their own reasons for their new habit of civic engagement. Finally, we must not forget the basic point that liberal policies toward voluntary associations certainly provide the conditions, or the space, for civic activities to flourish, but this space must be filled by the citizens themselves. Even if encouraging policies are put in place by the occupying powers, levels of civic engagement may increase, decrease, or remain the same. SCAP neither forced individuals to join associations nor brought the organizations into being by fiat. Without denying the role of occupation policies, a complete explanation for the actual extent to which participation rises in different countries or regions must examine the underlying social conditions in which those policies are implemented. Possible Explanation #3: Disaster-Driven Civic Engagement A third possible explanation for the postwar rise in civic engagement sees the amount of physical destruction as the chief causal variable. Sociological and psychological studies of disasters have found that voluntary participation often rises in the wake of natural as well as man-made disasters, such as major floods, earthquakes, and terrorist attacks (e.g., Fritz 1961; Hobfoll 1989; Kaniasty and Norris 1995; St. John and Fuchs 2002). From this perspective, the more intense the disaster, the greater the response or, in other words, the larger the rise in civic engagement in the wake of that disaster. The rise in participation in the wake of major disasters is intuitive. The rushing of volunteers to assist in rescue and relief efforts has by now become a familiar scene. More than one million units of blood were donated in the first weeks of the September 11, 2001 terrorist attacks (American Red Cross 2002: 10–11). More than 6 million volunteers are said to have volunteered in the relief efforts following the Kobe earthquake of January 1995 (Hyogo-ken 2002). Contributions may also be financial; U.S. relief organizations had raised $342 million just in the first three weeks after the tsunami disaster in Southeast Asia in December 2004 (Strom 2005). Indeed, one New York Times article described the charities as being “overwhelmed” by the inflow of disaster donations (Strom 2005). Perhaps, then, something similar happened in Japan in the wake of World War II. Scholars of civil war have also recently drawn on what has been termed “post-traumatic growth theory” to argue that witnessing more violence leads to heightened levels of participation in the wake of civil wars (Bellows and Miguel 2009; Blattman 2009). Perhaps, then, witnessing more violence also causes an analogous rise in civic engagement in the wake of international wars as well. In a sense, this argument turns the victory/defeat hypothesis on its head,

feasible practically to purge all leaders. Since, in most associations, households simply rotated taking leadership positions, most Japanese citizens would have had to be “purged.”

War and Civic Engagement: A Theoretical Framework

49

since it stresses the empowering, rather than enervating, effect of traumatic personal experiences. Disasters per se, however, do not provide a satisfying explanation for the rise in levels of civic engagement between the prewar and postwar periods, especially in cases of defeat. This is because disaster-driven participation is generally short-lived. A glimpse at some of the data on levels of civic engagement in the wake of disasters is indicative. The terrorist attacks on September 11 created a temporary surge in voluntarism, but the evidence indicates that this did not last for more than a few months (Putnam and Sander 2002). Similarly, figures compiled by Hyogo Prefecture reveal that although more than 20,000 volunteers participated in disaster relief during the first few days after the Kobe earthquake, in two months the number of volunteers had fallen to 5,000, and, in three months, less than 1,000 (Hanshin-Awaji Daishinsai Hisaichi no Shimin wo Oensuru Shimin no Kai 1996). Once the acute need for participation dissipates – or rather is believed to have dissipated – participation generally falls to pre-disaster levels.3 Disasters per se, then, do not produce a lasting rise in civic engagement. They may at best serve as a catalyst for short-lived surges in participation, but they do not provide a satisfactory account for the long-term rise in civic engagement that transpired in the wake of World War II, as documented in the previous chapter.

The Process of War: Wartime Mobilization War and Society The remainder of this chapter advances an alternative framework for understanding the relationship between war and civic engagement. Focusing on the processes, rather than the outcomes, of war, the study argues that the experiences of war, especially wartime mobilization, may often offer opportunities for social learning in a manner and to an extent that is rare during times of peace, thereby transforming both citizens’ preferences and costs of participation. Ceteris paribus, more extensive mobilization should lead to a general rise in levels of civic engagement in the wake of wars. At the same time, the study argues that the variations in the postwar rise in civic engagement within countries may be explained by path-dependency effects from the prewar period, and in particular by the legacies of past levels of participation. Ceteris paribus, the more extensive the prewar civic engagement, the greater the effect of wartime mobilization in boosting levels of civic engagement in the wake of war.

3

For a contrary view that disasters may elevate the civic-mindedness of citizens over the longer term, see Tatsuki and Hayashi (2000).

Civic Engagement in Postwar Japan

50

The Processes of War Political scientists and sociologists have long recognized the effects of war on state-building. Indeed, the state-building literature pioneered the idea that the domestic consequences of war, even when it ends in defeat, are not always regressive. As Charles Tilly (1975: 42) famously noted, “War made the state and state made war.” Scholars have subsequently agreed: Wars generally strengthen state capacities (Titmuss 1955; Levi 1988; Skocpol 1992; Rasler and Thompson 1985 Ertman 1997).4 The main reason for this is that “People will accept, in a period of crisis, tax levels and methods of raising revenue that in quieter times they would have thought intolerable …” (Peacock and Wiseman 1961: 27).5 As Tilly writes, “The general rule… has been for some contraction in governmental scale to occur after a war – but never a return to the prewar scale. Preparation for war has been the great state-building activity” (Tilly 1975: 74). It has often been assumed that the growth of the state, particularly in war, might come at the expense of civil society. As war intensifies, however, the growth of new wartime demands, both military and civilian, means that the state often cannot provide all of the needed services on its own. Societal actors must also be mobilized to provide the resources that are necessary for the war effort. As Skocpol, Muson, Karch, and Camp argue, “By definition, modern wars create a demand for joint action by state authorities and members of society” (Skocpol et al. 2002: 140; italics added). States may be able to extract at higher levels during war compared to times of peace, but they remain dependent on society to a substantial degree. Thus, the state must mobilize society; and, this study argues, the effects of that wartime societal mobilization may persist over the long run. The Merriam-Webster Dictionary defines mobilization as “The act of assembling and making ready for war duty.” Mobilization may take place at one or more of three levels: the war front, the workplace, and the neighborhood. ●

4

5

Mobilization into the armed forces has generally involved the recruitment or conscription of young males, and less frequently females, usually between the ages of around eighteen up to around forty (Flynn 1998). As wars intensify, recruitment typically expands to broader age groups, both older and younger. It may also extend to marginalized groups within society such as ethnic minorities (Krebs 2006). Moreover, other scholars have argued that the lack of wars led to a persistence of weak state capacities in places like sub-Saharan Africa (see Herbst 2000). But this point is not universally accepted; Centeno (1997) argues that if states start out weak, wars are likely to weaken them further. In recent years, some scholars have questioned the generalizability of the claim that wars promote state-building as a European anomaly. See, for instance, Centeno (1997).

War and Civic Engagement: A Theoretical Framework ●

●

51

The war economy, of course, represents the most obvious channel through which the state may mobilize societal actors on the home front; as military demand grows, production shifts to steel, aircraft, and other heavy production as government demand expands. As men are mobilized into the war front, in the twentieth-century, women, as well as other socially marginal groups, were increasingly mobilized into the war production system (Braybon and Summerfield 1987). Perhaps most consequential for civic engagement over the long term is the mobilization at the neighborhood or community level. The next paragraphs consider this point in some detail.

In what Skocpol et al. (2002) call “patriotic partnerships,” voluntary associations have played a particularly important role in providing wartime services during the modern era. The role of voluntary associations such as the International Red Cross during wars is well known (Hutchinson 1996). Groups such as the YMCA also provided a wide range of services to American soldiers serving in the war front during World Wars I and II, from health services and mail delivery to recreation. Oxfam, one of the most prominent disaster relief organizations today, was founded in the United Kingdom in 1942 to provide relief in war-torn Europe (Iriye 2002: 52). In Japan as well, the YMCA also provided similar services for Japanese soldiers during the Russo-Japanese War, World War I, and World War II (Nihon YMCA Domei 2003). As war intensifies, mobilization into para-military tasks or the home front also increasingly becomes coerced. Especially in undemocratic contexts, modern wars have given rise to what Gregory Kazsa (1995) has called “administered mass organizations.” Historically, these associations emerged in Germany and Russia in the wake of World War I and spread to Japan soon thereafter (Kazsa 1995). The well-known Hitler Jugend, for instance, mobilized more than 7 million German youths between the ages of ten and eighteen to assist the military in a broad array of tasks, from army hospital work or serving food to troops in transit to delivering messages between the party and the military in the field (Kazsa 1995: 78, 80). Many neighborhood associations had been in existence prior to the war in Japan as well, but they were increasingly incorporated into the governing apparatus to serve as the basic unit for air raid drills, firefighting, or mutual assistance during the interwar and wartime periods (Akimoto 1974; Akagi 1977; Kasza 1995). As the war wore on, these associations also served as conduits for food rationing as well as to promote higher levels of savings, which would then be funneled to fund the war (Garon 1997; Fujihira 2003). In short, major war produces major mobilization, including at the neighborhood level for tasks that are often taken care of in peace time by civic associations. This coercive mobilization at the community level is not necessarily experienced as an imposition by the state, so that when the coercion ends, so too does the participation. Rather, this study argues that the effects of this wartime mobilization may be more lasting than is often assumed.

52

Civic Engagement in Postwar Japan

Voluntary and Involuntary Participation A long line of literature, from Alexis de Tocqueville to Robert Putnam, has claimed that voluntary associations serve as a “school for democracy” or a local-level training ground for citizens to apprentice for broader citizenship (Tocqueville 2000; Putnam 1993). By interacting with fellow citizens at the grassroots level, it is argued, individuals acquire important skills of communication, deliberation, and compromise as well as developing a broader social and political awareness. To the extent that mobilization takes place through purely voluntary organizations such as those highlighted by the Skocpol group, then, we may certainly expect to see enduring social learning through those organizations, and thus a postwar rise in civic engagement even in defeated countries. However, purely voluntary participation is a very small part of overall wartime mobilization. And the underlying assumption of much of the literature is that while voluntary participation is conducive to the learning of civic skills and ultimately to civic engagement, involuntary participation is not. But this assumption needs to be problematized. Of course, at the extreme of involuntary participation, such as slave labor camps, the existing literature clearly has a point. However, most citizens do not experience such coercive extremes. For them, the distinction between voluntary and involuntary participation may not be stark as has often been portrayed in existing studies. Rather, the difference may be a matter of degree rather than kind. If this is the case, then learning may still be possible for citizens who are placed, particularly during their formative years, into some degree of involuntary participation. Indeed, the whole premise of compulsory education is that some degree of learning is possible under compulsion (Mill 1989). The line between voluntary and involuntary participation is fuzzy in at least two ways. First, what is “voluntary” after all? We all know from personal experience that our willingness to engage in certain activities is often the product of a mix of desire and external pressure. For instance, a son may take out the trash both because his mother demanded it and because he wanted to please her. Is this coerced or voluntary activity? Or take another example: I may fasten my seatbelt during takeoff both because regulations require it and also because I recognize that doing so would make it safer for both myself and the passengers around me. In short, the voluntary/involuntary distinction in the literature ignores the vast gray area in between. These examples are also relevant to the case of civic engagement. Some members of bowling leagues are there because they love the camaraderie, but others are there because they don’t want to be accused of being spoilsports. Second, the mix of voluntarism and involuntarism may also shift over time. For instance, it is not unusual for individuals to sour on activities that they had initially begun voluntarily but find themselves trapped by their prior commitment to serve. Just as often, they may surprise themselves by coming to enjoy

War and Civic Engagement: A Theoretical Framework

53

and learn from activities that they initially embarked on involuntarily. These occur for at least one of two reasons. First, individuals may lack full information on the activity at the time that they embark on the activity. Second, the conditions that had existed at the time that they had embarked on the activity, which had initially shaped their willingness or unwillingness to participate, may also subsequently change. War and Social Learning War, then, leads the state to mobilize citizens both at the workplace and in the community. In most cases, this mobilization is a mixture of voluntary and involuntary. But whatever this mix, through the process of mobilization, citizens come into contact with individuals, officials, and groups that they would not have encountered in a peacetime setting and interact with them to achieve various task objectives. In so doing, they may acquire important communication skills, political and social awareness, organizational savvy, the art of deliberation, and so on – in short, what scholars have termed “civic skills” (Verba et al. 1995). According to Verba et al. (1995: 304), civic skills are the “communications and organizational abilities that allow citizens to use time and money effectively in political life.” They involve the “ability to speak and write, the knowledge of how to cope in an organizational setting” (Verba et al. 1995: 305). The accumulation of practical knowledge, then, is the key to the acquisition of civic skills. As Verba et al. (1995: 305) argue, individuals are more likely to participate if they can participate well. That is, individuals who are capable of operating in an organizational context in an effective manner are much more likely than otherwise to participate because, on the one hand, the acquisition of civic skills is likely to vastly reduce citizens’ perceived costs of participation. At the same time, the fact that individuals can participate effectively should produce a feeling of subjective self-efficacy, which, in turn, is likely to also fundamentally alter citizens’ preferences toward participation. Whether in the workplace, voluntary associations, or administered mass organizations, the experience of mobilization opens up dramatic opportunities for the acquisition of practical knowledge about society, and thus civic skills. Participating in the disbursement of rationed food, for instance, brings citizens into greater contact with their neighbors, as well as with government officials. Citizens may need to negotiate various demands both amongst themselves as well as with officials, giving them a first-hand experience in civic engagement. The conscription of family members may also sensitize citizens to developments in remote lands. These skills play a crucial role in promoting the participation of citizens thereafter. While society as a whole is likely to be impacted by mobilization, the marginal members of society, or those who are generally least likely to participate in civic affairs in times of peace, such as women, youth, and other minorities, may be likely to benefit the most from these new opportunities for

54

Civic Engagement in Postwar Japan

participation (e.g., Braybon and Summerfield 1987). War involves a dramatic expansion of the public sphere. Citizens are transferred from less “strategic” industries to the more “strategic” industries; jobs and tasks that had hitherto not been considered to serve the public good take on greater public meaning; and groups and individuals who had scarcely participated in the public sphere, particularly minorities, become mobilized to enter this growing public sphere, not unusually for the first time. For such groups, the civic skills learning curve should be particularly steep. Once acquired, these skills will have lasting effects and are likely to dramatically reduce the perceived costs of as well as to enhance the preferences toward participation thereafter. This is not to say that these effects necessarily last forever. Civic skills are probably more like rowing than riding a bike. Rowing requires a certain muscle mass to be done effectively. If not practiced on a regular basis, the muscles are likely to dissipate, making it difficult to row in the first place. Moreover, effective coordination among multiple rowers also requires constant practice. Newly acquired civic skills, then, must be capitalized soon after the war is over, or they may be squandered. Empirical Illustrations The view that some limited coercion may still promote the learning of civic skills is supported by three bodies of literature that have examined the effects of involuntary participation in contemporary society: military service, community service, and jury duty. Each will be examined in detail herein. 1) Military Service. Scholars and practitioners alike have long debated the social and psychological effects of the draft. On the one hand, some argue that the draft instills teamwork, civic skills, political awareness, and the prioritizing of collective over individual goals, all of which are crucial to a healthy functioning of a democratic society. Others counter that military service leads to authoritarian attitudes, precisely the kinds of attitudes that would seem inconducive to a civic culture. While the literature is small and mostly focuses on the American case, existing studies generally agree that, contrary to the claims of some, military service does not appear to lead to authoritarian attitudes but rather reduces them (Campbell and McCormack 1957; Roghmann and Sodeur 1972). Elder and Clipp (1988) point to the strong horizontal ties, indeed lifelong comradeships, that are often forged over the course of military service. Jennings and Markus (1976: 183) find that veterans tend to be more politically aware than their civilian counterparts, typically tuning in more to the radio, television, and newspapers for news on public affairs, although the difference that they find in that study is not very large.6

6

For a contrary view that military service does not promote prosocial behavior, see Krebs (2004).

War and Civic Engagement: A Theoretical Framework

55

Perhaps for these reasons, studies have found that ex-draftees also tend to participate in politics at higher rates than non-draftees. In particular, Jennings and Markus (1976: 195–196) report that individuals with military service in Vietnam participated in community or neighborhood affairs at substantially higher rates compared to those with no experience; in fact, individuals with longer service experience in Vietnam tended to exhibit higher levels of participation.7 Based on a survey of Latino veterans, Leal (1999) argues that ceteris paribus, veterans, especially draftees, were not only more likely to vote in elections, whether presidential, congressional, or school board, compared to nonveterans, but also that they were more likely to participate in rallies, sign petitions, or to wear a button.8 Based on a survey of African Americans, Ellison (1992) finds that military service and/or combat experience are significant predictors of participation in “high-intensity” political activities, such as campaign activities and contacting officials, though not in “low-intensity” activities such as voting. While existing studies mainly examine the attitudes and behavior of veterans, the effects of mobilization may also apply to those who were mobilized into the home front as well. Mobilization requires civilians as well as soldiers to engage in collective endeavors on behalf of the war effort. Indeed, civilians who are mobilized into the home front may conceivably acquire an even wider repertoire of civic skills, since they are more likely to deal with more analogous problems in the postwar period.9 At work, they may be mobilized into warrelated factories or other tasks; in the neighborhood, they may assume responsibilities for food rationing, firefighting, organizing donation drives to raise various resources that are needed for the war effort, or all of the above. Many of the acquired skills will be directly transferable to the postwar context. 2) “Required” Community Service. A heated debate has emerged in recent years over “required” public service programs in secondary and postsecondary schools. While this trend raises many normative questions, the initial evidence seems to suggest that these programs can be beneficial for instilling enduring 7

8

9

Needless to say, veterans who had served in Vietnam for extended periods generally express a very high level of political cynicism (Jennings and Markus 1977: 137). In the United States, another pathway through which military service influences participation is through education. A large number of studies find that veterans are likely to achieve higher educational, occupational, and income attainment than non-veterans (Browning et al. 1973; Sampson and Laub 1996). All of these, in turn, are well known to also be conducive to higher rates of participation. However, these studies do not necessarily apply outside of the American context, since the availability of the G.I. Bill as well as ROTC programs means that in the United States citizens are likely to serve in the military precisely in order to attain educational and occupational ends. Homefront mobilization may be particularly important because it touches many previously socially marginalized groups. In particular, women may benefit most in terms of acquiring civic skills through wartime mobilization. War often forces women, including full-time housewives, to enter the work force in greater numbers, as well as to take on more unpaid civic chores. And as Schlozman et al. (1999) argue, women who work full-time have much higher rates of participation than those who do not.

56

Civic Engagement in Postwar Japan

patterns of civic engagement. Metz and Youniss (2003), for instance, find that “required” volunteering among high school students significantly increased the willingness of students to volunteer in the future. The school under study in their research required high school students to volunteer for only forty hours, yet this was enough to spark later volunteering behavior. Hunter and Brisbin (2000) also report similar positive effects for the case of a service learning requirement at a college. If required community service, which generally only gives brief exposure to the field, yields positive results for future volunteering, it would be reasonable to expect that extended mobilization over a period of several years, as in the case of major wars, is likely to exert even greater effects on the participatory behavior of citizens. Studies of community service requirements in schools must be interpreted with some caution since most of these studies only draw on single communities, which means that they may be vulnerable to selection effects on socioeconomic variables. Moreover, many of these studies fail to account for differences in requirements or the quality of the programs that the students participated in (Raskoff and Sundeen 1999). Nevertheless, taken together with studies on the effects of involuntary military service, the findings thus far are quite suggestive. 3) Jury Duty. Research on jury duty in the United States also suggests that this institution engenders similar effects. According to one study, almost 70 percent of both trial and non-trial jurors replied that they learned something positive or factual while serving as jurors (Diamond 1993), despite the fact that at least 40 percent of them had initially tried to get out of jury duty (Guinther 1998).10 Typically, more than 80 percent of survey respondents express a willingness to serve again on a jury in the future (Cecil et al. 1987; Cutler and Hughes 2001).11 If these studies are correct to argue that some limited compulsory mobilization may stimulate higher levels of participation, then we have a prima facie case that civic engagement may similarly be expected to rise in the wake of wartime mobilization. Again, if mobilization is completely coerced against one’s will, then individuals may not profit from the experience. However, as argued previously, wartime mobilization may not be entirely involuntary. Moreover, while citizens may resent mobilization on the part of the state, placing strains on the vertical relationship between state and society, in response to this they may forge strong horizontal bonds that may last beyond the end of the disliked activity.

10

11

The percentage of respondents who had not looked forward to serving is likely to have been even higher. This includes those who served in long trials (Cecil et al. 1987). It is possible that individuals who are more civic-minded are more likely to appear for jury duty in the first place, which may also in turn make them more willing to learn from the experience; but recent studies suggest that responders and nonresponders do not vary substantially in their attitudes toward jury service or in their demographic or socioeconomic characteristics (Boatright 1999).

War and Civic Engagement: A Theoretical Framework

57

War, Strong State, and Strong Society Not only do war-affected citizens acquire important civic skills through wartime work. Social learning via wartime mobilization may be further enhanced by the fact that as war intensifies, societal actors often gain a greater voice in public affairs. Indeed, the existing literature has sometimes exaggerated the extent to which states may become “predatory” during wars (Levi 1988). Citizens during wartime are not just the compliant and submissive subjects who rally around the flag the moment any war breaks out. On the contrary, society expects greater compensation for its sacrifices as mobilization intensifies. As British historian Arthur Marwick (1981: 9) has noted, “… in total war, when national energies must be maximized, underprivileged groups are placed in a strong bargaining position.” Social actors must be induced, cajoled, co-opted, and at times bribed into performing tasks that they would not otherwise perform during times of peace, especially as war intensifies and becomes prolonged. Citizens also see this. As citizens become aware of the value of their services, they also realize that they may extract a price for it, both during and after wars, and they have not hesitated from demanding it. Thus, as Shinju Fujihira (2003) demonstrates, contrary to conventional wisdom, states have not necessarily been able to raise taxes at will during wartime but have at times been forced to rely on monetary policy or other instruments to raise needed revenues.12 Wellknown provisions such as the G.I. Bill, as well as veterans’ pensions, indicate that states cannot unilaterally mobilize their citizens, but rather that citizens must be paid off for their troubles. Indeed, the fact that many of these benefits, such as the G.I. Bill and the Civil War veterans’ benefits (Titmuss 1955; Skocpol 1992: ch. 2; Kasza 2002), were introduced during, not after, wars, is indicative of the citizens’ bargaining power over the state during times of war. Rising discontent on the part of citizens during wars not infrequently results in social unrest; scholars have noted the increase in strike activities during World Wars I and II (Smith 1996: 10; Stein 1980: 47). When the state fails to provide adequate compensation to its citizens, the consequences for the state may be even more dire. As Charles Tilly points out, “[C]onnections among state-making, the building of armed forces, and the maintenance of internal control help account for the tendency of revolutions to occur in conjunction with the preparation and the termination of war” (Tilly 1975: 74). While states have certainly been able to extract more during wartime compared to times of peace, this extraction often comes at a considerable price, at times at the expense of the state itself (Tilly 1975: Skocpol 1979).

12

Indeed, there may be a selection bias in the existing scholarship here; states that have been relatively successful at extracting social resources are more likely to have survived into the present period, which makes those states more likely to become objects of study. But this masks the substantial variation in the extent to which states have successfully raised resources for war.

58

Civic Engagement in Postwar Japan

This need to induce citizens into performing war-related services often means giving them greater access to decisionmaking over the process of service production and delivery. In Britain, government and firms provided greater channels for labor unions to discuss issues of common concern during World War II (Donnelly 1999). The close consultative relationships between the Japanese state and industries, as well as between labor and management, have also been traced back to the World War II period (Noguchi 1995). Thus, not only does war present societal actors with opportunities to cultivate their civic skills; it also provides greater opportunities for those actors to exercise those skills, both individually and collectively, further reinforcing those capacities. Research in the field of industrial psychology has also noted how participatory decisionmaking is much more conducive to behavioral change as well as work satisfaction than decisions that are imposed from above (Lawler and Hackman 1969). If this is the case, it may be more realistic to believe that citizens would emerge from wars with a sense of accomplishment and efficacy, regardless of victory or defeat. The argument of this section may be summarized as follows. On the one hand, the acquisition of civic skills is likely to vastly reduce the affected citizens’ perceived costs of civic engagement. At the same time, the wartime broadening of opportunities for voice, which leads citizens to effectively experience participation first-hand, is likely to fundamentally alter citizens’ preferences toward civic engagement. Mobilization for war, then, may provide vast opportunities for social learning, thereby creating an entire generation of citizens who are both capable and willing to engage in collective endeavors. Ceteris paribus, the more extensive the wartime mobilization, the greater the opportunities for citizens to acquire civic skills and, by implication, the greater the likelihood that, if one assumes a postwar democratic regime type, voluntary civic engagement will rise once war is over. Seen from this perspective, some of the findings from the existing studies on participation may be viewed in a new light. Education, for instance, is generally considered to be the best predictor of any form of participation, but the loss of formal educational opportunities in the schools may not matter if it is compensated for by other means of civic education. The forced mobilization of citizens during war may have the paradoxical effect of providing on-the-job training for peacetime voluntary civic life and, in turn, democratic citizenship. This latter point is explored in more detail later. Another point to note is that the emphasis here on the effects of mobilization suggests that actual fighting is not what produces the beneficial effects on civil society. Therefore, nonbelligerent countries that are highly mobilized for extended periods of time may benefit equally from mobilization as countries that do fight. Finally, it should be noted here that there are two dimensions to the debate over “required” public service: the normative and the empirical. The empirical question concerns whether some degree of coercion may, under certain conditions, lead to effective learning, while the normative question concerns the issue

War and Civic Engagement: A Theoretical Framework

59

of whether it is morally justified to coerce individuals into participation even for the sake of socially desirable ends. While recognizing the importance of both issues, the present study hopes to draw a careful distinction between the empirical and the normative. This book is interested in the empirical issue of whether mobilization from above may lead to a more civic-minded citizenry and by no means intends to suggest such mobilization is desirable as public policy.

Path-Dependency in Civic Engagement Wartime mobilization, then, creates a large pool of citizens who are both willing and able to participate successfully in voluntary activities once the war is over, regardless of whether it ended in political victory or defeat. Mobilization both enhances citizens’ preferences toward participation and lowers the costs of participation. Ceteris paribus, societies should experience a marked rise in levels of civic engagement in the wake of wars. A key question remains, however. Chapter 2 demonstrated the considerable variations in the rise in civic engagement across different Japanese prefectures. Even if wartime mobilization explains the general rise in civic engagement in the wake of wars, what explains the variations in this rise across different societies and regions? Indeed, acquiring a particular set of skills does not automatically lead to the exercise of those skills. This book argues that, given a certain level of wartime mobilization, the variations in the growth of postwar voluntary participation across different societies and regions may be explained by path-dependency effects stemming from prewar organizational legacies. These legacies play a crucial role in reducing the costs of participation. The remainder of this section elaborates on the precise mechanisms. Cross-Regional Variation in the Rise of Civic Engagement: A Dynamic Approach Scholars have noted strong path-dependency effects in collective action in times of peace (e.g., Pierson 2004). Pierson argues that collective action exhibits both large start-up costs as well as adaptive expectations, both of which produce positive feedback (Pierson 2004: 33–34). Therefore, once institutionalized, patterns of mobilization are likely to persist over extended periods of time; Pierson cites the “freezing” of party systems in Western Europe, as discussed by Lipset and Rokkan (1967), as a prime example. The high start-up costs for an association is also a key insight of resource mobilization theory (Jenkins 1983). Leaders must be hired, and the organizational apparatus must be established; funds must be raised; rules and procedures must be determined; and members must be recruited. All of these represent substantial investments in time, money, and knowledge. Other things being equal, then, collective action is likely to be path-dependent. As Skocpol argues, associations with large numbers of members are likely

60

Civic Engagement in Postwar Japan

to continue to be able to attract large numbers of members for an extended period of time (Skocpol 1999). Likewise, as Putnam suggested for the case of Italy, societies or regions that exhibit high levels of participation at time t may continue to exhibit high levels of participation at time t + 1, and vice versa, even over a span of centuries (Putnam 1993). However, other things are not equal, and war is an extraordinary time. Should not the acquisition of new civic skills via wartime mobilization be sufficient to overcome the weight of the past? This section argues to the contrary. It claims the organizational legacies of prewar voluntary associational activities shape participation in the postwar period. In particular, it highlights two important variables affected by those legacies: organizational costs, or the costs of (re-)starting voluntary associations, and informational costs, or the costs of (re-)joining such associations. Each is discussed in more detail in the following paragraphs. Organizational Costs Wars inflict hardships on associations in a myriad of ways. While governments often collaborate with some organizations during war to mobilize needed resources, they also typically repress other groups that are deemed to threaten national unity and morale. Even when groups are not outright repressed, the expansion of the draft and the increasing scarcity of resources may make it difficult for organizations to continue to operate as they had before the war. Associations may incur varying degrees of physical damage to their facilities, especially in defeated countries, but also in victorious countries as well. Once the war is over, however, it is still generally easier to revive preexisting organizations than to create new ones from scratch. As Pierson notes, the initial start-up costs for associations are considerable, and preexisting organizations represent valuable “sunk costs,” even when their operation has been interrupted for a while. Indeed, even if physical facilities are destroyed, the memory of how the organization used to operate before the war represents important legacies of investments made in the past and facilitates the revival of associations. If the demand for participation is still there, then, ceteris paribus, it is usually less costly to accommodate such participation by restarting preexisting associations than by building an entirely new organizational apparatus. Indeed, this need to conserve resources should be particularly acute in the wake of defeat, when resources of every kind are likely to be scarce. In regions where levels of civic engagement had been high prior to the war, even if groups ceased to operate during wartime, the preexisting organizational apparatus will therefore be more easily revived or even scaled up once the war is over, opening up opportunities for citizens to participate. In contrast, where levels of associational activity had been low prior to the war, mobilization may create a large demand for participation on the part of citizens, but the supply of associations may be short. And the lack of obvious opportunities for participation may discourage citizens from doing so altogether.

War and Civic Engagement: A Theoretical Framework

61

In addition, even holding cost issues constant, high preexisting legacies of participation – or lack thereof – may affect the willingness of associational leaders to pay the high costs of (re-)starting those associations. In places where associations had been able to attract large numbers of members before the war, former group leaders may have a greater incentive to attempt to revive those associations because they have faith that their efforts will be welcomed. In contrast, where groups had struggled to recruit members before the war, associational leaders may choose to abandon the group altogether when the war is over. Of course, in any society, a certain number of enterprising individuals will prove willing to pay the high entry costs to establish a new voluntary association, even where one has not existed before. The number of such enterprising individuals probably rises in societies that have undergone extensive wartime mobilization. However, in societies or regions that had relatively low levels of associational activity during the prewar period, more of these individuals are necessary to accommodate the heightened levels of demand for participation, since the stock of preexisting associations is relatively small. There is no systematic reason to believe that regions that previously had low levels of associational activities will see such a boom in new organizations that they will be able to catch up with their better-endowed peers, especially given that war-ravaged regions often face a shortage of resources of every kind. Thus, preexisting relative levels of civic engagement across different regions are likely to be replicated in the postwar period, even if civic engagement rises in absolute terms across the board. Informational Costs Even if associations are successfully revived, information on the groups must be readily available to the broader citizenry if actual participation is to increase. This point leads to the second kind of costs, informational costs. Studies of participation, whether on civic engagement or voting, have long stressed the importance of information in shaping participatory behavior (Palfrey and Poole 1987; Verba et al. 1997). In order for citizens to participate, they must first acquire knowledge about what kinds of opportunities for participation are available. They may not need complete information about all of the opportunities that are available in their area; indeed, as studies of voting behavior suggest, citizens’ decisions to participate are often driven by incomplete information regarding candidates and their policy positions (Lupia 1994; Bartels 1996). Nevertheless, existing research suggests that greater levels of information do facilitate levels of participation. Indeed, in the aftermath of wars, when chaos reigns, it might be difficult to know which associations are in existence at all. The lower the costs of acquiring this information, the more rapidly membership may be expected to grow. If citizens know at least a few groups in terms of their organizational culture, membership fees, possible obligations, and so forth, they are more likely to find one that suits their tastes.

62

Civic Engagement in Postwar Japan

This kind of information is likely to be available in greater abundance where the levels of associational activities had been relatively high before the war. In areas or regions that had exhibited high levels of voluntary activities before the war, citizens seeking opportunities to participate are likely to know larger numbers of people with information about voluntary associations before the war, about the state of their postwar revival, and even about whom to contact at the organization. Citizens themselves may also have heard of particular names of groups and know where the groups used to be physically located, and personally go there to solicit information, even without directly knowing any previous group members. This framework suggests that associations in regions where levels of voluntary activity had been high before the war may experience rapid growth rates in membership, even when they do not actively engage in recruiting new members, whether by choice or due to a lack of resources. Since mobilization creates a large pool of citizens who are eager to participate in new organizations, and since citizens may be expected to have enough information to find these organizations, they will rush to join associations, even when associations lack the resources for a full-scale recruitment drive, which is quite likely to be the case in the wake of defeat. In short, the postwar rise in civic engagement may greatly exceed organizations’ initial desire simply to “recover” from their wartime distress. In contrast, information-gathering is likely to pose a greater challenge to citizens in areas where levels of associational activity had been low before the war, impeding the participation of individuals who are otherwise willing. In fact, even if information is available in these types of societies, individuals may think that there are no groups that meet their particular preferences, since the pool of organizations is relatively small to begin with. In this case, citizens would need to create organizations anew or, more likely, simply retreat from the field. Path-Dependency Effects: Summary In sum, the extent to which wartime mobilization produces a growth in civic engagement in the wake of wars is highly contingent on the legacies of prewar institutions. Preexisting groups represent valuable organizational “sunk costs” and substantially lower the informational costs of participation for potential participants once the war is over. Ceteris paribus, the greater the levels of participation from the prewar period, the greater the growth in participation in the wake of war. Ultimately, therefore, wartime mobilization should not close the inequalities between different regions or even whole societies in terms of levels of civic skills and norms of civic engagement, but may rather sustain or even exacerbate them. If this is the case, we may expect to see substantial variation in the extent to which civic engagement grows in the wake of wars. That is, regions or societies that had been endowed with higher levels of associational activities before the war should experience higher rates of growth, whereas those

War and Civic Engagement: A Theoretical Framework

63

that had been characterized by lower levels of associational activities should witness slower rates of growth. The latter are also likely to have increased their stock of civic skills above prewar levels, but they face greater obstacles in putting those new skills to use when the war is over. Consequently, in relative terms, they will not catch up and may even fall farther behind. The argument that prewar associational legacies should crucially shape the extent to which civic engagement rises in the wake of wars, however, is not a claim of historical determinism. The effects of organizational legacies should be mediated by the length and extent to which voluntary associational activity is interrupted. The longer that war suspends voluntary associational activities, and the more severely that voluntary associations were repressed during war, the higher the costs should become to revive those associations, and the weaker the effects of preexisting organizational legacies should be. This should be particularly true if the postwar period does not see a lasting liberalization of state policies toward civic associational life.

Summary This chapter has developed a two-step model to explain the relationship between wars and the trajectory of civic engagement in the wake of those wars. It has argued that the rise of civic engagement among war-affected countries should be conditioned by two factors: the extent of wartime mobilization and the levels of prewar civic activities. While wartime mobilization represents a form of “forced” participation, the distinction between “voluntary” and “involuntary” participation is a matter of degree rather than kind. Mobilization provides citizens with valuable opportunities for civic learning to a degree that is rare during times of peace, enabling them to acquire a wide range of civic skills. Citizens emerge from war with more intense preferences for civic engagement as well as lower costs. The acquisition of those skills, however, does not mean that citizens would automatically start to participate voluntarily in associations once the war is over. The extent to which those skills are subsequently exercised is conditioned by the availability of opportunities, which in turn are shaped by preexisting legacies of prewar participation. Preexisting legacies of participation lower both organizational and informational costs of participation. In short, the higher the levels of civic engagement in the prewar period, the more rapidly engagement may grow in the postwar period.

4 Quantitative Analysis The Rise of Civic Engagement across Forty-Six Japanese Prefectures

Overview Chapter 2 presented new data on the rise of civic engagement in Japan in the wake of World War II. It demonstrated that, contrary to standard expectations, Japan experienced a sharp rise in levels of civic engagement in the initial decade after the war across several different types of groups. This rise was well under way even before economic growth had begun to take off. At the same time, Chapter 2 also showed the presence of considerable variation in the extent of this rise across different Japanese prefectures. To explain these findings, Chapter 3 took a step back from the Japanese case per se to rethink the more general relationship between war and civic engagement. It offered a new theoretical framework that focuses on how experiences of mobilization during wartime equip citizens with new skills that they can apply with renewed vigor after war’s end, particularly if preexisting legacies from the prewar period permit the rapid revival of organizations and help to disseminate information on those groups. The next four chapters test the validity of the framework presented in Chapter 3. The present chapter uses quantitative analysis to assess the explanatory power of the hypothesis on the change in levels of civic engagement between the prewar and postwar periods in Japan at the prefectural level. Chapter 5 then moves beyond the Japanese case and again uses quantitative analysis to test the broader applicability of the framework on thirteen countries that were affected by World War II. Chapters 6 and 7 then return to the case of Japan and perform process-tracing analyses of specific groups in particular cities, illuminating the causal mechanisms that could not be captured by the quantitative analyses in Chapters 4 and 5. This chapter is organized as follows. The next section presents the model to be tested in this chapter, the operationalization of the dependent and independent variables, and the data sources that were used for this study. The third section presents the results of the quantitative analyses. The fourth section tests the validity of an alternative hypothesis that the severity of physical damage 64

Quantitative Analysis

65

incurred during war explains the growth of civic engagement between the prewar and postwar periods. The fifth and final section summarizes and discusses the findings. Data and Operationalization The present chapter seeks to establish the validity of the framework presented in the previous chapter using quantitative analysis. Building on the basic framework presented in Chapter 3, the study tests the following model: ∆Yt, t−3 = α + β1∆X1(t−1, t−3) + β2∆X2(t−1, t−3) + β3X3(t−2) + β4X4(t−3) + β5X3(t−2) X4(t−3) Note that the model consists of independent and control variables that are drawn from four different time periods: time t, or 1955 (1958 for youth groups); time t −1, or 1950, as a lagged control variable; time t − 2, or the war period; and time t − 3, for the prewar period. The dependent variable (∆Yt, t−3) is the relative change in levels of associational membership between the prewar (t − 3) and postwar (t) periods. The main independent variable is an interaction term between wartime (t − 2) mobilization and levels of prewar civic activities (X3(t−2) X4(t−3)). Two control variables are included, income growth per capita (∆X1(t−1, t−3)) and changes in levels of educational attainment (∆X2(t−1, t−3)). Dependent Variables The dependent variable for this chapter (∆Yt, t−3) is changes in the levels of participation at the prefectural level between the prewar and postwar periods. As noted in Chapter 2, data for social service and women’s groups are unfortunately insufficient to test the propositions advanced in this study at the prefectural level. Thus it is only possible to test the model against change in levels of memberships in youth groups and in Christianity at the prefectural level (N = 46). For the postwar period (Yt), membership data from 1955 was used for Christian churches; for youth groups, the closest available data were from 1958. For more on data sources, see Chapter 2. For the prewar period (Yt−3), youth group membership figures at the prefectural level are available from 1922 and 1931; for Christianity, they are available from 1936. Thus, the study used three different operationalizations of changes in membership levels between the prewar and postwar periods: between 1922 and 1958 and 1931 and 1958 for youth groups and between 1936 and 1955 for Christian churches. All three measures of membership change were standardized by population. While it is unfortunate that we cannot subject more types of groups to quantitative analysis, it is noteworthy to point out that youth groups in particular pose a hard case for the hypothesis presented in this study. After all, youth groups involve young people, and therefore many of those people who participated in them before the war would not also participate in them afterward, simply due to advancing age. If, despite this fact, we find that levels of prewar youth group activities, when interacted with levels of mobilization, strongly

66

Civic Engagement in Postwar Japan

conditioned postwar participation in youth groups, this would provide especially strong support for the framework presented in Chapter 3. Note, however, that there would be some overlap, since “youth,” as defined by many of these groups, is a broad category that includes people from childhood through their twenties or even thirties. Independent Variables Chapter 3 argued that the postwar rise in civic engagement may be explained by a combination of wartime mobilization and prewar legacies of participation. Testing for this kind of argument requires an interaction term. The main independent variable for this study thus is a multiplicative interaction term between two variables: a) wartime mobilization (X3(t−2)) and b) prewar levels of participation (X4(t−3)). The operationalization of each of these variables is discussed in the following. Wartime Mobilization. Existing studies operationalize mobilization in at least two ways. Jaggers (1992) uses the number of battle deaths. This, however, is problematic since the number of battle deaths is more likely to reflect the nature of the battles that are being fought rather than levels of mobilization per se; for instance, the United States experienced relatively moderate levels of mobilization compared to many other belligerent countries during World War II, but, as shown in Chapter 1, it incurred relatively high numbers of battle deaths owing to the particularly vicious nature of the Pacific War. This operationalization also does not allow for the measurement of the extent of mobilization for countries that remained neutral, which is often quite high. Bennett and Stam’s (1996) alternative operationalization of mobilization uses a more sophisticated Correlates of War (COW) composite capabilities index, which combines measures for a country’s share of total international military personnel, military expenditures, energy consumption, iron/steel production, total population, and urban population. This index is an improvement, since it captures mobilization on both the home and the war front. The COW index, however, conflates the presumed effects of mobilization with mobilization itself. Iron/steel production, for instance, measures the presumed effect of mobilization rather than mobilization per se, and similarly for energy consumption. But there is no necessary relationship between the number of mobilized workers, that is, the extent of mobilization, and the resulting levels of production. Cohen (1949: 271), for instance, argues that Japan’s industrial production actually suffered during World War II despite large-scale mobilization because skilled workers were not exempt from conscription.1

1

Recent historical evidence reveals that this was not entirely accurate. For instance, highly skilled workers in strategic industries such as aircraft manufacturing, mining, and shipbuilding were allowed to postpone their draft; even in 1945, an estimated 850,000 workers were permitted to postpone military service under this system (Oe 1988: 10–11). But the point remains that different societies may exempt skilled workers from military services to different degrees during wartime.

Quantitative Analysis

67

This study instead uses the total number of mobilized soldiers between 1937 and 1945 for each prefecture, standardized by the population of each prefecture as of 1940, as the measure of mobilization. Militaries may rely on manpower as opposed to technology to varying degrees, so this measure of overall mobilization is indirect. But using military mobilization as a proxy for both military and civilian mobilization is less problematic in the case of World War II, given the common time frame and the relative extremity of the threat. There is still much to be discovered about the precise history of conscription in Japan during World War II, but conscription data at the prefectural level are available in Oe (1988: 252–263). All of Japan was highly mobilized during World War II, but the data show the presence of substantial variation in the extent of mobilization across different prefectures. This provides a useful measure not only of military mobilization but also indirectly of civilian mobilization, since regions that mobilize heavily militarily may be assumed also to mobilize intensively on the home front. If more men are drafted into the armed forces, for instance, women should be more likely to leave the home to undertake tasks previously performed by men. Prewar Levels of Participation. As noted earlier, prewar levels of participation are available at the prefectural level for youth groups for 1922 and 1931, and for Christian groups for 1936. For youth groups, the study tests for the effects of levels of prewar civic activities using figures from both 1922 and 1931. Using figures from both 1922 and 1931 is particularly useful because the political context in Japan underwent considerable change between the early 1920s and the early 1930s. The year 1922 falls just inside the period known as “Taisho Democracy,” the most liberal era in prewar Japan. By 1931, this liberalism had begun to give way to rising militarism. Testing for the effects of membership figures from different points in time from the prewar period should thus help to illuminate not only whether but also which legacies of prewar voluntary activities most affected the growth of civic engagement in the postwar period. If, for instance, membership figures from 1931, in combination with wartime mobilization, are found to be strong determinants of membership growth between the prewar and postwar periods but the 1922 figures are not, this may suggest that the postwar rise in civic engagement reflected policies enacted by the increasingly militaristic regime of the 1930s. If, by contrast, membership figures from both baseline years, in conjunction with mobilization, are found to exert strong effects over the rise of membership between the prewar and postwar periods, we may speculate that the rise in civic engagement in the postwar period may have been shaped not just by the military regime’s policies but also by more enduring legacies from the period of Taisho Democracy, if not from an even earlier era. Interaction Term. In the framework presented in this study, the combined effects of wartime mobilization and prewar legacies should exert strong effects over the growth in membership levels between the prewar and postwar periods. This effect may be captured by a multiplicative interaction term (X3(t−2) X4(t−3)), in which prewar levels of civic engagement are multiplied by the levels

68

Civic Engagement in Postwar Japan

of mobilization. If the hypothesis presented in this study is valid, then, ceteris paribus, prefectures that exhibit both higher levels of mobilization and higher levels of prewar civic activities should see a more rapid rise in civic engagement during the initial decade after the war. Conversely, we should see much more limited growth in levels of civic engagement after the war among prefectures that experienced high mobilization but had seen low levels of prewar associational activities, those that experienced relatively limited mobilization but had seen high levels of associational activities during the prewar period, or those that underwent neither extensive mobilization during wartime nor had seen high levels of civic activities prior to the war. Control Variables While the relatively small sample size (N = 46) did not allow for the consideration of a large number of control variables, the study followed existing studies on participation and included two standard control variables: income (prefectural income per capita) and education. Since the dependent variable is dynamic, representing levels of change between the prewar and postwar periods, the analysis in this chapter also used the changes in levels of income per capita and education between the prewar and postwar periods as controls (∆X1(t−1, t−3), ∆X2(t−1, t−3), respectively). Ceteris paribus, the expectation is that the greater the increase in levels of income and/or educational attainment during this period, the more rapidly civic engagement should rise (Schlozman et al. 1994; Putnam 2000). Since the prewar membership figures that constitute the dependent variable are from 1922, 1931, and 1936, the study used these same three years for education and income per capita as the prewar baseline. For the postwar period, this study uses a lagged model for both of the control variables of income and education (t − 1). Given the high political, economic, and social uncertainty in immediate postwar Japan, it is reasonable to assume that citizens may not immediately have changed their participatory behavior as levels of income and/ or educational attainment changed, but that they may instead have embarked on new activities after a period of adjustment to the new postwar realities. This study assumes the presence of a roughly five-year lag between changes in socioeconomic variables and changes in the level of participation; the assumption here is that levels of participation in 1955/58 were influenced by socioeconomic factors in 1950. Since prefectural-level income per capita data that spanned the prewar and postwar periods were not available, this study uses prefectural-level industrial output and divided the figures by the population of each prefecture as a proxy for income per capita. Data for prefectural-level industrial output were obtained from the Nihon Choki Tokei Soran (Somucho 1988a: 414–418). Education is operationalized by the percentage of the population entering high school, standardized by population. This is a more useful measure compared to, for instance, the percentage of the population entering college, since the latter was exceedingly small during the prewar period and the early

Quantitative Analysis

69

postwar period, providing very little variation across prefectures. Moreover, while high school attendance has become nearly universal in Japan today, this was not the case in the early postwar period.2 Data from 1950 are available from the Ministry of Education’s Gakko Kihon Chosa Hokokusho (Report on Basic Statistics on Schools) (Monbusho 1950); data for the prewar period are provided in the Nihon Teikoku Monbusho Nenpo [Annual Report of the Japanese Imperial Ministry of Education] (Monbusho, various years). The study also checked for multicollinearity among the independent and control variables and found that there were no serious issues.3 Results: Explaining Changes in Prewar and Postwar Membership Levels Regression analyses using different measures of change in youth and Christian membership levels between the prewar and postwar periods largely confirm the validity of the framework presented in the previous chapter. Ceteris paribus, prefectures that had been more highly mobilized and had seen higher levels of civic activities prior to war saw more rapid rates of growth in participation between the prewar and postwar periods as compared to prefectures that were less highly mobilized or had less rich historical legacies of associationalism. Table 4.1 shows the results of the regression analyses using the rise in participation in youth groups between 1931 and 1958 as the dependent variable. The table shows the interaction term between wartime mobilization and prewar levels of civic activities to be a strong predictor of the changes in the levels in participation between 1931 and 1958. The results for Model 1 show that, ceteris paribus, the higher the levels of membership in youth groups in 1931 and the higher the levels of mobilization, the faster the growth rate in memberships was likely to be between 1931 and 1958. Note that the fact that the coefficients on both the mobilization variable on its own and the prewar youth group membership variable on its own are

2

3

Across the forty-seven prefectures, the percentage of the population entering high school as of 1950 ranged from 78.9% in Kumamoto Prefecture to 97.1% in Osaka Prefecture (Monbusho 1950). Conceivably, levels of conscription may have been higher in urban prefectures, where larger numbers of youth may have been found. Changes in levels of income per capita between the prewar and postwar periods may also be expected to be higher in more populous prefectures. Tests of multicollinearity, however, found no correlation between the extent of mobilization and changes in levels of income per capita. Conscription-aged males were indeed more numerous in the urban as opposed to rural prefectures, as the youth migrated to cities in search of educational and occupational opportunities; according to the 1940 census, males between the ages of fifteen and twenty-nine comprised 26.5% of the population nationwide, but the corresponding figure was 34.2% in Tokyo and 31.7% in Osaka (calculated from Sorifu Tokei-kyoku 1961: 306–311). Nevertheless, rates of conscription were not systematically higher in more urban prefectures; that is, there appears to be no correlation between the level of urbanization of a prefecture and the percentage of individuals conscripted from that prefecture.

