CONSTITUTIONAL OLIGARCHY: THE COMPLEX UNITY OF THE IMPERIAL JAPANESE STATE IN THE FACE OF CRISIS by NICHOLAS A. R. FRASER B.A., The University of Calgary, 2007 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES (Political Science) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) November 2009 © Nicholas A. R. Fraser ii ABSTRACT Under the Meiji Constitution, a political system designed to create an institutional framework that allowed for the sustained oligarchic rule of the Meiji Genrō, Japan experienced multiple crises generated by popular upheaval against the government during the interwar years. One was an economic crisis in 1918 triggered by Japan‟s participation in the First World War which generated an unprecedented level of popular protests in the form of nation-wide riots and some strikes. Known as the Rice Riots, this crisis threatened to unleash a confrontation of the Meiji Genrō by political parties holding seats in the Diet who sought to establish party-led cabinets. A second crisis occurred in 1936 when 1400 soldiers of the Imperial Army stationed in Tokyo occupied government buildings and assassinated several high-ranking government officials in an attempt to set up a an all-military cabinet. While both party-politicians and military officers had further expanded their influence over the policy-process after these crises neither set of actors suspended, revised or replaced the Meiji Constitutional system. It is the purpose of this thesis to explore the reason why the Imperial Japanese polity was not structurally altered as a result of power change that accompanied the Rice Riots and the 1936 Incident. This essay makes two arguments about the Meiji Constitutional system‟s sustainability during the prewar years. First, it argues that the Meiji Constitutional system due to institutional design and elite political culture functioned in practice as an oligarchic state. Second, it argues that the reason the Meiji Constitution was never revised, suspended or discarded during the course of regime change was because political parties and high ranking military officers ended up using the same strategies as the Meiji Genrō to successfully maneuver the institutional structure of the policy-process. Hence, in the process of learning how to master the institutional dynamics of the political system, they eventually overcame legislative deadlock and in the process stabilized the oligarchic state without having to reform it in order to expand their power within it. iii TABLE OF CONTENTS ABSTRACT………………………………………………………………………………………………......ii TABLE OF CONTENTS……………………………………………………………………………………..iii ACKNOWLEDGEMENTS………………………………………………………………………………......vi LIST OF TABLES…………………………………………………………………………………………….v Introduction……………………………………………………………………………………………….......1 I. Conventional Theoretical Approaches to Studying Prewar Japanese Politics………………….......5 Barrington Moore‟s Marxist Approach and its Theoretical Problems…………………………………...6 Richard J. Samuels‟ Political Leadership Approach and its Theoretical Problems…………………...8 J. Mark Ramseyer and Frances M. Rosenbluth‟s Rational Choice Approach and its Theoretical Problems……………………………………………………………………………………….10 Shūichi Kato‟s Institutionalist Approach and its Theoretical Problems…………………………….....12 II. Oligarchic Policy-Making Theory as an Approach to Studying Prewar Japanese Politics………14 R. P. G. Steven‟s Hybrid Constitutional Theory: a Basis for Conceptualizing the Meiji Constitutional System in Practice…………………………………………………………………………15 Expanding Steven‟s Hybrid Constitutional Theory to Explain the Long-Term Stability of the Meiji Constitutional System………………………………………………………………………………..20 Blending Political Culture with Institutions: Oligarchic Policy-Making Theory………………………..26 Observable Implications of Oligarchic Policy-Making Theory………………………………………….28 Methodological Approach……………………………………………………………………………….....35 III. The Development of Oligarchic Policy-Making in the Meiji Period………………………………...37 The Legacy of Passive Imperial Leadership………………………………………………………….....37 The Re-Emergence of Passive Imperial Rule during the Meiji Period………………………………..38 Oligarchic Policy-Making in Early Meiji Japan: The Product of Passive Imperial Rule and iv Oligarchic Institution-Building……………………………………………………………………………..41 The Institutional Framework of the Meiji Constitution…………………………………………………..46 The Early Years of the Meiji Constitutional System and the Persistence of Oligarchic Policy-Making and in the Imperial Japanese Polity……………………………………………………..48 IV. Test Case One: Assessing the Culture of Policy-Making in the Era of the 1918 Rice Riots……………………………………………………………………………………………………52 Understanding the Political Context of the Rice Riots……………………………………………….....52 The Rice Riots and the Long-term Political Consequences…………………………………………...55 The Marxist Interpretation of the Rice Riots and its Theoretical Problems…………………………..57 The Political Leadership Theory Interpretation of the Rice Riots and its Theoretical Problems……………………………………………………………………………………….58 The Rational Choice Theory Interpretation of the Rice Riots and its Theoretical Problems…….....60 The Institutionalist Interpretation of the Rice Riots and its Theoretical Problems…………………...62 Oligarchic Policy-Making Theory‟s Interpretation of the Rice Riots…………………………………...63 V. Test Case Two: Assessing the Culture of Policy-Making in the Era of the 1936 Incident……….78 Understanding the Political Context of the 1936 Incident………………………………………………78 The 1936 Incident and its Long Term Political Consequences………………………………………..82 The Marxist Interpretation of the 1936 Incident and its Theoretical Problems……………………….84 The Political Leadership Interpretation of the 1936 Incident and its Theoretical Problems………...85 The Rational Choice Interpretation of the 1936 Incident and its Theoretical Problems………….....86 The Institutionalist Interpretation of the 1936 Incident and its Theoretical Problems…………….....87 Oligarchic Leadership Theory‟s Interpretation of the 1936 Incident…………………………………..88 Conclusion………………………………………………………………………………………………….105 Works Cited………………………………………………………………………………………………..114 v TABLES Table 1: Typologies for classifying potentially destabilizing political crises…………………….