70

Civic Engagement in Postwar Japan

Table 4.1. Regression Analysis, Explaining the Growth in Youth Group Membership, 1931–1958, Using Participation Rates as of 1931 and 1922 Variable

Model 1

Model 2

Prewar membership per population (1931)* Mobilization per population (interaction term) Prewar Youth Group Membership per pop. (1931) Prewar membership per population (1922)* Mobilization per population (interaction term) Prewar Youth Group Membership per pop. (1922) Mobilization per population ∆income per capita ∆Education Constant

.10 (.048)**

−3.73 (1.516)** −.00 (.000) .29 (.204) 22.01 (7.827)***

−2.47 (1.428) .00 (.000) .51 (.256)* 17.61 (7.812)**

R2 N

.439 44

.266 42

−.55 (.237)** .00 (.001)**

−.01 (.005)**

Robust standard errors in parentheses. ***p ≤ .01, **p ≤ .05, *p ≤ .1.

negative does not refute the validity of the model proposed in this study. This is because in analyses using interaction terms, the coefficients on the individual variables that make up the interaction term cannot be interpreted on their own (Braumoeller 2004). Model 2 uses the same dependent variable as Model 1, that is, the change in membership figures between 1931 and 1958, but replaces the measure of prewar membership with youth group membership figures as of 1922. The results show that even when membership figures from 1922 are used as the measure of prewar levels of civic engagement, the effects of the interaction term between prewar membership and wartime mobilization are substantial. Concretely, prefectures that enjoyed higher levels of membership in 1922 and higher levels of mobilization during wartime were also more likely to exhibit higher rates of membership growth between 1931 and 1955. The magnitude of the effect is weaker compared to Model 1; the coefficient on the interaction term here is .00, as opposed to .02 in the previous two regressions. But it is still positive and highly significant. In this regression, the effect of educational gains, which had not been significant in Model 1, now becomes weakly significant. Results presented for Models 1 and 2 show support for the hypothesis advanced in this study. Moreover, the fact that not only the 1931 but also the 1922 baseline works suggests that the regional discrepancies in youth

Quantitative Analysis

71

Table 4.2. Regression Analysis, Explaining the Growth in Youth Group Membership, 1922–1958, Using Participation Rates as of 1922 Variable Prewar membership per population (1922)* Mobilization per population (interaction term) Prewar youth group membership per population (1922) Mobilization per population ∆income per capita ∆Education Constant R2 N

Regression Coefficient (Standard errors) .01 (.006)* .02 (.074) −3.04 (5.233) .00 (.000) −.29 (.396) 40.52 (23.144)* .306 42

Standard errors in parentheses. ***p ≤ .01, **p ≤ .05, *p ≤ .1.

group membership growth were longstanding, dating at least back to the era of Taisho Democracy. The wartime mobilization then exacerbated these discrepancies. Similar results are obtained when rates of membership growth between 1922 and 1958, instead of 1931 and 1958, are used as the dependent variable. Table 4.2 tests the hypothesis that the growth in youth group memberships between 1922 and 1958 was conditioned by a combination of wartime mobilization and levels of youth group memberships as of 1922. The results shown in Table 4.2 also indicate that, ceteris paribus, when extensive wartime mobilization occurred in regions that had seen higher levels of youth group memberships in 1922, this yielded higher levels of growth in youth group memberships between 1922 and 1958, although the magnitude of the effect is weaker than for Model 1; see Table 4.1. The p-value for the interaction term between prewar membership as of 1922 and mobilization is .051, very nearly significant at the p ≤ .05 level. The somewhat weaker effect of the interaction term in this regression is not surprising, since the time lag between the prewar experiences and the postwar growth in civic engagement is much longer in this regression compared to the previous analyses. As in Model 1 of Table 4.1, neither economic growth nor gains in educational attainment between 1922 and 1958 is found to be a significant predictor of growth in youth group memberships during this period. Finally, Table 4.3 shows the regression analysis to explain the growth in Christian membership. Since 1936 is the only year for which prewar religious membership is available at the prefectural level, this study merely reports results from the one regression analysis.

72

Civic Engagement in Postwar Japan

Table 4.3. Regression Analysis, Explaining the Growth in Religious Membership, 1936–1955, Using Participation Rates as of 1936 Variable

Regression Coefficient (Standard errors)

Prewar membership per population (1936)* Mobilization per population (interaction term) Prewar religious membership per population (1936) Mobilization per population ∆income per capita ∆Education Constant

.17 (.077)** −.90 (.387)** −.38 (.329) .000 (.000) .01 (.020) 3.68 (1.795)**

R2 N

.232 44

Standard errors in parentheses. ***p ≤ .01, **p ≤ .05, *p ≤ .1.

The interaction term between mobilization and levels of prewar membership in religious groups as of 1936 is found to be a strong determinant of growth rates in Christian membership between 1936 and 1955, and in the expected positive direction. It is also significant at the 5 percent level. That is, ceteris paribus, the higher the levels of Christian members as of 1936 and the more extensive the mobilization, the more rapid was the growth in Christian membership in that prefecture between 1936 and 1955. As with the analyses for youth groups, neither economic growth nor educational gains was found to be a strong predictor of gains in membership in religious organizations. Baby Boom and/or Demand for Child Care? One alternative hypothesis for the rise in youth group memberships needs to be addressed. For youth groups, it is conceivable that the rapid growth in membership reflected not so much the effects of wartime mobilization or prewar legacies of participation as the baby boom that occurred shortly after the war. Indeed, membership growth may even have been driven by a need for child care. First, to test for the possible impact of the baby boom, Kage (2005) tested for the effects of fertility, or number of births per unit of population, on both levels and the extent of growth in youth group memberships across prefectures and found no effect. That is, prefectures that had higher birth rates did not systematically exhibit higher levels of youth group memberships or membership growth. Similarly, the correlation between the growth in fertility rates and the growth in youth group memberships is also weak; for instance, the correlation between the change in fertility rates between the 1935 and 1947 censuses and the growth in youth group memberships between 1936 and 1958 was −.112;

Quantitative Analysis

73

between the change in fertility rates between the 1930 and 1947 censuses and the growth in youth group memberships between 1936 and 1958 it was −.131 (calculated from Somucho 1987). Thus, the baby boom per se does not appear to provide a satisfying explanation for the rise in youth group memberships in immediate postwar Japan. Second, a more subtle version of the hypothesis would suggest that the heightened levels of participation in youth groups may have been driven not so much by the baby boom per se as by the demand for child care services. A full assessment of this hypothesis would require data on the number of working mothers who did not have the support of extended families during this period for each of the forty-six prefectures, preferably for both the prewar and postwar periods. Unfortunately, these data are not available. Nevertheless, it is possible to perform some crude tests of the hypothesis by focusing on location: urban versus rural. In general, the percentage of working mothers who lacked support of an extended family to take care of children should conceivably have been higher in urban than in rural prefectures. Moreover, according to the 1950 census, prefectures that saw the highest rates of fertility around this time tended to be clustered among the urban and suburban prefectures such as Osaka, Saitama, and Chiba (Sorifu 1954).4 Thus, the demand for child care outside the home should have been highest in these more urban prefectures. The data in Chapter 2, however, showed that the highest rates of growth in youth group memberships tended to be concentrated among the rural prefectures such as Mie, Yamagata, Tottori, and Akita. Thus this child care hypothesis also seems incorrect, although it is possible that the assumption that rural households have a lower demand for child care than urban households is misplaced. Third, in June 1950, the Ministry of Health and Welfare conducted a national survey of elementary school pupils in need of care (Koseisho Jido-kyoku 1950). According to this survey, only 82,628 pupils nationwide, or .25 percent of all pupils in Japan, were found to be in need of day care due to working parents. In addition, there was no correlation between the percentage of children in need of care and the level of youth group membership or membership growth by prefecture.5 Unfortunately, similar data are not available for the prewar period, so it is impossible to test whether the trans-war growth in youth group membership was driven by a corresponding growth in children in need of day care between the prewar and postwar periods. None of this information provides definitive evidence that the rise in youth group memberships was not driven by a heightened demand for child care services. Nevertheless, while the data are somewhat sporadic, taken together, 4

5

Tokyo Prefecture was an exception, with the lowest rate of fertility among the forty-six prefectures. The correlation coefficient between the percentage of children in need of day care in 1950 and youth group membership as of 1955 was .059; between the percentage of children in need of day care in 1950 and youth group membership growth between 1936 and 1955 it was −.118.

74

Civic Engagement in Postwar Japan

the available information seems to cast doubt on the hypothesis that the baby boom and/or the demand for child care services was the primary impetus for the growth in youth group memberships between the prewar and postwar periods. Summary The foregoing quantitative analyses found that a combination of more extensive wartime mobilization and higher levels of prewar levels of civic engagement strongly conditioned the increase in memberships in youth and Christian organizations between the prewar and postwar periods. Ceteris paribus, the more extensive the mobilization during wartime and the higher the levels of prewar memberships, the more rapid was the rise in memberships between the prewar and postwar periods. For youth groups, this was true regardless of whether membership growth was measured in terms of change between 1922 and 1958 or between 1931 and 1958. Moreover, the regressions for youth groups also found that rates of membership growth between 1931 and 1958 were not merely shaped by levels of membership as of 1931 but also from 1922 as well; and similarly, the rise in memberships between 1922 and 1958 was also strongly affected by levels of memberships from 1922, although the effects did become weaker as the distance between start and end date widened. This suggests that, at least in the case of youth groups, the effects of preexisting legacies may be traced at least back to 1922, if not earlier. Unfortunately, the lack of surviving data does not allow us to probe how much farther we may go back into the prewar period before the path-dependency effects dissipate. However, the fact that membership levels as of 1922 still exerted strong effects over the growth in youth group memberships a quarter-century and a cataclysmic war later does suggest that relatively enduring sociocultural factors shape levels of civic activities. Because membership in labor unions in the postwar period was often not voluntary, the present chapter did not present results of the analysis of labor union membership. Nevertheless, analyses that I have conducted elsewhere suggest that the overall hypothesis advanced in this study also holds for labor union membership as well.6 Interestingly, in three of the four regressions presented, neither the incomegrowth figures nor gains in educational attainment, typically the most reliable predictors of voluntary participation, were found to be a strong predictor of membership growth in Japanese youth and Christian memberships between the prewar and postwar periods. While a full exploration of this finding remains beyond the scope of this project, it is possible to speculate on some reasons for why this might be. First, the income growth rates and gains in educational attainment were measured at the prefectural, rather than the individual, level. It is thus conceivable that had they been measured at the individual level, the 6

Results are available in Kage (2005) or on request.

Quantitative Analysis

75

regressions may have found stronger effects for both of the variables. Second, for education in particular, it is possible that at least for the period under study, formal schooling was perhaps less important than, and/or may have been augmented by, informal social learning, whether through wartime mobilization or otherwise. This view is in fact consistent with the hypothesis put forth in this project overall.

Alternative Hypothesis: The Effects of Wartime Destruction This study also tests the validity of one variant of the victory/defeat hypothesis. The victory/defeat hypothesis argues that defeat may cause psychological traumas among the affected citizenry that lead citizens’ to withdraw from civic life. All of Japan was defeated in war, but some parts of Japan experienced much higher levels of destruction than others. It is thus reasonable to extend the victory/defeat hypothesis to assert that trauma should have been more acute in prefectures that experienced higher levels of physical damage, and that the more trauma during the war, the less civic engagement after it. This section probes whether harder hit areas in fact witnessed slower rates of postwar growth in memberships. To this end, this study replaced the variables of wartime mobilization and prewar levels of civic engagement in the regressions with levels of damage, operationalized as the number of deaths per 1,000 citizens in each prefecture. Figures for wartime damage are available through the Economic Stabilization Board’s report on wartime damage (Keizai Antei Honbu 1949). The report provides data on both human casualties as well as physical damage (number of destroyed buildings) during World War II at the prefectural level.7 However, since, perhaps not surprisingly, quantitative analyses using both measures of wartime damage yielded very similar results, only the results from the regressions using wartime death figures are reported here.8 The number of deaths is, of course, merely one way of operationalizing the victory/defeat hypothesis; while it is certainly plausible that higher death figures should lead to more severe psychological trauma, and that their effects may overlap with the effects of victory and/or defeat, the correlation may or may not be perfect. Chapter 6 tests more explicitly for the effects of victory and/or defeat itself using cross-national analysis, but this chapter will test just one variant of the argument.

7

8

While the damage in Japan was mainly concentrated in the cities, there was no issue of multicollinearity between the number of war deaths per 1,000 citizens and differences in levels of income per capita. To some degree, this is to be expected, since levels of wartime deaths should be roughly correlated with levels of physical damages. Results from the analyses using wartime physical damages as the independent variable are available on request.

Civic Engagement in Postwar Japan

76

Table 4.4. Regression Analysis, Explaining the Growth of Youth and Christian Memberships Using Wartime Damage Figures Variable

Youth Group Membership, 1931–1958

Youth Group Membership, 1922–1958

War deaths per 1,000 citizens ∆income per capita ∆Education Constant

−.03 (.028)

−.22 (.166)

R2 N

.070 46

.00 (.000) .29 (.202) .45 (1.491)

Christian Membership, 1936–1955 .00 (.007)

.00 (.000) .56 (.889) 13.13 (12.101)

.00 (.000) .01 (.022) 1.37 (.370)***

.076 46

.007 46

Standard errors in parentheses. ***p ≤ .01, **p ≤ .05, *p ≤ .1.

An examination of the effects of deaths also helps to address another argument that has emerged among scholars of civil wars in recent years: the idea that witnessing more traumatic events actually elevates subsequent levels of participation (Bellows and Miguel 2009; Blattman 2009). This “post-traumatic growth” hypothesis turns the typical hypothesis about trauma on its head. Although the number of deaths is not the equivalent of the extent to which citizens witnessed violence, if more people were killed in a prefecture, then citizens in that prefecture should also have witnessed more violence first-hand as well. Table 4.4 shows the results of the regression analyses. It shows that, ceteris paribus, levels of damage during World War II do not provide a strong explanation, either for the growth in Christian memberships between 1936 and 1955 or for the growth in youth group membership between 1922 and 1958 or between 1931 and 1958. In all three regressions, the coefficients for wartime damage figures are in the negative direction, as the victory/defeat hypothesis would expect, but the standard errors are too large for the variable to be considered a reliable predictor of membership growth. Thus the regressions do not provide support for either the wartime trauma hypothesis or the revisionist “post-traumatic growth” hypothesis of Bellows and Miguel (2009) and Blattman (2009). As in the earlier analyses, economic growth and gains in educational attainment are also found not to exert strong effects over membership growth in youth and Christian memberships between the prewar and postwar periods.

Summary The quantitative analyses presented in this chapter provide considerable support for the hypothesis that the interaction of wartime mobilization and the

Quantitative Analysis

77

preexisting legacies of participation strongly affects the growth of civic engagement between the pre- and postwar periods. Ceteris paribus, prefectures that exhibited higher levels of associational membership during the prewar period and which then underwent more extensive mobilization saw the highest rates of membership growth in youth and Christianity between the prewar and postwar periods. Meanwhile, the analyses found little support for the baby boom hypothesis with respect to youth groups, or for at least one variant of the victory/defeat hypothesis, according to which levels of damage account for levels of civic engagement in the wake of war. This result, however, should not be viewed as spelling the end of the victory/defeat hypothesis, since generally that hypothesis is pitched at the level of nations rather than regions. The study will test more explicitly for the effects of overall national victory versus defeat in the next chapter.

5 The Long-Term Effects of Wartime Mobilization Cross-National Analysis

Overview The previous chapter tested, on Japanese datasets, the validity of the argument that levels of civic engagement may rise dramatically in the wake of wars a) when wartime mobilization has equipped citizens with the skills that are necessary for participation and b) when the preexisting associational legacies provided fertile ground for the rapid expansion of civic activities. Higher levels of membership growth between 1945 and 1955 were seen in areas of Japan that had been more highly mobilized and which had exhibited relatively high levels of civic activities prior to war. These factors indeed propelled civic engagement to new heights once the war ended, regardless of how much physical or economic damage the region had incurred. This chapter moves beyond the Japanese context and tests the broader applicability of the framework on data from thirteen countries that experienced World War II. Moving beyond the Japanese context also allows the study to test more explicitly for the validity of the victory/defeat hypothesis. Chapter 4 did test for the validity of one variant of the victory/defeat hypothesis: the idea that different levels of physical damage may produce the differential growth of civic engagement in the wake of wars. It found the empirical evidence for this hypothesis to be quite weak in the Japanese case. However, physical damage is not always synonymous with victory/defeat per se, which are national-level variables; while defeated countries may often incur higher levels of damage than victorious countries, the correlation is not perfect. Moreover, Japan is just one country. A full test of the effects of victory or defeat, which are nationallevel variables, requires cross-national analysis, and this is the task of the present chapter. An analysis of the long-term effects of World War II on levels of civic engagement across countries should also speak to the recent lively debate over the extent to which the younger cohorts in industrialized countries today “bowl alone” (Stolle and Hooghe 2005). U.S. data appear to show a dramatic decline in civic engagement over the last few decades (Putnam 1995a, 1995b, 2000), 78

The Long-Term Effects of Wartime Mobilization

79

while studies from certain Western European countries, such as Belgium, the Netherlands, and Denmark, seem to contradict this trend (e.g., Hooghe 2003; de Hart and Dekker, 1999; Torpe 2003). The argument advanced in this study may yield insights that may help to begin to resolve this debate. Within the literature on the United States, it is often argued that World War II prompted a surge in levels of civic engagement in the postwar period, and that this gave rise to the “golden age” of civic engagement in the United States (Skocpol 1999; Skocpol et al. 2002). The recent fall in civic engagement in the United States is attributed at least in part to the passing of this “greatest generation.” This study agrees and points out that since (for instance) Belgium and the Netherlands did not experience the extensive, long-term mobilization for war that the United States did, it would not be surprising if they also did not give rise to a “greatest generation” either. In short, if the framework offered in this book is valid, then, ceteris paribus, countries that were more extensively mobilized during World War II should have seen a more rapid growth in levels of civic engagement than compared to those that were mobilized less. If the victory/defeat hypothesis is valid, then ceteris paribus, countries that emerged victorious from World War II should have seen a more rapid growth in levels of civic engagement than those that saw defeat. A major challenge in assessing the effects of World War II on voluntary participation cross-nationally is that reliable time-series data on levels of civic engagement that go back to the pre–World War II era have not been compiled, and often do not exist, for many countries outside of the United States. Even less data have been compiled that are comparable across countries. Thus, as a second-best option, this study makes novel use of existing postwar surveys, leveraging patterns in participation that may have persisted since World War II. It finds that countries that were more highly mobilized during World War II, such as Germany and the U.K., indeed show much larger and enduring jumps in levels of civic engagement than other, less mobilized countries between the pre–World War II to the World War II cohorts. The data reveal this pattern regardless of whether that country was victorious or vanquished, and that this effect has persisted over subsequent decades. The World War II generation may indeed be the “greatest generation” in many countries (Mettler 2005), but its “greatness” is contingent on the intensity of wartime mobilization. This chapter proceeds as follows. The next section outlines the data and the methods that were used to test the framework advanced in this book on a broader set of country-cases. The third section lays out the results of the test of the argument offered in this volume as well as for two alternative hypotheses. The fourth section concludes.

Data and Methods This chapter tests the validity of the hypothesis presented in Chapter 3 beyond simply the Japanese context. Unfortunately, the data limitations for conducting

80

Civic Engagement in Postwar Japan

this kind of analysis are considerable. In particular, to the best of my knowledge, there is no reliable data source that provides information on countries’ levels of civic engagement prior to World War II in a comparable manner. This presents a serious problem for testing the effects of path-dependency from the prewar period. While this is quite unfortunate, the theory offered in Chapter 3 argued that mobilization on its own should produce at least some rise in levels of civic engagement, even though the extent of such a rise would depend on prewar legacies. Therefore, this chapter tests that proposition, and it also tests the victory/defeat hypothesis. The findings presented in this chapter, then, should be interpreted with some caution, as they do not provide a full test of the framework presented in this volume. Even in testing simply for the effects of mobilization, the data constraints are formidable. We cannot simply compare overall levels of civic engagement from the prewar period until 1955, because such data are not directly available, and certainly not in a cross-nationally comparable form. Given the limits of data availability, this study devises an alternative strategy. Specifically, it leverages surveys to gain information on the recent participatory behavior of citizens who spent their formative years at different moments in historical time. In short, we use contemporary surveys to gain insights on past shocks. If the framework presented in this volume is valid, then individuals who came of age around the time of war in more highly mobilized countries should participate at consistently higher rates throughout the course of their lives compared to individuals in the immediately preceding generational cohort. A large social science literature suggests that the socializing effect of mobilization should be particularly strong among the younger members of society who are in their “formative years” and who are therefore more susceptible to external influences. The experience of mobilization opens up opportunities for the acquisition of civic skills. Once acquired, these skills will have lasting effects, and they are likely to facilitate individuals’ participation in civic activities long after the war has ended. Thus, individuals who experienced intensive mobilization as young adults should continue to participate at high levels throughout their lives, whether their country emerged from war victorious or vanquished. In contrast, in less mobilized countries, individuals who came of age around the time of war should participate at rates that are more or less similar to individuals of the immediately preceding generational cohort. And this difference may manifest itself in present-day surveys as well. Therefore, we can compare postwar levels of participation in civic groups as reported by individuals belonging to the pre–World War II cohort vis-à-vis those who belong to the generational cohort that came of age during World War II. Similarly, if the victory/defeat hypothesis is valid, then individuals who came of age during their countries’ victories in World War II should exhibit markedly higher rates of participation throughout the course of their lives than their elders. By contrast, those who came of age just in time to see their countries’ defeat in the war should participate at lower levels than their elders.

The Long-Term Effects of Wartime Mobilization

81

This research strategy is admittedly a second-best option, in that events of the intervening years from World War II may have affected people’s rates of participation in one direction or another. At the same time, however, in some ways this method provides a hard test; if, in spite of the intervening years, individuals in generational cohorts that came of age around World War II in more highly mobilized countries were still participating at higher rates compared to their preceding cohorts decades after the war, this would provide considerable support for the hypothesis laid out in this book. Moreover, the assumption underlying this strategy – that experiences during youth persist for the remainder of individuals’ life-courses – has considerable backing from the literature on participation. Longitudinal studies of participation consistently find participation during youth to be a strong predictor of participation during adulthood (e.g., Beck and Jennings 1982; Verba et al. 1995; Putnam 2000; Jennings 2002; McFarland and Thomas 2006; Hart et al. 2007; Kjonniksen et al. 2009). These formative years are typically viewed as “critical junctures” during which individuals’ worldviews and habits are formulated and crystallized. Once in place, these worldviews in turn serve as the “predispositions” through which subsequent events are interpreted, thereby further shaping behavior (Sears and Funk 1999). The large number of studies that link community service during school years to higher levels of participation during adulthood noted in Chapter 3 also supports the claim that youth should be most strongly influenced by the experience of mobilization. Indeed, the effects of these external events have often been shown to trump parental effects (Jennings et al. 2001; Sears and Valentino 1997). This certainly does not suggest that no learning takes place after the formative years, but it is certainly plausible that the youth should be more susceptible to external events compared to other generational cohorts and therefore more likely to be shaped by those events thereafter. If this is the case, then individuals who experienced extensive mobilization during these crucial formative years should be especially likely to continue to participate at high levels throughout the course of their lifetimes. It should be noted that the effect of war on participation should manifest itself the most clearly between war-affected generational cohorts and their immediately preceding cohorts, but not necessarily between war-affected cohorts and their subsequent cohorts. This is because other exogenous factors may be at work that may either promote or dampen participation among cohorts which succeed war-affected cohorts. The war-affected cohort’s commitment to civic engagement may or may not be passed on to subsequent generational cohorts. Thus, in order for the hypothesis that war itself may accelerate civic engagement to be deemed valid, the pattern that Putnam identified on the basis of U.S. data, that the “long civic generation” exhibits higher rates of participation compared to cohorts that both precede and succeed it (Putnam 2000), need not hold elsewhere. Thus I limit my present analysis to the war-affected cohorts and their immediately preceding cohorts only. I now turn to a more technical description of the data used in this project.

82

Civic Engagement in Postwar Japan

Data and Case Selection This study analyzed data from the World Values Survey (WVS), waves 1981, 1990, and, when available, 1995–1997. Using three waves of surveys, each with identically worded questions, allows, to some degree, for the isolation of cohort effects from life-cycle effects: that is, to distinguish the parts of participatory behavior that are affected by the unique experiences of a particular cohort from those that are driven by normal changes within an individual’s life-cycle, such as entry into the workforce, beginning a family, or retirement. Country-case selection was based on availability of high-quality data (i.e., few missing values) from at least two of the three waves of the WVS. Thirteen countries met this criterion (Belgium, Canada, Finland, France, Germany, Ireland, Italy, Japan, Netherlands, Sweden, Switzerland, the U.K., and the United States). These countries included all of the major belligerents that were already democratic or became democracies in the immediate postwar period. The thirteen countries also provide useful variation for contrasting the mobilization hypothesis against the victory/defeat hypothesis as well. The United States, the U.K., and Canada offer clear-cut cases of victory; Germany, Italy, Japan, and Finland are clear cases of defeat; Ireland, Sweden, and Switzerland present cases of neutrality; and Belgium, France, and the Netherlands provide more ambiguous cases of “victory” and liberation after initial defeat. Dependent Variable All three waves of the World Values Survey include a battery of questions that asks: Now I am going to read off a list of voluntary organizations; for each one, could you tell me whether you are an (1) active member, (2) an inactive member, or (3) not a member of that type of organization? Church or religious organization; sport or recreation organization; art, music, or educational organization; labor union; political party; environmental organization; professional organization; charitable organization; any other voluntary organization.

Because “active” or “inactive” membership is a subjective judgment that may vary across cultures as well as individuals, all those who responded that they were “members” of a given type of organization were grouped together, as opposed to “non-members.” For each respondent, the number of memberships was then added up to create a variable between 0 (minimum) and 9 (maximum). This was used as the dependent variable. Table 5.1 shows the distribution of this dependent variable, by country: the mean number of associational memberships for the war-affected cohort and its immediately preceding cohort. Following Inglehart (1988), the study uses ten-year cohorts and compares levels of memberships for those born between 1911 and 1920 against those born between 1921 and 1930. The 1921–1930 cohort is particularly important because these are the individuals who were

The Long-Term Effects of Wartime Mobilization

83

Table 5.1. Mean Number of Associational Memberships: Pre–World War II and World War II Cohorts Country

1911–1920 Cohort

1921–1930 Cohort

Percentage Change

Belgium Canada Finland France Germany (West) Ireland Italy Japan Netherlands Sweden Switzerland United Kingdom United States

0.69 1.02 1.96 0.39 1.09 0.62 0.30 0.38 1.39 1.14 1.00 0.86 1.73

0.79 1.18 2.20 0.52 1.63 0.73 0.31 0.61 1.44 1.64 1.30 1.23 2.11

114.94 115.69 112.24 133.33 149.54 117.74 103.33 160.53 103.60 143.86 130.00 143.02 121.97

13-country mean

0.97

1.22

128.93

Source: World Values Surveys 1981, 1990, 1995–1997.

young enough, but not too young, to be molded for life by participation in the wartime mobilization effort; at the end of World War II, these individuals were aged between fifteen and twenty-four. This is slightly longer than Mannheim’s ([1928] 1952) definition of the “formative years” as those who are aged between eighteen and twenty-five. At first sight, those who were aged only fifteen at the end of World War II, and even younger during the war, may seem too young to be profoundly affected by the experiences during war. However, including the relatively young is also consistent with the fact that as World War II intensified, not only adults but also large numbers of school-aged children were also mobilized into the war effort in many countries as well. This demarcation of the war-affected cohort is also quite close to the “long civic generation” that has been identified by Putnam (2000). Putnam refers to the “long civic generation” as those born before 1934, and in particular he sees individuals who were born between 1925 and 1930 to be the “core” of this group (Putnam 2000: 254). The operationalization adopted in this study indeed includes this core group. In order to isolate the effects of World War II from other events that may have affected the levels of participation around the time, such as the Great Depression, this study added a lower bound of 1921.1 Table 5.1 shows that for all thirteen countries under study, individuals who belong to the World War II–affected cohort born between 1921 and 1

Goss (1999) operationalizes the “long civic generation” as those individuals who were born between 1910 and 1930. However, to isolate the effects of the Great Depression from the effects of World War II, this study takes just the latter half of the 1910–1930 cohort.

84

Civic Engagement in Postwar Japan

1930 still participated at higher rates compared to their immediately preceding cohorts during the 1980s and 1990s, when the World Values Surveys were conducted. The table also reveals substantial variation across countries in the extent of the increase in levels of participation between generational cohorts; individuals of the war-affected 1921–1930 cohorts in Japan, (West) Germany, Sweden, and the United Kingdom participate at much higher levels compared to their counterparts in the 1911–1920 cohort, indeed 140 percent more associations or more, while in Italy and the Netherlands, for instance, the difference in levels of participation between individuals of the two cohorts is negligible. At first glance, victory and/or defeat does not appear to provide the main explanation; among countries that show relatively large differences in levels of participation, Japan was a vanquished country, the United Kingdom was a victor, while Sweden remained neutral. The bottom half of the table is similarly comprised of a mixture of victorious, defeated, and neutral countries. However, the figures shown in the table need to be interpreted with caution, since by the time that the surveys were conducted, individuals in the 1911–1920 cohort were quite elderly and were therefore likely to participate in fewer associations than those in the 1921–1930 cohort. Thus, the effects of age are separated out more explicitly from the effects of generational cohort in Table 5.2. Table 5.2 breaks down the figures presented in Table 5.1 and presents the number of associational memberships for individuals of the 1911–1920 and 1921–1930 generational cohorts as they reached the same ages. Thus, for instance, a West German citizen who was born between 1911 and 1920 participated in an average of .68 associations when s/he was aged between sixtyone and sixty-five, whereas a West German citizen who was born between 1921 and 1930 participated in an average of 1.90 associations when s/he was also aged between sixty-one and sixty-five. The table thus shows that even when controlling for age, the difference in levels of memberships between the 1911–1920 and the 1921–1930 cohorts still persists. Across the thirteen countries under study, individuals who were born between 1911 and 1920 participated, on average, in .80, .88, and 1.01 groups when they were aged between sixty-one and sixty-five, sixty-six and seventy, and seventy-one and seventyfive, respectively, while those who were born between 1921 and 1930 were members of 1.26, 1.48, and 2.21 groups as they reached the same ages, respectively. The levels of memberships in the later generational cohorts at the ages of sixty-one to sixty-five, sixty-six to seventy, and seventy-one to seventy-five were thus 157.50 percent, 168.18 percent, and 218.81 percent, respectively, of what they were in the prior generational cohorts. Again, it would appear that there was a nearly across-the-board increase in levels of civic engagement from the 1911–1920 to the 1921–1930 cohorts. But what might explain the crossnational variations in the extent of this increase?

85

Table 5.2. Number of Associational Memberships at the Same Age Level: Pre–World War II and World War II Cohorts Country

Age 61–65 1911–20

1921–30

Difference (%)

Age 66–70 1911–20

1921–30

Difference (%)

Age 71–75 1911–20

1921–30

Difference (%)

Belgium Canada Finland France Germany (West) Ireland Italy Japan Netherlands Sweden Switzerland U.K. U.S.

0.52 0.91 N/Avail 0.53 0.68

0.92 1.01 1.95 0.58 1.90

176.92 110.99 – 109.43 279.41

0.84 1.13 1.88 0.37 1.03

0.90 1.34 2.53 0.73 1.98

107.14 118.58 134.57 197.30 192.23

0.70 0.96 2.04 0.11 1.41

N/Avail N/Avail 2.30 N/Avail 2.67

– – 112.75 – 189.36

0.82 0.31 0.36 1.33 1.24 N/Avail 0.64 1.46

0.40 0.25 0.67 1.81 2.03 1.42 1.31 2.15

48.78 80.65 186.11 136.09 163.71 – 204.69 147.26

0.74 0.36 0.36 1.27 0.86 0.63 0.80 1.18

0.65 0.23 0.93 1.51 1.67 2.61 1.32 2.83

87.84 63.89 258.33 118.9 194.19 414.29 165.00 239.83

0.49 0.15 0.24 1.94 2.06 0.58 0.96 1.55

N/Avail N/Avail 0.92 N/Avail 2.23 2.02 N/Avail 3.14

– – 383.33 – 108.25 348.28 – 202.58

13-country mean

0.80

1.26

157.50

0.88

1.48

168.18

1.01

2.21

218.81

Source: World Values Surveys 1981, 1990, 1995–1997, Table 2.

86

Civic Engagement in Postwar Japan

Index of Mobilization The key independent variable in this chapter is wartime mobilization. In the analyses conducted in Chapter 4, mobilization was operationalized by the total number of conscripted soldiers per population in each prefecture. At the crossnational level, unfortunately, much of these data are missing or, when available, questionable (e.g., Ellis 1993). As a next-best option, this chapter operationalized mobilization as the peak number of soldiers mobilized into the armed forces at any single point in time, a much more reliable figure in most cases. In addition to peak mobilization levels, since the civic skills learned through the process of mobilization should grow gradually over time, a measure of the duration of the country’s mobilization was also included. I thus created an index of mobilization by multiplying the percentage of citizens who served in the armed forces at its peak numbers by the number of years of mobilization. Table 5.3 shows the configuration of this index. Predictably, the main belligerents of Germany, Japan, and Britain rank high on the index, as, interestingly, do the highly mobilized neutrals such as Switzerland and Sweden. The bottom of the table consists of the less mobilized neutrals such as Ireland and the quickly vanquished Netherlands, France, and Belgium. The span from low to high mobilization here is of course much wider than that which was discussed in the intra-Japan comparison in Chapter 4. The relationship between the percent difference in the number of associational memberships between individuals in the 1911–1920 and 1921–1930 cohorts, as shown in Table 5.1, and wartime mobilization, as presented in Table 5.3, is plotted in Figure 5.1. At first glance, there does indeed appear to be a rough correlation between the two variables. Figure 5.1 shows that the difference in levels of participation between the 1911–1920 and the 1921–1930 cohorts does indeed appear to be larger among the more highly mobilized countries, such as Germany, Japan, and Switzerland, than in the less mobilized countries, such as the Netherlands, Ireland, and Belgium. But a more sophisticated quantitative analysis is required to confirm that this relationship indeed holds. This will be conducted in the next subsections. An Alternative Hypothesis: The Great Depression? First, however, it may be useful to address one other alternative hypothesis. One potential explanation stresses the effect of the Great Depression. The Great Depression may have dampened levels of participation among the pre– World War II cohorts. Thus, to the extent that a war-affected generation participates at higher levels compared to their immediately preceding cohort, this may reflect a return to pre-Depression levels of participation rather than the effects of the wartime experience per se.

The Long-Term Effects of Wartime Mobilization

87

Table 5.3. Index of Wartime Mobilization Country

Population Peak (in Strength thousands) of Armed Forces

Germany (West) Switzerland Japan United Kingdom Sweden Italy Finland United States Canada France Belgium Netherlands Ireland

Percent of Length of Population Full-Scale Mobilization (years)

Mobilization Index: Peak Strength * Duration

78,000

10,000

12.82

6

76.92

4,206 72,200 47,500

430 6,095 4,683

10.22 8.44 9.86

6 7 5

61.32 59.09 49.29

6,326 43,800 3,800 129,200

500 4,500 250 12,364

7.90 10.27 6.58 9.57

6 4 6 4

47.40 41.08 39.47 38.28

11,100 42,000 8,300 8,700 2,940

780 5,000 800 500 20

7.03 11.90 9.64 5.75 0.68

4 2 1 1 5

28.12 23.80 9.64 5.75 3.40

4.38

37.20

13-country mean

8.51

% Difference in Number of Memberships between Cohorts

Source: Ellis (1993).

Japan

160 150

(West) Germany

Sweden

140

United Kingdom France Switzerland

130

United States

120

Ireland

Canada Finland Belgium

110

Netherlands

Italy

100 0

20

40

60

80

100

120

Mobilization Index

Figure 5.1. Percentage Difference in Number of Associational Memberships between 1911–1920 and 1921–1930 Cohorts, by Level of Mobilization. Source: Table 5.3 and World Values Surveys 1981, 1990, 1995.

Civic Engagement in Postwar Japan

88

Table 5.4. Intensity of the Great Depression Country

Peak (year) (A)

Belgium

5139.50 (1928) 5171.59 (1928) 2716.82 (1929) 4710.00 (1929) 4051.41 (1929) 2971.70 (1931) 3093.24 (1929) 2025.73 (1929) 5719.60 (1928) 3937.03 (1930) 6331.73 (1929) 5503.32 (1929) 6672.53 (1929)

Canada Finland France Germany Ireland Italy Japan Netherlands Sweden Switzerland U.K. U.S. Mean Standard dev.

Trough (year) (B) 4607.45 (1932) 3370.41 (1933) 2549.76 (1932) 3959.12 (1932) 3361.68 (1932) 2800.14 (1933) 2876.67 (1931) 1837.34 (1931) 4804.91 (1934) 3666.01 (1932) 5709.98 (1932) 5138.42 (1931) 4776.92 (1933)

% Fall (100 × B/A, C)

Duration of Depression

10.35

1929–39

10

103.50

34.84

1929–39

10

348.40

6.15

1930–34

4

24.60

13.25

1930–38

8

17.80

1929–34

5

89.00

5.79

1932–34

2

11.58

6.98

1930–34

4

27.92

9.33

1930–32

2

18.66

16.00

1929–39

10

6.88

1931–33

2

13.76

9.82

1930–37

7

68.74

6.63

1930–33

3

19.89

30.76

1930–39

9

276.84

13.43 (9.424)

Length of Duration (years; D)

% Fall × Duration (C × D)

106.00

160.00

5.85 97.61 (3.262) (106.777)

Source: Maddison (2003).

Thus, the study also tested whether the Great Depression may have systematically undercut participation among the 1911–1920 cohorts. To this end, a measure for the severity of the Depression was created by taking the percentage fall in per capita income between the pre-Depression peak and the Depression-era trough, multiplied by the duration of the Depression in that country. Duration was operationalized as the number of years that the country required to recover to its pre-Depression peak income per capita levels. The configuration of this index is shown in Table 5.4.

The Long-Term Effects of Wartime Mobilization

89

The remainder of this section tests the hypothesis presented in this volume in three stages. First, it tests the effects of mobilization. Second, it tests the validity of the alternative hypothesis that victory and/or defeat may have crucially shaped the rise in levels of civic engagement between the World War II–affected cohort and the pre–World War II generational cohorts. Third, it tests a second alternative hypothesis that the relatively high levels of participation among individuals who came of age around the time of World War II compared to their preceding cohort may reflect not so much the effects of World War II as the dampening effect of the Great Depression on the participation of the pre– World War II cohort. Analysis (1): Cross-National Analysis If the framework presented in this book is valid, individuals in the war-affected cohorts should, ceteris paribus, exhibit higher rates of participation than those in the immediately preceding cohorts. This relationship should be more pronounced in countries that underwent higher levels of wartime mobilization. Ceteris paribus, if the hypothesis presented in this volume is valid, individuals who were born between 1921 and 1930 and subsequently experienced high levels of mobilization should exhibit higher levels of civic engagement than individuals born between 1911 and 1920 or in countries that were not highly mobilized. Thus, the main variable of interest in the analysis presented in this chapter is a multiplicative interaction term between the 1921–1930 cohort and wartime mobilization. The model uses ordinary least squares regressions using robust clustered standard errors to account for the correlations among observations within each country.2 Several control variables were included: education, income, gender, age, and survey year. The control variables were chosen on the basis of existing studies of voluntary participation. The results are shown under Model 1 of Table 5.5. As shown in Table 5.5 (Model 1), the coefficient on the mobilization × 1921–1930 cohort variable is positive and significant at the p ≤ .01 level, as expected on the basis of the framework advanced in this book. This suggests that, ceteris paribus, when individuals in the 1921–1930 cohort experienced higher levels of mobilization, this strongly served to elevate levels of participation thereafter. In Model 2, the study tested the victory/defeat hypothesis. The study replaced the variables relating to mobilization with variables relating to victory/defeat. Two dummy variables were created to measure victory (Britain, Canada, United States) and defeat (Finland, Germany, Italy, Japan). An alternative operationalization was also tested and will be discussed later. Since the victory/defeat variables may also be expected to interact with cohort variables, 2

The study also tested the model using a multilevel approach (Steenbergen and Jones 2002) and found the results to be largely similar. Results are available on request.

Civic Engagement in Postwar Japan

90

Table 5.5. The Determinants of Associational Membership

Education Income Gender Age Survey Year (1990) Survey Year (1995) Mobilization 1911–1920 Cohort 1921–1930 Cohort 1921–1930 Cohort*Mobilization Victory

Model 1 (Mobilization)

Model 2 (Victory/Defeat)

.08 (.017)*** .07 (.028)** −.19 (.059)*** .01 (.001)*** .52 (.177)** 1.89 (.253)*** .01 (.003)** −.12 (.058)* −.15 (.027)*** .003 (.000)***

.07 (.017)*** .05 (.031) −.20 (.055)*** .01 (.001)*** .53 (.176)** 2.05 (.217)***

.08 (.003)*** .08 (.003)* −.18 (.017)*** .01 (.001)*** .61 (.019)*** 1.90 (.032)***

−.10 (.051)* −.09 (.038)**

−.13 (.049)*** −.04 (.030)

.11 (.246) −.14 (.301) .05 (.046) .04 (.069)

Defeat Victory*1921–1930 Cohort Defeat*1921–1930 Cohort Depression

−.04 (.130) −.01 (.026)

Depression*1911– 1920 Cohort Log-Likelihood N Number of Countries

Model 3 (Great Depression)

−39045.482 26831 13

−40776.903 26831 13

Standard errors in parentheses. ***p ≤ .01, **p ≤ .05, *p ≤ .1.

−40798.652 24418 13

The Long-Term Effects of Wartime Mobilization

91

two interaction variables were created (victory × 1921–1930 cohort, defeat × 1921–1930 cohort). The results of this regression are shown under Model 2 in Table 5.5.3 As shown in Table 5.5, the coefficient for neither the victory × 1921–1930 cohort variable nor for the defeat × 1921–1930 cohort variable is statistically significant. Furthermore, the results in Model 2 do not change substantially when the configuration of the victory variable is changed to include Belgium, France, and the Netherlands, which were initially defeated by Germany but then were liberated; in fact, in that case both the victory × 1921–1930 cohort variable and the defeat × 1921–1930 cohort variable are not only statistically insignificant but both also point in the opposite direction from what would be expected on the basis of its logic. Models including dummy variables for the three countries in the dataset that were non-democracies prior to World War II, namely, Germany, Italy, and Japan, were also tested, but the inclusion of this variable also did not change the results of the analysis in substantial ways. Model 3 tested the effect of the Great Depression. As shown in Table 5.5, the coefficient for the Depression × 1911–1920 cohort variable is not statistically significant. The results show that, ceteris paribus, individuals in the 1911– 1920 cohort in countries that experienced more severe and longer Depressions did not systematically participate at lower rates compared to individuals of later generational cohorts. This suggests that the higher levels of participation among individuals of the 1921–1930 cohort compared to those of the 1911– 1920 cohort were not simply an artifact of the Depression-induced decline in levels of participation among the latter cohort. While this finding lends support to the analysis provided earlier in this chapter, given the strong relationship that studies of participation typically find between income and levels of participation, it is also somewhat surprising that the Depression does not appear to have left enduring marks on the levels of participation among individuals of the affected generational cohort. It is possible to speculate on why this might be. For instance, the adverse effects of the Depression on participation may have been overridden by the subsequent effects of mobilization. Or perhaps the different policies that different states adopted to deal with the Depression may have affected civic engagement in ways not correlated with economic performance.4 Finally, if the data were available, it would be interesting to compare the participation rates of the 1911–1920 cohort against its immediately preceding cohort, which experienced World War I as a formative experience. But this is beyond the scope of the present study. Suffice it to note here that on the basis of data that are

3

4

Analyses were also conducted using simply the variables of victory and/or defeat without the interaction terms and found the effects were not statistically significant. Many studies of social capital point to income equality as a strong determinant of social capital (e.g., Boix and Posner 1998; Uslaner 2000). If this is the case, then countries that adopted more social democratic policies during the Great Depression may have been able to mitigate the damaging effect of reduced incomes on participation among the affected generational cohort.

Civic Engagement in Postwar Japan

92

Predicted Number of Memberships

1.8

1.6 1921-30 cohort 1.4 1911-20 cohort

1.2

1

0.8 10

20

30

40

50

60

70

80

Mobilization Index

Figure 5.2. Predicted Number of Associational Memberships by Generational Cohort, Based on Cross-National Analysis. Source: Table 5.5, Model 1. Predicted values were generated using CLARIFY (Tomz et al. 2001).

available in the World Values Surveys, individuals who came of age in countries where the Great Depression was more severe do not appear to participate at systematically lower rates than their counterparts in countries where the Great Depression was less. Substantive Effects To illustrate the predicted effects of mobilization more concretely, I used the results under Model 1 in Table 5.5 to simulate, using the software CLARIFY (Tomz et al. 2001; King et al. 2000), the predicted difference in the number of associational memberships between individuals in the 1911–1920 and the 1921–1930 cohorts at different levels of mobilization, holding all other control variables constant at their means. One thousand simulations were conducted. The results are shown in Figure 5.2. Figure 5.2 shows that the effect of mobilization is indeed considerable. At the low end of mobilization, where the index of mobilization = 10, levels of participation between the individuals in the 1911–1920 cohort and the 1921– 1930 cohort are virtually identical; the former is predicted to participate in 1.01 associations, on average (standard error = .035), while the latter is predicted to participate in 1.00 associations, on average (standard error = .039), holding all other variables at their means. During World War II, countries such as Ireland, the Netherlands, and Belgium exhibited levels of mobilization of 10 or less. At the high end of the spectrum, where the index of mobilization = 80 (i.e., around Germany’s level), individuals in the 1911–1920 cohort are

The Long-Term Effects of Wartime Mobilization

93

predicted to be members of 1.51 associations, on average (standard error = .035), while individuals in the 1921–1930 cohort are predicted to participate in an average of 1.68 associations, on average (standard error = .049), holding all other variables at their means. That is a difference of more than 10 percent in levels of participation, still persisting decades after the war ended.

Conclusions This chapter has made several claims. First, it showed that war-affected cohorts in the more highly mobilized countries do indeed appear to participate more than their predecessors, even long after the end of war. In contrast, for the less mobilized countries, the difference in levels of participation between the generational cohort that came of age around World War II and the immediately preceding cohort is considerably smaller. Unfortunately, due to data constraints, this chapter could not test for the effects of path-dependency, so the results presented here should only be viewed as a partial test of the theoretical framework offered in this book. However, the initial results do provide encouraging support for the broader applicability of the perspective offered in this volume beyond simply the Japanese context. Second, the chapter also did not find support for the victory/defeat hypothesis. In the more highly mobilized countries, individuals in the war-affected cohorts appear to participate at higher rates than those in their immediately preceding cohorts, regardless of whether the country was ultimately victorious or vanquished. Finally, the analysis presented in this chapter also found some initial evidence that the systematically higher rates of participation among individuals who came of age around the time of World War II did not simply reflect lower rates of participation among the preceding cohort due to the Great Depression.5 To date, studies of the “long civic generation” have overwhelmingly focused on a single-country case, the United States. However, if the analysis presented in this book is valid, there are “long civic generations” also present outside of the United States. Indeed, among the belligerents of World War II, the United States, to which the “long civic generation” label is most often given, only falls around the middle of the spectrum in terms of the magnitude of the war-induced jump in civic engagement. The more highly mobilized countries such as Germany,

5

It is important to note that despite the long period between World War II and the World Values Survey, the findings presented in this chapter are not simply based on a small sample size. The sample size of a typical World Values Survey in one country is usually between 1,000 and 2,000 respondents. To separate out the effects of age from the effects of cohort, the analyses in this chapter drew on two or three waves of the survey, which multiplies the sample size. Thus, for instance, for Belgium, the overall sample size is 3,937 and there are 449 individuals who are born between 1921 and 1935 and were also aged between 60 and 69. Of these, 203 individuals belonged to one or more association. Similarly for Japan, the corresponding figures are 3,269, 230, and 89, respectively.

94

Civic Engagement in Postwar Japan

Japan, and the U.K. show much larger jumps in levels of participation between the pre–World War II and the World War II cohorts. The variations in the war-induced growth in civic engagement that were demonstrated in this chapter may have important implications for understanding the “bowling alone” phenomenon as well. Scholars of many Western European countries have disputed the claim that the postwar cohorts participate to a much lesser extent than the war-affected cohort. The findings presented in this chapter may help to account for both the recent decline in civic engagement in the United States and the lack thereof in some Western European countries. For instance, the insights of this study are entirely consistent with both Putnam’s findings and Hooghe’s (2003) recent work that shows that civic engagement does not appear to be on the decline in Belgium. Since World War II mobilization in Belgium was quite moderate, the warinduced “mobilization premium” for civic engagement was also quite limited. Assuming that levels of participation for the postwar cohorts returned to levels comparable to pre–World War II cohorts, there should indeed be a much smaller decline in Belgium compared to, for instance, the United States. A similar case may be made with respect to the Netherlands, where a decline has also not been found (de Hart and Dekker 1999). This does not, however, imply that the postwar cohorts in the more highly mobilized countries should all “bowl alone,” and certainly not to the same degree. The extent to which individuals of the generational cohorts that follow the “long civic generation” participate in voluntary activities should have been affected by the unique political and social contexts in which individuals of those cohorts came of age. For instance, the introduction of appropriate policies could have helped to sustain the war-induced jumps in levels of civic engagement among subsequent generational cohorts. Peter Hall notes how new educational policies that were introduced in the postwar period may have counteracted the potential fall in civic engagement in Britain (Hall 1999). A fuller assessment of the factors that shape the participatory behavior of individuals in younger cohorts needs to examine the unique formative experiences of the postwar cohorts of each country.

6 Repression and Revival of the YMCA Japan

Overview Chapter 4 used quantitative analysis to demonstrate that the combination of path-dependency effects from the prewar period and wartime mobilization crucially shaped the rise of civic engagement in Japan in the initial postwar decade. Participation in youth and religious organizations grew more rapidly in areas where levels of participation had been high prior to the war and that were highly mobilized during wartime. Conversely, areas that had seen low levels of participation during the prewar period and/or were less mobilized during the war saw lower rates of postwar membership growth. Chapter 5 then showed that mobilization had important effects on postwar levels of civic engagement in thirteen countries in Asia, Europe, and North America. The quantitative analyses, however, did not test the precise mechanisms through which mobilization and historical legacies would affect civic engagement in the postwar period. A separate investigation is required to confirm the causal relationship between prewar, wartime, and postwar levels of participation at the micro-level. The present chapter and the next seek to further test the validity of the hypothesis presented in this project through historical processtracing. Chapter 3 argued that in conjunction with mobilization, prewar levels of voluntary participation leave considerable historical legacies, which survive despite the disruption caused by war. The legacies of prewar associational activities, in particular, are hypothesized to operate via two distinct pathways. First, ceteris paribus, where there are legacies of groups to be revived, the costs of organization should be lower than when entrepreneurs need to establish a group from scratch. And, second, where there are preexisting channels via which information about these revived associations can easily diffuse, the easier it should be for citizens to learn about and to join them.

95

96

Civic Engagement in Postwar Japan

If the theoretical framework advanced in Chapter 3 is valid, in addition to the overall outcomes in terms of postwar participation rates, we should also observe the following patterns in the process of decline and recovery: ●

●

●

First, associations in places that had seen higher levels of civic activities before the war should show much greater resilience against the adverse social and political conditions brought on by war, compared to associations in places that had been less well-endowed. In areas that had not had high levels of civic activities in the first place, voluntary associations may suspend their activities once increasing numbers of members begin to be drafted, the city suffers physical damage, and/or when government officials begin to eye the associations’ activities with suspicion. In areas where associational activities had flourished, however, groups may be able to hold out for longer periods of time. The sunk costs are considerable, and the larger number of members means that groups are likely to be better integrated into the rest of society, which may in turn help them to draw on various resources, both material and social, that would help them to survive. Even when the groups’ activities are suspended, the suspension is likely to come later during the war. Second, once the war is over, the revival of associations should begin sooner in places that had enjoyed high levels of voluntary participation before the war, as compared to those that had not. Ceteris paribus, it is generally much less costly to revive an organization than to start one from scratch, and this is especially likely to be true if the period of interruption in associational activities was relatively short, as is likely to be the case in areas where associational activities had been vibrant before the war. This, combined with high demand from potential participants, should encourage leaders to seek to resume their group’s activities relatively quickly. Third, in areas where levels of voluntary activities had been high prior to war, associations may see a rapid rise in memberships in the wake of war, regardless of whether that group launched extensive recruitment drives. This is not to say that recruitment drives should not be effective in these societies or regions. However, if prior levels of participation are high, levels of memberships are likely to increase rapidly because citizens will seek out the associations at least as energetically as the associations are seeking participants, particularly when wartime mobilization was extensive.

This chapter and the next illustrate how these predictions are borne out empirically, using comparative case studies. In particular, the two chapters rely on structured, focused comparisons (George and Bennett 2005). The present chapter compares the revival of the YMCA in the cities of Kobe and Sapporo. Chapter 7 then compares the revival of judo in the cities of Fukuoka and Yokohama. The present chapter is structured as follows. The next section explains the case selection of the YMCA and judo to test the hypothesized mechanisms of

Repression and Revival of the YMCA Japan

97

the theory offered in this book. The third section traces the rapid rise in membership in the Kobe YMCA in the immediate postwar period. The fourth section contrasts this rise with the relatively weak growth of participation in the Sapporo YMCA. The fifth section summarizes and concludes.

The YMCA and Judo This study chose to focus on two types of groups: the YMCA and judo. For the purposes of this study, these offer particularly valuable cases. First, the YMCA and judo represent two of the largest voluntary activities in Japan. This means that they are important in their own right. It also means that both are likely to have reflected broad societal readiness to participate and even to have affected it by their own success or failures. In fact, limited surviving data compiled by the Ministry of Education on prewar youth groups suggest that the evolution of prewar YMCA membership and judo closely parallels broader national trends; in both cases, membership growth was slow but steady until the 1910s but accelerated rapidly during the 1920s (Monbusho Futsu Gakumu-kyoku 1924; Monbusho Shakai Kyoiku-kyoku 1937a). If one were instead to choose smaller associations, the resulting study may not be so representative of broader societal patterns. Because of their relatively large size, there is also a wealth of secondary literature on the histories of both groups.1 Second, both the YMCA and judo were well established throughout most of the populous prefectures of Japan during the prewar period. The number of voluntary associations that had chapters in many parts of Japan during the prewar period is not very large. The YMCA already had 13 chapters by 1938, which makes it possible to compare the trajectory of different YMCAs across different Japanese regions, and especially to compare those where levels of civic engagement have historically been high with those where they have been low. Similarly, Kodokan judo’s Yudanshakai association had 109 branches in 1936 across all forty-seven prefectures except for Tottori (Kodokan Bunkakai 1936: 84–91). Third, the YMCA and judo offer very useful differences. The first YMCA was founded in Britain in 1844 and expanded to Japan during the late nineteenth

1

Unlike most groups that have been in existence since the prewar period, the YMCA Japan has been prolific in publishing various histories of the association, both at the national and local levels. The Tokyo YMCA, for instance, has published the 100-year History of Tokyo YMCA; the Osaka YMCA has produced the 100-year History of the Osaka YMCA; the Hokkaido YMCA, the 100-Year History of Hokkaido YMCA; the Kobe YMCA, the 70-Year History of Kobe YMCA and the 100-year History of Kobe YMCA; the Nagoya YMCA, the 60-year History of YMCA Nagoya; and so forth (Yoshimura 1964; Kobe YMCA 1969; Saito 1980; Osaka YMCA 1982; Kobe YMCA Hyakunenshi Hensanshitsu 1987; Hokkaido YMCA 1997). There is also a wealth of secondary literature as well (e.g., Yoshikawa 1955; Nara 1958: Kimoto 1966; Davidann 1998). Similarly, the judo world has also published a large number of works, by judo insiders, journalists, and academics (e.g., Oimatsu 1966; Maruyama 1967; Yoshida 1969; Kudo 1972; Matsubara 2006; Todo 2007).

98

Civic Engagement in Postwar Japan

century. Judo, on the other hand, is much more indigenously rooted. The two groups’ experiences during wartime and the occupation period were also quite different. The YMCA was repressed by the Japanese government during the war for its Western origins but then was subsequently viewed favorably by the occupation for the same reason. By contrast, judo was favored by the militarist regime during World War II but then was viewed by the occupation forces as having propagated militaristic values during the war and was actively discouraged. While these different experiences between the 1930s and the 1950s might have been expected to affect postwar recruitment, in fact, by the early 1950s it was clear that they had not. As shown in Chapter 2, postwar membership grew in both the YMCA and judo. The growth was more rapid for the YMCA, at least partly since SCAP banned judo from schools between 1945 and 1950. Nevertheless, judo did continue to grow at a steady rate. Yet these national increases conceal important regional variations. The large variance in the growth rates in membership figures between different regions reflects regional differences rather than being a commentary on the overall national climate for the resurgence of the groups. Despite the difference in the Western origin of the YMCA and the much more domestically rooted judo, legacies from the prewar period, in conjunction with wartime mobilization, shaped the growth of participation in the immediate postwar period in very similar ways in the two cases. The rest of this chapter covers the case of the YMCA, and the following chapter examines the revival of judo.

Case Selection: The YMCA in Kobe and Sapporo This chapter compares the growth of membership in the YMCAs in the two cities of Kobe (Hyogo Prefecture) and Sapporo (Hokkaido Prefecture). Aside from varying quite substantially in levels of prewar YMCA membership between the two cities, case selection rested on three considerations. First, both Kobe and Sapporo represent relatively “new” cities that had fairly small populations prior to the Meiji Restoration (1868) but which grew rapidly thereafter. This sets the two cities apart from many of the other major cities in Japan, such as Tokyo, Kyoto, and Osaka, which had already developed into major urban centers by the time of the Meiji Restoration. The YMCAs in the two cities were also founded roughly around the same time: 1899 in Kobe and 1897 in Sapporo. This enables us to ignore, for analytical purposes, the effects of preexisting legacies from the pre-Meiji period and to focus primarily on the era following the Meiji Restoration. Second, the two cities were also similar in terms of socioeconomic indicators. Precise figures for the income levels for the two cities are not available for the prewar period, although it is possible to gain a glimpse into their economic development through labor statistics that are available in census figures. As of 1920, for instance, roughly one-third of the population was engaged in manufacturing in both cities (Naikaku Tokei-kyoku 1925: 123; 1927: ix), and this

Repression and Revival of the YMCA Japan

99

was still the case in the 1930 census as well (Naikaku Tokei-kyoku 1933: 15; 1934: 66). In terms of educational attainment, they are comparable as well; while Sapporo did slightly better in the early prewar period, with 99.44 percent of school-age children attending elementary school compared to 93.6 percent in Kobe, by 1930 they were more similar, with 99.61 percent in Sapporo as opposed to 96 percent in Kobe (Kobe-shi Kyoikushi Henshuu Iinkai 1966; Sapporo-shi Kyoiku Iinkai 2000: 546–547). Third, the two cities of Kobe and Sapporo were mobilized to similar extents during the war. In Kobe’s Hyogo Prefecture, roughly 146,000 men were conscripted between 1937 and 1945, or approximately 4.53 percent of the population (Oe 1988). A total of 158,240 men were drafted from Sapporo’s Hokkaido Prefecture between 1937 and 1945, or roughly 4.84 percent of the population (Oe 1988). This similarity in levels of mobilization allows us to control for the effects of mobilization and to focus on the effects of the legacies from prewar associational activities. Fourth, the comparison between Kobe and Sapporo is especially compelling because while Sapporo, a city with low levels of participation in the YMCA during the prewar period, suffered relatively little physical damage during the war, destruction in Kobe, a high-participation city in the prewar period, was extensive. Hyogo Prefecture (Kobe) suffered 3.49 deaths per 100,000 people, the fourth-highest number of deaths among all Japanese prefectures, following only Hiroshima, Nagasaki, and Tokyo (calculated from Keizai Antei Honbu 1949: 279–281). Hokkaido (Sapporo) was thirty-sixth out of forty-seven prefectures, with just .26 deaths per 100,000 people (Keizai Antei Honbu 1949: 279–281). In terms of property damage, 63.99 buildings per every 100,000 buildings were destroyed in Hyogo Prefecture, the third-highest level of damage in the country following Tokyo and Osaka; Hokkaido again ranked thirty-sixth, with 1.31 out of 100,000 (Keizai Antei Honbu 1949: 344–345). According to the victory/defeat hypothesis, then, conditions should have been favorable for a rapid growth in membership in Sapporo, while in Kobe recovery should have struggled. But, in fact, postwar membership growth in the YMCA was high in Kobe, while it stagnated in Sapporo. This is shown in Table 6.1. Table 6.1 compares the levels of YMCA membership per one million population in the prewar period against the postwar period. Because city-level membership figures are available only sporadically from the prewar period, prewar membership is calculated as the mean of membership levels in 1916, 1928, and 1932, the years for which figures are available, while postwar membership is calculated as the mean of the membership levels as of 1953, 1954, and 1956. The mean growth was 111.83 percent (standard deviation: 143.44), meaning that, on average, each YMCA roughly doubled its membership over prewar levels. Growth rates in four cities (Osaka, Nagoya, Kyoto, and Kobe) were above the mean, while in six (Yokohama, Tokyo, Fukuoka, Sapporo, Nagasaki, and Sendai) they were below the mean. Among the latter six, the last three cities actually declined in memberships per population between the prewar and

Civic Engagement in Postwar Japan

100

Table 6.1. The Growth of YMCA Membership, Postwar Period (vis-à-vis Prewar Period) City (Prefecture)

Number of Members per Million Population, Mean (1916, 1928, 1932)

Number of Members per Million Population, Mean (1953, 1954, 1956)

Membership Growth between Prewar and Postwar Means (%)

Osaka (Osaka) Nagoya (Aichi) Kyoto (Kyoto) Kobe (Hyogo) Yokohama (Kanagawa) Tokyo (Tokyo) Fukuoka (Fukuoka) Sapporo (Hokkaido) Nagasaki (Nagasaki) Sendai (Miyagi) 10-City Mean

499.78 618.42 905.74 951.81 1034.15

2763.30 1777.90 2554.36 2405.11 1966.45

452.90 187.49 182.02 152.69 90.15

471.77 341.05

744.62 466.64

57.84 36.83

371.69

354.73

−4.56

3136.87

2557.14

−18.48

1301.06 963.23

1059.84 1665.01

−18.54 111.83

Note: Average growth in membership for the years 1953, 1954, and 1956 in terms of average prewar membership levels of 1916, 1928, 1932. Figures are in numbers of members per million population.

postwar periods. Overall, however, Table 6.1 shows that the YMCAs that had relatively high levels of memberships in the prewar period, such as Yokohama, Kobe, and Kyoto, also grew rapidly in the postwar period, while those that had relatively small memberships, such as Fukuoka and Sapporo, grew at a much slower rate. Figure 6.1 shows the substantial divergence over time in the levels of YMCA memberships per million population in Kobe and Sapporo. While many dates are unfortunately missing, the general trend is clear: The difference in memberships in the two cities was already quite substantial prior to the war, but it became even more pronounced in the wake of the war. In 1928, the difference in memberships per population was just under two to one, at 611.97 YMCA members per one million people in Kobe versus 367.80 in Sapporo. In 1954, however, membership levels in Kobe outstripped those in Sapporo by about ten to one, at 3000.8 members per one million people in Kobe versus 304.7 in Sapporo. As noted in Chapter 2, then, the war did not close the gaps between high-participation and low-participation cities but rather exacerbated them.

Number of Members (per 1 Million Pop.)

Repression and Revival of the YMCA Japan

101

3000 YMCA Kobe YMCA Sapporo

2500 2000 1500 1000 500 0 1915

1925

1935

1945

1955

Year

Figure 6.1. Differences in Membership per Population, Kobe and Sapporo YMCAs, 1916–1956. Source: YMCA Japan Annual Reports (various years). Figures denote memberships per one million people.

This pattern of high relative levels of prewar memberships leading to higher relative levels of postwar memberships, and vice versa, is consistent with the pattern expected by the theory introduced in Chapter 3 and confirmed in the statistical analysis in Chapter 4. The rest of this chapter seeks to gain some insights into the mechanisms through which this path-dependency operated, by comparing the evolution of civic engagement in the YMCAs of Kobe and Sapporo. It pays special attention to how the different levels of prewar voluntary associational activities left very different legacies in the two cities, thereby promoting the rise of civic engagement to a much greater extent in Kobe compared to Sapporo.

The Recovery of the YMCA in Postwar Kobe Background: History of the City of Kobe Suma, a town located in what is now Western Kobe, appears in Lady Murasaki’s The Tale of Genji, written in the early eleventh century, and the area flourished as a retreat for aristocrats from Kyoto.2 In 1180, the area very briefly served as the capital of Japan (“Fukuharakyo”). More lastingly, due to its relative proximity to Kyoto, which was the capital of Japan until 1868, the Kobe area also served as the center of trade routes to China as trade expanded between the eleventh and sixteenth centuries. But the Tokugawa Shogunate’s policy of 2

Okahisa (1978) and Kusumoto (2007) provide a detailed account of the history of Kobe prior to the Meiji Restoration.

102

Civic Engagement in Postwar Japan

closing off Japan to the outside world after the early seventeenth century led to the rapid decline of Kobe during the Edo Period (1603–1868). In contrast to many of Japan’s other major cities, such as Tokyo and Osaka, Kobe did not become a major urban center again until the latter half of the nineteenth century. Around the time of the Meiji Restoration, it was still a small fishing port with a population of only around 20,000, which is miniscule compared to Tokyo, which had already surpassed 3 million people by that time. The opening of Kobe in 1868 as one of the first trading ports for Western vessels, however, led to dramatic growth in the late nineteenth century.3 The population grew to 130,000 in 1889 and passed the one million mark in 1941 (Kobe-shi n.d.). The United States was the first country to establish a consulate in Kobe, arriving in 1867. By 1878, twelve other countries, including Britain, France, Germany, the Netherlands, Belgium, Spain, and Austria-Hungary, also had established consulates in Kobe;4 by 1930, thirty countries had consulates (Kusumoto 2007: 68–70, 99). Initially, it was stipulated that these foreigners were to reside exclusively in the foreign settlement, although with construction delays in the foreign settlement area, the diplomatic missions of Britain, France, and the Netherlands lobbied for – and won – the right to reside in specified areas outside the foreign settlement as well (Kamiki 1993: 41–42). In 1877, the first railroad service between Kobe and Kyoto began operations, vastly improving the transportation infrastructure in the Kansai area (Kusumoto 2007: 71). By the end of World War I, Kobe had become the largest trading port in Japan (Okahisa 1978: 261). Trading in Kobe further expanded when the Great Kanto Earthquake (1923) destroyed Yokohama, Japan’s other preeminent trading port.5 Figures compiled by Shigenobu Ochiai indicate that Kobe handled about one-third of all exports from Japan and about one-half of imports into Japan for much of the prewar period (Ochiai 1975: 186). The expansion of trade also encouraged the growth of other industries in Kobe, such as shipbuilding, textiles, and steel production. Ochiai notes that most of the indigenous merchants in the Kobe area were initially skeptical of trading with the newly arrived foreigners in Kobe. Instead, ambitious and enterprising minds from all over Japan flooded into Kobe to take advantage of the new commercial opportunities (Ochiai 1975: 151). The city of Kobe was also progressive in terms of democratic rule. A democratically elected local assembly was founded as early as 1873, long before the founding of the national Diet, in 1891. Representatives were elected from districts determined by population size, and property-owning males over the age of sixteen were eligible to vote. It was also during the Meiji period that Kobe began to develop a reputation for its high-quality beef. 3

4

5

Initially, the port of Sakai in southern Osaka had been selected to become available to Western vessels. Kobe was chosen in the end because the Tokugawa Shogunate felt that Sakai would provide foreigners with too easy an access to Nara, the old capital of Japan (Ochiai 1975: 134). According to Kusumoto (2007: 73–74), the British Consulate also initially conducted affairs for the French, Italian, Spanish, and Austro-Hungarian Consulates. Imports exceeded exports in the port or Kobe until the early 1930s (Okahisa 1978: 261–262).

Repression and Revival of the YMCA Japan

103

With the opening of the port and the foreign settlement came a large influx of Westerners, giving Kobe a distinctly cosmopolitan flavor. The number of Westerners rose from around 200 in 1878 to 2,000 in the early 1890s (Ochiai 1975: 171–172). The majority of them were British, American, or German, with the British representing the largest contingent. These Westerners opened social clubs, theaters, and athletic facilities in Kobe, and many Western cultural activities became available for the Japanese to observe for the first time (Kusumoto 2007: ch. 3). The Kobe Regatta and Athletic Club, for instance, which was founded in 1870, introduced soccer, rugby, and cricket to the Kansai area (Kobe Gaikokujin Kyoryuchi Kenkyukai 2005: 142). The Western expatriates also included a large group of Western missionaries who were eager to proselytize this newly opened nation. The missionaries initially engaged covertly in their activities, since Christianity was still outlawed during the early years of the Meiji period. However, their activities expanded rapidly following the legalization of Christianity in 1873. The Christian contribution to Kobe was not merely spiritual; missionaries eagerly founded hospitals as well as private schools, especially for girls, including the prestigious Kobe College (Kobe Jogakuin), founded by Americans Eliza Talcott and Julia Dudley; the Palmore School, founded by American W. A. Palmore; and the Shoei School, founded by American A. L. Howe (Kobe-shi Kyoikushi Henshu Iinkai 1966: 247–249). The YMCA was another legacy of the Christian missionaries in Kobe. There is some debate over exactly when the Kobe YMCA was founded.6 Initially, it appears to have been established in 1886 by a group of Western missionaries, but this fledgling Y soon dissipated. The missionaries appear to have made little distinction between the YMCA and the church activities around this time, with the result that the YMCA did not take on a life of its own. In 1899, however, Christian-influenced business leaders in the Kobe area started a new YMCA that was more open to the general public. If we take 1899 as the genuine founding date, the Kobe YMCA is Japan’s fifth-oldest YMCA, after Tokyo (founded in 1880), Osaka (1882), Yokohama (1884), and Sapporo (1897). The Kobe YMCA’s evening English language school was founded in 1902, and in its early years the YMCA appears to have been better known as an English language school than as a religious organization. However, it soon expanded the scope of its activities to include bible-reading, arts, and recreation, as well as broader social services such as disaster relief. From early on, Kobe’s YMCA movement was comprised not only of Christian religious leaders but also intellectuals, successful businessmen, and college students. Kichitaro Muramatsu, for instance, was the owner of a large trading company when he became involved in the Kobe YMCA. He had been warmly welcomed into the YMCA in New York while studying business there, and as a result he became eager to promote its sister association in Kobe (Kobe 6

See Kobe YMCA Hyakunenshi Hensan Shitsu (1981a, 1981b), Kobe YMCA no Rekishi wo Kataru: Zadankai-hen, 1–2 [Roundtable Discussion on the History of the Kobe YMCA], pp. 1–31.

104

Civic Engagement in Postwar Japan

YMCA 1969). Suke Harada, also one of the founding leaders of the Kobe YMCA, later became president of the well-known Doshisha School in Kyoto. The fact that the Kobe YMCA’s networks extended well beyond Kobe into the other major cities of Osaka and Kyoto, especially at a time when the OsakaKobe-Kyoto metropolis was still much larger and vibrant than the new capital of Tokyo, was particularly important in channeling resources toward the fledgling Kobe YMCA. Surviving records show that these business leaders donated substantial sums to the association. Taisho Democracy in Kobe, 1905–1932: The Flourishing of Civic Life Associational activities flourished in Kobe during Taisho Democracy (1905– 1932), the period of brief liberal rule in Japan before militarism set in. Around the turn of the century, Kobe experienced a wave of new factory construction, from shipping to textiles and, later on, steel. This rapid industrialization, combined with the relatively liberal political atmosphere, spawned a large labor movement in Kobe. The first labor union in Kobe was founded in 1880 by a group of sailors (Ochiai 1975: 240), but with the heavy industrialization of Kobe, the labor movement expanded rapidly. Labor protests surged in the wake of the Russo-Japanese War (1904–1905), led by workers who were opposed to the terms of the Portsmouth Treaty. There were forty labor disputes in Kobe in 1907 alone, up from thirteen for the entire 1890s (Kobeshi 1965: 328). Labor activism intensified during Taisho Democracy under the leadership of Toyohiko Kagawa, a Princeton-educated Christian, labor leader, philanthropist, consumer advocate, and social reformer, and the Kobe labor movement wielded widespread influence over unions across the country (Kobe-shi 1965: 332). In 1921, Kagawa led a forty-five-day strike in Kobe that mobilized 40,000 workers, demanding the right to collective bargaining.7 The Kobe YMCA also thrived in this liberal climate. The Kobe YMCA Hall, completed in 1913, allowed for the expansion of the YMCA’s activities, and the YMCA quickly became an important intellectual, cultural, and athletic center of Kobe. It housed Japan’s first indoor gymnasium, where Japan’s first basketball and volleyball games were played (Ochiai and Arii 1967: 122–123). During this period, enrollment at the Y’s evening classes grew by roughly 20 percent every year (Kobe YMCA Hyakunenshi Hensanshitsu 1987: 163). Its lecture halls provided the facilities for numerous lectures by leaders of labor and agricultural movements as well as by writers and artists (Ochiai and Arii 1967: 154). While the Kobe YMCA itself is generally not considered to have been particularly active in politics itself, it nevertheless played a key role in encouraging and fostering these movements, often providing important physical space for movement leaders to deliver lectures and to hold rallies. This was an especially important function since, at the time that the YMCA building was completed, in 1913, very few other facilities in the area could accommodate 7

However, this movement was repressed by the military and ended in failure.

Repression and Revival of the YMCA Japan

105

crowds the size of more than 1,000 people; the YMCA Hall could hold up to 1,200 (Kobe YMCA Hyakunenshi Hensanshitsu 1981a: 20). Toyohiko Kagawa was a frequent speaker at the Kobe YMCA and is said to have donated substantial sums to it (Kobe YMCA Hyakunenshi Hensanshitsu 1981b: 7). In sum, prewar Kobe was richly endowed with associational activity, and the YMCA was a major hub of such activity. Kobe during World War II Mobilization for war in Kobe was extensive. According to one estimate, roughly 146,000 males from Kobe and the surrounding Hyogo Prefecture were conscripted into military service between 1937 and 1945 (Oe 1988). This represents roughly 4.5 percent of the 1940 population overall and 8.8 percent of the male population as of 1940. Reflecting Kobe’s status as a major trading port as well as steel producer, industrial mobilization also greatly intensified during the war. Mobilization was also extensive at the community level. As noted in Chapter 3, Japanese society was mobilized into World War II primarily through “administered mass organizations” rather than voluntary associations, and this was true in Kobe as well (Kasza 1995). By the time the war broke out in 1931, a considerable number of these associations had already been established in Kobe, and, as with the rest of Japan, mobilization into neighborhood associations accelerated in Kobe during World War II. As of 1942, there were an estimated 1,418 neighborhood associations in the City of Kobe, with an average of 159 households per association (Kobe-shi 1965: 6). The number of households per association ranged from 9 to 345 (Kobe-shi 1965), indicating the considerable variance in the sizes of the associations. Neighborhood associations were further organized into smaller units, called tonari-gumi, each of which had an average of 8 households within the City of Kobe; the number of households in the tonari-gumi ranged from 1 to 22 (Kobe-shi 1965: 7). These associations engaged in a broad range of war-related tasks, from savings drives, recycling drives, food rationing, assisting widows and/or the elderly, to other administrative tasks. As the threat of air raids on mainland Japan became imminent, the neighborhood associations also organized fire drills. As the fall in imports increasingly reduced food supplies, neighborhood associations often hosted cooking lessons for families to learn to make meals out of leaves and flowers (Kobe-shi 1965: 14). With the expansion in the tasks of the associations, the City of Kobe hired 200 new administrative staff to work full-time at the associations (Kobe-shi 1965: 477). Subsidies to the associations surged from 300,000 yen in 1942 to 820,000 yen in 1944 (Kobe-shi 1965). The strengthening of the state went in tandem with the strengthening of society. In anticipation of air raids, the evacuation of small children, the elderly, and other residents requiring assistance began in the spring of 1943 (Kobe-shi 1965: 490). Many evacuated to nearby Okayama, Himeji, or, in the case of the most unlucky ones, to Hiroshima. The evacuation of schoolchildren also began

106

Civic Engagement in Postwar Japan

in the summer of 1944; 17,190 school-aged children had been evacuated to the countryside as of October 1944 (Kobe-shi 1965: 491). In March 1945, all remaining pupils between the third and sixth grades were evacuated; all but about 5,000 students departed (Kobe-shi Kyoikushi Henshu Iinkai 1964: 22). At the secondary school level, instruction was suspended and all students were mobilized into war-related work. While depriving the youth of formal educational opportunities, mobilization did provide opportunities for socialization and the acquisition of civic skills. Ken Zako, a boy who grew up in Kobe during the war, recalls that when the children worked for the military, they would often each receive one anpan (sweet bun) as a reward for the hard day’s work. “What did we do with the anpan? We could have eaten it all by ourselves. But instead, we would bring it home carefully and share it equally with the others. We wanted to use the anpan as leverage to get more food from others in the future” (Zako 1990: 66–67). The young children were certainly acquiring norms of generalized reciprocity. Of mobilization into war work, Zako recalls, “Middle-school children like ourselves were no longer allowed to devote our time to studying in the schools [after 1940]. Every day, we were taken to different places to work, like day laborers” (Zako 1990: 63–64). However, the children of Kobe appear not to have minded too much; in fact, they seem to have found considerable enjoyment in the work. “We were given sickles and put to work in the fields. Being young, we quickly learned how to use the tools. We thought it was more fun than studying in the classrooms” (Zako 1990: 64). They were also mobilized to build launching bases for cannons (Zako 1990: 66), into civil defense groups to defend the schools in the event of air raids (Kobe-shi Kyoikushi Henshu Iinkai 1964: 21–22), and a variety of other tasks. Employers often welcomed these young workers since, in contrast to adults, who were becoming increasingly war-weary, the youth were still enthusiastic and eager to contribute to the war effort (Kobe-shi Kyoikushi Henshu Iinkai 1964: 101). At the same time, Zako also provides an indication that mandatory civic engagement may at times breed less, not more, trust in government: “Even back then, we thought: what kind of military uses children to build their cannon bases? We quickly became very skeptical of the military’s ability to defend us from attacking planes” (Zako 1990: 67). The intensification of the war effort also led to stronger restrictions on voluntary activities, especially Christian-influenced movements, which were seen by the government as potential spies for the Western Allies. This placed groups such as the YMCA in an especially difficult position. On the one hand, they could oppose the emerging ultranationalist regime and face persecution, or they could support it in the interest of organizational survival. The YMCA Japan chose the latter and provided a wide range of wartime services, such as sending care packages to soldiers serving in the front lines, assisting families where the breadwinner had been conscripted, and so forth.8 This cooperation 8

To some extent, the YMCA’s decision to cooperate with the war effort during World War II was a natural outgrowth of its cooperation in the Russo-Japanese War and, to a more limited degree, World War I as well.

Repression and Revival of the YMCA Japan

107

with the wartime regime led to considerable criticism toward the YMCA after the war (Kobe YMCA Hyakunenshi Hensanshitsu 1987: 268). Despite this cooperation, the YMCA nevertheless did not escape state surveillance. Kobe Seinen [Kobe Youth], the Kobe YMCA’s newsletter, came under military censorship (Kobe YMCA Hyakunenshi Hensanshitsu 1987: 285–286). YMCA meetings were watched closely by the police. Young women as well as men were increasingly drafted into war-related work after 1942, and by 1943 so many of the Kobe YMCA’s officials and staff had been drafted that the YMCA’s activities could no longer be sustained. In September 1942, Kobe Seinen ceased publication on orders from authorities to conserve paper (Kobe YMCA 1969: 26). The slowing of activities led the board of directors to rent out the Kobe YMCA building for military use in the spring of 1944, effectively putting an end to its public activities. But the board continued to meet elsewhere in the city, and surviving minutes of the Board of Kobe YMCA indicate that the last board meeting was held in April 1945 (Kobe YMCA, n.d.), when the annual report on the activities of 1944 was presented. The Kobe YMCA had in fact held out longer than many other cities. At the Osaka YMCA as well, all classes were closed by December 1944. And in Kyoto, despite not being bombed, all recreational activities had come to a virtual halt by 1942 (Nomura 1975: 205). An undated wartime Kobe YMCA minute provides a vivid description of the association’s condition toward the end of the war: After October 1944, the intensification of air raids made it practically impossible to continue with member activities. Mr. Nara left for Beijing, Mr. Nakagawa was mobilized to work in food procurement, Mr. Yamakawa was drafted to serve in Southeast Asia. The operations of the Kobe YMCA, then, came to a virtual halt. For these reasons, it also became impossible to hold regular board meetings. In the meantime, Mr. Honjo remained in touch with the other members and waited for the war to end. (Kobe YMCA, n.d.)9

In addition to wartime privations, Kobe also suffered from direct attacks. Air raids on Kobe began early, on April 18, 1942. They accelerated toward the end of the war, especially after the United States gained control over Saipan in July 1944, which could be used as a launching base for air raids over the main islands of Japan. Major air raids hit Kobe on January 3 and March 17, 1945, destroying much of western Kobe, and later on June 5, which destroyed much of eastern Kobe. Surviving records indicate more than 100 air raids before the end of the war that continued until August 14, only one day before Japan’s surrender (Kobe-shi 1965: 504). Overall, an estimated total of 6,235 people were killed, 15,331 were injured, and 128,181 buildings were damaged in Kobe over the course of the war (Kobe Kushu wo Kiroku suru Kai 1972: 250). The Kobe YMCA building was 9

The author of this passage is unknown, although since Honjo appears to have been the only remaining leader of the Kobe YMCA, he is likely to have been responsible.

108

Civic Engagement in Postwar Japan

destroyed by fire raids in June 1945. Indeed, many YMCAs around Japan lost their facilities during the war. The Kobe YMCA in the Wake of World War II The wartime period was a time of hardship for voluntary associations like the YMCA. In Kobe, however, the postwar revival of the YMCA was quite swift, and indeed it quickly surpassed its prewar size. Leaders of the Kobe YMCA met for the first time after the war on October 28, 1945, roughly two and a half months after defeat. Since the YMCA building had been destroyed in the war, seven board members assembled in Kobe Church, just across the street from the former YMCA. Both in Kobe and in Sapporo, the local Christian community provided assistance as the YMCA sought to rebuild. Surviving minutes from these early meetings indicate that the seven leaders were committed to reviving the YMCA and never considered closing it down; they immediately moved to re-open the Sunday schools by the following spring and thereafter met monthly to prepare (Kobe YMCA, n.d.). In 1946, English language classes resumed instruction at the YMCA Kobe for the first time in many years, since the use of English had been banned during the war. A total of 2,454 students enrolled during the first year (Kobe YMCA Hyakunenshi Hensanshitsu 1987: 323). Initially, the Kobe YMCA was able to rent out space for these activities inside the Kobe Yamate Church, but as the classes became increasingly popular, the classes expanded into the cafeteria and even the warehouse of the Yamate Church until the new YMCA building was completed in 1953 (Kobe YMCA 1969: 29). That same year, in 1946, children’s events attracted another 1,850 people (Kobe YMCA 1969: 324), YMCA-sponsored Easter events 800 people, and bible-reading classes 635 people (Kobe YMCA 1947: 2–3). The Kobe YMCA also organized visits to assist families in the cities who had lost their homes and children who had lost parents (Kobe YMCA 1969: 29). As the years went on, the YMCA’s activities further expanded to include chorus groups, haiku-composition sessions, and socializing groups for singles (Kobe YMCA 1969: 31). Membership doubled every year, from 124 in 1945 to 240 in 1946 to 472 in 1947 (Kobe YMCA Hyakunenshi Hensanshitsu 1987: 323, 339). While membership continued to grow thereafter, it is quite notable that the growth in membership began before systematic efforts at recruiting new members began in the fall of 1947 (Kobe YMCA Hyakunenshi Hensanshitsu 1987: 328). Even after recruitment efforts began, the YMCA was able to recruit members in numbers that far exceeded their expectations; in 1948, for instance, 649 new members joined, far more than the targeted goal of 500 (Kobe YMCA 1969: 29). In a later recruitment drive that was organized on a national scale, the Kobe YMCA sought to recruit 400 members and brought in almost 1,300 new members (YMCA Japan Annual Report, 1954). Of course, it is possible that the Kobe YMCA set modest recruitment targets so that targets could be met with relative ease. As will be seen later, however, the Sapporo YMCA was unable to

Repression and Revival of the YMCA Japan

109

meet its much more modest recruitment targets, suggesting that groups may not simply have set goals that they saw to be easily attainable. The impetus for the growth in participation in the wake of World War II appears to have been primarily bottom-up, rather than a result of top-down concerted recruitment efforts. Aspiring members appear to have come to the groups on their own accord. As one member who joined the YMCA in the early postwar years recalls: “Young people who were starved for culture, recreation, or knowledge all flooded into the small quarters of the YMCA. In those days, we were both materially and spiritually impoverished, but the YMCA was seen as a place where personal relationships were very important” (Kobe YMCA Hyakunenshi Hensanshitsu 1985: 8). Indeed, there is some evidence that the YMCA leadership did not necessarily welcome, and was somewhat baffled by, this sudden surge in membership. As Tetsuo Suekane, a senior leader of the YMCA, noted in October 1947: “It is true that the number of members is growing rapidly. However, many of them seem to come to the YMCA without really knowing what the YMCA stands for. Some of the senior officers of the YMCA are quite concerned” (“Kirisutokyo Seinen Undo wa Do Yukuka”: 26). This bottom-up postwar surge in participation is consistent with the hypothesis that organizational legacies from the prewar period and civic skills acquired during the war significantly affect postwar levels of civic engagement. Before the war, the YMCA was a bustling center of activity in Kobe. Then, during the war, individuals like Ken Zako acquired crucial civic skills through wartime mobilization despite their material privations. Once the war ended, the Kobe YMCA offered an outlet for citizens like Zako to exercise those newly acquired skills without expending high costs of becoming associational entrepreneurs. Process-tracing of the revival of the Kobe YMCA also suggests that the levels of prewar membership may have affected the timing that activities became suspended. That is, YMCAs that had larger memberships, like the Kobe YMCA, were able to operate for a longer period, even as members were increasingly drafted into the war effort and authorities increasingly viewed Western groups with suspicion. And the shorter the interruption in associational activities, the more likely that the organization would be revived sooner. The YMCA leaders appear to have taken it for granted that the YMCA would be revived, and the thought of closing it down does not even seem to have occurred to them. Of course, this pattern was not limited to the Kobe YMCA. Many groups that had been in existence from before the war were revived, allowing citizens to easily find a good match. Indeed, the fact that associations like the YMCA were able to recruit members with relative ease after the war probably encouraged other group leaders to try to revive their organizations as quickly as they could once the war was over. A virtuous cycle seems to have been set in motion whereby the early revival of the organization facilitated the rise in memberships, which in turn accelerated the rebuilding of the wider associational organizational apparatus. In sum, the happy story of the Kobe YMCA amply confirms the theoretical model of postwar civic engagement being advanced in this book. But what

110

Civic Engagement in Postwar Japan

of the postwar faltering of the YMCA in the very similar city of Sapporo? Does process-tracing of this case also point to the importance of prewar associational size?

The Revival of the YMCA in Sapporo Background: History of the City of Sapporo Interactions between Hokkaido, then known as Ezo, and the rest of Japan were sporadic until the early eighteenth century. Around this time, however, the Tokugawa Shogunate learned that the Russians had exhibited an increasing interest in Hokkaido (Sapporo-shi Kyoiku Iinkai 1989: 415–416). British ships had also been seen in the area. Concerned that a seizure of Hokkaido by foreign powers would pose a serious threat to Japan’s security, the Shogunate dispatched a series of missions to conduct an extensive study of the geography of Hokkaido in 1785, 1786, and an 180-person mission in 1798 (Sapporo-shi Kyoiku Iinkai 1989: 417–420). On the basis of these studies, and as Russian ships became increasingly active in the area, the Shogunate imposed direct rule over eastern Hokkaido in 1799 and on western Hokkaido in 1807 (Baba 1998: 11–12). Prior to the Meiji Restoration, however, Hokkaido was viewed as an alien land and remained populated mostly by the indigenous Ainu people. Sapporo itself is an Ainu name, meaning “dry valley,” as it lay on the Toyohira River valley. It was still a vast area of undeveloped land during this period. In the wake of the Meiji Restoration, however, Sapporo was designated as the capital of Hokkaido, owing primarily to its geographical position in the center of Hokkaido (Sapporo-shi Kyoiku Iinkai 1989: 1015). Due to Hokkaido’s strategic location vis-à-vis Russia, the new Meiji government actively promoted the development of this new frontier (Ebina 1991: 12). In 1870, Kiyotaka Kuroda, one of the emerging leaders of the former Satsuma Domain, which had engineered the Meiji Restoration, was charged with the responsibility of developing Hokkaido. Kuroda would later serve as prime minister between 1888 and 1889, and the fact that the new Meiji government would charge one of its brightest young leaders with this task reveals the urgency with which the new regime sought to develop Hokkaido (Baba 1998: 39–40). Policies to encourage immigration from the mainland to Hokkaido had already begun during the late Edo Period, in 1854, but they accelerated markedly during the Meiji era. In 1869, the Meiji regime allotted plots of land in Hokkaido to each prefectural government, each of which then provided various incentives for landless citizens to relocate there. Willing immigrants were usually provided with housing, moving expenses, stipends, and agricultural equipment (Sapporo-shi Kyoiku Iinkai 1991: 85). The former samurai class, which had been stripped of its privileges after the Meiji restoration, comprised a substantial portion of immigrants, especially during the early Meiji period (Baba 1998: ch. 6).