…….33 Table 2: Typologies for conceptualizing different responses by regime to potentially destabilizing crises……………………………………………………………………………..34 vi ACKNOWLEDGEMENTS I would like to thank my mother for giving me the diligent work ethic required to see this project through. I would like to express my deepest gratitude to my supervisor, Dr. Yves Tiberghien, for taking the time to help me adjust to the hyper competitive positivist academic climate that is modern comparative political science and for being extremely encouraging as well as generous with constructive criticism. I would also like to express sincere thanks to Dr. Benjamin Nyblade acting as my second-reader and contributing to the betterment of my thesis. Like Dr. Tiberghien, Dr. Nyblade was instrumental in helping me think like a comparative political scientist—special thanks on this matter also go to Dr. Alan Jacobs whose methods class almost killed me but made me stronger as well as Dr. Lisa Sundstrom whose advice in the early stages of my work was crucial. I would also like to thank Go Murakami for the countless hours he put in helping me perfect my craft as well as Rhea McCarroll, Daria Boltokova and Konrad Kalicki for their boundless moral support. I would also like to thank Beth Schwartz, Anastasia Shesterina, Augustin Goenaga, Aim Sinpeng and Serbulent Turan for taking valuable time out of their schedules to help me with drafts and prepare to defend this body of work. Special thanks go to my friends and family for supporting me through what has been a relentless task of immense proportions. Finally, I would like Robert O. Paxton, Mike Patton, Henry Rollins, and Greg Ginn, whose passion and genius have inspired me during the course of this project. - 1 - INTRODUCTION During the interwar years many European countries experienced political crises that resulted in the revision, replacement and/or collapse of prevailing political institutions. In contrast to these, Asia‟s first modern state and the twentieth century‟s only non-western imperial power, Japan experienced high levels of political upheaval in the interwar years but never underwent regime change before 1945. Between 1917 and 1919, Russia, Italy, Germany and Japan all experienced economic crises related to participation—either as combatants or non-combatants—in the First World War. This combined with other factors sparked revolution and varying degrees of civil war in Russia and Germany, and replacement of democratic institutions with a fascist dictatorship in Italy by the mid 1920‟s. Japan experienced nation-wide riots for several weeks, known as the Rice Riots, in 1918 and experienced a change in government but not a change in the structure of its political system. In the 1930‟s, the Great Depression triggered increases in violent attacks from and a rise in popular support for the radical right in Germany, Spain and Japan. Spain‟s existing political institutions were destroyed by a civil war and the establishment of a military dictatorship. Germany‟s were revised to facilitate the building of a fascist state under the Nazi Party. Though the military increased its influence in politics and there were many attempted coups by military officers, Japan never experienced regime change in the 1930‟s. Many Japanese specialists have argued that Japan experienced periods of democratic rule in the 1920‟s and military or fascist dictatorship in the 1930‟s and 40‟s. If this were true then why didn‟t party politicians or military officers in their expansion of power suspend, revise or replace the Meiji Constitutional system? It will be the purpose of this essay to explore the nature of the Meiji Constitutional system in terms of its structural stability throughout the prewar period. Specifically the question that this essay seeks to explore is as follows: why didn‟t power change during the prewar period—that is the rise of party-led and then military-led cabinets cause the replacement, revision or suspension of the - 2 - Meiji Constitution? This essay makes two arguments about the Meiji Constitutional system‟s sustainability during the prewar years. First, it argues that the Meiji Constitutional system due to institutional design and elite political culture functioned in practice as an oligarchic state. Second, it argues that the reason the Meiji Constitution was never revised, suspended or discarded during the course of regime change was because political parties and high ranking military officers ended up using the same strategies as the Meiji Genrō1 to successfully maneuver the institutional structure of the policy-process. Hence, in the process of learning how to master the institutional dynamics of the political system, they eventually overcame legislative deadlock and in the process stabilized the oligarchic state without having to reform it in order to expand their power within it. This essay will utilize process-tracing to examine the ways in which elite political actors within the Meiji Constitutional system managed conflict between actors from different policy-affecting institutions. Specifically this essay will focus on two cases, that is the Rice Riots of 1918 and the 1936 Incident. Both of these events represented or created significant legitimacy crises for the government as there were members of particular policy-affecting institutions who sought to have their institution gain more control over the political