Repression and Revival of the YMCA Japan

111

Sapporo was primarily a commercial, rather than industrial, city for much of the nineteenth century. The first factories in Sapporo were established between 1871 and 1880, in machineries, textiles, and food processing (Sapporo-shi Kyoiku Iinkai 1985: 50). These first factories in Sapporo were entirely stateowned, although they were all privatized in 1886 (Sapporo-shi Kyoiku Iinkai 1985).10 Because of Sapporo’s relative proximity to Russia, the Russo-Japanese War served as a crucial catalyst for the industrialization of the city (Sapporo-shi Kyoiku Iinkai 1994: 414). Industrialization proceeded at a rapid pace between 1910 and 1918, around which time industrial output grew to one-half of the city’s income (Sapporo-shi Kyoiku Iinkai 1994: 289). The main industries were light industries such as food processing, which produced more than 50 percent of Sapporo’s total income until 1928 (Sapporo-shi Kyoiku Iinkai 1994: 394). There were also some machine tools industries, as well as textiles, which produced around 20 percent of Sapporo’s total output (Sapporo-shi Kyoiku Iinkai 1994: 230–235). But heavy industrialization was relatively slow compared to Kobe. Even at its prewar peak in 1937, total output in heavy industries was only 16 percent of total income (Sapporo-shi Kyoiku Iinkai 1994: 394). In the same year, heavy industries comprised 77.3 percent of total income in Kobe (Kobe-shi Keizaibu Sangyo Kenkyusho 1939: 6–9). Like Kobe, Sapporo was a city with a distinctly international flavor. The Meiji regime hired a large number of Westerners to provide technical expertise in the task of cultivating the new frontier. Hokkaido’s cold climate was quite different from the rest of the Japanese archipelago, and the cultivation of the land there required different expertise. In 1871, Kuroda personally traveled to the United States to request President Ulysses S. Grant to send advisers to Hokkaido (Ebina 1991: 15). The first American agricultural adviser, Horace Capron, who had served as Commissioner of Agriculture in the Andrew Johnson and Grant administrations, arrived later that year (Baba 1998: 23), and records indicate that seventy-eight Westerners were employed by the Meiji government between 1868 and 1889 (Sapporo-shi Kyoiku Iinkai 1991: 208).11 These individuals assisted in the rise of “Western” industries in Hokkaido that had hitherto not existed in Japan, such as dairy, meat processing, and beer. Many of the companies that were founded in Sapporo during the prewar period became household names in Japan, such as Snow Brand Milk Products (Yukijirushi) and Sapporo Beer, and these companies played a key role in bringing Western-style cuisines to the average Japanese household. These Western advisors in Hokkaido were also instrumental in spreading Christianity there. Especially influential was William S. Clark, the president 10 11

This was when the Hokkaido-cho (the Prefectural Office) was founded. Among these seventy-eight Westerners, there were forty-eight Americans, thirteen Chinese, five Russians, four English, four Germans, three Dutch, and one French (Sapporo-shi Kyoiku Iinkai 1991: 208). These experts brought a diverse range of expertise in fields including agriculture, mining, geology, botany, civil engineering, railroads, and education (Ebina 1991: 18). In contrast to Kobe, where the British were most numerous, American influence was relatively strong in Sapporo and in the Hokkaido area more generally.

112

Civic Engagement in Postwar Japan

of Massachusetts Agricultural College,12 who the Meiji government invited to teach agriculture in the Sapporo Agricultural College. Due to the historical ban on Christianity and the continued hostility of the Meiji government toward the penetration of Western thought, Clark’s use of the Bible in the classrooms was not without initial controversy (Ebina 1991: 38–41; Baba 1998: 187–192). But once the introduction of the Bible was permitted, Christianity spread rapidly. The Sapporo Independent Church (Sapporo Dokuritsu Kyokai) was founded in 1884 by a group of Christians that included Kanzo Uchimura and Inazo Nitobe, who later served as under-secretary general of the League of Nations (both men were graduates of the Sapporo Agricutural College); the Kita Ichijo Church was founded in 1890; and the Sapporo Methodist Church was founded in 1892. Sapporo, along with Kumamoto and Yokohama, came to be known as one of the three main centers of Protestantism in Japan (Nagai and Matsuzawa 1987: 11). The percentage of Christians in Hokkaido generally hovered between 1.8 and 2.5 percent during the first two decades of the twentieth century, roughly double the national average around the same time (Sapporo-shi Kyoiku Iinkai 1991: 865). The percentage of Christians is said to have been particularly high among professionals and local elites; according to one estimate, Christians comprised 20 percent of local bureaucrats, 12 percent of professors and local university officials, and 20 percent of students (Nagai and Matsuzawa 1987: 15). These figures suggest that the social influence of Christianity extended well beyond the mere 2–3 percent of the local citizenry that it comprised. Christianity grew rapidly again around the time of World War I, driven largely by an international recruitment drive that was launched by the Christian church. The number of churches in Sapporo doubled during the first two decades of the twentieth century, from ten to nineteen (Hokkaido YMCA 1997: 889). All of the students from Clark’s two graduating classes later became Christians (Sapporo-shi Kyoiku Iinkai 1991: 436),13 and many became founding members of the Sapporo YMCA. Several, including Kanzo Uchimura, also became active labor leaders. These Christian-influenced leaders often played important roles in the early social movements in Hokkaido. In the wake of a particularly bad harvest in 1914, Christian leaders in Sapporo organized a series of fundraising drives to benefit the farmers (Sapporo-shi Kyoiku Iinkai 1991: 884). They were also active in the prohibition movement. During the Taisho era, many of these Christians also became socialists and became leaders in the fledgling local labor movement (Sapporo-shi Kyoiku Iinkai 1991: 885). The Sapporo YMCA, another outgrowth of the Sapporo Christian community, was founded in 1897 by a group of Japanese students at the Sapporo Agricultural School who were influenced by Clark, including Uchimura 12

13

Founded as a land-grant university, the college was later renamed the University of Massachusetts at Amherst. As Ebina (1991) notes, Clark himself was not a missionary – he was trained as a botanist – but was deeply influenced by Puritanistic thought.

Repression and Revival of the YMCA Japan

113

(Kawamura 1987: 203). It was only the fourth YMCA to be founded in Japan.14 When the Russo-Japanese War broke out, Christian groups and leaders in Hokkaido largely sided with the state. Anti-war sentiment was virtually nonexistent. Christian leaders in Sapporo, for instance, held two fundraising concerts, the benefits from which were donated to the Red Cross (Sapporo-shi Kyoiku Iinkai 1994: 881). These concerts were greatly successful, and more than 1,700 people attended the two concerts (Sapporo-shi Kyoiku Iinkai 1994: 881). The YMCA actively participated in these activities as well; as early as the Sino-Japanese War (1894–1895), the YMCA held tributary events for families of the war dead and engaged in fundraising for the families as well (Hokkaido YMCA 1997). Taisho Democracy in Sapporo Associational activities expanded further in Sapporo during Taisho Democracy. Indeed, there is no evidence that the voluntary sector overall was weaker in Sapporo compared to Kobe around the turn of the century. As in Kobe, industrialization led to the rise of the labor movement. The first labor union in Sapporo was founded by twenty-one skilled workers at the Hokkaido Linen Company (Hokkaido Seima Kaisha), in 1899, and the labor movement grew rapidly thereafter. This coincided with the spread of socialist and communist thought in Hokkaido; many of the intellectual leaders, such as Takeo Arishima, were based at Hokkaido University, which had been renamed from the Sapporo Agricultural College (Ebina 1991: 379–380). The Ainu movement also emerged around the same time. The leader of this movement was Zenichiro Oyabe, who became aware of minority issues while studying at Howard and Yale Universities (Sapporo-shi Kyoiku Iinkai 1994: 512). He founded the Hokkaido Kyudojin Kyuikukai [Hokkaido Indigenous Peoples’ Relief Association] in 1900. A key pillar of their activities was to provide education to the Ainu people so that they could escape their poverty. Reflecting the rise of the middle class in Sapporo, a variety of cultural organizations were founded during the mid- to late 1920s (Sapporo-shi Kyoiku Iinkai 1997: 924). Notable examples include the Sapporo Music Association (1923), the Hokkaido Association for the Arts (1925), the Nakajima Orchestra (1928), and the Ikenobo Flower Arrangement Society (1924). These associations represented a break from most of the existing associations in Sapporo, which had been primarily led by (and served) intellectuals and Christians. The first newspaper of Sapporo, Hokkai Taimuzu, was also founded in 1901 (Ebina 2000: 35).

14

As noted earlier in the chapter, this depends on when the founding date of the Kobe YMCA is set. If it was in 1886, then the Sapporo YMCA becomes the fifth-oldest YMCA in Japan.

114

Civic Engagement in Postwar Japan

Christian groups continued to be active during this era. Toyohiko Kagawa, the Kobe-based Christian leader, was enormously popular in Sapporo, and his two lectures in Sapporo in 1931 attracted a total of 4,640 people (Sapporo-shi Kyoiku Iinkai 1997: 1094). At 5.83 members per 1,000 population as of 1937, Hokkaido still had the fifth-largest membership in Christianity in Japan, following Nagasaki, Tokyo, Kyoto, and Miyagi Prefectures. But, despite the strong Christian influence in Sapporo, the Sapporo YMCA remained relatively small. At its prewar peak, membership only reached 62, in 1928; in 1933 and 1940, it declined to 46. This pales in comparison to the Kobe YMCA, which already had 502 members in 1923 and 677 in 1936 (Kobe YMCA Hyakunenshi Hensanshitsu 1987). In contrast to Kobe, the YMCA in Sapporo remained primarily an elite-based movement that failed to extend to the local grassroots. The small number of members forced the Sapporo YMCA to suspend its activities quite early in the war, around 1937 (Hokkaido YMCA 1997: 224). Part of the reason for the weakness of the YMCA may owe to political repression. Due to its strategic location, repression of left-leaning activities appears both to have begun sooner in Hokkaido than other areas of Japan and to possibly have been more severe. Authorities were particularly nervous about social movements in Sapporo because of its proximity to now-Communist Russia, and because of the militancy of the large coal miners’ unions in the Hokkaido area. The tokko (secret police) was established in Hokkaido in 1923, earlier than in most areas of Japan (Sapporo-shi Kyoiku Iinkai 1997: 609). This may provide at least a partial explanation for the weakness of the YMCA in prewar Sapporo. On March 15th, 1928, in the so-called “3–15 Incident,” authorities raided and arrested 250 communist, socialist, and anarchist leaders (Sapporo-shi Kyoiku Iinkai 1997: 611–612), which dealt a severe blow to the labor movement in Hokkaido. The first May Day parade in Sapporo was planned in 1927 but was canceled because only 11 people showed up (Sapporo-shi Kyoiku Iinkai 1997: 642). The first May Day parade in Sapporo was finally held in 1930 with just 50 participants (Sapporo-shi Kyoiku Iinkai 1997). Repression, however, does not appear to have thwarted the growth of the local Christian community in Sapporo, despite the fact that many Christian leaders had close connections to the labor movement, so at best it provides a partial explanation for the weakness of the Sapporo YMCA during the prewar period. A second, possibly related explanation for the relatively slow growth of the YMCA in Sapporo in the prewar period is that the Christian community in Sapporo may have been more of an elite-centric movement than its counterpart in Kobe. One indication of the more elite-centric orientation of the Sapporo Christian community has to do with schools. As noted earlier, the Christian community in Kobe actively founded schools and hospitals in the area. In Sapporo, the founding of Christian elementary and secondary schools appears to have been somewhat slower compared to Kobe. During the early twentieth century, the only Christian-influenced schools were the Hokusei Jogakko

Repression and Revival of the YMCA Japan

115

(Sapporo-shi Kyoiku Iinkai 1991: 886), which was founded in 1887, and the Fuji Women’s School, which was founded in 1924. In terms of the YMCA itself, as seen earlier, the completion of the large Kobe YMCA Hall helped to catapult the YMCA into a major cultural center in the area and helped to boost its membership. But, despite the success of Kagawa’s lectures in Sapporo, the YMCA Sapporo made no move to establish a similarly large-capacity facility. Another reason for the relative weakness of the YMCA in Sapporo compared to the one in Kobe may owe to Sapporo’s relative geographical isolation. As noted earlier, early leaders of the YMCA in Kobe were well connected with business and Christian leaders in other nearby major cities, such as Osaka and Kyoto. Sapporo, on the other hand, was relatively far from other cities and therefore had more difficulty drawing on resources beyond its immediate vicinity. In sum, after a promising beginning around the turn of the century, the Sapporo YMCA grew only fitfully during the Taisho period. In contrast to the case of southern Italy that Putnam (1993) depicts, this was not because civic life in Sapporo was weak overall. Rather, there is at least some evidence to suggest that a combination of early repression, the more elite-centric orientations of the local Christian community, and geographical isolation may have slowed the growth of the YMCA in Sapporo. Sapporo during World War II Studies of Sapporo’s history concur that the war effort, which expanded government control over the economy, served to catapult Sapporo into the most prominent city in Hokkaido (e.g., Ebina 2000; Hokkaido Shimbunsha 1972). Until World War II, Sapporo had been overshadowed by Otaru and Hakodate, port cities that also were centers of heavy industry, and had been more populous than Sapporo since the nineteenth century. Indeed, it was not until 1940 that the population of Sapporo exceeded those of Hokkaido’s two major port cities (Ebina 2000: 43). The first branch of the Bank of Japan in Hokkaido had also initially been established in Otaru, not Sapporo (Ebina 2000). The extension of government auspices over the economy during wartime served to concentrate government functions in Sapporo and spurred rapid growth of the city; the Sapporo branch of the Bank of Japan, for instance, opened in April 1941 (Hokkaido Shimbunsha 1972: 281). As with the rest of the country, the expansion of war with China during the 1930s led to the gradual intensification of mobilization in Sapporo. On the economic side, workers in Sapporo were shifted into mining, agriculture (to cultivate new areas of land for increased food production), and, to a lesser extent, heavy industries such as aircraft production (Yamazaki 1992: 422– 433). After the U.S. oil embargo began in 1941, coal became a crucial source of energy in Japan; and because of Sapporo’s proximity to several major coal mines, many workers and students were mobilized into the mines (Sapporo-shi 1997: 469–470). When these were not enough, workers were forcibly brought

116

Civic Engagement in Postwar Japan

in from Korea and China (Sapporo-shi 1997: 437; Hokkaido Shimbunsha 1972: 278–279). New graduates were often also sent out of Hokkaido to work in war-related industries such as aircraft. Students of Hokkaido University were mobilized according to their fields of specialization; records indicate that students in the Fisheries Department, for instance, were sent to help in fisheries, medical students were sent to military hospitals, engineering students were sent to assist in railroad building, and so forth (Sapporo-shi Kyoiku Iinkai 1997: 470). Mobilization became extensive at the community level as well. As early as 1937, neon lights in the main commercial streets of Sapporo were turned off at night in an effort to conserve energy (Hokkaido Shimbunsha 1972: 290). Competition among women’s organizations led to particularly enthusiastic participation among women’s groups. The Aikoku Fujinkai [Patriotic Women’s Association] had 1,229 members in Sapporo as of 1938, and the Kokubo Fujinkai had 11,752 (Sapporo-shi Kyoiku Iinkai 1997: 817, 819).15 As with their counterparts across the country, these associations actively engaged in seeing off departing soldiers, assisting families of departed soldiers, recycling, fundraising drives, sending care packages to soldiers serving in the front, and so forth. As with other parts of the country, however, the main conduit for mobilization was the neighborhood associations. As of 1940, 311 neighborhood associations with 1,966 sub-associations (han) were in existence in Sapporo, enrolling a total of 42,007 households (Sapporo-shi Kyoiku Iinkai 1997: 126– 127, 130). Because Sapporo had already had a base of numerous administered mass organizations by the time that mobilization intensified, the task of reorganizing them into neighborhood associations with uniform by-laws and rules was not exceedingly complicated. As in Kobe, mobilization did provide citizens with on-the-job training in civic skills. Kyozo Takahashi, who was a ninth grader at the end of the war and who had been mobilized to load food supplies for the military onto cargo trains, attests: “The work was hard. … We were undernourished middle-school students but we had to carry bags of rice that were heavier than we were (60 kilograms). At the same time, we were proud to be doing the work and felt like we had grown up. Perhaps we had a sense of mission that we were contributing to the war effort” (Takahashi 1980: 38). Not all citizens of Sapporo shared this enthusiasm, however. Kiyoko Nishida, a housewife, recalls the air raid drills in Sapporo: “We looked very impressive. But many people complained that the drills only made us hungry. Morale was quite low” (Nishida 1980: 79). Nishida also looks back grimly on the behavior of her compatriots. She recalls that as food became increasingly scarce, she began to grow vegetables on the sidewalks. The people would sometimes come 15

While the Kokubo Fujinkai overwhelmed the Aikoku Fujinkai in terms of sheer numbers, the Aikoku women came from the upper and upper-middle classes and were endowed with far better resources than the Kokubo Fujinkai, which drew primarily on working-class women.

Repression and Revival of the YMCA Japan

117

with money and would ask her to sell the vegetables. “The less conscientious ones, however, would simply steal the vegetables, just as they became ripe. … This led me to harvest the crops early, even when it seemed a bit too early” (Nishida 1980: 77). With the intensification of the war effort from the 1930s onward, Christianity increasingly came to be viewed as an enemy creed, and government pressure became increasingly frequent during the early 1930s. Church services were monitored by police for “subversive” content, and Christian newsletters came under censorship. Many of the leading figures of the Sapporo Christian community were arrested and newsletters were suspended (Nishida 1980: 70–71). Much in the same manner as in Kobe, some of the larger churches sought to respond to this increasing pressure by actively contributing to the war effort on their own accord. The Sapporo Church, for instance, organized a civil defense group in the event of air raids on the church (Sapporo Kyokai Hyakunen no Ayumi Henshu Iinkai 1992: 233). The Sapporo Church also raised funds to be donated for the purchase of aircraft (Sapporo Kyokai Hyakunen no Ayumi Henshu Iinkai 1992: 243). In 1945, as labor increasingly became scarce, religious groups dispatched 2,000 workers to engage in construction work (Sapporo Kyokai Hyakunen no Ayumi Henshu Iinkai 1992: 249). YMCA membership began a gradual decline after the early 1930s as well, much sooner than in many other cities such as Kobe and Osaka, where memberships did not begin to decline until very late in the 1930s (“Sapporo Kyokai Hyakunen no Ayumi Henshu Iinkai” 1992: 200; Kaneda 1987: 62; Kobe YMCA Hyakunen shi Hensanshitsu 1987; Osaka YMCA 1982). Indeed, the Sapporo YMCA building was sold off to the Sapporo Church in 1937 to be converted into a dormitory for Hokkaido University students. The reasoning was that “The building has become so old that it is not practical to continue to use it.… Moreover, given the current [social and political] environment, the future of the YMCA is quite bleak” (Hokkaido YMCA 1997: 215). Under the terms of the sale, the headquarters of the Sapporo YMCA was to be housed in the new dormitory building. But as Tomoo Nakano, former president of the Hokkaido Electric Power Company, recalls, “In converting the old YMCA building into college housing, the Sapporo YMCA was to have an office in the dormitory building. However, I do not recall ever seeing any signs showing that the YMCA was there” (cited in Hokkaido YMCA 1997: 217). The operations of the Sapporo YMCA appear to have come to a virtual halt after this sale (Hokkaido YMCA 1997: 235). This was much earlier than the suspension of the Kobe YMCA’s activities in 1944. Meanwhile, compared to most of the other major Japanese cities, the physical damage to Sapporo during the war was quite limited. Hokkaido was not hit by air raids until the very end of the war, in July of 1945, and when it was, the damage was much more limited compared to Tokyo, Osaka, Kobe, and other major cities. Most of the attacks on Hokkaido were concentrated on the Pacific coast, away from Sapporo (Kikuchi 2007: 8). Compulsory evacuation for the elderly, the physically challenged, and school-aged pupils only began on

118

Civic Engagement in Postwar Japan

August 10, 1945, only five days before defeat (Sapporo Kyokai Hyakunen no Ayumi Henshu Iinkai 1992: 848). It is estimated that only 20 percent of pupils, most of whom attended schools in the city center, were evacuated (Yamazaki 1992: 434). An estimated 0.3 out of 1,000 Hokkaido residents died in the war, compared to the national average of 2.7, 43.9 in Hiroshima, and 17.6 in Nagasaki. Only 1.3 per 1,000 buildings were destroyed in Hokkaido, a markedly lower figure compared to the national average of 20.6, 98.1 in Tokyo and 77.9 in Osaka (calculated from Keizai Antei Honbu 1949). The social disruptions from war, then, were much more limited compared to the other major cities in Japan. In economic terms as well, the damage in Hokkaido was quite limited compared to the rest of the country. As of 1939, industrial output in Hokkaido had still lagged behind the national mean (Somucho 1988a). But in 1949, its income per capita was 29,238 yen, well above the national average of 18,996 yen. This does not mean that life was easy; as in the rest of the country, food shortages and inflation were severe. The official history of the Sapporo Fire Department, for instance, recounts that members of the fire department began to grow their own crops in 1946 to alleviate the food shortage (Sapporo-shi Shobo Enkakushi Hensan Iinkai 1971: 268). Nevertheless, the income per capita figures for Hokkaido in the late 1940s exceeded even those for Tokyo, which had the highest income per capita in the prewar period. At 1.46 percent, the unemployment rate in Hokkaido was also well below the national average of 1.87 percent (Keizai Antei Honbu 1950). The Weak Revival of the Sapporo YMCA The wartime difficulties of Christian groups including the YMCA notwithstanding, Sapporo’s relatively low level of material damage and high level of mobilization would seem to favor a rapid growth of the Sapporo YMCA in the postwar period. In reality, this was not the case. The blow itself was not severe, but the weakness of the prewar Sapporo Y meant that even the soft blow that it incurred during war was enough to knock it down at a relatively early stage. The early suspension and the long period of inactivity, in turn, made the revival of the YMCA after the war that much more difficult. Thus, even though many Sapporo residents had gained civic skills during wartime mobilization, the YMCA was not prepared to provide them with the opportunity to exercise those skills. Citizens turned to other activities instead. As we have seen, in the case of the Kobe YMCA, by early 1946 both English classes and Sunday school had been revived. And Kobe was not alone. In Tokyo, which had incurred particularly extensive levels of damage, the YMCA resumed its activities almost immediately upon the end of the war, by re-opening its English classes in September 1945 (YMCA Nihon Domei 2003: 119). Similarly, the Yokohama YMCA re-opened in November of 1945 (YMCA Nihon Domei 2003: 119). In Nagoya, where, like in Kobe, the YMCA had lost its building to bombing, activities resumed in January of 1946. But

Repression and Revival of the YMCA Japan

119

the Sapporo YMCA officials could only report that they had begun preparing for the re-opening of Sunday school classes in August 1946 (Sapporo- shi Kyoiku Iinkai 2002: 970). And those “preparations” took several years to bear fruit. The Sapporo YMCA did not officially resume its activities until January 1950 (Hokkaido YMCA 1997: 239–240), much later than many of the other Japanese YMCAs around the country. The Sapporo YMCA did eventually limp back to life, but, interestingly, many of the members of the newly revived YMCA Sapporo had not been YMCA members prior to the war. Indeed, only three members of the prewar Sapporo YMCA returned to push for the revival of the organization (Ueda, cited in Hokkaido YMCA 1997: 239). This marks a stark contrast to the case of Kobe, where officials from the prewar and wartime periods reconvened quickly to discuss plans for revival. Indeed, the initial impetus for the revival of the Sapporo YMCA in the postwar period appears to have come not from within the local community but top-down from the National Federation of YMCAs. In 1946, Toshio Suekane, a leader of the national YMCA federation, visited Sapporo and urged the revival of the local YMCA. To revive the movement, Suekane argued, it was necessary to mobilize support not only from the local Christian community for this idea, but also from business leaders. To gain the support of the local Christian community, Suekane and local proponents sought the support of Rinzo Onodera, an influential leader of the Sapporo Kita Ichijo Church. Suekane’s initiative slowly took root in Sapporo. Masao Tokito, one of the founding members of the postwar Sapporo YMCA, had not been a member of the YMCA prior to the war. But he became involved through his church friends who had been active members (Hokkaido YMCA 1997: 247). The Tokito family, including Masao, had long been active in the Sapporo Christian community during the prewar period (Nihon Kirisuto Kyokai Sapporo Kita Ichijo Kyokai Hensan Iinkai 2000: 153, 330–332), and Masao himself continued to serve as one of the leaders of the Sapporo Kita Ichijo Church (Nihon Kirisuto Kyokai Sapporo Kita Ichijo Kyokai Hensan Iinkai 2000: 312), one of the larger churches in the Sapporo area, in the postwar period, which meant that he could draw on a network of sympathizers to the YMCA. He recounts his involvement in the YMCA as follows: I became involved in the YMCA because Shigeaki Sasaki, who ran a hospital in Sapporo, Shini Nakamura, President of Fukido, and Kyuzo Nishimura, of Nishimura Pastries, all of whom, like myself, were active in the Kita Ichijo Church, were involved. … (Hokkaido YMCA 1997: 247)

After the initial top-down push by Suekane, some prewar members of the Sapporo YMCA also helped its revival in the wake of the war. For instance, Sonjiro Ueda had been an active leader of the Sapporo YMCA during the prewar period and became one of the association’s chief fundraisers in the postwar period (Hokkaido YMCA 1997: 272–273). He also took advantage of his prewar connections with YMCA leaders outside of Sapporo to seek

120

Civic Engagement in Postwar Japan

suggestions as to how the YMCA in Sapporo might be revived (Hokkaido YMCA 1997: 248). Since the postwar YMCA did not initially have its own building, a member of the prewar Sapporo YMCA, Sadatoshi Sukekawa, offered his house to be used as the temporary YMCA office free of rent (Hokkaido YMCA 1997: 249). Records also indicate that Sadatoshi’s son, Toshio, actively contributed to the initial fundraising efforts by taking part-time jobs to sell detergents and honey and donating the proceeds to the YMCA (Hokkaido YMCA 1997). Gen Miura, president of a radio station in Sapporo, had been a member of the YMCA in China before and during the war and agreed to broadcast recruitment ads for the YMCA for free (Hokkaido YMCA 1997: 250–251). These preexisting networks of reciprocity were particularly important because, in the wake of defeat, material resources were severely limited. In order for groups to expand and to recruit new members, a network of individuals who were willing to donate time and money was imperative. Since the Sapporo YMCA was essentially re-starting from scratch, the startup costs were much higher than in the Kobe case, even though the Kobe YMCA also had lost its building. The Sapporo Y had sold off its building in 1937 and had no permanent physical facility of its own. Led by Ueda, efforts to build a new building were well underway by early 1951, but fundraising efforts were fraught with difficulties. The small prewar presence of the YMCA in Sapporo meant that local businesses were often reluctant to contribute toward the construction of a new building. For this reason, leaders of the Sapporo Y turned to the national federation as well as to Y’s in the United States for assistance. But by the time that the Sapporo YMCA was prepared to request assistance from outside, there was a long line of YMCA’s elsewhere in the country that had also lost their buildings during the war and which already were seeking assistance from the national and international YMCAs. The new building in Sapporo was finally completed in November 1955. While Sapporo had been the site of one of Japan’s earliest YMCAs, it was one of the last to be revived in the postwar period. The difference with Kobe in membership recruitment is stark. As noted earlier, the Kobe YMCA consistently surpassed its recruitment targets during the initial decade after the war. The Sapporo YMCA, in contrast, set a recruitment target of 50 new members in 1953, a much more modest target than Kobe, which sought 400; and Sapporo still failed to meet it, increasing only from 118 to 130 members between 1953 and 1954 (YMCA Japan Annual Report 1954: 78). The virtuous cycle seen in Kobe, in which the revival of the association and membership growth proceeded in tandem, never materialized in Sapporo. Again, this is not to say that postwar growth in voluntary participation in Sapporo was weak overall. Winter sports, for instance, had always been popular in Sapporo, owing to its northern geographic location, and a series of ski competitions were held between January and March of 1946. In particular, the Hokkaido Ski Competition, held on March 1–3, 1946, attracted more than

Repression and Revival of the YMCA Japan

121

420 skiers (Sapporo Ski Renmei 1979: 158). One account of the history of skiing in Hokkaido recounts that “The ski community in Hokkaido had completely recovered from the war by 1947” (Sapporo Ski Renmei 1979: 160). Skiing was not an exception. Other sports that had also been popular during the prewar period, such as sumo and track and field, were also relatively quick to recover. The Hokkaido Track and Field Association had revived by 1947 and the Hokkaido Sumo Association also revived in 1947 (Hokkaido Taiiku Kyokai 1968). “A series of tournaments were held beginning in July 1946,” according to one account of the Hokkaido Track and Field Association, “and quickly attracted many more athletes than in the prewar period” (Hokkaido Taiiku Kyokai 1968: 66). In sum, the story of the Sapporo YMCA, like the case of Kobe, confirms the theoretical model presented in Chapter 3. As in the case of Kobe, thanks to wartime mobilization, the demand for associations such as the YMCA was strong, but due to prewar weakness, the YMCA could not provide the opportunities. Associational entrepreneurs in Sapporo ended up paying the high costs of the YMCA’s startup, but by then most people had filled up their schedules with other pursuits. The small and weak state of the YMCA in Sapporo during the prewar period led to a slow postwar recovery.

Summary This chapter has assessed the validity of the hypothesis that, along with mobilization, cultural and organizational legacies from the prewar period shaped the trajectory of civic engagement in postwar Japan in crucial ways. The hypothesis presented in this project generated three predictions concerning the process of civic engagement growth. First, in areas or regions where civic engagement had flourished before the war, associations were expected to survive longer in the face of adverse social and political conditions brought on by the war. Second, in such areas, the revival of associations was expected to begin sooner. And third, civic engagement was expected to grow in a bottom-up manner, regardless of whether associations could launch extensive recruitment drives. Finally, this model also expects that the effects of the aforementioned factors will trump the effects of physical damages on the groups’ recovery. Historical process-tracing of the cases of the Kobe and Sapporo YMCAs appears largely consistent with these expectations. In Kobe, where the YMCA had flourished and then persisted through hardships until finally bowing to its fate in 1944, its leaders convened soon after the end of the war to begin the process of resuming its activities. Records indicate little evidence that the officials ever considered closing down the association, even though its building lay in ruins. However, while the early revival of the Kobe YMCA was crucial for its postwar success, the real impetus for its spectacular membership growth appears to have been bottom-up; membership surged even before the YMCA had an opportunity to systematically launch its recruitment drives, beginning in 1947, and even then the Kobe YMCA consistently outperformed

122

Civic Engagement in Postwar Japan

its recruitment targets. Wartime mobilization had produced a strong demand for the kind of opportunities that the Kobe YMCA supplied. The situation was quite different in Sapporo. The weakness of the prewar YMCA had led to its early postwar demise, and its revival after the war required a push from the national federation. Even when plans for revival finally got underway, the small size of the YMCA, the new set of leaders on whom it had to rely, as well as the lateness of the revival process itself hampered the amassing of necessary resources. This was despite the relatively limited war damage that Sapporo had suffered. Thus, when the group finally did reemerge, it was too little, too late; and the Sapporo YMCA consistently failed to meet its more modest targets for membership recruitment.

7 Wartime Endorsement and Postwar Repression of a Traditional Art

Overview This chapter continues to process-trace the mechanisms through which prewar legacies of participation and wartime mobilization shape the growth of postwar civic engagement. In particular, it compares the postwar revival of participation in judo in the two cities of Fukuoka and Yokohama. Judo offers a useful contrast with the YMCA that was the focus of the previous chapter. Whereas the YMCA was repressed by the Japanese government during the war due to its Western origins but then was subsequently viewed favorably by the U.S. occupation for the same reason, judo, with its more “indigenous” origins, was favored by the militarist regime during World War II but then was actively discouraged by the U.S. occupation forces, as they believed it propagated militaristic values. Despite this adverse political climate, participation in judo grew during the postwar period across Japan, albeit at a slower rate than the YMCA (see Chapter 2, Figure 2.6). The number of new members rose every year, from 3,033 in 1946, to 4,301 in 1947, to 6,550 in 1948, to 8,103 in 1949. The fact that judo grew at all during this period highlights the strong bottom-up impetus for growth in civic engagement in Japan in the immediate postwar period; while occupation policies may have had some effect in discouraging this growth, they were not able to quench it. As this chapter will demonstrate, a major source of this bottom-up surge in participation was wartime mobilization. But mobilization could only produce these benefits in places where sufficient organizational legacies allowed for a rapid revival of this martial art. As with the YMCA, there was considerable variation in the rate at which participation in judo grew across different regions of Japan in the immediate postwar years. The two cases of Fukuoka, where judo went from strength to strength, and Yokohama, where judo never became more than a marginal avocation, illustrate how preexisting legacies of participation shaped the postwar growth of participation. The chapter proceeds as follows. First, it provides a brief overview of the history of judo in Japan. Then it traces the trajectory of participation in judo 123

124

Civic Engagement in Postwar Japan

in Fukuoka, followed by Yokohama. As with the previous chapter, the discussions will focus on the timing of revival as well as the sources of the growth in membership: bottom-up or top-down.

The History of Judo Despite its current position as one of Japan’s national sports, judo is actually quite a recent invention (e.g., Oimatsu 1966; Maruyama 1967; Yoshida 1969; Kudo 1972; Inoue 2004; Matsubara 2006; Todo 2007).1 Judo was developed into its current form in the late nineteenth century by Jigoro Kano, who integrated elements of different streams of jujitsu, or a traditional Japanese martial art. An important difference between jujitsu and judo, according to Kano, was that jujitsu was intended for use in combat and therefore aimed at defeating the opponent, whereas judo was meant to represent more of a way of life and a means for training the mind: the do in judo means “way.” Despite its relatively late founding, judo spread rapidly in Meiji Japan. To a considerable degree, this owed to Kano’s privileged position within Meiji society. Kano was born to a wealthy merchant family in Kobe and graduated from the second graduating class of the University of Tokyo, in 1881. After graduation, Kano served as principal of a series of prestigious secondary schools, including Gakushuin, which is typically attended by members of the Imperial family, and the Tokyo Higher School for Education (Tokyo Shihan Gakko), while also devoting himself to the martial arts. In 1882, he founded the Kodokan in Tokyo, a judo dojo as well as an umbrella organization to promote judo. Here, with his disciples, Kano developed judo into a full-fledged martial art form. From there, judo spread quickly. A major boost for the new sport came from the fledgling Japanese national police. During the late 1880s, Kano’s Kodokan wrestlers were invited to engage in a series of matches against some jujitsu wrestlers in the Tokyo Metropolitan Police Department; Kodokan wrestlers reportedly won all but a few of the matchups. Kano parlayed these victories into a close institutional relationship between judo and the police forces in Japan (Todo 2007). Soon thereafter, Michitsune Mishima, superintendent general of the Tokyo Metropolitan Police Department, adopted Kodokan-style judo as a required course for all members of the Metropolitan Police forces. This decision was quickly copied by the police forces in the rest of the country. The police then began hosting judo classes for local youth, and this became a major engine for the growth of judo across Japan. The Japanese military also soon warmed to judo. Indeed, judo, along with kendo and kyudo (Japanese archery), have often been criticized for serving 1

Matsubara (2006) notes that while a large number of histories have been written on judo, these should be read with caution since many of them have been written by Kodokan-affiliated individuals and tend to present narratives that are heavily biased in favor of the Kodokan. He notes that this reflects the legacy of factional strife between the Kodokan and the Butokukai.

Wartime Endorsement and Postwar Repression of a Traditional Art

125

militaristic ends in prewar and wartime Japan. The military did promote the three martial arts, especially after the 1920s, making them a required part of the middle school curriculum from 1931, and for all boys in the fifth grade and above from 1939 (Oimatsu 1966; Inoue 2004). Judo and kendo were also part of the required basic training program for Japanese military recruits. It is important, however, to distinguish between the political uses of judo and the ideas and attitudes of the Kodokan itself. Indeed, Kano himself was an internationalist who served as Japan’s first representative to the International Olympic Committee (IOC) and worked hard to attract the Olympic Games to Tokyo. (Tokyo actually won the right to host the 1940 Games in 1936, only to concede the rights to Helsinki in 1938 as the war in China escalated.2) Nonetheless, judo’s close relationship with the prewar military led the Civil Information and Education Section (CIE) of the Allied Occupation to view judo as a “military art” that had been part and parcel of Japan’s militarist regime. SCAP disbanded the Butokukai, the other umbrella organization for Japan’s various martial arts, including judo, for having allegedly propagated militaristic and ultranationalist values.3 Although the judo community protested this portrayal of the role of judo in Japanese militarism, stressing that Kano conceived of judo as a way of life rather than a means for combat (Oimatsu 1966: 335), the CIE held fast in its views and told the Ministry of Education (Monbusho) to ban judo, kendo, and kyudo from both the formal school curriculum as well as school extracurricular activities in November 1945.4 At the adult and community levels, however, judo was permitted to continue, as long as competition was encouraged “as a sport,” so the Kodokan’s doors remained open. Repeated petitions from the judo community, such as school judo coaches as well as the Kodokan and several national Diet members, led the Monbusho to make a concerted effort to lobby the CIE to reinstate judo in the schools throughout the late 1940s. Perhaps in quintessentially judo-esque fashion, the Kodokan also more subtly contributed to this effort by hosting a series of judo classes and “shows” for American soldiers serving in the occupation (Maruyama 1967: 428–9). These became extremely popular among the soldiers. In fact, it is estimated that more than 500 American soldiers serving in the occupation forces enrolled in dojos around the country between 1945 and 1948 (Todo 2007: 182). As a result of these lobbying efforts, the CIE gradually softened, though it did continue to insist that judo operate as a competitive sport only, rather than as a means to instill values. The CIE formally approved the restoration of judo in the junior high schools and above in September 1950.

2 3

4

In the end, the Helsinki Games were also canceled due to the expansion of war in Europe. During the prewar period, the Butokukai had been active in the preservation and promotion of traditional Japanese martial arts. Yamamoto (2003) provides a useful overview of the history of the Butokukai and its breakup by the U.S. occupation. The Ministry Education had lobbied the CIE to keep judo as part of the middle school physical education curriculum, although as an elective (Oimatsu 1966: 332–333).

126

Civic Engagement in Postwar Japan

Judo thereafter became an elective sport within the schools’ physical education curriculum.5

The Puzzle The postwar growth of judo despite the initially strong repression by the U.S. occupation highlights the importance of bottom-up factors in explaining postwar levels of civic engagement. But it is also important to note that, despite the overall nationwide resurgence of judo by 1955, there was substantial variation among Japanese regions, even among those with similar socioeconomic profiles. Consider, for instance, the comparison between Yokohama, a major industrial and port city near Tokyo, and Fukuoka, a major industrial city in Western Japan. Both cities emerged after the Meiji Restoration as major industrial centers. The population of both cities’ metropolitan areas stood at roughly one million around 1940. Both Yokohama and Fukuoka built strengths in heavy, rather than light, manufacturing. Levels of industrial output were roughly similar in both areas around the turn of the century and grew in tandem through the first half of the twentieth century; between 1909 and 1940, Yokohama’s Kanagawa Prefecture grew from 21 million yen to 2,268 million yen, while Fukuoka Prefecture grew from 24 million yen to 2,178 million yen (Somucho 1988a: 415–417). The numbers of firms and workers were also roughly comparable; Kanagawa Prefecture had 1,360 establishments employing 74,958 people as of 1935, compared with Fukuoka’s 1,587 establishments employing 83,964 workers in 1935 (Somucho 1988a: 415–417). As sports historians have demonstrated many times over, industrialization is key to the development of mass sports activities. Yet despite their similar size and socioeconomic profiles, Yokohama and Fukuoka’s levels of participation in judo before the war were quite different. Fukuoka had developed a strong grassroots judo community by the mid-1910s and attracted aspiring wrestlers from across the country to train there, while participation in judo in Yokohama was weak for much of the prewar period. Indeed, Fukuoka has been a national powerhouse in judo since the early twentieth century, when judo began to spread throughout Japan. The area has produced numerous great judo athletes, including Ryoko Tani, Kazuhiro Ninomiya, and Kenzo Nakamura. It is home to the Kinshuki annual Japanese national high school judo tournament, founded in 1916, and, also until recently, the prestigious Fukuoka Women’s International Judo Tournament, which was held annually between 1983 and 2006. Key to the longstanding success of judo wrestlers from Fukuoka has been the vibrant judo community at the grassroots level. 5

Note that the revival of kendo came much later than judo; the CIE viewed it as having been much more closely intertwined with the wartime regime and the instilling of militaristic values than judo or kyudo. Kendo was not restored in junior high schools until 1957 (kyudo was restored in 1951).

Wartime Endorsement and Postwar Repression of a Traditional Art

127

By contrast, perhaps because its port was a major gateway to the West, Yokohama gravitated more toward Western sports such as baseball, tennis, and yachting in the early part of the twentieth century. While it is true that during the 1930s Yokohama-based judo wrestlers, who had struggled in national competition for much of the prewar period, began to win national titles, this was due in large part to the Kanagawa Prefectural Police judo team’s recruitment of wrestlers who were born in Fukuoka and its neighboring areas. And even with the success of the local police at the national level, judo was slow to trickle down to the Yokohama grassroots. Some statistics can tell the story well. While the Kodokan had six branches across Fukuoka Prefecture in 1936, in Yokohama’s Kanagawa Prefecture it only had two (Kodokan 1936: 84–91). Similarly, according to a survey conducted in 1940 by the Ministry of Health and Welfare, there was only one major dojo (judo training center) in Kanagawa Prefecture, compared with thirteen in Fukuoka (Koseisho Tairyoku-kyoku 1940).6 The difference in levels of participation in judo is also reflected in the number of students electing to take judo in the two prefectures during the prewar period. According to figures compiled by the Kodokan in 1932, 9,500 students at the middle schools and teacher-training schools [shihan gakko] in Fukuoka Prefecture, or 37.8 of every 10,000 people in Fukuoka, elected judo (Kodokan Bunkakai 1933: 71–95). In contrast, the corresponding figure in Kanagawa Prefecture for the same year was only 4,117 students, or roughly 25.4 of every 10,000 people in Kanagawa.7 This difference persisted into the postwar period. According to a survey conducted by the Japan Sports Association, there were seventy-eight dojos in Fukuoka as of 1968, an increase of sixty-five over the prewar number; but there were only thirty-seven in Kanagawa. Although this was certainly an improvement over the lone major dojo Kanagawa had had in the prewar period, the absolute growth there was thus much less than in Fukuoka (Nihon Taiiku Kyokai 1968: 95). Moreover, as will be shown later, Fukuoka in the wake of World War II produced leaders such as Shigetaka Takemura, who spearheaded the movement to revive judo nationwide in Japan, while Yokohama was very much a follower in this process. The drive to revive judo in Yokohama after the war remained within the police corps and did not greatly spill over to the local grassroots. We now take a closer look at the process via which judo in Fukuoka grew from prewar strength to postwar strength. After that, we turn to the process

6

7

The national mean was 2.0 per 1,000,000 population (standard error = 1.78). Of course, the number of dojo does not directly reflect the population engaged in judo, as each dojo could have had any number of members. Moreover, the definition of “major” also seems somewhat arbitrary. Nevertheless, it seems reasonable to assume that had there been greater enthusiasm toward judo in Kanagawa, especially in Yokohama, more dojo would have qualified for the list. The national mean across 46 prefectures was 29.3 per 10,000 population (standard error = 7.88).

128

Civic Engagement in Postwar Japan

that led judo in postwar Yokohama to grow as well, but always more tentatively than in Fukuoka.

The Revival of Judo in Fukuoka Background: Judo in Fukuoka during the Prewar Period Located in Kyushu, the westernmost major island of the Japanese archipelago, Fukuoka was a major port city prior to the Edo Period (1603–1867). But, as in the case of Kobe, the sakoku policy of closing off Japan from external trade except through the port of Nagasaki led to the subsequent decline of Fukuoka (Maruyama and Cho 2004). During the Edo Period, nearby Nagasaki flourished as the only port open to the Western world. Thus at the beginning of the Meiji Restoration, Fukuoka was far from a thriving urban center. Whereas activists from nearby Kagoshima and Saga orchestrated the overthrow of the Tokugawa shogunate and the Meiji Restoration and continued to wield strong influence over the fledgling Meiji regime, Fukuoka’s presence paled. Indeed, the Fukuoka domain in fact actually supported the Shogunate in opposition to the Meiji Restoration and thus was disfavoured by the new regime from its inception. Many former samurai from the Fukuoka area then took part in the Saga Rebellion (1874) and the Seinan War (1877), in which former samurai in the Kyushu area unsuccessfully rebelled against the Meiji government (Kawazoe et al. 1997: 167). In short, during the early Meiji years, the relationship between Fukuoka and the new Meiji government was tepid at best. Fukuoka’s fortunes changed, however, when the requirements of industrialization cast coal into the spotlight. Fukuoka’s Chikuho and Miike coal mines were among the largest in Japan, and this led the new Meiji government to invest heavily in the area. In 1888, the mines were privatized and sold to the Mitsui zaibatsu. In 1891, one of the first railroads in Japan was built to connect the coal mines from the Chikuho area to Kokura, just outside Fukuoka city. By 1900, the Mitsui-Miike coal mines produced 54 percent of the country’s coal output (Kuroki 2004: 134), and, combined with the neighboring Chikuho coal mines, Fukuoka Prefecture produced two-thirds of the country’s coal (Kawazoe et al. 1997: 277). Private businesses also began to establish in the Fukuoka area to take advantage of its proximity to the mines; during the course of the 1880s, for instance, a large number of textile companies were founded in the Fukuoka area. In 1901, Japan’s first steel plant was built in Yawata, also just outside Fukuoka, with great fanfare. This plant was seen in Japan as the symbol of the rise of Japan’s industrial power. Fukuoka’s industrial prominence rose further with the great increase in industrial production during the Sino-Japanese and the Russo-Japanese Wars. The rise of Fukuoka as a major industrial center also had spillover effects into the cultural realm. For instance, Fukuoka successfully beat out Nagasaki and Kumamoto in the competition for Kyushu’s first Imperial University. First, Kyoto University opened its medical school there in 1903; then, in 1911, this

Wartime Endorsement and Postwar Repression of a Traditional Art

129

medical school was renamed Kyushu Imperial University and subsequently added other academic fields. Civic engagement also grew rapidly during this period. Fukuoka became one of the centers of the universal male suffrage movement that had initially emerged in Kochi Prefecture in the early 1870s. The movement successfully led to the founding of the Imperial Diet in 1890. Moreover, the growth of the coal mine industry and heavy manufacturing produced an active labor movement in the region. In 1920, unions in the area launched the Yawata labor dispute, in which between 18,000 and 23,000 workers went on strike demanding improved pay and reduced hours. This strike was severely repressed and ended with mass arrests, but it did lead to some improvements in wages and working hours. Also, the farmers’ protests that erupted all over Japan in 1918 were especially fierce in Fukuoka, which counted the highest number of indictments of any prefecture (Kawazoe et al. 1997: 285). Judo spread to Kyushu during Kano’s 1891–1893 stint as principal of the Fifth National High School in Kumamoto, not far from Fukuoka. At the time of Kano’s arrival, the school did not have a dojo, so Kano converted a student lounge into a dojo. Many of the teachers that he had brought from Tokyo had also been his disciples at the Kodokan there, and judo spread rapidly among the students. Kano was only in Kumamoto for seventeen months, but the seeds that he sowed there blossomed. After Kano’s departure, Kodokan-affiliated judo wrestlers continued to serve as judo coaches of the Fifth National High School (Yoshida 1969: 6). The Kumamoto Kodokan was founded by Sumiomi Arima in September 1893. It was the fifth branch of the Kodokan nationwide and the first in Kyushu (Yoshida 1969: 6).8 However, with the rise of Fukuoka as a major industrial city, Fukuoka, rather than Kumamoto, subsequently became the center of judo in Kyushu. Instructors from the Tokyo Kodokan were often dispatched to Fukuoka for several years at a time to train the youth there. Particularly notable was the role of Kunisaburo Iizuka, one of the first wrestlers to attain judan status, the highest rank attainable by Kodokan wrestlers. Iizuka was posted in Fukuoka between 1898 and 1904 and contributed much to popularizing Kodokan-style judo among the youth there (Iizuka 1990: ch. 3).9 Building on this popularity, the inaugural Kinshuki judo tournament was held in 1916 as a tournament for high school judo wrestlers in Fukuoka. It subsequently opened its doors to wrestlers from other prefectures in Japan and also from Korea, Manchuria, and Shanghai. Despite the steady growth of the tournament, Fukuoka high schools continued to dominate the competition, winning all but seven of the twenty-seven tournaments that were held between 1916 and 1942 (Nishi Nihon Shimbunsha 1997). Fukuoka’s Shuyukan High 8

9

The remaining Kodokan branches were located in Nirayama (Shizuoka Prefecture), Edajima (Hiroshima Prefecture, within the Naval Academy), Kyoto, and Kojimachi (Tokyo) (Yoshida 1969: 9). Iizuka went on to serve as judo coach for Keio University (Iizuka 1990: 116–121).

130

Civic Engagement in Postwar Japan

School and Kurume Commercial High School, in particular, won thirteen of those twenty titles between the two of them. The key to Fukuoka’s success in judo was the fact that many of its citizens participated in it. During the late nineteenth and early twentieth centuries, local judo leaders established numerous judo dojos. These produced many prominent judo athletes, such as Sadao Hisanaga, Rokuro Ono, and Kentaro Furuno. The early success of Fukuoka-based wrestlers, in turn, drew more local youth into judo, creating a virtuous cycle of success in judo competition that led to growing membership, which in turn spurred more competitive success. By 1940, as noted earlier, thousands of Fukuoka residents were judo participants, and there were thirteen major dojos in the area, dwarfing the number in most other major metropolitan areas, including Yokohama. Fukuoka during World War II During the 1930s, in the early stages of Japan’s offensive into China, the western-located Fukuoka benefited enormously from heightened industrial demand. Demand for steel and coal was especially strong, and industrial output in Fukuoka tripled between 1935 and 1940 (Somusho 1988: 417). During this period, Fukuoka also became a major center of munitions production. Many workers were mobilized from other prefectures, and also from Korea and China to work in factories and mines in Fukuoka. However, toward the end of the war, production began to decline as raw materials became increasingly scarce. Judo continued to be popular in the area, and it also had the military government’s strong support, but as the war effort intensified tournaments were increasingly canceled. The Kinshuki national high school judo tournament, for instance, was suspended after 1942, and the last national judo championship was also suspended after that year. In Fukuoka, as with the rest of Japan, conscription and industrial mobilization made it increasingly difficult for the youth in Fukuoka to practice judo as the war wore on. The first U.S. attacks on Fukuoka were launched in June 1944. Thereafter, the evacuation of students from the urban centers began. The major air raid on Fukuoka, known as the Great Fukuoka Air Raid, occurred on June 19 and 20 of 1945. More than 1,000 people were killed or missing, and roughly onethird of all houses are said to have been destroyed. By the end of the war, more than 5,700 people had been killed or missing as a result of the raids. Severe as the destruction was, Fukuoka narrowly missed suffering much worse. Kokura, the next-door city to Fukuoka, is known to have been the initial target for the second atomic bombing of Japan on August 9, 1945, but it narrowly escaped because on that morning thick clouds were seen to hang over the city. The aircraft headed for Nagasaki instead. The Revival of Judo in Fukuoka after World War II In September 1945, the first occupation forces arrived in Fukuoka, and by December 28,000 U.S. troops were stationed in the area. As Japan’s

Wartime Endorsement and Postwar Repression of a Traditional Art

131

gateway to East Asia, the port of Hakata, in Fukuoka, became the center of repatriation of Japanese soldiers and civilians returning to Japan from Korea and China; more than 1.4 million people are said to have returned to Japan from the continent through Hakata Port (Kawazoe et al. 1997: 306). In turn, many Koreans and Chinese sailed back to their homelands via Hakata. The local economy struggled during the initial postwar years until the Korean War broke out in 1950. The war revived production at the Yawata Steel plant and in the local coal mines, and by 1951 production levels had returned to prewar peak levels. Despite SCAP’s antagonism, judo recovered swiftly in Fukuoka and the rest of Kyushu in the wake of World War II. As with judo in the schools, SCAP had initially banned judo in the police, but the Fukuoka Prefectural police shrewdly found a way around this ban. It simply re-named judo as “arrest techniques” or “self-defense techniques” and continued to teach the martial art despite the SCAP’s skepticism (Yoshida 1969: 200). By 1947, “arrest techniques” had become a formal part of police training throughout Japan. During the ban on school judo, neighborhood dojos were still permitted to operate. The Sekiryukan, a major dojo in Fukuoka that had produced numerous influential figures during the prewar period, had not been completely destroyed by the air raids, and it functioned continuously during the dark days of the end of the war and the immediate postwar period. It attracted numerous youth in the wake of the war at a time when students were not permitted to practice in schools. Registration of new members during the immediate postwar months often exceeded 50, and sometimes even 100 per month (Yoshida 1970: 190). As with the YMCA in Kobe, no systematic recruitment drive appears to have been undertaken; indeed, local judo leaders were wary of launching large-scale recruitment efforts that could draw the ire of the occupation forces. Rather, the growth was very much bottom-up in nature. The popularity of judo at the grassroots level pushed leaders to revive judo associations quickly, and the Fukuoka Prefectural Judo Association was reborn in the fall of 1946. As will be shown later, this revival occurred much sooner after the war than was the case in Yokohama. But this was merely a halfstep toward the creation of an organizational framework for judo’s postwar growth in the city. At the time, the judo community in Fukuoka was riddled with factional strife between the Kodokan-affiliated judo groups and former Butokukai-affiliated groups. This division threatened to put a damper on the martial art’s incipient recovery. In February 1947, however, an influential judo leader from the prewar period, Shigetaka (“Moko”) Takemura returned to Fukuoka from Manchuria. Takemura became a key figure in unifying the judo factions in Fukuoka. He had trained extensively at the Butokukai dojo while attending Kyoto University and during his years at Sumitomo Bank in Osaka, where he had also established close relations with the local Butokukai leaders. On the other hand, he was a graduate of Fukuoka’s elite Shuyukan High School, from which most of the Fukuoka Kodokan leaders hailed. Takemura

132

Civic Engagement in Postwar Japan

thus was one of the few individuals who had ties to both communities – and was also well-liked by both (Yoshida 1970: 186). Upon returning from Manchuria, Takemura quickly embarked on the task of reconciling the competing factions. Thanks to his efforts, a new, unified Fukuoka Prefectural Judo association was founded on May 24, 1947, only five months after his return to Japan. Takemura became its president shortly thereafter. On July 1, 1947, the Western Japan Judo Championships were held, with Takemura playing an important role in fundraising for this event. Wrestlers and fans alike awaited this event eagerly, and more than 100 judo wrestlers participated, with more than 5,000 fans in attendance in the 3,000capacity hall (Yoshida 1969: 192). Still more fans who could not secure entry climbed on ceilings and rooftops to get a glimpse of the competition (Yoshida 1969: 192). Not only did judo revive swiftly in Fukuoka, but leaders of the Fukuoka judo community, especially Takemura, also played an important role in laying the organizational, logistical, and financial groundwork for the revival of judo nationwide. Takemura led the movement to re-establish the Kyushu Judo Association, which was founded in January 1948. In March 1948, a tournament was held in Fukuoka between Kyushu-based wrestlers and wrestlers from the Kansai area of central Japan, and Takemura took advantage of the opportunity to host a meeting of judo leaders from Kansai and Kyushu as well as from the Tokyo Kodokan to discuss the founding of a new national-level umbrella organization for judo (Fukuoka-ken Taiiku Kyokai 1976: 337). Takemura served as the chair of this meeting. This initial meeting led to another meeting in Tokyo, in May 1949, where members of the Yudansha Kai, another national-level judo association that had existed since the prewar period (the other was the Butokukai, which had been disbanded by SCAP), including Takemura, met at the Tokyo Marunouchi Kogyo Club to discuss the founding of a new national judo association. The idea was to include not only wrestlers who had been certified at a certain level of accomplishment, but all judo participants. The new All Japan Judo Federation was founded in May 1949. The case of judo in Fukuoka highlights the importance of prewar legacies in reviving participation in the postwar period. Because judo had been popular in Fukuoka prior to the war, citizens wanted to return to it immediately after the war’s end. This provided an impetus for dojos and associations to be revived with relative ease. Judo leaders in Fukuoka found it relatively easy to raise funds and other resources that were necessary for revival. Moreover, individual leaders such as Takemura drew heavily on the logistical and financial support and good will of connections made during their high school, college, and prewar working years, both in planning the revival of judo organizations and in the implementation stage. In similar fashion to the Kobe YMCA, something of a virtuous cycle appears to have been set in motion whereby the high levels of demand for judo accelerated the revival of the associations, and the revival of the associations in turn further facilitated membership growth. In short, by 1955 judo in Fukuoka was stronger than ever.

Wartime Endorsement and Postwar Repression of a Traditional Art

133

The Revival of Judo in Yokohama Background: Yokohama during the Prewar Period As with Fukuoka, Yokohama is a relatively new city that emerged as a major urban center after the mid-nineteenth century. Like Kobe, it was designated as one of the first ports to be opened to the West under the Treaty of Amity and Commerce, signed between the United States and Japan in 1858. The Americans had wanted to gain access to Tokyo (then Edo), but the Shogunate, fearing the influx of Western forces into the political center, insisted that they locate in Yokohama instead, ostensibly because it had a better natural port (which was not untrue). Thereafter, Yokohama developed rapidly. Yokohama served as Japan’s largest trading port for much of the prewar period. Throughout the early Meiji period, more than half of the imports and exports to Japan passed through the port of Yokohama. In 1868, 86.6 percent of all exports and 82.6 percent of all imports passed through Yokohama; in 1889, Yokohama still handled 59.8 percent of all exports and 51.9 percent of all imports to and from Japan (Yokohama-shi 1963: 197). In the late 1880s and early 1890s, the port of Kobe began to grow more rapidly than Yokohama in the handling of imports into Japan (Yokohama-shi 1963: 215), but Yokohama has remained a major port to the present day. The existence of such a major port also stimulated the growth of light and heavy industry. The rapid growth of the industry started after the City of Yokohama offered generous tax benefits to attract industrial investment in an effort to recover from the Great Kanto Earthquake of 1923; and when space ran out in Yokohama, the city financed major landfill efforts to create more land so that more industries could be lured. By the late 1930s, Yokohama became a major center of industrial production in Japan and came to be known (at least in Japan) as the “Manchester of Japan.” According to Takamura (2006: 46–48), industrial production in Yokohama doubled between 1929 and 1936, tripled between 1929 and 1937, and quadrupled between 1929 and 1938. Heavy industrial production comprised 49 percent of Yokohama’s industrial production in 1929 but surged to 75 percent in 1935 and 83 percent in 1938 (Takamura (2006: 46). This industrial growth was supported by, and also led to, a flood of new residents into the city; Yokohama’s population soared from 620,306 in 1930 to just over one million in 1942 (Yokohama-shi n.d.). Yokohama was also quick to benefit from Western technologies. Japan’s first telegram stations were opened between Tokyo and Yokohama in 1870, with the help of British engineers and technology; Japan’s first steam locomotive also began service between Tokyo (Shinagawa) and Yokohama in 1872, also with British assistance (Yokohama-shi 1961: 325). With all of this economic activity, the city of Yokohama boomed. Most the growth in population in Yokohama was the result of internal migration rather than natural births, a phenomenon that remained true all the way into the 1970s. In the prewar

134

Civic Engagement in Postwar Japan

period in particular, the population growth was so rapid that it greatly strained the educational system; elementary schools often conducted classes in two shifts, one that began in the morning and another that began in the afternoon (Yokohama-shi Kyoiku Iinkai 1976). In a similar manner to Kobe, the growth of trade also led to the settling of a large population of foreigners in Yokohama, giving the city a distinctly international flavor. Christian missionaries opened churches, schools, orphanages, and other social services (Mainichi Shimbunsha Yokohama Shikyoku 1968; Tamura 1989). Along with Yokohama’s economic and population growth during the Meiji era came a boom in associational life and even political activism. For instance, like Fukuoka, Yokohama was a major center for the movement to demand the opening of Japan’s national legislature during the 1870s (Jiyu Minken Undo). As for sports, sixteen municipal associations were founded between 1868 and 1932: baseball, Japanese rubber baseball, tennis, soft tennis (which uses a rubber ball), swimming, table tennis, Japanese archery (kyudo), soccer, rugby, track and field, sumo, handball, alpine climbing, volleyball, and yachting (Yokohama-shi Taiikushi Kikaku Kanko Iinkai 1989: 384–385). Note that judo is not on this list. Despite Yokohama’s proximity to the Kodokan headquarters in Tokyo, judo was relatively slow to expand. The Kanagawa Prefectural Police, which covers Yokohama, was one of the first police forces in Japan to adopt judo as part of its training program (Kanagawaken Keisatsushi Hensan Iinkai 1972: 317), but judo spread to the grassroots only haltingly. The first surviving records of ordinary citizens practicing judo in Yokohama are from an 1897 newspaper report (Yokohama-shi Taiikushi Kikaku Kanko Iinkai 1989). This is much later than Kumamoto or Fukuoka, even though the latter two cities are geographically much farther away from the Kodokan than Yokohama. The Yokohama Judo Association was finally only founded in 1933, the seventeenth sports association to be founded in Yokohama. Due to the strong Western influence, the popular sports in Yokohama were the more Western activities such as baseball, tennis, and yachting. Despite judo’s slow start in the city, Yokohama’s relative wealth, along with its proximity to Tokyo, allowed it to recruit influential judo leaders in the 1930s. These efforts paid off in terms of national competition results in the 1930s, when wrestlers representing Yokohama, and Kanagawa Prefecture more broadly, won eight national titles in various weight divisions over a threeyear span from 1934 to 1936. In 1936 alone, wrestlers representing Kanagawa even took four of the eight weight division titles. But it should be noted that this success owed in large part to the recruiting of wrestlers from Kyushu, especially Fukuoka (Kanagawa-ken Judo Renmei 1994: 31). The Kanagawa Prefectural Police recruited a large number of judo stalwarts from the Fukuoka area around this time, including Suekichi Tanaka and Masayuki Nakajima, among others (Kanagawa-ken Keisatsu-shi Hensan Iinkai 1972: 328). Tanaka went on to win the national judo title in his weight group in 1934 and 1937;

Wartime Endorsement and Postwar Repression of a Traditional Art

135

Nakajima also won national titles in his weight group in 1934 and 1935. In short, unlike Fukuoka, Yokohama’s national success was not founded on a strong local base of activity. Nor did it cause one to come into being. In contrast to Kyushu, no concerted effort appears to have been made in Yokohama to encourage judo at the high school level during the prewar period, for reasons that are unclear. This may have led to the relatively little enthusiasm among the youth in Yokohama toward judo. One indication of this is reflected in the surviving photographs from Yokohama Commercial High School, the major athletic school in the area, from the 1930s, which shows only four members of the judo team, compared to swimming (fourteen members) and tennis (about thirty members) (Kanagawa-ken Keisatsu-shi Hensan Iinkai 1972: 126–7). And when the Yokohama Sports Association was founded in 1929, the founding members were the local associations in baseball, tennis, and Alpine climbing, not judo (Yokohama-shi Taiikushi Kikaku Kanko Iinkai 1989: 144). Yokohama during World War II Economically, World War II initially contributed to the rapid growth of the Yokohama region. Major mobilization for war in Yokohama steadily accelerated starting in the late 1930s. For much of the prewar period, in terms of industrial production, the Keihin Industrial Belt, which stretches from Tokyo to Yokohama, had lagged behind the Hanshin Industrial Belt, which extends from Osaka to Kobe. But the Keihin surpassed the Hanshin Belt for the first time in 1939. Toward the end of the war, however, Yokohama incurred heavy damages (Yokohama-shi n.d.: 40; Tamura 1989: 252–253; Imai 2007: 302). A total of more than thirty air raids hit Yokohama by the end of the war. The largest of these came on May 29, 1945, known as the “Great Yokohama Air Raid,” when, according to official figures, 3,650 people were killed, 10,197 were injured, and 100,091 homes, or 48.5 percent of all homes in Yokohama, as well as 34 percent of all structures of any kind, were destroyed (Yokohama-shi Kyoiku Iinkai 1978: 400, 445; Yokohama-shi n.d.: 40; Tamura 1989: 252–253; Imai 2007: 302). In fact, the actual level of damage is said to have been more than double these official figures (Tamura 1989: 252–253; Imai 2007: 302). Seventyseven percent of the city’s railroad tracks also incurred damage (Takamura 2006: 113). In contrast to Fukuoka, where the industrial facilities were the main targets, most of the destruction in Yokohama was concentrated in the residential neighborhoods, perhaps because these raids were intended more to diminish civilian morale than to undermine industrial production (Yokohamashi 1993: 40–41; Takamura 2006: 74). As in Fukuoka, judo also suffered in Yokohama during World War II. Young males, who were the main judo constituency, were increasingly drafted into the military services, and by 1943 judo activities in Yokohama came to a halt.

136

Civic Engagement in Postwar Japan

The Weak Revival of Judo in the Postwar Period General Douglas MacArthur arrived in Japan and initially established his headquarters in Yokohama on August 30, 1945. While the occupation headquarters moved to Tokyo only two weeks later, Yokohama was arguably the most “occupied” of all major cities in Japan during the occupation; according to one source, more than 60 percent of the facilities confiscated by the occupation forces in Japan outside of Okinawa were in Yokohama (Tamura 1989: 257).10 The Allied occupation proceeded to confiscate a large portion of the city’s public facilities, occupying everything from corporate offices and the ports to schools, hotels, and parks. According to one estimate, at one point 90 percent of all port facilities in Yokohama were being used by the United States (Takamura 2006: 85). The large U.S. footprint in the city delayed the urban planning and reconstruction of Yokohama considerably. On the other hand, the arrival of a large number of occupation forces also created jobs for local residents. This economic lifeline was particularly important because the number of factories in operation in Yokohama had fallen from 1,565 at the end of the war in August 1945 to 826 just two months later; and the number of employed workers also fell from more than 229,000 to a little more than 87,000 over the same period (Takamura 2006: 94). Nevertheless, the occupation hired a large number of unemployed workers and at least mitigated the social and economic blows of demobilization. Despite the confiscation of public facilities, the city of Yokohama began slowly but steadily. Yokohama had lost more than one-third of its population during the war to bombings and to evacuation, but by 1952 it had recovered to 1943 levels (Takamura 2006: 151). As in Fukuoka, the outbreak of the Korean War in 1952 heightened demand for manufactured products and further spurred Yokohama’s recovery. In contrast to other major sports, such as baseball, tennis, and yachting, judo also requires no equipment, which would seem to favor a rapid revival in the wake of defeat. In addition, because judo had been endorsed by the military during war, there had been a surge in the number of individuals who had been exposed to judo during the 1930s and 1940s. Potentially, there was thus a vastly expanded market for judo in the wake of war. Moreover, the Kanagawa Judo Association had not lost its building during the war despite the heavy bombing. But its headquarters was confiscated for use by the U.S. occupation between 1946 and 1958. As we have seen in the case of the Kobe YMCA, however, the loss of physical facilities need not pose an insurmountable constraint on the revival of associations. Yet the revival of judo in Yokohama was relatively slow compared to Fukuoka. The weak revival of judo in Yokohama does not appear to owe to a lack of leadership; indeed, in contrast to the case of the Sapporo YMCA, there 10

Many buildings in the city still remain in the hands of Allied forces today (Kawaguchi 1983: 123–126).

Wartime Endorsement and Postwar Repression of a Traditional Art

137

was considerable continuity in leadership across the prewar and postwar periods. Many of the postwar leaders of judo in the area, such as Akira Sasaguchi and Masami Takasaki, had been active in the Yokohama judo scene since the prewar period. Within the Kanagawa Police as well, many of the prewar leaders, including Shigeya Kakisaki and Masayuki Nakajima, also remained into the postwar period. Nevertheless, these leaders appear to have lacked strong linkages to the local community. As with the prewar period, the enthusiasm for judo appears to have been limited to the police forces and did not spread to the rest of Yokohama. Judo within the local police force revived early, around November 1945 (Kanagawa-ken Keisatsushi Hensan Iinkai 1974: 1203–1204). The police, however, mainly appear to have participated in tournaments amongst themselves; in December 1947, for instance, they held a judo tournament pitting them against other prefectural police forces from the Kanto region (Kanagawaken Keisatsushi Hensan Iinkai 1974: 1206). In contrast to the case of Fukuoka, a vicious cycle appears to have transpired, whereby weak demand at the grassroots level slowed the revival of judo associations in Yokohama, which in turn discouraged demand. Indeed, the few individuals who wished to practice judo appear to have traveled to neighboring cities such as Kamakura and Yokosuka (Kanagawa-ken Judo Renmei 1994). The first dojo in Yokohama was not reopened until 1948, when it was built by businessman Soichiro Kondo. And, as previously noted, while in 1968 Yokohama had sixty-eight dojos, Fukuoka had twice that number. This is not to say that the revival of sports in Yokohama generally was slower; to the contrary, by 1946, for instance, the Kanagawa Cycling Federation had already been revived; a Kanto-wide rugby tournament was held during the same year; and the yacht harbor was very active. Baseball associations were quick to revive as well; the Association of Yokohama Middle School Baseball Teams was revived in February 1946, and the Yokohama Baseball Association, as well as the Yokohama Soft Baseball Association, were also revived the following month (Yokohama-shi Taiikushi Kikaku Kanko Iinkai 1989: 425). In short, sports that had been popular in Yokohama before the war became even more popular after it; but sports that had been less popular before the war, such as judo, grew only incrementally in the postwar period. This pattern was not only limited to sports. A comparison between judo and the Yokohama YMCA is revealing. The Yokohama YMCA had been a popular and successful organization in the prewar period, with the third-largest membership in the country as of 1932 (Owaku 1984: 262). The Yokohama YMCA building, like the judo association building, was also confiscated by the occupation forces after the war, but membership in the YMCA still grew rapidly, from 200 members in 1945 to 600 members in 1948 to 1,593 in 1950 (Nara 1959: 370). It sponsored a range of social and recreational activities that included both non-religious activities such as theater and sports as well as religious activities such as bible-reading. The Yokohama Y also temporarily provided office space for the Kanagawa Prefectural Basketball Association and

138

Civic Engagement in Postwar Japan

the Kanagawa Volleyball Association and helped the two associations resume their activities as well (Owaku 1984 306–307). In contrast to the YMCA, the recovery of judo in Yokohama was painstakingly slow. Nevertheless, in light of the general rise in associational activities, judo in Yokohama did grow in the postwar period. As previously noted, a survey conducted in 1968 reports 14 dojos in the city of Yokohama and 37 in Yokohama’s Kanagawa Prefecture. This was a big change from the lone major dojos of the prewar era; but still, in relative terms, after the war Yokohama fell even further behind Fukuoka, which now had 78 (Nihon Taiiku Kyokai 1968: 95). On a per-population basis, Fukuoka had 1.94 dojos per 100,000 population, while Kanagawa had only .68. Again, this is a judo-specific phenomenon. The ratio is reversed, for instance, for the number of tennis facilities; in the same survey, Kanagawa reportedly had 133 tennis facilities (33 in Yokohama proper), while Fukuoka only had 54; this was the equivalent of 2.47 tennis facilities per 100,000 population in Yokohama, compared with 1.34 for Fukuoka (Nihon Taiiku Kyokai 1968: 93). It is possible that the large presence of the occupying forces in Yokohama and their initial opposition toward the Japanese martial arts may have discouraged the rapid recovery of judo in the area. The weakness of judo’s recovery in Yokohama, however, was due more to other factors. For instance, as with Yokohama, Tokyo was also home to a large number of occupation forces, but the occupation forces gradually began to express an interest in judo, and, for instance, wrestlers were even invited to teach judo sessions to members of SCAP who were working in the Ministry of Finance building in Tokyo (Inoue 2004: 183). SCAP officials’ views on judo turned out to be quite malleable, and, sensing this, judo leaders in Tokyo also lobbied hard to promote it. Indeed, as noted earlier, many American soldiers became attracted to judo during their service in Japan. According to one source, the number of foreigners who became new Kodokan members nationwide rose steadily during the occupation era, from 4 in 1945 to 61 in 1946 to 372 in 1949 (Inoue 2004). Thus, the presence of a large number of Allied forces should not have posed an insurmountable obstacle to the local growth of judo. The relative weakness of local grassroots judo tradition in Yokohama provides a more plausible explanation for the relatively low level of efforts to revive it in the postwar era.

Conclusions As with the YMCA, legacies from the prewar period played a key role in shaping the growth of judo in the two cities of Fukuoka and Yokohama in the immediate postwar period. Judo began to thrive in Fukuoka around the turn of the century; it attracted large memberships throughout the prewar period and even larger memberships in the postwar era. Citizens in Yokohama, in contrast, were more drawn to “Western” sports such as baseball, tennis, and yachting than to judo. Yokohama-based wrestlers began to enjoy considerable success in national judo competition during the 1930s, as the Kanagawa Prefectural

Wartime Endorsement and Postwar Repression of a Traditional Art

139

Police recruited wrestlers from Fukuoka, but the growth of judo among the grassroots remained sluggish during the prewar period, and this pattern of merely incremental growth continued into the postwar. Compared to the YMCA, which had Western roots and therefore was repressed during the war, judo was viewed favorably by the wartime regime and survived through the wartime years. Nevertheless, the discrepancy in levels of involvement in judo in Fukuoka and Yokohama persisted, and they were then exacerbated during the immediate postwar period. This is the same pattern seen for the case of the YMCAs in Kobe and Sapporo. Judo in Fukuoka was quick to revive, and willing wrestlers flocked to the dojo at a time when judo leaders hesitated to launch active recruitment drives for fear of offending the occupation forces. Encouraged by this popular demand, leaders from the area not only rushed to revive the judo organizations in the Fukuoka and Kyushu areas but also took the lead in the movement to revive judo at the national level as well. Preexisting personal networks from the prewar period played a crucial role in shaping this process. In Yokohama, in contrast, the revival of judo trailed well behind other sports such as baseball and rugby. Demand for participation in judo may have been present, and indeed some devout judo wrestlers even commuted long distances to practice the martial art, but because judo did not have an extensive following in the local grassroots during the prewar period, the opportunities for participation as well as the information about those the opportunities were few and far between. While neither Fukuoka nor Yokohama judo groups and dojos launched extensive recruitment drives, membership sprang up spontaneously in the former while it saw much slower growth in the latter. The occupation forces discouraged judo nationwide, but the two cities responded very differently and their responses were strongly shaped by the preexisting legacies from the prewar era.

8 Civil Society and Reconstruction in Postwar Japan

Overview1 Until now, this book has focused on highlighting the factors that shape civic engagement in the wake of wars, both within and across countries. This chapter addresses the “so what?” question. Civic engagement grew rapidly in the wake of World War II in Japan, but what were some of the macro-level consequences of this growth? Existing studies of civil society suggest that a growing civic life in Japan should have had important beneficial effects for postwar Japanese reconstruction. How well is this claim substantiated by the empirical evidence? To address this question, this chapter assesses the relationship between Japan’s civil society in the immediate postwar period, on the one hand, and postwar reconstruction, on the other. A voluminous political science literature now examines the factors that determine the success or failure of peace-building, both for civil and for international wars (e.g., Doyle and Sambanis 2006; Fortna 2004; Hartzell et al. 2001; Humphreys and Weinstein 2007; Paris 2004; Werner 1999), but the majority of studies of postwar reconstruction per se consist of in-depth qualitative inquiries of single cases that have been conducted by historians and/or historical sociologists. This work is valuable in its own right, but there is a need for political scientists to build and test a systematic, middle-range theory of the determinants of successful postwar reconstruction. This chapter argues that civil society serves as a key variable in facilitating the society-wide coordination that makes reconstruction work. By generating greater amounts of information, a strong civil society helps to identify the areas of most acute need, to facilitate implementation, and to provide effective monitoring. All of these functions lead to better organization of reconstruction efforts and, ultimately, assuming adequate economic resources, to more 1

The remainder of this chapter draws substantially on Rieko Kage (2010), “Making Reconstruction Work: Civil Society and Information after War’s End,” Comparative Political Studies, vol. 43, no. 2: 163–187. © 2010 SAGE/SOCIETY. Publications DOI: 10.1177/0010414009332128 http:// cps.sagepub.com. All rights reserved. Used by permission.

140

Civil Society and Reconstruction in Postwar Japan

141

rapid societal recovery. Indeed, without a strong civil society, reconstruction efforts are likely to flounder or collapse even when infused with large financial resources. Existing studies of civil society have typically stressed its capacity to serve as a “school for democracy,” where citizens learn to cooperate and to become more civic-minded, thereby improving the overall macro-level health of their societies (Putnam 1994). In recent years, this view of civil society has come under increasing attack as being theoretically and/or empirically inadequate (e.g., Encarnación 2006). But another function of civil society that has been somewhat underappreciated in the political science literature is its capacity for information-generation and -transmission. Sociologists argue that at its root, civil society comprises a dense and complex set of networks among individual citizens (Foley et al. 2001: 12; see also Gibson 2001). This view complements and deepens the typical understanding of civil society within the field of political science. According to the sociological perspective, civil society networks provide crucial information that, in turn, promotes collective action. This function seems particularly relevant for the question of postwar reconstruction. Since more formal communications networks have often been disrupted in immediate postwar environments, and since successful reconstruction is by nature information-intensive, the informational function of civil society may be crucial for explaining why rates of reconstruction may differ markedly from region to region, even when other basic socioeconomic conditions are relatively similar. The informational function of civil society also suggests that a vibrant formal associational sector of the kind typically highlighted in the civic engagement literature, and indeed in prior chapters of this book, may not be the whole story. In addition, extensive interpersonal networks consisting of “weak ties” also effectively transmit information that can facilitate collective action, thereby accelerating the process of reconstruction. The rest of this chapter proceeds as follows. The next section defines the concept of reconstruction. The following two sections develop a theory of reconstruction that stresses the flow of information among citizens through informal networks. The fifth section tests the hypothesis presented in this chapter drawing on newly assembled data from Japan. The final section summarizes and discusses the findings.

Conceptualizing Reconstruction Political leaders, international organizations, and policy analysts increasingly recognize the importance of post-conflict reconstruction in the broader task of peace-building. Reconstruction, it is claimed, may turn peacebuilding from a zero-sum game to a positive-sum game. Post-conflict reconstruction is also increasingly viewed as an important end in itself (World Bank 1998, 2005). International donors pledged more than $100 billion in postwar reconstruction aid across three dozen nations during the 1990s (Forman and Patrick 2000: 1).

142

Civic Engagement in Postwar Japan

But what precisely is “reconstruction”? Much of the recent work on postwar reconstruction has been conducted by policy specialists (e.g., Azimi et al. 2003; Kumar 1997; World Bank 1998, 2005). Existing studies in this field, however insightful, have increasingly conflated the concept of reconstruction with other concepts such as peace-building, state-building, and democratization. This literature typically defines reconstruction as a broad process that involves not only the rebuilding of societies in the wake of war but also the building of new political and social institutions that can prevent future conflicts (e.g., World Bank 1998: 14; Orjuela 2003). For instance, the World Bank, long a champion of postwar reconstruction, defines “reconstruction” as follows: Reconstruction does not only refer to the reconstruction of physical infrastructure. Nor does it necessarily signify a rebuilding of the socioeconomic framework that existed in a country prior to the onset of confl ict. Confl ict, particularly long-lasting confl ict, transforms societies, and a return to the past may not be possible or desirable. … In such cases what is needed is a reconstruction of the enabling conditions for a functioning peacetime society. … (World Bank 1998: 14)

From this perspective, reconstruction involves both reconstructing and constructing anew. Similarly, in Kumar’s view, for instance, the objective of reconstruction “is not to go back to precrisis conditions but to move in a different direction” (Kumar 1997: 2), ultimately “redefining and reorienting relationships between political authority and the citizenry” (Kumar 1997). As with the World Bank’s definition, Kumar’s concept of reconstruction also encompasses both elements of recovery and the introduction of new institutions. For instance, he argues: “[A] war-torn society that lacked adequate legislation to protect human rights needs new legislation for human rights and administrative mechanisms to enforce it” (Kumar 1997: 3). In reality, of course, the process of reconstruction is often inextricably linked with the processes of peace-building and/or state-building; indeed, the success or failure of peace arrangements may, to a significant degree, depend on the success of reconstruction, or vice versa (Doyle and Sambanis 2006: 5). Analytically, however, the three processes of reconstruction, peace-building, and state-building are distinct and should be kept so as well (Fukuyama 2006: 3–8). Moreover, even if the challenges of reconstruction, peace-building, and state-building may often be present simultaneously, each is likely to be present to varying degrees in different contexts. Thus, while it is important to explore how the three tasks may be addressed as a whole, it is first necessary to investigate their determinants in isolation from one another. Thus, this chapter relies on the more traditional definition of reconstruction as “the rebuilding of the physical infrastructure and socioeconomic framework of the society” (World Bank 1998: 14). By examining the issue of reconstruction in isolation from the processes of peace-building and state-building, the chapter hopes to better understand the factors that shape the success or failure of postwar reconstruction per se. This, in turn, may also help to more effectively illuminate the determinants of successful peace-building and/or state-building as well.

Civil Society and Reconstruction in Postwar Japan

143

Reconstruction as a Collective Process Reconstruction, of course, is more arduous the more damage a society has incurred. Even when levels of damage are comparable, however, there is often considerable variation in the speed with which societies recover from war. Existing studies of peace-building implicitly offer two hypotheses regarding the factors that shape the extent of postwar reconstruction: I term these the “economic hypothesis” and the “state-centric hypothesis.” First, the economic hypothesis suggests that, holding levels of damage constant, societies or regions that have higher absolute levels of economic resources at the end of war should recover more quickly than those with less (Fortna 2004; Doyle and Sambanis 2006). Economic resources may come from indigenous or from international sources, such as foreign governments, international organizations (e.g., the World Bank), or non-government organizations (NGOs). Second, the statecentric hypothesis suggests that the presence of a cohesive and autonomous state leads to more effective targeting of needs and implementation, both of which promote more rapid and efficient reconstruction (e.g., Johnson 1982). Of course, the two hypotheses may also work in tandem. Economic resources and a cohesive, autonomous state probably do indeed facilitate reconstruction. But as a process, reconstruction requires much society-wide coordination; it is inherently collective and political in nature. Reconstruction, then, cannot simply be bought by large infusions of cash, and even a strong, authoritative state may not have enough knowledge of the complexities of the production system to be able to program funding efficiently. For example, steel mills may be rebuilt, but the mills will not fulfill their production potential if the energy sector does not provide them with power and if the manufacturing sector does not buy what they produce. Even Japan, typically viewed as the prototypical “strong state” and working with the well-prepared U.S. occupation authorities, failed to allocate raw materials appropriately to key economic sectors in the early postwar years; this failure is often noted as an important cause of Japan’s initially stagnant recovery during the immediate postwar period (Kosai 1986). Coordination poses a particularly difficult challenge in times of political, economic, and social turmoil, as immediate postconflict situations tend to be. In the presence of such high uncertainty, societal actors are likely to choose to reap short-term gains rather than to cooperate with others. For instance, industries may hoard resources, creating a bottleneck in the materials that are needed for recovery in the broader economy. Worse, they may engage in outright corruption. Cases of corruption have in fact been reported in the reconstruction of Cambodia, Mozambique, and Lebanon, among others (Forman and Patrick 2000), and even within the World Bank (Hurtic et al. 2000: 350–353). For reconstruction to work, then, the expectations of various actors within the economy must be coordinated, either by the state or among private actors themselves, and in a manner that expands the time-horizons of all parties. While economic resources are no doubt necessary for effective reconstruction,

144

Civic Engagement in Postwar Japan

they by no means offer a full explanation. And although state authority may play a key role in coordinating societal expectations and behavior toward the long-term goal of reconstruction, there is no reason to assume a priori that strong states will in fact devote themselves to that end. Nor can we assume that even strong, well-intentioned states will have sufficient information to make the right choices on behalf of their societies.

Reconstruction, Information, and Civil Society A strong civil society can provide crucial information that facilitates the coordination of reconstruction efforts among key societal actors. As noted earlier, this informational function of civil society has often been pushed into the background in the field of political science by the emphasis on civil society as a “school for democracy,” but information-provision has long been the focus of the civil society literature within the field of sociology (Edwards and Foley 2001). Specifically, a strong civil society allows societal actors to coordinate perceptions regarding areas of the most urgent needs, the efficient delivery of needed resources, and the monitoring of implementation, all of which are essential pillars of the reconstruction process. Indeed, much of what has been argued in the existing literature regarding social capital and civil society is implicitly underpinned by this informational dimension of civil society. Putnam (1993), for instance, argues that in societies with higher levels of social capital, interactions are seen in the context of repeated games rather than one-shot games. But this function of social capital depends on dense networks of individuals as well as associations efficiently transmitting information regarding free-riders, and ultimately leading to effective punishment of such behavior. Moreover, the expectation that information regarding violators would become readily available often preempts the need to exercise those sanctions at all. Again, the notion of “civil society” as conceptualized here is much broader than the formal voluntary associational sector typically emphasized in recent scholarship and in earlier chapters of this book. Voluntary associations surely do provide the functions that we highlight here. But formal voluntary associations comprise just one component of civil society, and perhaps not even the most important one in many non-Western contexts (Haddad 2006). After all, most activities that occur in civic life are not formally organized ones. As Sandel (1996) argues, what is necessary for communities to thrive is for them to be connected, whether those connections are formal or informal. From this perspective, this chapter follows Gibson (2001) and tries to look beyond formal associations to institutionalized social networks that are composed mostly of “weak ties.” Some of these networks take the form of formal membership associations, but others may be much more informal in nature. Of course, there probably is a correlation between the degree of formal associational life and the extent of informal social networks, particularly if one

Civil Society and Reconstruction in Postwar Japan

145

is looking longitudinally at the evolution of civic engagement in a particular country. But it is the latter that is doing the theoretical heavy lifting when it comes to explaining postwar reconstruction. The key lies not so much in the formality of the ties but rather in their efficacy as transmitters of information. As the seminal work of Granovetter (1973) argues, weak ties serve as powerful channels for disseminating information. Individuals who are connected through “strong ties” (as between family members, for instance) are likely to share acquaintances, while those who are linked through weak ties are less likely to have large numbers of acquaintances in common (Granovetter 1973: 1362). Information that is transmitted across strong ties is of course also valuable, but people’s strong ties are relatively few and thus likely to stay within the same social networks, whereas information that travels via networks of weak ties is more likely to be dispersed broadly to different segments of society. Recent studies have largely confirmed Granovetter’s initial insight that weak ties are much more effective than strong ties at diffusing information across broad segments of a society (Diani 2003: 8). In short, the key to promoting collective action lies not so much in the existence of voluntary associations per se as in the presence of informational networks among individuals. Formal voluntary associations play an important role in shaping and reinforcing these networks, but they are just the tip of the iceberg. This is particularly true in societies that have been ravaged by war. War may have destroyed the crucial physical infrastructure of a society, such as communication, transportation, and utilities, but interpersonal networks of weak ties are usually much more resilient (Elder and Clipp 1988). This chapter argues that information that flows through interpersonal networks in the wake of wars serves to coordinate perceptions regarding where resources may be most needed, speeds the delivery of resources to those areas, and monitors the delivery of resources. Without a large amount of information being shared among a broad group of actors within society, these crucial pillars of reconstruction will not materialize. Identifying Needs. Before societies can successfully embark on the task of reconstruction, key actors must share perceptions regarding where the most critical needs lie. Frequently, the needed information may lie beyond firms, the state, or the mass media, especially because war may have destroyed the capacities of such actors to collect and disseminate information. But local people do know what their needs are, and where societal actors are better connected by interpersonal networks, information regarding the most acute needs can travel more efficiently to those who are in a position to address those needs. The importance of identifying needs in the wake of disaster was shown quite vividly, for instance, in the Kobe earthquake of 1995, when national and local governments had no shortage of relief items but where damage to the communications and transportation systems made it impossible for them to identify where those items were most needed. While that was a case of a natural disaster as opposed to wartime destruction, the situation is analogous to postconflict reconstruction.

146

Civic Engagement in Postwar Japan

Implementation. Once actors come to share basic perceptions regarding areas of most crucial need, the next stage is implementation: the provision of resources to those needed areas. A strong civil society, and the information that flows through it, can play a key role at this stage as well. If actors are better informed about what others have done or are currently doing, they can build on each other’s efforts or even actively cooperate. As noted earlier, this coordinating function of civil society is particularly important in times of high uncertainty, because otherwise actors are likely to simply pursue their short-term interests. Indeed, a well-connected society may even generate more resources than those that are less well connected, because information regarding the availability of various resources may be more efficiently transmitted to responsible agents. The case of Japan again highlights the importance of broad societal coordination in the implementation of reconstruction. As Okimoto (1988) notes, the Japanese business sector is characterized by dense networks of communication. The relationship is one of generalized reciprocity nested within contexts of long-term relationships. Yet even in Japan, bottlenecks often became a problem within the Japanese economy as it sought to rebuild from the destructions of World War II until well into the mid-1950s (Kosai 1986). This underscores the importance of information in the successful implementation of reconstruction; if there was almost not enough coordination among actors even in a relatively well-connected society such as postwar Japan, the informational requirements for successful implementation are considerable, and extensive networks of communication must be present to provide the needed information. Monitoring. Civil societies not only can help to coordinate the reconstruction efforts of different societal actors ex ante, but they can also monitor the progress of reconstruction-related measures as they are being undertaken, whether by the state or by private enterprise. Because networks generate information, networked actors can closely monitor other actors within society to make sure that their behavior conforms to initial expectations; and if it does not, they can choose either to adjust their own behavior to compensate and/or demand that the other actors alter theirs. In particular, where actors are better connected, information regarding corruption and mismanagement – typical banes of reconstruction projects – disseminates quickly. When they learn of these abuses, various interest groups, either formal or informal, may in turn lobby governments or mobilize public opinion against the offending party. But perhaps even more importantly, societal actors directly involved in the reconstruction effort may also choose to impose sanctions of their own. In anticipation of this effect, actors who are engaged in reconstruction work in such densely networked societies are likely to complete their projects more efficiently and effectively than actors who conduct similar work in less connected societies. It should be noted here that this study does not seek to suggest that all information flowing within and across these networks is equally accurate or wellintentioned. This chapter does not subscribe to the oft-implied idea that civil society is inherently virtuous and/or socially beneficial. Scholars are correct to

Civil Society and Reconstruction in Postwar Japan

147

note that civic networks may just as easily contribute to antisocial activities, such as mafia operations, as to activities that serve the broader public good (Encarnación 2006). In societies where citizens are well-connected, as in those where they are less, there is bound to be a subset of citizens and groups who attempt to manipulate information to serve their own personal ends. Where citizens, governments, and firms have access to greater amounts of information, however, ceteris paribus, actors should be better equipped to assess the validity of each piece of information. Where flows of information are limited or suppressed, as in dictatorial societies, actors may be more prone to acting on misinformation. The argument being presented here also does not contradict the state-centric hypothesis (Johnson 1982) referred to earlier, but rather complements and nuances it. States in the immediate aftermath of war, especially but not only if the war ended in defeat, are often quite weak. Therefore, the ability to control not only other societal actors but even their own agents may be at a low ebb. In this situation, a strong civil society can strengthen the state. As Tsai (2007) has recently argued, informal bottom-up networks of accountability may promote the provision of public goods by government officials, even in authoritarian contexts. Moreover, information that flows through interpersonal networks cannot only constrain the state to provide assistance where it is needed, but also do so in a way that is well-coordinated with the behavior of other societal actors as well. In sum, more vibrant, well-connected civil societies generate and disseminate more information, which in turn helps to accelerate the reconstruction process by shaping shared perceptions among societal actors regarding the most urgent needs, speeding the delivery of appropriate resources, and monitoring the progress of the work for any deviations from original expectations. Although this argument does not deny the validity of the traditional view that greater amounts of economic resources and autonomous states certainly are key to reconstruction, it also points to the crucial role that civil societies play in generating the information without which even well-endowed states’ reconstruction efforts will falter. In short, ceteris paribus, societies that have higher levels of economic resources, stronger governments, and dense networks of “weak ties” among individuals should see more rapid postwar reconstruction; conversely, societies where at least one or more of these factors is missing should see considerably slower rates of reconstruction.

Testing the Hypothesis The remainder of this chapter tests the theoretical framework laid out in the previous section using quantitative analysis, drawing on the case of post–World War II Japan. Japan in the wake of World War II presents a particularly useful empirical test of the question posed in this chapter for at least three reasons. First, Japan experienced exceedingly high levels of damage, by many scales following only the USSR and Germany (Ellis 1993). This makes Japan an important case to study.

148

Civic Engagement in Postwar Japan

Second, Japan exhibited substantial variation in the extent to which the country’s different prefectures recovered from damage over the initial decade after the war. Often, the Japanese reconstruction story is framed in international terms: Generous reconstruction aid flows in from the U.S. government, international organizations such as the World Bank, and philanthropic organizations such as the Ford Foundation (Yamamoto et al. 2006), and, as a result, “Japan” recovers. While this frame is legitimate, it overlooks considerable disparities in the pace and extent of reconstruction within Japan over the first postwar decade. The fact of Japan’s uneven reconstruction suggests that domestic and, indeed, local variables represent a crucial element of the Japanese reconstruction story. Third, another relatively unique characteristic of the case of Japan in relation to many of the postconflict societies is that it was a historically strong nation-state relatively free from ethnic or religious strife defeated by a clearly external enemy. As argued at the outset of this chapter, it is important to distinguish between reconstruction, on the one hand, and peace-building and statebuilding, on the other. In this sense, the case of Japan is valuable precisely because domestic peace-building and state-building were largely non-issues. This allows the analysis to focus narrowly on the sources that shape the relative success of reconstruction narrowly defined. The Dependent Variable: Reconstruction in Forty-Six Japanese Prefectures To begin with, how might one operationalize the concept of reconstruction, the dependent variable in this study? As noted earlier, this chapter defines reconstruction as “the rebuilding of the physical infrastructure and socioeconomic framework of the society.” Following the United Nations, it focuses on four aspects of material reconstruction: employment, health, education, and culture (Azimi et al. 2003). Specifically, the study uses four measures: number of jobs, number of hospitals, number of elementary schools, and number of library books.2 Jobs, hospitals, schools, and libraries represent some of the most basic aspects of the social infrastructure. Admittedly, this operationalization does not compare the possible change in the quality of the jobs or of services provided by schools, hospitals, and libraries between the prewar and postwar periods. Nevertheless, before one can begin to discuss the quality of social infrastructure, these quantitative measures represent an important starting point. Both public and private institutions were included. A second issue concerns the question of periodization: For how long are countries considered to be “in reconstruction,” and at which point do they 2

Conceivably, focusing on libraries themselves, rather than library books, could provide a better measure of physical infrastructure. However, the surviving data on the number of libraries in Japan are highly uneven and unreliable. Available data for the number of library books are much more reliable and therefore will be used for purposes of this study.

Civil Society and Reconstruction in Postwar Japan

149

move beyond it? There is no clear-cut rule for delineating the period of recovery from the period of postrecovery. To some degree, the duration of the period of recovery may also be subjective. However, it should certainly be true that the farther one gets from the end of the war, the less one may be said to be rebuilding from it. This chapter measures reconstruction at ten years after the end of the war, that is, as of 1955. Reconstruction on each of the four dimensions was calculated as follows. By definition, reconstruction is a relative concept that captures the extent of recovery relative to the level of damage. It is not sufficient, for instance, to simply measure the growth in the number of jobs over the initial decade after the end of the war, as such a measure does not capture the severity of destruction prior to postwar growth. In this study, the extent of damage was operationalized as the difference in levels of employment, education, health, and library holdings between the immediate prewar period, 1940, and the immediate postwar period. I chose 1940 as the prewar starting point since, while Japan was engaged in war in China throughout the 1930s, full-fledged mobilization in Japan did not begin until late in 1940 (for hospitals, the chapter uses data from 1937 as these are the last available figures from the prewar period). The extent of recovery was operationalized as the difference in the levels of employment, education, health, and library holdings between the immediate postwar period and ten years after the war. I then created an index of reconstruction by taking the ratio of the two measures. Thus, the Index of Reconstruction was calculated as R = (% change in Yi during recovery phase) / (% change in Yi during damage phase) where Yi consists of numbers of jobs, hospitals, schools, or library books, respectively.3 For instance, if the number of schools declined by 20 percent from the start to the end of the war and then increased by 20 percent between the end of the war and 1955, the Index of Reconstruction would be calculated as R = 120/80 = 1.5.4 Intuitively, the larger the value of R, the greater the extent of reconstruction. Note that, as in the example given previously, the score may be greater than 1 and still not represent a full return to the prewar level. 3

4

An alternative way to measure reconstruction would be to use the numbers of jobs, schools, hospitals, and library books by unit of population. However, the population of postwar Japan surged as a result of a major baby boom. Thus, this study focuses on absolute levels rather than numbers per population since the former is more likely to represent a more accurate measure of reconstruction. In this example, it is conceivably possible to calculate the index of reconstruction as .2/.2 = 1. However, the problem with this operationalization is that it does not allow for the distinction between prefectures that gained in jobs, schools, hospitals, or library books during the war and then lost them after the war, on the one hand, and those that gained during the war and then lost; the former would divide a negative by a positive, while the latter would divide a positive by a negative.

150

Table 8.1. Descriptive Statistics for Damage and Recovery, by Prefecture: Jobs, Hospitals, Schools, and Library Books Damage

Max Min Mean Standard error

Index of Reconstruction

Jobs

Hospitals

Schools

Library Books

Jobs

179.3 40.7 98.9 28.22

231.3 42.6 104.4 35.46

339.0 45.8 108.3 54.99

233.7 18.8 99.6 48.57

7.02 0.98 2.12 1.084

Hospitals 6.31 0.49 2.28 1.294

Schools

Library Books

2.27 0.32 1.15 0.404

9.98 0.09 2.03 2.191

Note: Figures for damage are as of 1947, except for elementary schools, which are from 1946, calculated on the basis of 1940 = 100 in each prefecture for jobs, schools, and library books and 1937 = 100 for hospitals. Index of Reconstruction is percentage change between 1946 and 1955 for hospitals and schools and 1947 and 1955 for jobs and library books divided by damage (in the left-hand columns). Refer to the text for precise operationalization. N = 46. Source: Somucho (1987, 1988c).

Civil Society and Reconstruction in Postwar Japan

151

A note is in order regarding the data on damage and recovery. Periods immediately following wars are not the best times for data collection, particularly in cases of defeat. In Japan as well, postwar data for schools was not available until 1946; and for jobs, hospitals, and library books, they were not available until 1947. Had the data been available, it would, of course, have been desirable to use data as of the end of the war, that is, August 1945. While less than ideal, it is also not entirely farfetched to assume that figures as of 1946/47 should be highly correlated with figures as of the summer of 1945. Thus, in assessing the levels of damage, this chapter compares the numbers of jobs, hospitals, schools, and library books as of 1946/47 against 1940 levels (except for hospitals, for which the last available figures are from 1937). Similarly, in measuring recovery, 1955 levels are compared with 1946/47 levels. It should be emphasized that since the figures are in total numbers, they may obscure the true extent of wartime destruction. For instance, a potentially large segment of schools, hospitals, and libraries were partially, or even substantially, damaged but were nevertheless considered still standing in light of the general devastation at the end of the war. Similarly, figures for jobs may include many workers who were formally employed but who were nevertheless significantly underemployed. Nevertheless, these figures may still represent a useful proxy for the relative level of damage from prefecture to prefecture. The overall levels of damage and recovery that Japanese jobs, hospitals, schools, and library books incurred, respectively, are presented in Table 8.1. 1) Damage. The first four columns of Table 8.1 present the number of jobs, hospitals, schools, and library books per unit of population across the fortysix Japanese prefectures at the end of the war, relative to the prewar period. Figures are calculated as a ratio of 1940 = 100, except for hospitals, which are in terms of 1937 = 100. At first glance, the level of destruction does not appear to be very large; the number of jobs and library books indicate a very small drop compared to 1940 levels, at 98.9 and 99.6 percent of 1940 levels, respectively, while the number of hospitals and elementary schools actually appear to increase, at 104.4 and 108.3 percent of prewar levels, respectively. But it is important to note that such stasis at the level of national figures represents a significant change from the prewar pattern of dynamic growth. Moreover, the national figures also mask substantial variation across the forty-six Japanese prefectures. Damage tended to be concentrated in the more urbanized prefectures, where Allied bombing was the heaviest, but there were also large differences in levels of change among the rural prefectures as well. For elementary schools, for instance, Shimane, a rural prefecture, had only 68 percent of the number schools as in 1940, while Kagoshima, another rural prefecture, saw the largest increase, with the number of schools rising by 339 percent. Note that this large discrepancy across prefectures is not simply due to a small number of outliers; seventeen prefectures lost schools between 1940 and 1946, while twenty-seven gained and two remained the same. This general pattern of a mix between gainers and losers is repeated on the other dimensions of

152

Civic Engagement in Postwar Japan

reconstruction. Twenty-two prefectures lost jobs over the course of the war, while twenty-four actually gained. Similarly, twenty-two prefectures lost hospitals between 1937 and 1946, while twenty-four gained. And twenty-two prefectures lost library books between 1940 and 1947, while twenty-four gained. 2) Recovery. The right-hand side of Table 8.1 shows the large variance in Japan’s recovery from World War II both across regions and across different measures of recovery. Nationally, recovery was fastest for hospitals and slowest for elementary schools. Between 1947 and 1955, for instance, the number of employed workers doubled, from just over 2.5 million to just over 5.1 million, a 100 percent increase. This amounted to a growth from 55,949 workers per prefecture in 1947 (standard error = 53,719) to 111,550 in 1955 (standard error = 135,953). All forty-six prefectures gained in number of jobs over the course of this decade. Between 1946 and 1955, the number of hospitals in Japan also almost doubled, increasing 87.7 percent from 2,727 to 5,119, or an increase from 59.3 per prefecture in 1946 (standard error = 55.3) to 111.3 in 1955 (standard error = 87.0). Between 1946 and 1955, the number of elementary schools in Japan increased 8.3 percent, from 20,514 schools nationwide to 22,225. This was the equivalent of a rise from an average of 446.0 schools per prefecture in 1946 (standard error = 266.4) to 483.2 in 1955 (standard error = 306.3). Finally, between 1947 and 1955, library books in Japan fell from a total of just over 14.2 million to just over 12.1 million, or a loss of about 15.5 percent. This was the equivalent of a change from a little more than 308,800 books per prefecture in 1947 (standard deviation = 305,669) to just under 265,000 books in 1955 (standard deviation = 259,024). On average, prefectures only gained back 50.0 percent of the books that they lost between 1940 and 1947 (standard deviation = 170.6). However, these figures are influenced by a few outliers; sixteen of the twenty-two prefectures that lost books between 1940 and 1947 actually gained in number of books during 1947–1955, while a small number of prefectures lost a very large number of books. (There may be a few data issues here.) Correlations: Jobs, Education, Health, and Culture It should be noted here that the measures for damage and reconstruction for jobs, schools, hospitals, and library books are not highly correlated. This is shown in Table 8.2. Table 8.2 shows that the prefectures that lost the most schools during the war were not necessarily those that also lost the most numbers of jobs, hospitals, or books, and vice versa. This should not be particularly surprising; indeed, cities also often incur heavier damage in some neighborhoods than in others. Similarly, prefectures that recovered more quickly in terms of numbers of books, for instance, were often not also those that recovered most swiftly in terms of jobs, hospitals, or schools. The low levels of correlation across different dimensions of reconstruction suggest that it is more appropriate to

Civil Society and Reconstruction in Postwar Japan

153

Table 8.2. Correlations across Measures Damage Jobs:Schools Jobs:Hospitals Jobs:Library Books Schools:Hospitals Schools:Library Books Hospitals:Library Books

−.107 .066 −.096 −.144 −.193 .335

Recovery −.061 .191 −.135 −.041 −.110 .336

Note: Figures are Pearson’s correlation coefficients. Source: See Table 8.1.

analyze each measure of reconstruction separately rather than to construct a single, composite measure of reconstruction for each prefecture. This will be the approach used in the rest of this chapter. Quantitative Analysis This section tests the theoretical propositions advanced in this chapter. a) Dependent Variables. The dependent variables in this study are the Indices of Reconstruction for the numbers of jobs, schools, hospitals, and library books, as presented previously. b) Independent Variable. This chapter has argued that neither economic resources nor a “strong state” offer a sufficient explanation for successful postwar reconstruction. Successful reconstruction also requires a vibrant civil society, which, by transmitting more information, allows the state as well as societal actors to address the challenges of reconstruction in a more effective manner. But a vibrant civil society without resources will not be able to take reconstruction very far. In essence, then, this argument points to the crucial role of both economic resources and the amount of information that flows across interpersonal networks. (The strength of the state is also likely important, but for this single-country study there is not sufficient variation in state strength at the national level to test that proposition.) Testing for this kind of argument requires an interaction term. Thus, the main independent variable for this study is a multiplicative interaction term between levels of economic resources and levels of interpersonal communications in society. How can the levels of such communications be operationalized? Until now, this study has used levels of membership in voluntary associations as the measure of civic engagement. Unfortunately, however, as noted in Chapter 2, good measures of overall figures for levels of civic engagement for the immediate postwar period are not available for the period under study. Moreover, this

154

Civic Engagement in Postwar Japan

chapter has attempted to push beyond the focus on formal voluntary associations. An alternative measure needed to be devised. The literature on modernization pioneered an alternative measure of the strength of civil society with its many studies on “communication development.” Along with economic development, rising levels of literacy, and urbanization, the modernization literature avidly measured the levels of communication flows within society, which it viewed as an important component of development. Typically, these studies employed some combination of measures of newspaper readership, radio access, and volumes of postal mail to measure of the density of communication (e.g., Cutright 1963; Pye 1963). Among these measures, mail is most crucial for this chapter as only it incorporates not only the transmission of information from central societal institutions down to the grassroots, but also from grassroots to grassroots as well as up to the decisionmaking institutions. In an era when the television and the telephone were still luxury items in Japan, postal mail was the primary means of interpersonal communication beyond the neighborhood level. (It may also be pointed out that, as Skocpol [1997: 460– 461] notes, the rise of extensive distribution of newspapers also often depended on the availability of a well-developed and well-functioning postal mail system, so the mail variable actually captures much media traffic as well.) This chapter thus uses mail volumes, standardized by population, as the measure of the levels of interpersonal communication, which is the hypothesized mechanism for the impact of civil society on postwar reconstruction. For Japan during the immediate postwar period, surviving documents of the Ministry of Communications provide data starting in June 1948 on how much regular postal mail was sent from each prefecture (Teishinsho 1949: 68–69). Admittedly, mail volume captures not only private but also official communications. However, it is fair to assume that the proportion of “official” mail was relatively small across different prefectures so overall mail volumes may still serve as a useful proxy for the level of interpersonal communications. We also do not know how much of the mail that was sent stayed within the prefecture, and how much went out to other parts of Japan. But this is not debilitating; what we are interested in here is the strength of civil society, and in most cases the degree of activity is a good measure of that strength. If the hypothesis presented in this chapter is valid, then, ceteris paribus, prefectures that have both higher volumes of mail and higher levels of income per capita should see more rapid reconstruction. Conversely, prefectures that have only one or the other, or neither, should see less swift recovery. The amount of mail may also reflect levels of economic activity, so that the more economically prosperous a prefecture, the more mail may be sent by its residents. This view assumes that much of postal mail is business-related. In order to probe this possibility, this study calculated the correlation between industrial output per capita as of 1947 and mail volume per capita as of 1948, and the correlation coefficient was found to be only .48. This is not an exceedingly high level of correlation and thus reinforces the notion that mail can be used as a proxy for interpersonal communication rather than simply economic activity.

Civil Society and Reconstruction in Postwar Japan

155

Figures for mail volume were standardized by dividing the mail volume by the population of each prefecture as of 1948. These figures were then multiplied by industrial output per capita figures from 1947 to produce the key interaction term to be tested in this study. As with the figures for damage and postal mail, industrial output per capita figures are unfortunately not available for 1945, but the first postwar figures at the prefectural level are available from 1947 (Somucho 1988a: 414–418), so these are the figures that are used here. Here again, the assumption is that income per capita levels for 1947 should be highly correlated with 1945 levels. c) Control Variables. A number of control variables were also included in the regressions. 1) Wartime damage. Ceteris paribus, it may be expected that the higher the absolute levels of destruction, the slower the postwar reconstruction may be. Damage was measured using the same measure as presented in Table 8.1. 2) Numbers of jobs, schools, hospitals, and library books in the immediate prewar period. While wartime damage may affect levels of reconstruction, the absolute numbers of jobs, schools, hospitals, and library books prior to destruction may also influence the rate of recovery. That is, prefectures that had more jobs, schools, hospitals, or books before the war may also recover faster in the wake of war. Figures are from 1940, except for hospitals, which are from 1937. 3) Industrial output per capita as of 1947. This variable allows us to test for the economic hypothesis that the amount of resources available at the end of the war should exert a strong effect over a region’s subsequent levels of recovery. Figures used are in millions of yen per population and are from Somusho (1988a: 414–418). 4) State size. This variable allows for the indirect testing of the statecentric model. To test for the hypothesis that state size may accelerate postwar reconstruction, this study includes the sum of the annual prefectural government expenditures between 1947 and 1954, in millions of yen (Somusho 2007), standardized by population. d) Results. For each of the four measures of reconstruction, the study used OLS regression to test four models: the economic model, the statecentric model, the social networks model, and the interactive social networks–economic resources model. The results of the regression analyses for all four dependent variables are provided in Table 8.3. For all four dependent variables, Model 1 shows the results of the regression for the resource-centric hypothesis, that is, that the level of resources at the end of war should strongly affect the rate of postwar reconstruction. Model 2 shows the results for the state-centric model. This tests the hypothesis that larger budgetary expenditures should lead to swifter recovery. Model 3 shows the results for the social networks model, which tests the hypothesis that social networks and economic resources should affect

Civic Engagement in Postwar Japan

156

Table 8.3. The Determinants of Reconstruction in Japan: Jobs, Hospitals, Schools, and Library Books Jobs

Hospitals

Model 1 Model 2 Model 3 Model 4 Mail Volume per Capita* Industrial output per Capita Mail Volume per Capita Industrial output per Capita

Model 1

Model 2

Model 3

6616.36***

150.59

40.47

(71.637) (30.654) Budget, .00 1947–1954 (.180) (per Capita) War Damage −.03*** −.02*** (per (.005) (.004) Capita) Prewar Levels −.01 −0.01 (.007) (.005) Constant 4.41*** 4.36*** (.526) (.683) N 46 46

Model 4 6051.71***

19.63 (13.01) 58.04

(1508.766) −15.50* (9.066) −450.21*** −13.80

43.071

(127.499)

−4.72

−11.37 (.646) 20.20

(1831.047) −27.55** (11.710) −289.62***

(53.976)

(62.303) −.18 (.193)

(8.752)

(172.892)

−.03*** −.03*** (.004) (.004)

−.03*** (.006)

−.03*** (.006)

−.03*** (006)

10..03*** (1.224)

−.01 (.008) 3.15*** (.959) 46

−1.23 (.786) 6.03*** (.773) 46

−1.23 (.848) 6.51*** (.869) 46

−1.24 (.895) 6.62*** (.969) 46

−2.47*** (.629) 4.23*** (.849) 46

.00 (.009) 5.58*** (.738) 46

Entries are OLS coefficients. Standard errors in parentheses. ***p ≤ .01, **p ≤ .05, *p ≤ .1.

levels of reconstruction independently from one another. Model 4 shows the results of the regression when an interaction term between interpersonal networks and postwar levels of economic resources is included in the model. This tests the hypothesis that networks and economic resources should combine to shape the relative success or failure of postwar reconstruction. First, Table 8.3 shows that, in all four models, the coefficient for damage is negative and significant at the 1 percent level, suggesting that higher levels of damage should, ceteris paribus, slow the rate of recovery in levels of employment. However, neither economic resources alone nor prefectural budgetary expenditures appear to exert a significant effect over recovery in levels of employment. The coefficient on industrial output per capita as of 1947 (Model 1) and budgetary expenditures per population (Model 2) are both statistically insignificant at the 10 percent level. Networks alone also do not appear to affect levels of reconstruction in a statistically significant manner (Model 3). In Model 4, by contrast, the coefficient for the interactive effect between the density of networks and economic resources is positive and significant at the 1 percent level.

Civil Society and Reconstruction in Postwar Japan

157

Table 8.3 (continued) Schools Model 1

Model 2

Model 3

Library Books Model 4

Model 1

Model 2

Model 3

2017.07***

−2.14 (3.696) 26.77**

(635.753) −13.04** (5.505) −120.95**

Model 4 10504.09*

132.47

80.77

12.29 (20.050) 118.16

(5818.771) −31.46 (31.112) −623.78

20.09

24.22

(13.528)

(17.534) −.05 (.06)

(14.316)

(46.784)

(113.970) (72.206) .07 (.432)

(117.193) (426.519)

.00*** (.001)

.000*** (.001)

−.00*** (.001)

.00*** (.001)

−.04*** (.005)

.03*** (.007)

−.04*** (.006)

−.04*** (.005)

1025.32*** (307.301) 1.12*** (.198) 46

1090.60*** (305.141) 1.24*** (.246) 46

1054.14*** (319.691) 1.24*** (.250) 46

1269.65*** (308.286) 1.84*** (.347) 46

.00 (.000) 5.77*** (.566) 46

.00 (.000) 5.26*** (1.459) 46

.00 (.000) 5.09*** 1.254 46

.00** (.000) 8.50*** (2.252) 46

This suggests that, ceteris paribus, prefectures with both denser interpersonal networks as well as higher levels of economic resources at the end of war were likely to see faster recovery in employment than otherwise.5 (Note that the fact that the coefficient for the mail volume variable on its own in Model 4 is negative does not refute the validity of this model; when a multiplicative interaction term is included in a model, the coefficients for the variables that constitute the interaction term cannot be interpreted on their own [Braumoeller 2004]). Second, for hospitals, Models 1, 2, and 3 fail to show statistically significant effects for economic resources or budgetary expenditures. By contrast, the coefficient on the interaction term in Model 4 is again positive and significant at the 1 percent level. Thus, ceteris paribus, prefectures that are endowed with 5

The relatively large size of the coefficient, at more than 6,000, owes to the fact that the figures for mail volume, when standardized by population, are relatively small, with a mean of .7 pieces of mail per resident. This relatively small figure owes in part to the limited capacity of mail service and in part to war damage. The figure becomes even smaller when it is multiplied by income per capita, which, in millions of yen per population, is also quite small; the mean for the latter figure is .003 and the standard deviation is also .003.

158

Civic Engagement in Postwar Japan

dense civil society networks as well as higher levels of economic resources at the end of war are more likely to recover faster in terms of numbers of hospitals, as with the employment regressions. Model 4 for hospitals does reveal one difference vis-à-vis the regression for jobs. The effect of wartime damage on the rebuilding of hospitals is found to be positive; that is, higher levels of damage appear to lead to faster, not slower, rebuilding. This may be due to the particular nature of hospitals; regions that suffer more damage may face higher demand for medical services, leading to more rapid reconstruction of hospitals. Third, for elementary schools, Models 1, 2, and 3 again fail to lend support to the economic, state-centric, and network hypotheses. The coefficients for industrial output per capita as of 1947 (in Model 1), budgetary expenditures per capita (in Model 2), and mail volume per population (in Model 3) are all statistically insignificant. By contrast, in Model 4, the coefficient for the interaction term between mail volume per population and industrial output per capita is again strongly positive and significant at the 1 percent level. As with employment, but in contrast to hospitals, the coefficient on the damage variable is negative and significant at the 1 percent level; that is, higher levels of damage appear to be associated with slower postwar reconstruction. Finally, for library books, Models 1, 2, and 3 once again fail to provide credence to the economic, state-centric, or network hypothesis. The analysis again finds strong effects for the interaction term between the density of interpersonal networks and the availability of economic resources at the end of war. In addition, as with employment and schools, the coefficient on wartime damage is negative and highly significant, suggesting that, ceteris paribus, higher levels of damage lead to slower rates of recovery. In sum, the regression analyses provide strong support for the hypothesis advanced in this chapter, that, ceteris paribus, economic resources combined with dense interpersonal networks substantially accelerate a region’s pace of postwar reconstruction. Meanwhile, they lend little support for the hypotheses that, on their own, higher levels of economic resources at war’s end produce swifter recovery, or that more state spending leads to more rapid recovery. They also disconfirm the hypothesis that dense social networks alone may promote more effective reconstruction. Analyses for the four dependent variables also find that, ceteris paribus, higher levels of destruction were usually detrimental to reconstruction. This finding is consistent with what one would normally expect, since higher levels of destruction should, in general, lead to slower reconstruction. The only exception was for hospitals, which appear to be rebuilt faster in areas that experienced more extensive damage. This is also not surprising; indeed, destruction may not only generate heightened demand for medical help in the short run but may also cause various chronic conditions that require professional care over the long run, leading to higher demand for hospital services over the medium to longer term as well.

Civil Society and Reconstruction in Postwar Japan

159

Discussion This chapter has argued that levels of civic engagement among citizens in the immediate postwar period have important consequences for the success of societal efforts to rebuild from the ashes. Reconstruction requires actors to share information as to where the most pressing needs may lie, how those needs may be addressed, and then to coordinate implementation. Economic and statecentric explanations provide at best a partial account of this crucial pillar of reconstruction efforts. Civil society helps to fill this theoretical gap. Networks among citizens help to shape shared perceptions regarding needs, to coordinate reconstruction efforts, and to monitor each other’s activities, even when formal channels of communication have been compromised or destroyed by war. Quantitative analysis using data from Japan confirmed the validity of the framework presented in this chapter, at least in that case. The chapter showed that, ceteris paribus, a combination of higher levels of industrial output per capita and higher levels of communication among citizens consistently led to faster reconstruction, whether in terms of employment, hospitals, schools, or library books. If the analysis presented in this chapter is valid, that more extensive communications among citizens in the immediate aftermath of World War II served to accelerate the reconstruction of Japan, then it also makes sense to conjecture that the rapid rise in civic engagement in postwar Japan (as documented in Chapter 2), spurred on both by the experience of wartime mobilization as well as legacies of prewar civic activities, should also have played an important role in accelerating the rebuilding of Japan in the wake of World War II as well. In sum, although reconstruction cannot take place without economic resources, providing material resources may only provide part of the answer for promoting the recovery of war-torn societies. Encouraging effective statebuilding, by itself, may also be insufficient. Actors seeking to assist societies in the task of physical recovery from war may also need to facilitate communications among societal actors, as well as between those actors and the state. There is some preliminary evidence that this relationship between high levels of communications and more rapid postwar reconstruction holds not only in the case of postwar Japan but also in other contexts as well. For instance, the Netherlands also incurred devastating levels of damage but recovered relatively swiftly during the initial decade after World War II (Kage 2009). The Netherlands had a high level of civic engagement dating back to the prewar period, so, although it did not see the large-scale mobilization that produced Japan’s postwar civic boom, its pre-existing networks were dense enough to carry it through the postwar period. Moreover, recent studies suggest that the insights of this chapter may be relevant beyond simply postwar reconstruction but also for understanding recovery from manmade and natural disasters more generally. Aldrich (2010b) finds that neighborhoods that were endowed with higher levels of social capital

160

Civic Engagement in Postwar Japan

recovered faster from the Great Kanto Earthquake of 1923 than those endowed with less. Aldrich (2008, 2010a) also argues that better organized neighbourhoods in Kobe, Japan, and Tamil, India, recovered faster in the wake of their respective natural disasters compared to the more atomized neighbourhoods. Similarly, another recent study finds that areas characterized by higher levels of social capital responded more effectively to the Red River Floods of 1997 in Manitoba, Canada (Buckland and Rahman 1999). Drawing on the post-earthquake recovery in Kobe, Japan, and Gujarat in India, Nakagawa and Shaw (2004) also show that neighborhoods with more social capital recovered more rapidly than those with less. These studies of post-disaster recovery in different parts of the world suggest that the general thrust of the argument presented here, that greater amounts of information should, ceteris paribus, accelerate postwar reconstruction, may well apply not only to relatively homogeneous societies like Japan but also to the ethnically divided societies that have been the more typical cases of postwar reconstruction in recent decades. In such societies, actors are typically more inclined to believe, and to act on, information that comes from their own ethnic networks and more prone to discounting information that comes from other ethnicities. This is a significant hurdle to reconstruction. However, this fact probably does not change the fact that more information still probably helps to speed reconstruction. Yet it should be acknowledged that in applying the findings from this study to ethnically divided societies, more careful attention may need to be paid not only to the amount of information that flows among societal actors but also to the breadth of dissemination of that information.

9 Conclusions

Conventional accounts of the impact of war on civil society typically paint a bleak picture of the prospects for civic life in the wake of wartime defeat. Defeat usually reduces incomes, interrupts education, and inflicts psychological trauma on a citizenry, all of which should lead citizens to withdraw from public life. By contrast, in victorious countries, it is argued, as these damages tend to be much more limited, there should be no fall, and possibly even a rise, in levels of civic engagement in the immediate postwar period. This project has questioned this conventional wisdom. As argued in Chapter 1, Japan in the wake of World War II provides a key case for examining the relationship between war and the postwar growth of civic engagement. Since Japan was so thoroughly defeated by 1945, if the conventional political science wisdom were correct, it would be an especially unlikely case in which to find any increase in postwar levels of civic engagement. Wartime destruction was severe, postwar economic conditions deteriorated precipitously, and educational opportunities were cut short on a large scale. Moreover after the war, the experience of foreign occupation, as well as the series of occupation reforms that sought to thoroughly remake not only Japan’s preexisting political and economic institutions but even the Japanese mind, could easily have added a new layer of trauma and even a demobilizing culture of defeatism. Yet, despite all of these adverse conditions, civic life actually blossomed in immediate postwar Japan. The conventional focus on the demobilizing consequences of defeat needs to be reconsidered. Chapter 2 documented the dramatic rise in civic engagement in Japan during the initial decade after the war across different types of groups, including youth, women’s, social service, and religious organizations. This takeoff in civic engagement occurred before the takeoff of Japan’s economic “miracle.” In the face of total defeat, citizens reached out and sought to re-connect with each other with renewed vigor. The evidence presented in this book is by no means comprehensive and further research is necessary, but the initial evidence strongly suggests that the revival and growth in civic engagement in immediate postwar Japan was very strong. 161

162

Civic Engagement in Postwar Japan

This study’s basic descriptive finding of a surge in Japanese civic engagement complements and reinforces the findings of Tsujinaka (2002). Drawing on possibly the most comprehensive survey of civic and interest groups ever conducted, whether in Japan or elsewhere, he argues among other points that a major surge in the number of associations occurred in Japan in the wake of World War II. But since the retrospective method used in the survey could not interrogate associations that had ceased operations, it was not ideal for measuring levels of associational birth in different eras (Kage 2003c). The historical figures compiled in this book, however, in fact lend strong support to Tsujinaka’s findings that civic activities indeed rose dramatically in the wake of World War II. Levels of participation in civic activities surged among established groups in the immediate aftermath of war, and most likely this was also accompanied by a marked increase in the number of associations themselves. How can this resilience and resurgence in civic engagement be explained? And what accounts for the intra- and cross-national variations in the extent of this rise? Chapter 3 offered a new two-step model that argued that a combination of two variables – wartime mobilization and legacies of prewar associational activities – explains how major international wars promote a rise in civic engagement in their wake, regardless of whether they ended in victory or defeat. As war intensifies, citizens, both on the war and the home front, come to assume increasingly large public responsibilities. They come into contact with individuals, officials, and groups that they would not have encountered otherwise, and in so doing they acquire important communication abilities, political and social awareness, and organizational savvy, or, in short, “civic skills.” They emerge from the war both willing and able to participate effectively in voluntary groups. The extent to which these citizens actually participate in voluntary associations after wars, however, depends crucially on the availability of opportunities to do so. The availability of opportunities, in turn, depends on the legacies of voluntary activities prior to war. Where associational activities flourished before the war, larger numbers of groups should revive quickly after the war and be able to handle the influx of willing participants. Moreover, where larger numbers of people participated in voluntary groups before the war, more information on the opportunities for participation should be available for potential participants, further facilitating civic engagement. The study tested this theoretical model through both quantitative analysis and historical process-tracing, drawing on both intra-national comparisons from different Japanese prefectures and a cross-national analysis of thirteen countries affected by World War II. In Chapter 4, quantitative analysis showed that, ceteris paribus, a combination of more extensive wartime mobilization and higher levels of civic activities in the prewar period led to more rapid growth in civic engagement in Japan during the initial decade after the war. Chapter 5 expanded the analysis to a broader set of countries and presented preliminary evidence that, ceteris paribus, individuals who came of age around the time of World War II in more highly mobilized countries

Conclusions

163

continued, even decades after the shooting had ended, to participate at higher rates compared to individuals in their immediately preceding generational cohort. Chapters 6 and 7 presented case studies of the YMCA Japan and the Kodokan judo to illustrate, through detailed historical narratives, the precise mechanisms through which prewar legacies shaped the growth of civic engagement. Where there had been more vigorous associational activity in the prewar period, the wartime interruption of civic engagement tended to be shorter, the postwar revival tended to be sooner, and the growth in membership tended to be driven by bottom up demand and exceeded organizational leaders’ expectations. Finally, Chapter 8 highlighted the broader significance of the surge in civic engagement in postwar Japan by presenting results from a preliminary analysis that suggests that areas that had higher levels of social connectedness enjoyed more rapid postwar reconstruction of the economy and of social services.

Broader Implications There are at least five broader theoretical implications of the findings of this project. Japanese Politics: Continuity or Discontinuity? First, this study speaks to the important debate in the field of Japanese politics and history over whether postwar Japanese politics represents a break from, or a continuation of, prewar Japanese politics.1 Many scholars from the early postwar era, such as Tsuji (1958), argued that postwar Japanese politics continued to exhibit many of the characteristics of prewar politics. In particular, these scholars stressed the persisting dominance of the Japanese bureaucracy from the prewar into the postwar Japanese political system. From this perspective, occupation reforms were viewed to be either ineffective or incomplete, perhaps because although the institutional reforms were successful on the surface, Japanese bureaucrats and politicians had failed, whether deliberately or otherwise, to implement the new institutions to their full potential. In contrast, scholars such as Michio Muramatsu (1979, 1988) argue that constitutional change from the Meiji Constitution to the new postwar Constitution marked a clear break in Japanese political history.2 According to this view, the new Constitution produced real changes in actors’ norms and, in turn, practices. For instance, the national Diet emerged as the central locus of Japanese politics. Not only did this have the effect of empowering legislators 1

2

For a comprehensive review of how Japanese scholars have conceptualized civil society, see Yamaguchi (2004). Ishikawa (1978: ch. 1) stresses the disjuncture in the composition of the Diet between the prewar/wartime and postwar periods; according to Ishikawa, only about 30% of Diet members in the 1942 Lower House elections were successfully reelected to the postwar Diet.

164

Civic Engagement in Postwar Japan

and undermining the power of bureaucrats.3 Muramatsu et al. (1986) contend that the rise of the Japanese Diet, along with the new Constitution’s guarantee of the freedom of association, served as a revolutionary shock that transformed the configuration of Japanese interest groups, in terms of both numbers and types. Drawing on the most comprehensive survey of associations in Japan to date, Tsujinaka (2002) also argues that democratization served as a critical juncture that spurred a rapid growth in civic and interest groups in Japan. Yet a third position holds that the postwar Japanese system can be understood as a direct outgrowth of the institutions and practices that were put in place during World War II. Yukio Noguchi, for instance, argues that many of the institutions that were introduced for purposes of pursuing the war effort endured into the postwar period, comprising what he terms the “1940 system” in Japan (Noguchi 1995).4 From a broader historical perspective, Mitani (1993) argues that, since the Meiji Restoration, Japan’s total wars have served to sequentially transform Japanese society not only by drastically altering the economic system, as Noguchi argues, but also by reconfiguring the lines between state and society and between Japan and its external environment. Among U.S. scholars, Chalmers Johnson (1982) also stresses the continuity between the wartime and postwar periods, highlighting how learning during wartime enabled bureaucrats in the Ministry of International Trade and Industry (MITI) to pursue particularly effective industrial policies in the wake of World War II. Some scholars also emphasize the combined effect of wartime changes and the postwar context as well; Moriguchi and Saez (2006), for instance, argues that World War II induced a dramatic leveling of income inequalities in Japan, giving rise to the relatively egalitarian income distribution that was reinforced by occupation reforms such as land reform and became characteristic of the country for much of the postwar period. The insights from the present study suggest a synthesis of these different strands of thinking. At the most basic level, the freedoms of association and expression that were stipulated in the new postwar constitution provided the basic legal framework for associational activities to flourish in the wake of war. Thus the postwar framework had undoubtedly been crucial. But, as noted earlier in this book, the new freedoms of association and expression only provide the basic enabling conditions for the growth of associational activities; they do not by themselves promote voluntary participation. The impetus for the growth in voluntary participation came from citizens’ experiences during the war. Through their participation in civic activities, sometimes rather involuntarily, citizens acquired a broad range of civic skills that they could continue to exercise once the war was over. Citizens thus emerged from war both capable and willing to participate in voluntary activities. In this respect the wartime experience was crucial. But the availability of 3

4

Tsujinaka (1986) also points to the discontinuity between the prewar and postwar in the Japanese associational sector. Sawai (2002) offers a useful review of the continuity/discontinuity debate in the field of economic history.

Conclusions

165

opportunities for participation in the wake of war was shaped by the legacies of voluntary associations from the prewar period. Thus, the configuration of civil society in Japan in the initial decade after the war was shaped by the new postwar context, the wartime experience, as well as the legacies of voluntary activities from the prewar era. This is not simply a claim that “everything matters” – rather, the study argues that each of the three periods contributed to the size and shape of the postwar civic engagement boom in specific ways. A full understanding of the evolution of not only Japan, but also other societies in the wake of war more generally requires a consideration of the intricate interactions among factors from the prewar, wartime, and postwar periods. The “Greatest Generation”? Second, the study sheds new light on the recent debate over whether, to what extent, and why civic engagement is on the decline among citizens of advanced industrialized countries. Scholars of the United States have generally argued that it is indeed on the decline, while scholars of certain Western European countries, such as Belgium and the Netherlands, have contended that it is not. Recent studies of Japan also find evidence of a considerable decline in social capital in Japan at least since the early 1990s, as the Japanese “greatest generation” passes (Naikakufu 2007; Japan Research Institute 2008; Kage 2008). The framework presented in this volume helps to account for both the fall in civic engagement in the United States and Japan and the lack thereof in some other countries. For instance, as seen in Chapter 5, World War II mobilization in Belgium was quite moderate and therefore the war-induced growth in civic engagement was also quite limited, unlike in the United States. If (and this is admittedly a big if) levels of participation for the postwar cohorts have a tendency eventually to return to levels comparable to pre–World War II cohorts, then it is no surprise that the decline in civic engagement in the United States has been much more dramatic than in Belgium. The postwar cohorts in the more highly mobilized countries are not inevitably fated to “bowl alone,” however, and certainly not to the same degree. The extent to which the generational cohorts that follow the “long civic generation” participate in voluntary activities should depend on the unique formative experiences of the postwar cohorts of each country. Appropriate policies may help to sustain the war-induced jumps in levels of civic engagement among subsequent generational cohorts, or at least mitigate the potential decline in participation that may occur as the “greatest generation” ages and subsequently passes away. Peter Hall notes, for instance, that new educational policies that were introduced in the postwar period have counteracted the potential fall in civic engagement in postwar British generational cohorts (Hall 1999). Society-Centric or State-Centric Perspectives on Civil Society Third, the present study also highlights the role of the state in shaping the configuration and evolution of civil society. Scholars such as Putnam (1993,

166

Civic Engagement in Postwar Japan

1995a, 1995b, 2000) have advanced a primarily societal-based account for the patterns and changes in civil society. Levy (1999), Skocpol (2000), and Pharr (2003), among others, have argued that it is primarily state policies that shape the configuration of civil society, as well as changes therein. The analysis presented in this project points to a more nuanced balance between the society-centric and state-centric perspectives. It suggests that state policies may play a key role in instilling civic skills into its citizenry, particularly in times of war. But at the same time, the long-term effectiveness of those lessons may be constrained to a substantial degree by the preexisting social context within which the state operates. In other words, the state can plant the seed of civic engagement, but it cannot control the richness of the soil. This view that the state may strengthen and empower civil society in significant ways (and vice versa) challenges the conventional wisdom that the relationship between state and society is inherently zero-sum. Scholars such as Peter Drucker and Everett Carl Ladd have argued that an expansion in the role of the state invariably crowds out the voluntary sector (Drucker 1993: 9; Ladd 1999). Even Alexis de Tocqueville, to some degree, subscribed to this view. From this perspective, the ideal way to promote civic activities would be for the state, to the extent possible, to retreat from the private sphere. The findings from this study, however, suggest that state capacities and societal capacities may expand in tandem. Indeed, the view that the state may empower societies by equipping them with crucial civic skills echoes some of the recent studies on human capital in the field of political economy (e.g., Iversen and Stephens 2008). In addition, not only may the relationship between state and society be “synergistic”; this study suggests that the effects of this state–society synergy on the mobilized generation may persist long after the mobilization is over. It should be noted, however, that not all states may want to promote the accumulation of civic skills or view it as a happy by-product of wartime activities. When such states find themselves stretched to the limit, as in wartime, they begin to ask their societies not only for compliance but also for help. But later, if the war has not led to the toppling of the regime, they may often try to roll back the concessions they made earlier. In so doing, of course, they are likely to encounter resistance from below. But they may succeed, in which case the post– World War II Japanese pattern would not be repeated. For instance, despite their taste for mass choreographed demonstrations, communist regimes in fact generally demobilized society and fostered distrust and mutual surveillance among citizens after World War II (Howard 2002). Pekkanen (2006: 162–163) also argues that the Japanese state in the Meiji era sought to deliberately discourage the formation of civil society groups to focus societal energies on militarization and industrialization. Path-Dependency Effects Fourth, the study advances a slightly different understanding of how legacies from the past may shape the evolution of civic engagement than that offered

Conclusions

167

by Robert Putnam. Putnam (1993) argued that cross-regional patterns in participatory behavior are at their root cultural phenomena that may persist over centuries.5 This study argues that levels of civic engagement may be much more malleable than Putnam suggests, and that, in terms of the absolute levels of participation, the participatory behavior of citizens can change dramatically perhaps over a course of a few years, particularly as a result of wartime mobilization. At the same time, in relative terms mobilization tends to exacerbate preexisting differences in levels of civic engagement. Given a certain “distribution of wealth” of civic activities across regions, war causes the “rich” to get even richer, in relative terms. This finding leads to some thoughts on potential mechanisms for overcoming the effects of path-dependency in historically low-participation countries or regions in the wake of wars. According to Putnam’s framework, boosting levels of civic engagement in such societies may require the wholesale re-education of a region or country, and it is not entirely clear from the argument if even this would be effective. By contrast, the present project suggests that one potentially promising way to promote participation in postwar societies would be to reduce the costs for participation, whether organizational or informational. Since these low costs are the primary mechanisms via which historical legacies of associationalism impact postwar civic engagement, it may be possible to find other ways to lower these costs even in situations where historical legacies are weak. For instance, international NGOs appearing immediately after the shooting stops to help local citizens set up new associations might well be of real benefit here. In short, the argument offered in this volume is more optimistic than Putnam about the ability of societies to increase their levels of civic engagement, at least among wartime generational cohorts. Voluntary and Involuntary Participation Fifth, in emphasizing the role of forced wartime mobilization in the accumulation of civic skills, this study speaks to the recent debate among scholars and practitioners over whether it is possible to promote civic engagement through “required” public service.6 The analysis of the effects of wartime mobilization suggests that, empirically, learning of civic skills may in fact be possible under some degree of coercion. At some level, this finding is hardly surprising; the whole notion of mandatory education rests on the assumption that it fosters some degree of fruitful learning. But some of the social science literature has 5

6

To be fair, the argument in Putnam’s Bowling Alone relies less on path-dependency to explain the levels of participation and more on exogenous societal-level changes, such as the advent of the TV (Putnam 2000). One aspect of mobilization that the present study did not fully address was the issue of different styles of mobilization. Societies may mobilize for war in different ways in different countries and in different historical eras. Kier (2010) has begun to conduct important work in this direction, and how variations in qualitative differences in mobilization affect social outcomes presents a promising avenue for further research.

168

Civic Engagement in Postwar Japan

made something of a cult of voluntary participation while castigating mandatory participation as simply evil. Of course, if participation is completely coerced against one’s will, as in the case of forced labor in concentration camps, the level of learning of skills useful for civic engagement would not be very impressive. However, our daily experiences also indicate that in the less extreme cases, some degree of coercion or external pressure may not necessarily impede learning; rather, it may at times, though not always, promote learning by presenting individuals with new and unanticipated challenges and opportunities. In short, it is high time that the literature problematized the distinction between “voluntary” and “involuntary” participation, and this study has made an effort to do just that. More issues will need to be addressed in the future, such as to what extent does participation need to be “voluntary” before effective learning may take place? And how might varying degrees of voluntarism be measured with a reasonable degree of accuracy? It must again be emphasized here that these empirical questions should not be confused with normative support for coercion. The empirical claim of this study, that forced mobilization may at times promote civic skills, should not be confused as implying or suggesting, at a normative level, that such coercion is desirable as public policy. Empirically, it is possible that mobilization may yield positive benefits for civic engagement but that its effects may be less sanguine for other social outcomes. Moreover, it may also be argued that involuntary mobilization is morally unjustified even if it yields overall positive social outcomes. This book is interested in the empirical issue of whether mobilization from above may lead to a more civic-minded citizenry, not in the normative question of whether such mobilization is desirable. The findings from this study need to be interpreted with much caution.

Contemporary Applications As noted in Chapter 1, democratization in the wake of wars has become increasingly common in the postwar era. Indeed, industrialized countries such as the United States increasingly wage war expressly to bring about democratization, at least on the level of rhetoric. Afghanistan and Iraq are prime examples. A voluminous literature has found a wide range of factors that affect the success or failure of democratization, from economic development to income equality to the process of democratic transition itself, but scholars in recent years increasingly argue that a vibrant and democratically minded civil society is also crucial (Diamond 1999). While the present study has focused primarily on historical cases, the framework advanced in this study also yields important insights for understanding the relationship between war and civic engagement for more contemporary cases as well. The findings from this book suggest that the legacies of the prewar and wartime periods are crucial determinants of the chances for the rise of a vibrant civil society in the wake of wars, and therefore also for the prospects for successful democratization as well.

Conclusions

169

One might ask to what extent the case of World War II offers relevant lessons for understanding the impact of the modal type of war today, the civil war. This is an important subject for further research. Civil wars may be much more disruptive to a society’s social fabric than international wars. But they may also foster civic skills via mobilization within ethnic or religious groups. For instance, rapid growth in the NGO sector has been reported in postwar Cambodia (Persson 2003). Recent studies by Blattman (2008) as well as Bellows and Miguel (2009) have also found a growth in political participation in the wake of civil war in Uganda. Although much more systematic research is necessary, the expectations derived from World War II do appear to have some bearing on the messier mix of civil and international wars that are more common in the post–Cold War era. To take another example, Tito’s Yugoslavia was by far the most liberal communist regime in history and permitted a relatively wide array of types of civic engagement. Then in the 1990s it became engulfed in war. The combination of a history of a relatively developed associational sector with substantial wartime mobilization during the 1990s, especially in countries such as Serbia and Croatia, appears to have provided fertile ground for an explosion of voluntary activities in the postwar period. In Serbia, for instance, the number of NGOs doubled between 1997 and 2000, from 695 to 1,344 (NGO Policy Group 2001). Democratization of the former Yugoslav republics remains incomplete, but the strength of civil society provides at least some basis to believe that democracy is possible in those new countries. In contrast to Yugoslavia, many other countries such as Afghanistan and Iraq lacked flourishing civic associations during their prewar periods, at least in the sense of formal organizations. This means that although these countries underwent wartime mobilization, there were few if any preexisting structures of voluntary associational activities that could provide the basis for a postwar boom in civic engagement. Much of the civic skills acquired during those countries’ wars during the 1980s, 1990s, and 2000s are therefore unlikely to find an outlet in voluntary associations after the shooting stops. The case of Iraq is particularly noteworthy here in light of the tremendous efforts made by the United States to build democracy there, indeed explicitly with reference to the U.S. experience in Japan after World War II. Of course, the U.S. occupation of Iraq made many mistakes, but as perhaps the best history of contemporary Iraq written to date, Ali Allawi’s The Occupation of Iraq (2007), argues, among the key mistakes was to assume that Iraqi society was ready and able to coalesce and engage in all of the variety of activities, political and otherwise, that make democracy work. In fact, the Bush administration’s view of Iraq as an educated middle-class, civic-minded society was largely a myth (Allawi 2007: 127–131). Instead, the invading forces found a society exhausted and demobilized by decades of dictatorial rule, combined with economic and social hardships, to such an extent that an Iraqi psychiatrist estimated that a quarter of the Iraqi population suffered from psychological distress symptoms (Allawi 2007: 130). Ethnic tensions had been exacerbated,

170

Civic Engagement in Postwar Japan

especially since 1991 (Allawi 2007: 127). Such a “physical and psychological ghetto,” in Allawi’s words, was hardly conducive to the re-emergence of a thriving civic life in the wake of war. The key insight of the present study for cases like Iraq is that an assessment of the chances for successful democratization requires not only the introduction of democratic institutions and large amounts of economic aid, but also a careful consideration of the recent history of the country in question, in particular whether the society has sufficient associational traditions to rise to the level of civic engagement necessary to sustain those new democratic institutions.

References

Aldrich, Daniel P. (2008), “The Crucial Role of Civil Society in Disaster Management and Recovery,” Japan Aktuell, vol. 3: 81–96. (2010a), “The Power of People: Social Capital’s Role in Recovery from the 1995 Kobe Earthquake,” Working Paper. (2010b), “Social, Not Physical, Infrastructure: The Critical Role of Civil Society in Disaster Recovery,” Working Paper. Allawi, Ali A. (2007), The Occupation of Iraq: Winning the War, Losing the Peace. New Haven, CT: Yale University Press. Almond, Gabriel, and Sidney Verba (1963). The Civic Culture: Political Attitudes and Democracy in Five Countries. Princeton, NJ: Princeton University Press. American Red Cross (2002), Corporate Annual Report. Washington, DC. Andolina, Molly W., Krista Jenkins, Cliff Zukin, and Scott Keeter (2003), “Habits from Home, Lessons from School: Influences on Youth Civic Engagement,” PS: Political Science and Politics, vol. 36, no. 2: 275–280. Archer, Dane, and Rosemary Gartner (1984), Violence and Crime in Cross-National Perspective. New Haven, CT: Yale University Press. Azimi, Nassrine, Matt Fuller, and Hiroko Nakayama, eds. (2003), Post-Conflict Reconstruction in Japan, Republic of Korea, Cambodia, East Timor and Afghanistan. Proceedings of an International Conference Hiroshima, 2002. New York and Geneva: United Nations. Baker, William D., and John R. Oneal (2001), “Patriotism or Opinion Leadership? The Nature and Origins of the Rally ’Round the Flag Effect,” Journal of Conflict Resolution, vol. 45, no. 5: 661–687. Bartels, Larry M. (1996), “Uninformed Votes: Information Effects in Presidential Elections.” American Journal of Political Science, vol. 40, no. 1: 194–230. Beck, Paul Allen, and M. Kent Jennings (1982), “Pathways to Participation,” American Political Science Review, vol. 76, no. 1: 94–108. Belloni, Roberto (2001), “Civil Society and Peacebuilding in Bosnia and Herzegovina,” Journal of Peace Research, vol. 38, no. 2: 163–180. Bellows, John, and Edward Miguel (2009), “War and Local Collective Action in Sierra Leone,” Journal of Public Economics, vol. 93, no. 11–12: 1144–1157. Bendor, Jonathan, Daniel Diermeier, and Michael Ting (2003), “A Behavioral Model of Turnout,” American Political Science Review, vol. 97, no. 2: 261–280. 171

172

References

Bennett, D. Scott and Allan C. Stam III (1996), “The Duration of Interstate Wars, 1816– 1985,” American Political Science Review, vol. 90, no. 2: 239–257. Berinsky, Adam J. (2007), “Assuming the Cost of War: Events, Elites, and American Public Support for Military Conflict,” Journal of Politics, vol. 69, no. 4: 975–997. Bermeo, Nancy (2003), “What the Democratization Literature Says – or Doesn’t Say – About Postwar Democratization,” Global Governance, vol. 9, no. 2: 159–177. (2007), “War and Democratization: Lessons from the Portuguese Experience,” Democratization, vol. 14, no. 3: 388–406. Blattman, Christopher (2009), “From Violence to Voting: War and Political Participation in Uganda,” American Political Science Review, vol. 103, no. 2: 231–247. Boatright, Robert G. (1999), “Why Citizens Don’t Respond to Jury Duty and What Courts Can Do About It,” Judicature, vol. 82: 156–64. Boix, Carles, and Daniel N. Posner (1998), “Social Capital: Explaining the Origins and Effects on Government Performance,” British Journal of Political Science, vol. 28, no. 4: 686–693. Bracken, Patrick J., Joan E. Giller, and Derek Summerfield (1995), “Psychological Responses to War and Atrocity: The Limitations of Current Concepts,” Social Science & Medicine, vol. 40, no. 8: 1073–1082. Brady, Henry E., Kay Lehman Schlozman, and Sidney Verba (1999), “Prospecting for Participants: Rational Expectations and the Recruitment of Political Activists,” American Political Science Review, vol. 93, no. 1: 153–68. Brandes, O. Jean (1950), “The Effect of War on the German Family,” Social Forces, vol. 29, no. 2: 164–173. Braumoeller, Bear. F. (2004), “Hypothesis Testing and Multiplicative Interaction Terms,” International Organization, vol. 58, no. 4: 807–820. Braybon, Gail, and Penny Summerfield (1987), Out of the Cage: Women’s Experiences in Two World Wars. London and New York: Pandora Press. Brehm, John, and Wendy Rahn (1997), “Individual-Level Evidence for Causes and Consequences of Social Capital,” American Journal of Political Science, vol. 41, no. 3: 999–1023. Browning, Harley L., Sally C. Lopreato, and Dudley L. Poston, Jr. (1973), “Income and Veteran Status: Variations Among Mexican Americans, Blacks, and Anglos,” American Sociological Review, vol. 38, no. 1: 74–85. Buckland, Jerry, and Matiur Rahman (1999), “Community-Based Disaster Management During the Red River Flood in Canada,” Disasters, vol. 23, no. 2: 174–191. Campbell, Donald T., and Thelma H. McCormack (1957), “Military Experience and Attitudes Toward Authority,” American Journal of Sociology, vol. 62, no. 5: 482–490. Cecil, Joe S., E. Allan Lind, and Gordon Bermant, eds. (1987), Jury Service in Lengthy Civil Trials. Washington, DC: Federal Judicial Center. Centeno, Miguel Angel (1997), “Blood and Debt: War and Taxation in NineteenthCentury Latin America,” American Journal of Sociology, vol. 102, no. 6: 1565–1605. Cohen, Jerome B. (1949), Japan’s Economy in War and Reconstruction. Minneapolis: University of Minnesota Press. Crowley, Jocelyn B., and Theda Skocpol (2001), “The Rush to Organize: Explaining Associational Formation in the U.S. States, 1860s-1920s,” American Journal of Political Science, vol. 45, no. 3: 813–829.

References

173

Cutler, Brian L., and Donna M. Hughes (2001), “Judging Jury Service,” Behavioral Sciences and the Law, vol. 19, no. 2: 305–20. Cutright, Philip (1963), “National Political Development,” American Sociological Review, vol. 28, no. 2: 253–64. Davidann, Jon Thares (1998), A World of Crisis and Progress: The American YMCA in Japan, 1890–1930. Bethlehem, PA: Leigh University Press. Dahl, Robert A. (1971), Polyarchy: Participation and Opposition. New Haven, CT: Yale University Press. “Developments in Japanese Education in terms of the Report of the United States Education Mission to Japan,” August 1950; Box 5402, Folder 8; Developments in Japanese Education, 1945–51; Administrative Files, compiled 1945–1951; Supreme Commander for the Allied Powers, Civil Information and Education Section, Education Division, Administration Branch, (1945–1952); Records of Allied Operational and Occupation Headquarters, World War II, 1907–1966, Record Group 331; National Archives at College Park, College Park, MD. Diamond, Larry (1996), “Is the Third Wave Over?” Journal of Democracy, vol. 7, no. 3: 20–37. (1999), Developing Democracy: Toward Consolidation. Baltimore: Johns Hopkins University Press. Diamond, Shari Seidman (1993), “What Jurors Think,” in Robert E. Litan, ed., Verdict: Assessing the Civil Jury System. Washington, DC: The Brookings Institution. Diani, Mario (2003), “Introduction,” in Mario Diani and Doug McAdam, eds., Social Movements and Networks. Oxford: Oxford University Press. Donnelly, Mark (1999), Britain in the Second World War. London and New York: Routledge. Dower, John W. (1999), Embracing Defeat: Japan in the Wake of World War II. New York: W. W. Norton & Co./New Press. (2005), “Occupation: A Warning from History,” in Lloyd C. Gardner and Marilyn B. Young, eds., The New American Empire. New York: New Press. Doyle, Michael W., and Nicholas Sambanis (2006), Making War and Building Peace: United Nations Peace Operations. Princeton, NJ: Princeton University Press. Drucker, Peter F. (1993), The Ecological Vision: Reflections on the American Condition. New Brunswick, NJ: Transaction Publishers. Durkheim, Emile (1951), Suicide: A Study in Sociology. New York: Free Press. Edwards, Michael W., and Bob Foley (2001), “Civil Society and Social Capital: A Primer,” in Bob Edwards, Michael W. Foley, and Mario Diani, eds., Beyond Tocqueville: Civil Society and the Social Capital Debate in Comparative Perspective. Hanover, NH: Tufts University/University Press of New England. Elder, Glenn H. Jr., and Elizabeth C. Clipp (1988), “Wartime Losses and Bonding: Influences Across 40 Years in Men’s Lives,” Psychiatry: Interpersonal and Biological Processes, vol. 51, no. 2: 177–198. Elder, Glenn H. Jr., and Yoriko Meguro (1987), “Wartime in Men’s Lives: A Comparative Study of American and Japanese Cohorts,” International Journal of Behavioral Development, vol. 10, no. 4: 439–466. Ellis, John (1993), World War II: A Statistical Survey. New York: Facts on File. Ellison, Christopher G. (1992), “Military Background, Racial Orientations, and Political Participation among Black Adult Males,” Social Science Quarterly, vol. 73. no. 2: 360–378.

174

References

Ember, Carol R., and Melvin Ember (1994), “War, Socialization, and Interpersonal Violence: A Cross-Cultural Study,” Journal of Conflict Resolution, vol. 38, no. 4: 620–646. Encarnación, Omar. G. (2006), “Civil Society Reconsidered,” Comparative Politics, vol. 38, no. 3: 357–376. Ertman, Thomas (1997), Birth of the Leviathan: Building States and Regimes in Medieval and Early Modern Europe. New York: Cambridge University Press. Evans, Peter, ed. (1997), State-Society Synergy: Government and Social Capital in Development. Berkeley: International and Area Studies, University of California at Berkeley. Flynn, George Q. (1998), “Conscription and Equity in Western Democracies, 1940–75,” Journal of Contemporary History, vol. 33, no. 1: 5–20. Foley, Bob, Michael W. Edwards, and Mario Diani (2001), “Social Capital Reconsidered,” in Bob Edwards, Michael W. Foley, and Mario Diani, eds., Beyond Tocqueville: Civil Society and the Social Capital Debate in Comparative Perspective. Hanover, NH: Tufts University/University Press of New England. Forman, Shepard, and Stewart Patrick (2000), “Introduction,” in Forman and Patrick, eds., Good Intentions: Pledges of Aid for Postconflict Recovery. Boulder, CO: Lynne Rienner. Fortna, Virginia Page (2004), “Does Peacekeeping Keep Peace?” International Studies Quarterly, vol. 48, no. 2: 269–292. Freeman, Jo (1972), “The Women’s Liberation Movement: Its Origins, Structures, and Ideas,” in Hans Peter Dreitzel, ed., Recent Sociology No. 4: Family, Marriage, and the Struggle of the Sexes. New York: MacMillan. Fritz, Charles E. (1961), “Disaster,” in Robert K. Merton and Robert A. Nisbet, eds., Contemporary Social Problems: An Introduction to the Sociology of Deviant Behavior and Social Disorganization. New York: Harcourt, Brace, and World, Inc. Fujihira, Shinju (2003), “Financing Warfare: Lessons from Imperial Japan,” USJP Occasional Paper 03–03, Program on U.S.–Japan Relations, Harvard University, Cambridge, MA. Fukuyama, Francis (1995), Trust: The Social Virtues and the Creation of Prosperity. New York: Free Press. (2006), “Nation-Building and the Failure of Institutional Memory,” in Fukuyama, ed., Nation-Building: Beyond Afghanistan and Iraq. Baltimore: Johns Hopkins University Press. Garon, Sheldon M. (1997), Molding Japanese Lives: The State in Everyday Life. Princeton, NJ: Princeton University Press. George, Alexander L., and Andrew Bennett (2005), Case Studies and Theory Development in the Social Sciences. Cambridge, MA: MIT Press. Gibson, James. L. (2001), “Social Networks, Civil Society, and the Prospects for Consolidating Russia’s Democratic Transition,” American Journal of Political Science, vol. 45, no. 1: 51–69. Ghobarah, Hazem, Paul Huth, and Bruce Russett (2003), “Civil Wars Kill and Maim People – Long after the Shooting Stops,” American Political Science Review, vol. 97, no. 2: 189–202. Goldsmith, Benjamin E., Yusaku Horiuchi, and Takashi Inoguchi (2005), “American Foreign Policy and Global Opinion: Who Supported the Iraq War?” Journal of Conflict Resolution, vol. 49, no. 3: 408–429.

References

175

Goralski, Robert (1981), World War II Almanac: 1931–1945. New York: Putnam. Goss, Kristin A. (1999), “Volunteering and the Long Civic Generation,” Nonprofit and Voluntary Sector Quarterly, vol. 28, no. 4: 378–415. Granovetter, Mark (1973), “The Strength of Weak Ties,” American Journal of Sociology, vol. 78, no. 6: 1360–80. Guinther, John (1998), The Jury in America. New York: Facts on File. Gurr, Ted (1980), Handbook of Political Conflict: Theory and Research. New York: Free Press. Haddad, Mary Alice (2006), “Civic Responsibility and Patterns of Voluntary Participation Around the World,” Comparative Political Studies, vol. 39, no. 10: 1220–1242. Hall, Peter (1999), “Social Capital in Britain,” British Journal of Political Science, vol. 29, no. 2: 417–461. Hara, Akira (1998), “Japan: Guns Before Rice,” in Mark Harrison, ed., The Economics of World War II. Cambridge: Cambridge University Press. Hart, Daniel, Thomas M. Donnelly, James Youniss, and Robert Atkins (2007), “High School Community Service as a Predictor of Adult Voting and Volunteering,” American Educational Research Journal, vol. 44, no. 1: 197–219. de Hart, Joep, and Paul Dekker (1999), “Civic Engagement and Volunteering in the Netherlands: A ‘Putnamian’ Analysis,” in Jan W. Van Deth, Marco Maraffi, Ken Newton, and Paul F. Whiteley, eds., Social Capital and European Democracy. London and New York: Routledge. Hartzell, Caroline, Matthew Hoodie, and Donald Rothchild (2001), “Stabilizing the Peace After the War,” International Organization, vol. 55, no. 1: 183–208. Havens, Thomas R. H. (1978), Valley of Darkness: The Japanese People and World War II. New York: W. W. Norton & Co. Herbst, Jeffrey (2000), States and Power in Africa: Comparative Lessons in Authority and Control. Princeton, NJ: Princeton University Press. Hobfoll, Stevan E. (1989), “Conservation of Resources: A New Attempt at Conceptualizing Stress,” American Psychologist, vol. 44, no. 3: 513–524. Hooghe, Marc (2003), “Why Should We Be Bowling Alone? Results from a Belgian Survey on Civic Participation,” Voluntas, vol. 14, no. 1: 41–59. Howard, Marc Morjé (2002),“Postcommunist Civil Society in Comparative Perspective,” Demokratizatsiya, vol. 13, no. 1: 157–69. Humphreys, Macartan, and Jeremy M. Weinstein (2007), “Demobilization and Reintegration,” Journal of Conflict Resolution, vol. 51, no. 4: 531–567. Hunter, Susan, and Richard A. Brisbin Jr. (2000), “The Impact of Service Learning on Democratic and Civic Values,” PS: Political Science and Politics, vol. 33, no. 3: 623–626. Hurtic, Zlatko, Amela Sapcanin, and Susan L. Woodward (2000), “Bosnia and Herzegovina,” in Shepard Forman and Stewart Patrick, eds., Good Intentions: Pledges of Aid for Postconflict Recovery. Boulder, CO: Lynne Rienner. Hutchinson, John F. (1996), Champions of Charity: War and the Rise of the Red Cross. Boulder, CO: Westview. Ichino, Andera, and Rudolf. Winter-Ember (2004), “The Long-Run Educational Costs of World War II,” Journal of Labor Economics, vol. 22, no. 1: 57–86. Inglehart, Ronald (1988), “The Renaissance of Political Culture,” American Political Science Review, vol. 82, no. 4: 1203–1230.

176

References

Iriye, Akira (2002), Global Community: The Role of International Organizations in the Making of the Contemporary World. Berkeley: University of California Press. Iversen, Torben, and John D. Stephens (2008), “Partisan Politics, the Welfare State, and the Three Worlds of Human Capital Formation,” Comparative Political Studies, vol. 41, no. 4–5: 600–637. Jacobson, Paul H. (1959), American Marriage and Divorce. New York: Rinehart. Jaggers, Keith (1992), “War and the Three Faces of Power: War Making and State Making in Europe and the Americas,” Comparative Political Studies, vol. 25, no. 1: 26–62. Japanese National Women’ s Organizations, July 1, 1947; Box 5250, Folder 21; Women’s Organizations – General, 1945–50; Women’s Affairs Activities, compiled 1945– 1950; Supreme Commander for the Allied Powers, Civil Information and Education Section, Information Division, Policy and Program Branch (1947–1952); Records of Allied Operational and Occupation Headquarters, World War II, 1907–1966, Record Group 331; National Archives at College Park, College Park, MD. Jenkins, J. Craig (1983), “Resource Mobilization Theory and the Study of Social Movements,” Annual Review of Sociology, vol. 9: 527–553. Jennings, M. Kent (2002), “Generation Units and the Student Protest Movement in the United States: An Intra- and Intergenerational Analysis,” Political Psychology, vol. 23, no. 2: 303–324. Jennings, M. Kent, and Gregory B. Markus (1976), “Political Participation and Vietnam War Veterans: A Longitudinal Study,” in Nancy L. Goldman and David R. Segal, The Social Psychology of Military Service. Beverly Hills and London: Sage Publications. (1977), “The Effect of Military Service on Political Attitudes,” American Political Science Review, vol. 71, no. 1: 131–147. Jennings, M. Kent, Laura Stoker, and Jake Bowers. 2001. “Politics across Generations: Family Transmission Reexamined.” Institute of Governmental Studies. Jentleson, Bruce W. (1992), “The Pretty Prudent Public: Post-Vietnam American Opinion on the Use of Military Force,” International Studies Quarterly, vol. 36, no. 1: 49–74. Johnson, Chalmers (1982), MITI and the Economic Miracle: The Growth of Industrial Policy, 1925–1975. Stanford: Stanford University Press. Kage, Rieko (2003a), “Embracing Democracy: The Promotion of Civic Engagement in Occupied Japan,” Paper Prepared for Presentation at the Asian Studies Japan Conference, Sophia University, Japan, June 22. (2003b), “How Does War Influence Civil Society? A Review of the Literature,” Paper Prepared for Presentation at the Annual Meeting of the American Political Science Association, Philadelphia, PA. (2003c), “Review: Yutaka Tsujinaka, ed., Civic and Interest Groups in Contemporary Japan,” Japanese Journal of Political Science, vol. 4, no.1: 164–167. (2005), Defeat and Reconstruction: Explaining the Rise of Civic Engagement in the Wake of Wars. Ph.D. Dissertation, Department of Government, Harvard University, Cambridge, MA. (2006), “The Impact of World War II on Voluntary Participation: Britain compared with Japan and Sweden,” Paper Presented at the 20th Annual CCBH Summer Conference, Institute of Historical Research, Senate House, University of London, United Kingdom, June 29.

References

177

(2008), “The Effects of Neoliberalism on Social Solidarity: A Seventeen- Country Analysis,” Paper Presented at the Annual Conference of the Association for Research of Nonprofit Organizations and Voluntary Action (ARNOVA), Philadelphia, PA, November 20–22. (2009), “Rebuilding from World War II: A Preliminary Assessment of Damage and Recovery,” Paper Presented at the Social Science History Conference, Long Beach, CA, November 15. Kaniasty, Krzysztof, and Fran H. Norris (1995), “In Search of Altruistic Community: Patterns of Social Support Mobilization Following Hurricane Hugo,” American Journal of Community Psychology, vol. 23, no. 4: 447–477. Kasza, Gregory J. (1995), The Conscription Society: Administered Mass Organizations. New Haven, CT, and London: Yale University Press. (2002), “War and Welfare Policy in Japan,” Journal of Asian Studies, vol. 61, no. 2: 417–435. Kaufman, Joyce P. (1983), “The Social Consequences of War,” Armed Forces and Society, vol. 9, no. 2: 245–264. Kernell, Samuel (1978), “Explaining Presidential Popularity,” American Political Science Review, vol. 72, no. 2: 506–522. Kier, Elizabeth (2010), “War and Reform: Gaining Labor’s Compliance on the Homefront,” in Ron Krebs and Elizabeth Kier, eds., In War’s Wake: International Conflict and the Fate or Liberal Democracy. Cambridge: Cambridge University Press. King, Gary, Michael Tomz, and Jason Wittenberg (2000). “Making the Most of Statistical Analyses: Improving Interpretation and Presentation.’” American Journal of Political Science, vol. 44, no. 2: 347–361. Kjonniksen, Lise, Norman Anderssen, and Bente Wold (2009), “Organized Youth Sport as a Predictor of Physical Activity in Adulthood,” Scandinavian Journal of Medicine and Science in Sports, vol. 19, no. 5: 646–654. Kosai, Yutaka (1986), The Era of High-Speed Growth. Tokyo: University of Tokyo Press. Krebs, Ronald R. (2004), “A School for the Nation? How Military Service Does Not Build Nations, and How It Might,” International Security, vol. 28, no. 4: 85–124. (2006), Fighting for Rights: Military Service and the Politics of Citizenship. Ithaca, NY: Cornell University Press. Kulka, Richard A., John A. Fairbank, B. Kathleen Jordan, and Daniel Weiss (1990), Trauma and the Vietnam War Generation. New York: Brunner/Mazel. Kumar, Krishna (1997), “The Nature and Focus of International Assistance for Rebuilding War-Torn Societies,” in Kumar, ed., Rebuilding Societies after Civil War: Critical Roles for International Assistance. Boulder, CO: Lynne Rienner. Kume, Ikuo (1998), Disparaged Success? Labor Politics in Postwar Japan. Ithaca, NY: Cornell University Press. Ladd, Carl Everett (1999), The Ladd Report. New York: Free Press. Lawler, Edward E., and Hackman, Richard J. (1969), “Impact of Employee Participation in the Development of Pay Incentive Plans: A Field Experiment,” Journal of Applied Psychology, vol. 53, no. 6: 467–471. Leal, David L. (1999), “It’s Not Just a Job: Military Service and Latino Political Participation,” Political Behavior, vol. 21, no. 2: 153–174. Levi, Margaret (1988), Of Rule and Revenue. Berkeley: University of California Press.

178

References

Levy, Jonah (1999), Tocqueville’s Revenge: State, Society, and Economy in Contemporary France. Cambridge, MA: Harvard University Press. Lian, Bradley, and John Oneal (1993), “Presidents, the Use of Military Force, and Public Opinion,” Journal of Conflict Resolution, vol. 37, no. 2: 277–300. Lipset, Seymour Martin, and Stein Rokkan (1967), “Cleavage Structures, Party Systems, and Voter Alignments: An Introduction,” in Lipset and Rokkan, eds., Party Systems and Voter Alignments: A Comparative Perspective. New York: Free Press. Lupia, Arthur (1994), “Shortcuts versus Encyclopedias: Information and Voting Behavior in California Insurance Reform Elections.” American Political Science Review vol. 88, no. 1: 63–76. McAdam, Doug (1982), Political Process and the Development of Black Insurgency, 1930–1970. Chicago: The University of Chicago Press. McFarland, Daniel A., and Reuben J. Thomas (2006), “Bowling Young: How Youth Voluntary Associations Influence Adult Political Participation,” American Sociological Review, vol. 71, no. 3: 401–425. Maddison, Angus (2003), The World Economy: Historical Statistics. Paris: Development Centre of the Organisation for Economic Co-operation and Development. Mannheim, Karl ([1928] 1952), Essays in the Sociology of Knowledge. London: Routledge and Kegan Paul. Marwick, Arthur (1981), “Problems and Consequences of Organizing Society for Total War,” in Nandor F. Dreiszinger, ed., Mobilization for Total War: The Canadian, American, and British Experience, 1914–1918,1939–1945. Waterloo, Ontario: Wilfrid Laurier University Press. “Membership of Active Women’s Organizations,” May 10, 1946; Box 5250, Folder 21; Women’s Organizations – General, 1945–50; Women’s Affairs Activities, compiled 1945–1950; Supreme Commander for the Allied Powers, Civil Information and Education Section, Information Division, Policy and Program Branch (1947– 1952); Records of Allied Operational and Occupation Headquarters, World War II, 1907–1966, Record Group 331; National Archives at College Park, College Park, MD. Merriam-Webster (2005), “Mobilization,” retrieved April 5, 2005 from http://www.m-w. com/cgi-bin/dictionary?book=Dictionary&va=mobilizing. Mettler, Suzanne (2002), “Bringing the State Back In To Civic Engagement: Policy Feedback Effects of the GI Bill for World War II Veterans,” American Political Science Review, vol. 96, no. 2: 351–365. (2005), Soldiers to Citizens: The G.I. Bill and the Making of the Greatest Generation. Oxford: Oxford University Press. Metz, Edward, and James Youniss (2003), “A Demonstration that School-Based Required Service Does not Deter – But Heightens – Volunteerism,” PS: Political Science and Politics, vol. 36, no. 2: 281–286. Mill, John Stuart (1989), On Liberty. Cambridge: Cambridge University Press. Moriguchi, Chiaki, and Emmanuel Saez (2006), “The Evolution of Income Inequality in Japan, 1886–2002: Evidence from Income Tax Statistics,” NBER Working Paper 12558. Mueller, John E. (1970), “Presidential Popularity Ratings from Truman to Johnson,” American Political Science Review, vol. 64, no. 1: 18–34. Nakagawa, Yuko, and Rajib Shaw (2004), “Social Capital: A Missing Link to Disaster Recovery,” International Journal of Mass Emergencies and Disasters,” vol. 22, no. 1: 5–34.

References

179

NGO Policy Group (2001), “Third Sector in Serbia: Status and Prospects.” Belgrade: Center for the Development of the Nonprofit Sector and the NGO Policy Group. Norris, Fran H. (1992), “Epidemiology of Trauma: Frequency and Impact of Different Potentially Traumatic Events on Different Demographic Groups,” Journal of Consulting and Clinical Psychology, vol. 60, no. 3: 409–418. O’Donnell, Guillermo, and Philippe Schmitter (1986), Transitions from Authoritarian Rule: Tentative Conclusions about Uncertain Democracies. Baltimore: Johns Hopkins University Press. Okimoto, Daniel I. (1989), Between MITI and the Market. Stanford, CA: Stanford University Press. Oliver, Pamela (1984), “If You Don’t Do It, Nobody Else Will: Active and Token Contributors to Local Collective Action,” American Sociological Review, vol. 49, no. 5: 601–610. Oliver, J. Eric (2000), “City Size and Civic Involvement in Metropolitan America,” American Political Science Review, vol. 94, no. 2: 361–373. Orjuela, Camilla (2003), “Building Peace in Sri Lanka,” Journal of Peace Research, vol. 40, no. 2: 195–212. Palfrey, Thomas R., and Keith T. Poole (1987), “The Relationship between Information, Ideology, and Voting Behavior,” American Journal of Political Science, vol. 31, no. 3: 511–530. Paris, Roland (2004), At War’s End. Cambridge: Cambridge University Press. Parker, Melissa (1996), “The Mental Health of War-Damaged Populations,” Institute of Development Studies Bulletin, vol. 27, no. 3: 77–85. Peacock, Alan T., and Jack Wiseman (1961), The Growth of Public Expenditures in the United Kingdom. London: George Allen and Unwin Ltd. Pekkanen, Robert (2006), Japan’s Dual Civil Society: Members Without Advocates. Stanford, CA: Stanford University Press. Pharr, Susan J. (2003), “Conclusion: Targeting by an Activist State: Japan as a Civil Society Model,” in Frank J. Schwartz and Susan J. Pharr, The State of Civil Society in Japan. Cambridge: Cambridge University Press. Pierson, Paul (2004), Politics in Time: History, Institutions, and Social Analysis. Princeton, NJ: Princeton University Press. Putnam, Robert D. (1993), Making Democracy Work: Civic Traditions in Modern Italy. Princeton, NJ: Princeton University Press. (1995a), “Bowling Alone: America’s Declining Social Capital,” Journal of Democracy, vol. 6, no. 1: 65–78. (1995b), “Tuning In, Tuning Out: The Strange Disappearance of Social Capital in America.” PS: Political Science and Politics, vol. 28, no. 4: 664–683. (2000), Bowling Alone: The Collapse and Revival of American Community. New York: Simon and Schuster. Putnam, Robert D. and Thomas Sander (2002), “Walking the Civic Talk Together,” Christian Science Monitor, February 19th. Pye, Lucian (1963), Communications and Political Development. Princeton: Princeton University Press. Raskoff, Sally A. and Richard A. Sundeen (1999), “Community Service Programs in High Schools,” Law and Contemporary Problems, vol. 62, no. 4: 73–111. Rasler, Karen A., and William R. Thompson (1985), “War Making and State Making: Governmental Expenditures, Tax Revenues, and Global Wars,” American Political Science Review, vol. 79, no. 2: 491–507.

180

References

Rodgers, Willard L., and Arland Thornton (1985), “Changing Patterns of First Marriage in the United States,” Demography, vol. 22, no. 2: 265–279. Roghmann, Klaus, and Wolfgang Sodeur (1972), “The Impact of Military Service on Authoritarian Attitudes: Evidence from West Germany,” American Journal of Sociology, vol. 78, no. 2: 418–433. Rose, Richard (2000), “How Much Does Social Capital Add to Individual Health? A Survey Study of Russians,” Social Science & Medicine, vol. 51, no. 9: 1421–1435. St. John, Craig, and Jesse Fuchs (2002), “The Heartland Responds to Terror: Volunteering After the Bombing of the Murray Federal Building,” Social Science Quarterly, vol. 83, no. 2: 397–415. Sampson, Robert J., and John H. Laub (1996), “Socioeconomic Achievement in the Life Course of Disadvantaged Men: Military Service as a Turning Point, Circa 1940– 1965,” American Sociological Review, vol. 61, no. 3: 347–367. Sandel, Michael J. (1996), Democracy’s Discontent. Cambridge, MA: Harvard University Press. Sanger, David E., and Eric Schmitt (2002), “Threats and Respones: A Plan for Iraq; U.S. Has a Plan to Occupy Iraq, Officials Report,” The New York Times, October 11. Schwartz, Frank J., and Susan J. Pharr (2003), The State of Civil Society in Japan. Cambridge: Cambridge University Press. Schlozman, Key Lehman, Nancy Burns, and Sidney Verba (1994), “Gender and the Pathways to Participation: The Role of Resources,” Journal of Politics, vol. 56, no. 4: 963–990. Schlozman, Kay Lehman, Nancy Burns, and Sidney Verba (1999), “’What Happened at Work Today?’ A Multi-Stage Model of Gender, Employment, and Political Participation,” Journal of Politics, vol. 61, no. 1: 29–54. Schofer, Evan, and Marion Foucarde-Gourinchas (2001), “The Structural Contexts of Civic Engagement: Voluntary Association Membership in Comparative Perspective,” American Sociological Review, vol. 66, no 6: 806–828. Sears, David O., and Carolyn Funk (1999), “Evidence of the Long-Term Persistence of Adults’ Political Predispositions,” Journal of Politics, vol. 61, no. 1: 1–28. Sears, David O., and Nicholas A. Valentino (1997), “Politics Matters: Political Events as Catalysts for Preadult Socialization,” American Political Science Review, vol. 91, no. 1: 45–65. Siisiänen, Martti (1999), “Voluntary Associations and Social Capital in Finland,” in Jan W. Van Deth, Marco Maraffi, Kenneth Newton, and Paul F. Whiteley, eds., Social Capital and European Democracy. London and New York: Routledge. Sivard, Ruth Leger (1986), World Military and Social Expenditures. Leesburg, VA: WMSE Publications. Skocpol, Theda (1979), States and Social Revolutions: A Comparative Analysis of France, Russia, and China. Cambridge: Cambridge University Press. (1992), Protecting Soldiers and Mothers: The Political Origins of Social Policy in the United States. Cambridge, MA: Harvard University Press. (1997), “The Tocqueville Problem,” Social Science History, vol. 21, no. 4: 455–79. (1999), “How Americans Became Civic,” in Theda Skocpol and Morris P. Fiorina, Civic Engagement in American Democracy. Washington DC: The Brookings Institution Press. (2000), “A Nation of Organizers: The Institutional Origins of Civic Voluntarism in the United States” (with Marshall Ganz and Ziad Muson), American Political Science Review, vol. 94, no. 3: 527–246.

References

181

Skocpol, Theda, Ziad Munson, Anderw Karch, and Bayliss Camp (2002), “Patriotic Partnerships: Why Great Wars Nourished American Civic Voluntarism,”in Ira Katznelson and Martin Shefer, eds., Shaped by War and Trade: International Influences on American Political Development. Princeton, NJ: Princeton University Press. Smith, David Horton (1994), “The Determinants of Voluntary Association Participation and Volunteering: A Literature Review,” Nonprofit and Voluntary Sector Quarterly, vol. 23, no. 3: 243–264. Smith, Harold L., ed. (1996), Britain in the Second World War: A Social History. Manchester and New York: Manchester University Press. South, Scott J. (1985), “Economic Conditions and the Divorce Rate: A Time-Series Analysis of Postwar United States,” Journal of Marriage and the Family, vol. 47, no. 1: 31–41. “Statistics on Religions in Japan (as of 31 December 1947)” (n.d.), Box 5164; Statistics on Religions in Japan, 1945–1952; Topic File, compiled 1945–1952; Supreme Commander for the Allied Powers. Civil Information and Education Section. Administrative Division. (10/02/1945–04/28/1952); Records of Allied Operational and Occupation Headquarters, World War II, 1907–1966, Record Group 331; National Archives at College Park, College Park, MD. “Statistics on Religions in Japan (as of 31 December 1949)” (n.d.), Box 5164; Statistics on Religions in Japan, 1945–1952; Topic File, compiled 1945–1952; Supreme Commander for the Allied Powers. Civil Information and Education Section. Administrative Division. (10/02/1945–04/28/1952); Records of Allied Operational and Occupation Headquarters, World War II, 1907–1966, Record Group 331; National Archives at College Park, College Park, MD. Steenbergen, Marco R. and Bradford S. Jones (2002), “Modeling Multilevel Data Structures,” American Journal of Political Science, vol. 46, no. 1: 218–37. Stein, Arthur A. (1980), The Nation at War. Baltimore, MD: Johns Hopkins University Press. Stein, Arthur A. and Bruce M. Russett (1980), “Evaluating War: Outcomes and Consequences,” in Ted Gurr, ed., Handbook of Political Conflict: Theory and Research. New York: Free Press. Stohl, Michael (1980), “The Nexus of Civil and International Conflict,” in Ted Gurr, ed., Handbook of Political Conflict: Theory and Research. New York: Free Press. Stolle, Dietlind, and Marc Hooghe (2005), “Review Article: Inaccurate, Exceptional, One-Sided or Irrelevant? The Debate about the Alleged Decline of Social Capital and Civic Engagement in Western Societies,” British Journal of Political Science, vol. 35, no. 1: 149–67. Strom, Stephanie (2005), “Asia’s Deadly Waves: The Donations; U.S. Charities Overwhelmed by Disaster Aid,” The New York Times, January 13. Tatsuki, Shigeo, and Haruo Hayashi (2000), “Family System Adjustment and Adaptive Reconstruction of Social Reality among the 1995 Earthquake Survivors,” International Journal of Japanese Sociology, vol. 9: 81–100. Taeuber, Irene B. (1958), The Population of Japan. Princeton, NJ: Princeton University Press. Tilly, Charles (1975), “Reflections on the History of European State-Making,” in Charles Tilly, ed., The Formation of National States in Western Europe. Princeton, NJ: Princeton University Press. (2004), Contention and Democracy in Europe, 1650–2000. Cambridge: Cambridge University Press.

182

References

Titmuss, Richard M. (1955), “War and Social Policy,” in Richard M. Titmuss, Essays on the Welfare State. London: George Allen and Unwin Ltd. Tocqueville, Alexis de (2000), Democracy in America (Translated, Edited, and with an Introduction by Harvey C. Mansfield and Delba Winthrop). Chicago: University of Illinois Press. Tomz, Michael, Jason Wittenberg, and Gary King (2001). CLARIFY: Software for Interpreting and Presenting Statistical Results. Version 2.0 Cambridge, MA: Harvard University, June 1 from http://gking.harvard.edu. Torcal, Mariano, and José Ramón Montero (1999), “Facets of Social Capital in New Democracies: The Formation and Consequences of Social Capital in Spain,” in Jan. W. Van Deth, Marco Maraffi, Kenneth Newton, and Paul F. Whiteley, eds., Social Capital and European Democracy. London and New York: Routledge. Torpe, Lars (2003), “Social Capital in Denmark: A Deviant Case?” Scandinavian Political Studies, vol. 26, no. 1: 27–48. Trainor, Joseph C. (1983), Education Reform in Occupied Japan: Trainor’s Memoir. Tokyo: Meisei University Press. Tsai, Lily L. (2007), “Solidary Groups, Informal Accountability, and Local Public Goods Provision in Rural China,” American Political Science Review, vol. 101, no. 2: 355–372. United Nations Secretariat, Statistical Office (various years), Monthly Bulletin of Statistics. Uslaner, Eric M. (2000), The Moral Foundations of Trust. Cambridge: Cambridge University Press. Verba, Sidney, Nancy Burns, and Kay Lehman Schlozman (1997), “Knowing and Caring About Politics: Gender and Political Engagement,” Journal of Politics, vol. 59, no. 4: 1051–72. Verba, Sidney, Kay Lehman Schlozman, and Henry E. Brady (1995), Voice and Equality: Civic Voluntarism in American Politics. Cambridge, MA: Harvard University Press. Werner, Suzanne (1999), “The Precarious Nature of Peace,” American Journal of Political Science, vol. 43, no. 3: 912–934. Wilson, John (2000), “Volunteering,” Annual Review of Sociology, vol. 26: 215–240. World Bank (1998), Post-Conflict Reconstruction. Washington, DC: The World Bank. (2005), Reshaping the Future: Education and Postconflict Reconstruction. Washington, DC: The World Bank. Wynn, Neil A. (1971), “The Impact of the Second World War on the American Negro,” Journal of Contemporary History, vol. 6, no. 2: 42–53. Yamamoto, Tadashi., Akira Iriye and Makoto Iokibe eds. (2006), Philanthropy and Reconciliation: Rebuilding Postwar US-Japan Relations. Tokyo: Japan Center for International Exchange. Japanese Language Sources Abe, Tamaki (2009), Hiroshima; Toshi to Kyujo no Monogatari [Hiroshima: A Story of a City and a Stadium]. Tokyo: PHP Kenkyujo. Akagi, Suruki (1977), Tokyo Tosei no Kenkyu: Fusenka no Tokyo Shisei no Kozo [Local Politics in Tokyo: Local Rule in Tokyo Under Universal Suffrage]. Tokyo: Miraisha. Akimoto, Ritsuo (1974), Senso to Minshu: Taiheiho Sensoka no Toshi Seikatsu [War and the Citizen: Japanese Urban Life During the Pacific War]. Tokyo: Gakuyo Shobo.

References

183

Baba, Hiroaki (1998), Taishi no Keifu: Ichiko to Sapporo Nogakko [The First Upper High School and the Sapporo Agricultural College]. Tokyo: Hokusensha. Chuo Shakai Jigyo Kyokai (various years), Nihon Shakai Jigyo Nenkan [Almanac of Social and Charitable Activities in Japan]. Tokyo: Chuo Shakai Jigyo Kyokai. Ebina, Kenzo (1991), Sapporo Nogakko [Sapporo Agricultural College]. Tokyo: Shinhyoron. (2000), Sapporo-shi no Toshi Keisei to Ikkyoku Shuchu [The Formation of the City of Sapporo]. Tokyo: Nishida Shoten. Fukuoka-ken Taiiku Kyokai (1976), Fukuokaken Taiiku Kyokai-shi. Fukuoka: Fukuokaken Taiiku Kyokai. Gakuto Doin (n.d.), “Gakuto Doin,” retrieved June 12, 2003 from http://www.ne.jp/ asahi/k/h/fujiwara/history/h08/dic/dic-gakutodouin.htm. Girl Scout Nihon Renmei (1971), Girl Scout Hanseiki no Ayumi [The Fifty-Year History of the Girl Scouts Japan]. Tokyo: Girl Scout Nihon Renmei. (2000), Nihon no Girl Scout Undo: Fu Shiryo Hen [The Girl Scout Movement in Japan: Statistical Appendix]. Tokyo: Girl Scout Nihon Renmei. Hanshin-Awaji Daishinsai Hisaichi no Shimin wo Oensuru Shimin no Kai (1996), “Hanshin- Awaji Daishinsai Hisaichi no Shimin wo Oensuru Shimin no Kai” Zen Kiroku [Support Group for Victims of the Hanshin-Awaji Earthquake: Records], retried October 5, 2004 from http://www.lib.kobe-u.ac.jp/html/kos/7–113/ eqb11_007.html. Hokkaido Shimbunsha, ed. (1972), Hokkaido Hyakunen: Gekan, Taisho-Showa Jidai Hen [The One-Hundred Year History of Hokkaido, vol. 3: The Taisho and Showa Periods]. Sapporo: Hokkaido Shimbunsha. Hokkaido Taiiku Kyokai (1968), Hokkaido Taiiku Kyokaishi [History of the Hokkaido Amateur Sports Association]. Sapporo: Hokkaido Taiiku Kyokai. Hokkaido YMCA Hyakunenshi Hensan Iinkai, ed. (1997), Subete no Waza niwa Tokiga aru: Hokkaido YMCA Hyakunenshi [The One Hundred-Year History of the Hokkaido YMCA]. Sapporo: Hokkaido YMCA. Hyogo-ken (2002), Hanshin-Awaji Daishinsai Ichinen no Kiroku [Records of the Hanshin-Awaji Earthquake: The First Year]. Kobe: Hanshin-Awaji Daishinsai Hyogoken Taisaku Honbu, retrieved August 20, 2003 from http://web.pref.hyogo. jp/syoubou/daishinsai/ichinen.html. Iizuka, Ichiyo (1990), Judo wo Tsukutta Otoktachi [The Men Who Made Judo]. Tokyo: Bungei Shunjusha. Imai, Seiichi (2007), Yokohama no Kanto Daishinsai [The Great Kanto Earthquake in Yokohama]. Yokohama: Yurindo. Inoue, Shun (2004), Budo no Tanjo [The Birth of the Martial Arts]. Tokyo: Yoshikawa Kobunkan. Ishikawa, Masumi (1978), Sengo Seiji Kozoshi [A Structural History of Japanese Postwar Politics]. Tokyo: Nihon Hyoronsha. Japan Research Institute (2008), Nihon no Social Capital to Seisaku [Social Capital in Japan and Policy]. Tokyo: Japan Research Institute. Kamiki Tetsuo (1993), “Kobe Gaikokujin Kyoryuchi no 4-bunno-3 Seiki {The SeventyFive Year History of the Kobe Foreign Settlement,” in Tetsuo Kamiki and Masahiro Sakiyama, eds., Kobe Gaikokujin Kyoryuchi no 4-bunno-3 Seiki. Kobe: Kobe Shimbun Sogo Shuppan Senta. Kanagawa-ken Judo Renmei (1994), Kanagawa-ken Judo Renmei 45shunen Kinenshi [The 45-year History of the Kanagawa Prefecture Judo Federation]. Yokohama: Kanagawa-ken Judo Renmei.

184

References

Kanagawa-ken Keisatsushi Hensan Iinkai, ed. (1972), Kanagawa-ken Keisatsushi [History of the Kanagawa Prefectural Police, vol. 2. Yokohama: Kanagawa-ken Keisatsu Honbu. (1974), Kanagawa-ken Keisatsushi [History of the Kanagawa Prefectural Police, vol. 3. Yokohama: Kanagawa-ken Keisatsu Honbu. Kaneda, Ryuichi (1987), “Junan to Fukko [Challenges and Revival],” in Sapporo-shi Kyoiku Iinkai, ed., Saporo to Kirisutokyo [Sapporo and Christianity]. Sapporo: Sapporo-shi. Kawaguchi, Masahide (1983), Kouto Yokohama: Sono Utsuri Kawari [Yokohama: Evolution of a Port City]. Yokohama: Books Ninomiya. Kawamura, Miyuki (1987),“Seinen Undo, Joshi Kyofukai, Kyuseigun [Youth Movements, Women’s Movements, and the Salvation Army],” in Sapporo-shi Kyoiku Iinkai, ed., Saporo to Kirisutokyo [Sapporo and Christianity]. Sapporo: Sapporo-shi. Kawazoe, Shoji, Junichi Takesue, Yoshitake Okafuji, Masahiro Nishitani, Yoshinori Kajiwara, and Etsuro Orita (1997), Fukuokaken no Rekishi [The History of Fukuoka Prefecture]. Tokyo: Yamakawa Shuppansha. Keizai Antei Honbu Sosai Kanbo Kikakubu Chosaka (1949), Taiheiho Senso ni yoru Wagakunino Higai Sogo Hokokusho [Summary Report: Damages to Japan as a Result of the Pacific War]. Tokyo: Keizai Antei Honbu Sosai Kanbo Kikakubu Chosaka Keizai Antei Honbu Kokudo Sogo Kaihatsu Jimsuho Henja, ed. (1950), Fukenbetsu Genkyo Bunseki Sogo Zuhyo [Illustrative Analyses of the State of the Prefectural Economies]. Tokyo: Kokumin Keizai Kenkyu Kyokai. Kikuchi, Keiichi (2007), Kataritsugu Hokkaido Kushu [Remembering the Air Raids on Hokkaido]. Sapporo: Hokkaido Shimbunsha. Kimoto, Mosaburo (1966), Showa on Tokyo YMCA [The Tokyo YMCA in the Showa Era]. Tokyo: Tokyo YMCA. “Kirisutokyo Seinen Undo wa Do Yukuka [Where is the Christian Youth Movement Headed?] (Roundtable Discussion)” (1947), Kaitakusha, vol. 41, no. 7: 26–30. Kobe Gaikokujin Kyoryuchi Kenkyukai, ed. (2005), Kobe to Kyoryuchi [Kobe and the Foreign Settlement]. Kobe: Kobe Shimbun Sogo Shuppan Senta. Kobe-shi (n.d.), “Shiiki Hensen zu [Changes in the Geographic Boundaries of Kobe],” retrieved January 9, 2004 from http://www.city.kobe.lg.jp/information/about/ energy/rekishi/index_menu.html. (1965), ed., Kobe-shi Shi, Dai 3 Shu: Shakai Bunka Hen [History of the City of Kobe, vol.3: Culture and Society]. Kobe: Kobe-shi. Kobe-shi Keizaibu Sangyo Kenkyusho (1939), Kobe-shi Sangyo Tokei Yoran [Statistical Abstract of the City of Kobe.] Kobe: Kobe-shi Keizaibu Sangyo Kenkyusho. Kobe-shi Kyoikushi Henshu Iinkai, ed. (1964), Kobe-shi Kyoikushi Dai 2 shu [The History of Education in the City of Kobe, vol. 2]. Kobe: Kobe-shi Kyoikushi Kanko Iinkai. (1966), Kobe-shi Kyoiku-shi Dai 1shu [The History of Education in the City of Kobe, vol. 1]. Kobe: Kobe-shi Kyoikushi Kanko Iinkai. Kobe Kushu wo Kiroku suru Kai (1972), Kobe Daikushu [The Great Air Raids on Kobe During World War II]. Kobe: Nojigiku Bunko. Kobe YMCA (n.d.), Minutes of the Monthly Board Meetings. Internal Documents. (1943), Showa 17nendo Jigyo Narabini Kaikei Hokoku [Annual Report of the Kobe YMCA: 1942]. Kobe: Kobe YMCA. (1947), Showa 21nendo Jigyo Narabini Kaikei Hokoku [Annual Report of the Kobe YMCA: 1946]. Kobe: Kobe YMCA.

References

185

(1969), YMCA 70yen no Ayumi: 1899–1969 [The 70-Year History of the Kobe YMCA: 1899–1969]. Kobe: Kobe YMCA. Kobe YMCA Hyakunenshi Hensanshitsu (1981a), Kobe YMCA no Rekishi wo Kataru: Soritsu Hyakunen ni Mukete, Zadankai-hen 2 [The History of Kobe YMCA, Roundtable Discussion: Towards the 100-year Anniversary, vol. 2]. Kobe: Kobe YMCA Hyakunenshi Hensanshitsu. (1981b), Kobe YMCA no Rekishi wo Kataru: Soritsu Hyakunen ni Mukete, Zadankaihen 3 [The History of Kobe YMCA, Roundtable Discussion: Towards the 100-year Anniversary, vol. 3]. Kobe: Kobe YMCA Hyakunenshi Hensanshitsu. (1985), Kobe YMCA no Rekishi wo Kataru: Soritsu Hyakunen ni Mukete, Zadankaihen 19-II [The History of Kobe YMCA, Roundtable Discussion: Towards the 100-year Anniversary, vol. 19, Part II]. Kobe: Kobe YMCA Hyakunenshi Hensanshitsu. (1987), Kobe to YMCA Hyakunen (Kobe and One Hundred Years of the YMCA). Kobe: Kobe YMCA. Kodokan Bunkakai (1933), Judo Nenkan [Judo Almanac]. Tokyo: Kodokan. (1936), Judo Nenkan [Judo Almanac]. Tokyo: Kodokan. Koseisho (1952), Shakai Fukushi Tokei Nenpo [Almanac of Social Welfare]. Tokyo: Koseisho Daijin Kanbo Tokei Chosabu. Koseisho Jido-kyoku (1950), Zenkoku Yohogo Jido Chosa Kekka [Results of the National Survey on Children in Need]. Tokyo: Koseisho Jido-kyoku. Kosei Rodo Sho (Ministry of Health, Labor, and Welfare) (n.d.), “Fuhyo 1: Rodo Kumiai Shurui Betsu Rodo Kumiaisu, Rodo Kumiaiinsu oyobi Suitei Soshikiritsu no Suii (Kakunen 6gatsu 30nichi Genzai) (Showa 22~Heisei 14nen) [Appendix 1: Number of Labor Unions and Labor Union Members, as of June 30th of Every Year, 1947– 2002],” retrieved January 30, 2003 from http://www.mhlw.go.jp/toukei/itiran/roudou/roushi/kiso/02/fuhyo1.html. (various years), Jinko Dotai Tokei [Vital Statistics]. Tokyo: Kosei Tokei Kyokai. Koseisho Tairyoku-kyoku (1940), Undojo Oyobi Undo Koen [Athletic Facilities and Athletic Parks in Japan]. Tokyo: Koseisho Tairyoku-kyoku. Kosai, Yutaka (1981), Kodo Seicho no Jidai: Gendai Nihon Keizaishi Noto [The Era of High-Speed Growth: Notes on the Postwar Japanese Economy]. Tokyo: Nihon Hyoronsha. Kudo, Raisuke (1972), Hiroku Nihon Judo [The Unknown History of Japanese Judo]. Tokyo: Tokyo Sports Sha. Kuroki, Morifumi (2004), “Tetsudo Fusetsu to Miike-Chikuho Tanden [Construction of the Railroad and the Miike-Chikuho Coal Mines,” in Yasunari Maruyama and Yoichi Cho, eds., Hakata-Fukuoka to Saikaido. Tokyo: Yoshikawa Kobunkan. Kusumoto, Toshio (2007), Kokusai Toshi Kobe no Keifu [History of an International City: Kobe]. Tokyo: Kojin no Tomo Sha. Mainichi Shinbunsha (various years), Mainichi Nenkan [The Mainichi Almanac]. Osaka and Tokyo: Mainichi Shinbunsha. Mainichi Shimbunsha Yokohama Shikyoku (1968), Kanagawano Hyakunen Gekan [The One-Hundred Year History of Kanagawa Prefecture, vol. 2]. Yokohama: Yurindo. Maruyama, Sanzo, ed., (1967), Sekai Judoshi [The World History of Judo]. Tokyo: Koyusha. Maruyama, Yasunari, and Yoichi Cho (2004), Hakata-Fukuoka to Saikaido [HakataFukuoka and the Saikaido Artery]. Tokyo: Yoshikawa Kobunkan. Matsubara, Ryuichiro (2006), Budo wo Ikiru. Tokyo: NTT Shuppan.

186

References

Matsubara, Takaomi (2009), “Uketsugareta Tamashii: Hiroshima Shin Stadium ni Komerareat Mono [Enduring Passion: The New Stadium in Hiroshima],” Number, vol. 30, no. 23: 88–93. Matsunaga, Ikuko (2001), Carp: Kunan wo Norikoeta Otoko tachino Kiseki [Hiroshima Carp: History of Men Who Overcame Adversity]. Tokyo: Takarajimasha. Mitani, Taichiro (1997), “Senji Taisei to Sengo Taisei [Wartime Systems and Postwar Systems],” in Taichiro Mitani, ed., Kindai Nihon no Senso to Seiji. Tokyo: Iwanami Shoten. Monbusho (various years), Shukyo Nenkan [Annual Report on Religions]. Tokyo: Okurasho Insatsu-kyoku. (various years), Nihon Teikoku Monbusho Nenpo. Tokyo: Monbusho. (1950), Gakko Kihon Chosa Hokokusho [Basic Survey on Schools]. Tokyo: Monbusho Chosa-kyoku Tokeika. (1953), Shukyo Dantaisu, Kyoshisu, Shinjasu Ichiran [Comprehensive Statistics on the Number of Religious Organizations, Religious Employment, and Religious Membership]. Tokyo: Monbusho Chosa-kyoku. (1954), Monbusho Dai 82 Nenpo, Showa 29-Nendo [Annual Report of the Ministry of Education, 1954]. Tokyo: Monbusho Chosa-kyoku Tokeika. (1972), Kyoiku Hyakunen no Ayumi [The 100-Year History of Education]. Tokyo: Monbusho. Monbusho Futsu Gakumu-kyoku (1922), Shonen Dantai no Gaikyo [The Current State of Youth Organizations]. Tokyo: Monbusho Futsu Gakumu-kyoku. (1924), Shonen Dantai no Gaikyo [The Current State of Youth Organizations]. Tokyo: Monbusho Futsu Gakumu-kyoku. Monbusho Shakai Kyoiku-kyoku (1937a), Wagakuni Shounendan Gaikyo [The Current State of Youth Organizations in Japan]. Tokyo: Monbusho Shakai Kyoiku-kyoku. (1937b), Doitsu Seishonen-dan Undo [Youth Movements in Germany]. Tokyo: Monbusho Shakai Kyoiku-kyoku. Monbusho Shakai Kyoiku-kyoku, ed. (1952), Shakai Kyoiku no Genjo, Showa 27-nendo [The State of Social Education in Japan, 1952], Tokyo: Monbusho Shakai Kyoikukyoku. (1953a), Fujin Kyoiku Shiryo [Sources on Women’s Education]. Tokyo: Okurasho Insatsu-kyoku. (1953b), Shakai Kyoiku no Genjo, Showa 28-nendo [The State of Social Education in Japan, 1953], Tokyo: Monbusho Shakai Kyoiku-kyoku. (1954a), Shakai Kyoiku no Genjo, 1954: Shakai Kyoiku Ho Shiko 5 Shunen Kinen [The Current State of Social Education in Japan: Commemorating the 5th Anniversary of the Adult Education Law]. Tokyo: Okurasho Insatsu-kyoku. (1954b), Fujin Kyoiku Shiryo [Sources on Women’s Education]. Tokyo: Okurasho Insatsu-kyoku. (1955), Shakai Kyoiku no Genjo, Showa 30-nendo [The State of Adult Education in Japan, 1955], Tokyo: Monbusho Shakai Kyoiku-kyoku. (1956), Shakai Kyoiku no Genjo, Showa 31-nendo [The State of Adult Education in Japan, 1956], Tokyo: Monbusho Shakai Kyoiku-kyoku. (1959a), Shakai Kyoiku 10nen no Ayumi: Shakai Kyoiku-ho Shiko 10shunen Kinen [The Ten Year-History of Social Education in Japan: Commemorating the 10th Anniversary of the Social Education Law]. Tokyo: Monbusho.

References

187

(1959b), Fujin Kyoiku Shiryo, Showa 34nendo [Facts and Figures on Women’s Education, 1959]. Tokyo: Okurasho Insatsu-kyoku. (1961), Fujin Kyoiku 15-nen no Ayumi [15 Years of Women’s Education]. Tokyo: Okurasho Insatsu-kyoku. (1969), Fujin Kyoiku no Genjo [The Status of Women’s Education in Japan]. Tokyo: Okurasho Insatsu-kyoku. (1976), Fujin Kyoiku oyobi Katei Kyoiku ni Kansuru Shisaku no Genjo [Current Policies on Women’s and Family Education]. Tokyo: Okurasho Insatsu-kyoku. (1981), Fujin Kyoiku oyobi Katei Kyoiku ni Kansuru Shisaku no Genjo [Current Policies on Women’s and Family Education]. Tokyo: Okurasho Insatsu-kyoku. (1986), Fujin Kyoiku oyobi Katei Kyoiku ni Kansuru Shisaku no Genjo [Current Policies on Women’s and Family Education]. Tokyo: Okurasho Insatsu-kyoku. (1991), Fujin Kyoiku oyobi Katei Kyoiku ni Kansuru Shisaku no Genjo [Current Policies on Women’s and Family Education]. Tokyo: Okurasho Insatsu-kyoku. Muramatsu, Michio (1979), Sengo Nihon no Kanryosei [The Bureaucracy in Postwar Japan]. Tokyo: Toyo Keizai Shimposha. (1988), Chiho Jichi [Local Autonomy]. Tokyo: Tokyo Daigaku Shuppankai. Muramatsu, Michio, Mitsutoshi Ito, and Yutaka Tsujinaka (1986), Sengo Nihon no Atsuryoku Dantai [Pressure Groups in Postwar Japan]. Tokyo: Toyo Keizai Shimposha. Murata, Tetsusaburo (1939), Shukyo Nenkan: Showa 14nen Ban [Annual Report on Religions: 1939 Edition]. Tokyo: Yukosha. Nada Kobe Seikatsu Kyodo Kumiai (1971), Nada Kobe Seikyo 50-nen no Ayumi [The 50-year History of the Nada-Kobe Consumer Cooperatives]. Kobe: Nada Kobe Seikatsu Kyodo Kumiai. Nagai, Hideo and Hiroaki Matsuzawa (1987), “Josho: Sapporo to Kirisutokyo eno Shikaku [Perspectives on Christianity in Sapporo],” in Sapporo-shi Kyoiku Iinkai, ed., Sapporo to Kirisutokyo [Sapporo and Christianity]. Sapporo: Sapporo-shi. Naikaku Tokei-kyoku (1925), Taisho 9nen Kokusei Chosa Hokoku: Fuken no Bu, Dai 5 kan, Hyogo [Census Statistics, 1920: Prefectural Figures, vol. 5, Hyogo]. Tokyo: Tokyo Tokei Kyokai. (1927), Taisho 9nen Kokusei Chosa Hokoku: Fuken no Bu, Dai 47 kan, Hokkaido [Census Statistics, 1920: Prefectural Figures, vol. 47, Hokkaido]. Tokyo: Tokyo Tokei Kyokai. (1933), Showa 5nen Kokusei Chosa Hokoku: Fuken no Bu, Dai 4 kan,Hyogo [Census Statistics, 1930: Prefectural Figures, vol. 4, Hyogo]. Tokyo: Tokyo Tokei Kyokai. (1934), Showa 5nen Kokusei Chosa Hokoku: Fuken no Bu, Dai 4 kan, Hokkaido [Census Statistics, 1930: Prefectural Figures, vol. 4, Hokkaido]. Tokyo: Tokyo Tokei Kyokai. Naikakufu (2007), Kokumin Seikatsu Hakusho [White Paper on the National Lifestyle, Fiscal Year 2007]. Tokyo: Naikakufu. Nakamura, Takafusa (1989), “Keikaku-ka” to “Minshu-ka [Planning and Democratization]. Tokyo: Iwanami Shoten. Nakamura, Takafusa and Miyazaki Masayasu, eds. (1995), Shiryo: Taiheiyo Senso Higai Chosa Hokokusho [Historical Sourcebook: Reports on the Extent of Damages from the Pacific War]. Tokyo: Tokyo Daigaku Shuppankai. Nara, Jogoro (1958), Nihon YMCA-shi [History of the YMCA Japan]. Tokyo: Nihon YMCA Domei.

188

References

Nara, Tsutae (1974), Senrino Michi [One Thousand Miles]. Osaka: Nara Tsutae Shuki Zuihitsushu Kankokai. Nara Kobe Seikatsu Kyodo Kumiai (1971), Nada Kobe Seikyo 50nen no Ayumi [The Fifty-Year History of the Nada Kobe Consumer Cooperatives]. Kobe: Nada Kobe Seikatsu Kyodo Kumiai. National Diet Library [Kokuritsu Kokkai Toshokan Chosa Rippo Kosa-kyoku] (1955), Meiji Iko Kyoiku Bunka no Tokei to Zuhyo [Figures and Statistics on Education and Culture Since the Meiji Period]. Tokyo: Kokuritsu Kokkai Toshokan. Nihon Bunka Chuo Renmei (1938), Nihon Bunka Dantai Nenkan [Yearbook of Cultural Groups in Japan]. Tokyo: Nihon Bunka Chuo Renmei. (1939), Nihon Bunka Dantai Nenkan [Yearbook of Cultural Groups in Japan]. Tokyo: Nihon Bunka Chuo Renmei. (1943), Nihon Bunka Dantai Nenkan [Yearbook of Cultural Groups in Japan]. Tokyo: Nihon Bunka Chuo Renmei. Nihon Kaiin Ekisaikai, ed. (1981), Ekisai Hyakunen [One Hundred Years of Seafarers’ Relief]. Tokyo: Nihon Kaiin Ekisaikai. Nihon Kirisuto Kyodan Sapporo Kyokai (1962), Kahanno Sento: Sapporo Kyokai 75nenshi [The Tower by the River: The 75-year History of the Sapporo Church]. Sapporo: Nihon Kirisuto Kyodan Sapporo Kyokai. Nihon Kirisuto Kyokai Sapporo Kita Ichijo Kyokai Rekishi Hensan Iinkai, ed. (2000), Nihon Kirisuto Kyokai Sapporo Kita Ichijo Kyokai Hyakunenshi: 1890–1995 [The One Hundred-Year History of the Sapporo Kita Ichijo Church: 1890–1995]. Sapporo: Nihon Kirisuto Kyokai Sapporo Kita Ichijo Kyokai. Nihon Sangakukai (2007), Nihon Sangakukai Hyakunenshi [The One Hundred-Year History of the Japan Alpine Club]. Tokyo: Meikeido. Nihon Taiiku Kyokai (1968), Supotsu Shisetsu Chosa Hokokusho [Report on Sports Facilities in Japan]. Tokyo: Nihon Taiiku Kyokai. Nihon YWCA (1987), Mizuwo Kazewo Hikariwo, Nihon YWCA Hachijunen [Water, Wind, and Light: The Eighty-Year History of the YWCA Japan]. Tokyo: Nihon Kirisutokyo Joshi Seinenkai. (2005a), Nihon YWCA Hyakunenshi [The One-Hundred Year History of the YWCA Japan]. Tokyo: Nihon Kirisutokyo Joshi Seinenkai. (2005b), Nihon YWCA Hyakunenshi Nenpyo [The One-Hundred Year History of the YWCA Japan, Timeline]. Tokyo: Nihon Kirisutokyo Joshi Seinenkai. Nishi Nihon Shimbunsha, ed. (1997), Kinshuki 70-nenshi [The Seventy-Year History of the Kinshuki Judo Tournament]. Fukuoka: Nishi Nihon Shimbunsha. Nishida, Kiyoko (1980), “Doro no Hatake Dukuri [Farming on the Side of the Streets,]” in Sapporo-shi Kyoiku Iinkai, ed., Showa Nijunen no Kiroku [Records from 1945]. Sapporo: Hokkaido Shimbunsha. Nisseikyo Soritsu 50 Shunen Kinen Rekishi Hensan Iinkai (2001), Gendai Nihon Seikyo Undoshi Shiryoshu [Records of the Contemporary Cooperative Movement in Japan]. Tokyo: Nihon Seikatsu Kyodo Kumiai Rengokai. Noda, Masaaki (1998), Senso to Zaiseki [War and Guilt]. Tokyo: Iwanami Shoten. Noguchi, Yukio (1995), 1940-nen Taisei [The 1940 System]. Tokyo: Toyo Keizai Shinposha. Nomura, Takeo (1975), Kyoto YMCA 70-nen Shi [The 70-Year History of the Kyoto YMCA]. Kyoto: Kyoto YMCA. Ochiai, Shigenobu (1975), Kobe no Rekishi: Kodai kara Kindai made [The History of Kobe: From Ancient Times to Modernity]. Kobe: Goto Shoten.

References

189

Ochiai, Shigenobu and Hajime Arii (1967), Kobe Shiwa: Kindaika Urabanashi [The History of Kobe: Modernization and Anecdotes]. Osaka: Sogensha. Oe, Shinobu (1988), Shina Jihen Daitoa Sensokan Doin Gaishi: Jugonen Senso Gokuhi Shiryoshu Dai 9 Shu [History of Wartime Mobilization in Japan During the War with China and the “Great East Asian War”: Secret Documents from the “Fifteen Years’ War, volume 9]. Tokyo: Fuji Shuppan. Ohara Shakai Mondai Kenkyusho (various years), Rodo Nenkan [Labor Almanac]. Tokyo: Shakai Mondai Kenkyusho. Oimatsu, Shinichi (1966), Judo Hyakunen [The One Hundred-Year History of Judo]. Tokyo: Jiji Tsushinsha. Okahisa, Kosaburo (1978), Kobe Shishi Gaisetsu: Kobe Monogatari [Introduction to the History of Kobe: The Story of Kobe]. Tokyo: Rekishi Toshosha. Osaka YMCA (1982), Osaka YMCA Hyakunen-Shi [The 100-Year History of the Osaka YMCA]. Osaka: Osaka YMCA. Otake, Hideo (1996), Sengo Nihon no Ideorogi Tairitsu [Ideological Conflicts in Postwar Japan]. Tokyo: San-ichi Shobo. Owaku, Yasutaro (1984), Yokohama YMCA Hyakunen-Shi [The 100-Year History of the Yokohama YMCA]. Yokohama: Yokohama YMCA. Rodosho (1955), Rodo Kumiai Chosa Hokoku [Annual Report on Labor Unions]. Tokyo: Rodosho. Rotary Club of Japan (n.d.), “Rotary Kakudai Nenpyo (Nihon) [The Annual Expansion of the Rotary Club (Japan),” retrieved January 31, 2003 from http://www.rotary. or.jp/data/toukei/kakudai-j.html. Saito, Minoru (1980), Tokyo Kirisutokyo Seinenkai Hyakunenshi [The One-Hundred Year History of the Tokyo YMCA]. Tokyo: Tokyo YMCA. Sapporo Joseishi Kenkyukai, ed. (1986), Kitano Joseishi [Women’s History in Hokkaido]. Hokkaido: Hokkaido Shinbunsha. Sapporo Kyokai Hyakunen no Ayumi Henshu Iinkai, ed. (1992), Sapporo Kyokai Hyakunen no Ayumi [The One-Hundred Year History of the Sapporo Church]. Sapporo: Nihon Kirisuto Kyodan Sapporo Kyokai. Sapporo-shi Kyoiku Iinkai (1985), Sapporo Seikatsu Bunkashi [The Cultural History of Sapporo]. Sapporo: Sapporo-shi. (1989), Shin Sapporo-shi shi, Dai 1 kan [The New History of Sapporo, vol. 1]. Sapporo: Sapporo-shi. (1991), Shin Sapporo-shi shi, Dai 2 kan [The New History of Sapporo, vol. 2]. Sapporo: Sapporo-shi. (1994), Shin Sapporo-shi shi, Dai 3 kan [The New History of Sapporo, vol. 3]. Sapporo: Sapporo-shi. (1997), Shin Sapporo-shi shi, Dai 4 kan [The New History of Sapporo, vol. 4]. Sapporo: Sapporo-shi. Sapporo-shi Kyoiku Iinkai (2000), Shin Sapporo-shi shi, Dai 8 kan: Tokei-hen [The New History of Sapporo, vol. 8: Statistics]. Sapporo: Sapporo-shi. Sapporo-shi Kyoiku Iinkai (2002), Shin Sapporo-shi shi, Dai 5 kan [The New History of Sapporo, vol. 5]. Sapporo: Sapporo-shi. Sapporo-shi Shobo Enkakushi Hensan Iinkai (1971), Sapporo Shobo Hyakunen no Ayumi [The One Hundred-Year History of the Sapporo Fire Department]. Sapporo: Sapporo-shi Shobo-kyoku. Sapporo Ski Renmei (1979), Miyasama Ski Taikai 50-nenshi [The Fifty-Year History of the Imperial Ski Competition]. Sapporo: Zaidan Hojin Sapporo Ski Renmei.

190

References

Sawai, Minoru (2002), “Senso niyoru Seido no Hakai to Kakushin [The Effect of War on the Destruction and Renewal of Institutions],” in Shakai Keizaishi Gakkai, ed., Shakai Keizaishigaku no Kadai to Tenbo [Studies in Socio-Economic History, 1991–2001]. Tokyo: Yuhikaku. Seikatsu Kyodo Kumiai Coop Kobe 70 shunen Kinen Jigyo Suishinshitsu (1991), Hyakumannnin no Wa: Coop Kobe 70 nen no Ayumi [One Million Members: The Seventy-Year History of the Kobe Consumer Cooperatives]. Kobe: Seikatsu Kyodo Kumiai Coop Kobe. Shonendan Nihon Renmei (various years), Shonendai Nihon Renmei Kameidan Meibo. Tokyo: Shonendan Nihon Renmei. Shukyo Nenkan, Showa 14nendo ban [Annual Report on Religions: 1939]. Tokyo: Yukosha. Somucho (1987). Nihon Choki Tokei Soran [Historical Statistics of Japan], vol. 1. Tokyo: Nihon Tokei Kyokai. (1988a). Nihon Choki Tokei Soran [Historical Statistics of Japan], vol. 2. Tokyo: Nihon Tokei Kyokai. (1988b). Nihon Choki Tokei Soran [Historical Statistics of Japan], vol. 4. Tokyo: Nihon Tokei Kyokai. (1988c). Nihon Choki Tokei Soran [Historical Statistics of Japan], vol. 5. Tokyo: Nihon Tokei Kyokai. Somusho (n.d.[1]), “Waga kuni no Suikei Jinko, Taisho 9~Heisei 12 [Estimated Population of Japan, 1920~2000,” retrieved May 16, 2004 from http://www.stat. go.jp/data/jinsui/wagakuni/index.htm. (n.d.[2]). Ordinary Accounts of Local Government [Online], retrieved August 28, 2007 from http://www.stat.go.jp/data/chouki/zuhyou/05–19.xls. . (n.d.[3]), Historical Statistics of Japan [Online], retrieved June 20, 2008 from http:// www.stat.go.jp/data/chouki/24.htm. Sorifu Tokei-kyoku (1961), Showa 15-nen Kokusei Chosa Hokoku Dai 1 Kan [Report of the 1940 Census, vol. 1]. Tokyo: Sorifu Tokei-kyoku. Takahashi, Kyozo (1980), “Rikugun Ryomatsusho deno 8-gatsu 15-nichi [August 15th at the Army Food Supply Center],” in Sapporo-shi Kyoiku Iinkai, ed., Showa Nijunen no Kiroku [Records from 1945]. Sapporo: Hokkaido Shimbunsha. Takamura, Naosuke (2006), Toshi Yokohama no Hanseiki: Shinsai Fukko kara Kodo Seicho Made [The Half-Century History of Yokohama: From Post-Earthquake Recovery to High-Speed Economic Growth]. Yokohama: Yurindo. Tamura, Akira (1989), Toshi Yokohama Monogatari [The Story of Yokohama]. Tokyo: Jiji Tsushinsha. Tanaka, Haruhiko (1988), Gakko-gai Kyoiku Ron [Education Outside the Schools]. Tokyo: Gakuyo Shobo. (1995), Boy Scout. Tokyo: Chuo Koronsha. (1999), Shonendan Undo no Seiritsu to Tenkai [The Rise and Evolution of the Boy Scout Movement in Japan]. Fukuoka: Kyushu Daigaku Shuppankai. Teishinsho (1949), Yubin Tokei Yoran, Showa 23-nendo [Yearbook of Postal Mail, Fiscal Year 1948]. Tokyo: Teishinsho. Todo, Yoshiaki (2007), Judo no Rekishi to Bunka [The History and Culture of Judo]. Tokyo: Fumaido Shuppan. Tokyo Rengo Fujinkai (various years), Fujin Nenkan [Almanac on Women]. Tokyo: Tokyo Rengo Fujinkai.

References

191

Tsuji, Kiyoaki (1958), “Kanryo Kiko no Onzon to Kyoka [Continuity and Reinforcement in the Japanese Bureaucracy],” in Yoshitake Oka, ed., Sengo Nihon no Seiji Katei [Political Process in Postwar Japan]. Tokyo: Iwanami Shoten. Tsujinaka Yutaka (1986), Rieki Shudan [Interest Groups]. Tokyo: Tokyo Daigaku Shuppankai. Tsujinaka, Yutaka, ed. (2002), Gendai Nihon no Shimin Dantai Rieki Dantai [Civic and Interest Groups in Contemporary Japan]. Tokyo: Bokutakusha. Tsujinaka, Yutaka and Jaeho Yeom (2002), “Rekishiteki Keisei [Historical Evolution],” in Tsujinaka, Yutaka, ed., Gendai Nihon no Shimin Dantai Rieki Dantai. Yaguchi, Tetsuya (2008), Joshi Hododan: Nihonnno Girl Scout Zenshi [Joshi Hododan: A Pre-History of the Girl Scouts of Japan]. Tokyo: Seibundo. Yamaguchi, Yasushi (2004), Shimin Shakairon: Rekishiteki Isan to Shintenkai [Civil Society: Historical Inheritance and New Developments]. Tokyo: Yuhikaku. Yamamoto, Reiko (2003), Beikoku Tainichi Senryou Seisaku to Budou Kyoiku [US Occupation Policies Towards the Japanese Martial Arts]. Tokyo: Nihon Tosho Center. Yamazaki, Chokichi (1986), Sapporo Kyoikushi, Jokan [The History of Education in Sapporo, vol. 1]. Tokyo: Daiichi Hoki Shuppan. (1992), Sapporo Kyoikushi, Chukan [The History of Education in Sapporo, vol. 2]. Tokyo: Daiichi Hoki Shuppan. YMCA Japan (various years), Annual Reports. Tokyo: YMCA Japan. YMCA Shigakukai Henshyu Iinkai (2003), Shinpen Nihon YMCA-shi [The History of YMCA Japan: New Edition]. Tokyo: Nihon YMCA Domei. Yokohama-shi (n.d.), “Yokohama-shi Tokei Portal Site [Portal Site: Statistics on the City of Yokohama],” retrieved February 19, 2010 from http://www.city.yokohama. jp/me/stat/jinko/ayumi/01/ayumi-j.html. (1961), Yokohama-shi-shi [History of the City of Yokohama], vol. 3, Book 1. Yokohama: Yurindo. (1963), Yokohama-shi-shi [History of the City of Yokohama], vol. 3, Book 2. Yokohama: Yurindo. (1993), Yokohama-shi-shi II [The Second History of the City of Yokohama], vol. 1, Book 1. Yokohama: Yokohama-shi. Yokohama-shi Kyoiku Iinkai (1976), Yokohama-shi Kyoikushi [The History of Education in the City of Yokohama], vol. 1. Yokohama: Yokohama-shi Kyoiku Iinkai. (1978), Yokohama-shi Kyoikushi [The History of Education in the City of Yokohama], vol. 2. Yokohama: Yokohama-shi Kyoiku Iinkai. Yokohama-shi Taiikushi Kikaku Kanko Iinkai (1989), Yokohama Sports Hyakunen no Ayumi [The One-Hundred Year History of Sports in Yokohama]. Yokohama: Yokohama-shi Kyoiku Iinkai Jimukyoku Taiikuka. Yoshida, Masaaki (1969), Oukoku no Keifu: Kyushu Judo no Nagare [History of a Dynasty: Judo in Kyushu]. Fukuoka: Nishi Nihon Shinbunsha. (1970), Takemura Shigetaka Judo ni Ikiru [Shigetaka Takemura’s Dedication to Judo]. Fukuoka: Kyushu Judo Kyokai. Yoshikawa, Takeshi (1955), Nanaju-Gonen Tokyo no Seinen to Tomoni: Tokyo YMCA no Ayumi [Seventy-Five Years with Tokyo’s Youth: The History of the Tokyo YMCA]. Tokyo: Tokyo YMCA. Yoshimura, Kinji (1964), Nagoya YMCA 60nen-Shi [The 60-Year History of the Nagoya YMCA]. Nagoya: Nagoya YMCA.

192

References

Zaidan Hojin Boy Scout Nihon Renmei (1978), Showa 53 Nendo Jigyo Hokokusho Bessatsu: Kakusho Tokei Shiryo [Annual Report, 1978, Statistical Appendix], Tokyo: Zaidan Hojin Boy Scout Nihon Renmei. Zako, Ken (1990), “Jugo” no Kiroku: Kobe Sensaiki [Memoirs of the Home Front: Wartime Destruction in Kobe]. Tokyo: Shinjuku Shobo. Zenkoku Tomo no Kai Chuobu (2000), Zenkoku Tomo no Kai Shoshi: 70 shunen Kinen [A Brief History of the Tomo no Kai: The 70-Year Anniversary]. Tokyo: Zenkoku Tomo no Kai.

Index

administered mass organization, 51, 53, 105, 116 Afghanistan, 168 Aikoku Fujinkai, 116 Ainu, 110, 113 air raids, 51, 105–107, 116–117, 130–131, 135 All Japan Judo Federation, 132 Arishima, 113 authoritarianism, 45, 54, 147 baby boom, 72, 74, 149 Belgium, 79, 82–83, 85–88, 91–94, 102, 165 bowl, 52, 78, 94, 165 Brady, Henry E., 53, 81 Cambodia, 143 Canada, 82–83, 85, 87–89, 160 casualt, 75 censorship, 107, 117 Centeno, Miguel, 50 child care, 73 China, 101, 115, 120, 125, 130–131, 149 Beijing, 107 Shanghai, 129 civic engagement, 44–49, 51–53, 56, 58–68, 70–71, 74–75, 78–81, 84, 89, 91, 93–95, 97, 101, 106, 109, 121, 123, 140, 145, 153, 159, 161–163, 165–166, 168 civic skills, 52–55, 57–58, 60, 62–63, 80, 86, 106, 109, 116, 118, 162, 166 civil war, 48, 76 Clark, William S., 111–112 community service, 54, 56, 81 conscription, 54–55, 60, 67, 96, 99, 107, 109, 135

corruption, 143, 146 culture, 54, 61, 74, 82, 103–104, 109–113, 115, 121, 128, 148, 167 damage war damage, 60, 64, 75–76, 78, 96, 99, 107, 117–118, 121–122, 135, 143, 145, 147–152, 155–159 war death, 66, 75–76, 99 war destruction, 45, 48, 75, 99, 130, 135, 145–146, 149, 151, 155, 158 day care, 72–73 defeat, 44, 48, 50, 52, 58–60, 62, 75, 78–80, 82, 84, 89, 91, 93, 99, 108, 118, 120, 124, 136, 147–148, 151, 161–162 vanquished, 79–80, 84, 86, 93 democratization, 45–47, 51, 54, 58, 82, 91, 102, 142, 168 Denmark, 79 Diamond, Larry J., 56, 168 Diet, 102, 125, 129 disaster, 49, 51, 103, 145, 159. See also damage Drucker, Peter, 166 education, 44, 52, 55, 58, 65, 68–72, 74, 76, 82, 89, 94, 99, 106, 111, 113, 125–126, 134, 148–149, 165 informal social learning, 75 on-the-job training, 116 schools, 47, 52, 55–56, 58, 68–69, 73, 75, 81, 83, 98–99, 103, 105–106, 108, 114, 116–119, 124–132, 134–136, 141, 144, 148–153, 155, 158–159 Ertman, Thomas, 50 evacuation, 105, 117, 130, 136

193

194 fertility, 72 Finland, 46, 82–83, 85, 87–89 formative years, 52, 80–81, 83 France, 82–83, 85–88, 91, 102 Fujihira, Shinju, 51, 57 Fukuoka, 96, 99–100, 123–124, 126–135, 137–139 G. I. Bill, 57 Germany, 46, 51, 79, 82–89, 91–93, 102, 147 Girl Scouts, 47 Goss, Kristin A. 83 Granovetter, Mark, 145 Great Depression, 83, 86, 88–89, 91–93 Great Kanto Earthquake, 102, 133, 160 Haddad, Mary Alice, 144 Hall, Peter A. 94, 104, 115, 165 Harada, Suke, 104 health, 51, 54, 141, 148–149 high-speed growth, 52 Hiroshima, 99, 105, 118, 129 Hokkaido, 97–100, 110–120 home front, 51, 55, 67, 162 hospital, 51, 103, 114, 116, 119, 148–153, 155, 157–159 human capital, 166 human rights, 142 Iizuka, Kunisaburo, 129 income, 55, 68, 89, 91, 98, 111, 168 industrial production 66, 128, 133, 135 industrial output, 111, 118, 126, 130 inequality, 62 inflation, 118 information, 47, 53, 60–63, 73, 80, 95, 139– 141, 144–147, 153–154, 159–160, 162 Inglehart, Ronald, 82 Ireland, 82–83, 85–88, 92 Italy, 60, 82–85, 87–89, 91, 115 Iversen, Torben, 166 Ito, Mitsutoshi, 187, Jennings, M. Kent, 54–55, 81 Johnson, Chalmers, 143, 147 judo, 46–47, 96–98, 123–127, 129–132, 134–139 jury duty, 54, 56 Kagawa, 104–105, 114 Kanagawa, 100, 126–127, 134, 136–138 Kano, Jigoro, 124–125, 129 Kier, Elizabeth, 167

Index Kobe, 48–49, 96–109, 111, 113–121, 124, 128, 131–136, 139, 145, 160 Kodokan, 97, 124–125, 127, 129, 131–132, 134, 138, 163 Kokubo Fujinkai, 116 Kondo, Soichiro, 137 Korea, 116, 129–131, 136 Krebs, Ron R., 50, 54 Kuroda, Kiyotaka, 110–111 Kyoto, 98–102, 104, 107, 114–115, 128–129, 131 labor union, 46, 58, 74, 82, 104, 113 Levi, Margaret, 50, 57 Levy, Jonah, 166 library, 148–153, 155, 158–159 Long Civic Generation, “greatest generation”, 79, 165 MacArthur, Douglas, 136 Manchuria, 129, 131–132 Mannheim, Karl, 83 Meiji Restoration, 98, 101–102, 110, 126, 128 membership association, 144 Ministry of Education [Monbusho], 47, 69, 97, 125 mobilization, 44–45, 49–53, 55–63, 65–67, 69–72, 74–75, 78–83, 86–87, 89, 91–92, 94–96, 98–99, 105–106, 109, 115–116, 118, 121–123, 130, 136, 149, 159, 162, 165–167 Nagasaki, 99–100, 114, 118, 128, 130 Nagoya, 97, 99–100, 118 neighborhood organization neighborhood association, 47, 51, 105, 116 tonari-gumi, 105 Netherlands, 79, 82–88, 91–92, 94, 102, 159, 165 Nitobe, Inazo, 112 occupation Allied occupation, 125 SCAP, 45–48, 55, 69, 98, 123, 125, 130–131, 136–139, 143 Osaka, 69, 97–100, 102–104, 107, 115, 117–118, 131, 135 participation involuntary, 52, 54 voluntary 44, 48, 52, 59, 74, 79, 89, 95–96, 120 path-dependency, 44, 49, 59, 74, 80, 93, 95, 101, 167

Index legacies, 44, 49, 59–63, 66–67, 72, 74, 80, 95, 98–99, 101, 103, 109, 121, 123–124, 132, 138–139, 159, 162–163, 166, 168 Pekkanen, Robert, 46, 166 Pharr, Susan, 46, 166 Pierson, Paul, 59–60 political party, 51, 59, 82, 146 Putnam, Robert D., 49, 52, 60, 68, 78, 81, 83, 115, 141, 144, 165, 167 reciprocity, 106, 120, 146 reconstruction reconstruction, 136, 140–150, 152–156, 158–160, 163 recovery, 96, 99, 121, 131, 136, 138, 141–143, 149, 151–152, 154–160 resources, 46–47, 50, 55, 57, 59–62, 96, 104, 115–116, 120, 122, 132, 140, 143–147, 153, 155, 157–159 Rotary Club, 47 Russo-Japanese War, 51, 104, 106, 111, 113, 128 Sapporo, 96–101, 103, 108, 110–122, 136, 139 SCAP, Civil Information and Education Section, 125 Schlozman, Kay Lehman, 53, 55, 61, 68, 81 Seinan War, 128 Sino-Japanese War, 113 Skocpol, Theda, 50–52, 57, 59–60, 79, 154, 166 social capital, 91, 144, 160, 165 social entrepreneur, 95, 109, 121 social learning, 49, 52, 58 social movement, 103–104, 106, 112–114, 119, 127, 129, 132, 134, 139 South Korea, 45. See also Korea state-building, 50, 142, 148, 159 “state-society synergy,” 166 Sweden, 82–88 Switzerland, 82–83, 85–88 Taisho Democracy, 67, 71, 104, 113 Takemura, Shigetaka, 127, 131–132 tax, 50, 57, 133 Tilly, Charles, 50, 57

195 Tocqueville, Alexis de, 52, 166 Tokyo, 69, 97–100, 102–104, 114, 117–118, 124–126, 129, 132–136, 138 trauma, 48, 75–76 Tsai, Lily L., 147 Tsujinaka, Yutaka, 45, 162 Uchimura, Kanzo, 112 United Kingdom, 46, 51, 57–58, 83–84, 86–87, 89, 94, 97, 102–103, 110–111, 133 United States, 46–48, 51, 54–55, 66, 79, 83, 87, 93, 103, 111, 123, 125, 130, 133, 136, 138, 147–148, 165, 168 Verba, Sidney, 53, 55, 61, 68, 81 veterans, 54–55, 57 victory/defeat hypothesis, 75, 78 violence, 48, 76 voluntary association, 46–47, 51–53, 60–63, 96–97, 101, 105, 108, 144–145, 153, 162 war. See also Seinan War; Sino-Japanese War; Russo-Japanese War; World War I; World War II Civil War, 57 Korean War, 131, 136 Vietnam, 55 “weak ties”, 141, 144–145, 147 women’s organizations, 116 World War I, 44, 46, 48–49, 51, 58, 64, 66–67, 75–76, 78–81, 83, 85–86, 89, 91–94, 98, 102, 105–106, 108–109, 112, 115, 123, 127, 130–131, 135, 140, 146–147, 152, 159, 161–162, 165 World War II, 44, 46, 48–49, 51, 58, 64, 66–67, 75–76, 78–81, 83, 85–86, 89, 91–94, 98, 105–106, 108–109, 115, 123, 127, 130–131, 135, 140, 146–147, 152, 159, 161–162, 165 YMCA, 51, 95–101, 103–110, 112–115, 117–123, 131–132, 136–139, 163 Yokohama, 96, 99–100, 102–103, 112, 118, 123–124, 126–128, 130–131, 133–139 youth group, 65, 67, 69–74, 76, 97