system. In the case of the Rice Riots, an economic crisis caused popular upheaval throughout the country and could have been marshaled by party-politicians in the Diet who sought to oust the existing Genrō-led cabinet (see chapter four). In the 1936 Incident, military officers seized control of Tokyo for several days in an attempt to establish active imperial leadership in the establishment of a military dictatorship. Although these 1 Genrō means „elder statesman‟ and is a term used to describe those individuals who served in the oligarchic government that emerged after the overthrow of the Tokugawa feudal polity in the late 1860‟s. The oligarchy contained several hundred individuals but as time wore on many of its members were absorbed into the new state institutions which said government had mandated. By the 1880‟s the oligarchy was considerably smaller and collectively ran the state and manipulated politics from behind the scenes once the Meiji Constitution was promulgated in 1889. These individuals were known as Genrō. Richardson, Bradley M. and Flanagan, Scott C. Politics in Japan. Little, Brown and Company, 1984., 6-15 - 3 - low- and mid-ranking officers had sympathizers in the Army High Command, the latter did not support the former (see chapter five). In each crisis, party-politicians in the first and military officers in the second, sought to expand their power to unprecedented levels over the political system. These crises might have provided them with an incentive to circumvent institutional structures to increase their power at the expense of other institutional actors, yet in both cases neither pursued this course of action. In the end, political parties were able to gain more political influence by working with and mimicking the Genrō rather than fighting them. Similarly, Imperial Japan‟s military‟s leadership did use the 1936 Incident to attain more influence in government but by compromising with the Imperial Diet rather than coercing it. Hence, party-politicians and military officers were able to use crisis to expand their power over the political system in a way that did not circumvent but utilized existing norms to successfully maneuver the institutional structure of the policy process. The significance of this is that the policy-process was never altered in such a way that it removed or introduced new institutional actors, and for this reason the oligarchic regime itself was never removed or changed. This thesis will be broken into five major sections. The first section will first evaluate existing theoretical approaches to studying the prewar political system and assess why they have failed to adequately address the stability of the Imperial state during the Rice Riots and the 1936 Incident. The second section will establish a theoretical and methodological framework for analyzing the Meiji Constitutional system as it operated in practice. This section will outline a new theoretical approach based upon R.P.G. Steven‟s Hybrid Constitutional model of prewar Japanese politics and proceed to explain the process-tracing methodology pursued in this work. The third section will explain the development of oligarchic policy-making as an entrenched norm before the Meiji Constitutional system was established (1889-1890). The passivity of imperial rule and the late development of an institutionalized policy-process will be highlighted and it will be argued that - 4 - these combined to create an oligarchic state in practice. The fourth section will examine the Rice Riots of 1918 and focus on how this crisis could have but did not destabilize the Meiji Constitutional system. This section will focus analysis on how the entrenched norm of oligarchic policy-making created low incentives for ascending party politicians to attack the regime directly. The fifth section will assess why the 1936 Incident did not destabilize the Meiji Constitutional system. Analysis will focus on how the entrenched norm of oligarchic policy-making created low incentives for high ranking military officers to fully support a coup d‟état carried out by the Young Officers‟ Movement. This will be followed by a thoughtful conclusion. - 5 - I. CONVENTIONAL THEORETICAL APPROACHES UNDERSTANDING PREWAR JAPANESE POLITICS The nature and trajectory of prewar Japanese politics is a subject that has been debated amongst sociologists, historians and political scientists. In the sub field of comparative politics, much of the scholarly literature in English has tended to focus on the so-called rise of democratic party-politics in the Taishō era and the subsequent „militarization‟ of politics in the Shōwa period. This existing body of literature, though diverse in terms of theoretical and methodological approaches, has failed to identify the complex relationship between political culture and institutions that led to norms of oligarchic policy-making between elites. As will be demonstrated below, much of the existing comparative politics explanations of the nature and trajectory of the Meiji Constitutional system cannot fully account for this. Politics under the Meiji Constitutional system between was characterized by an oligarchic state that was plagued by high cabinet turnover and the gradual increase of intra-state clashes over policy and the balance of power between institutions. Yet, existing theoretical models cannot adequately explain why the Rice Riots or the 1936 Incident did not lead to the break down of the Meiji Constitutional order for one of two reasons. Conventional theoretical approaches (structure- or actor-centered) either assume that the Imperial state structure is stronger and more unified than it actually was or fail to explain with precision and accuracy how the contradiction between clashing institutional actors was reconciled in terms of overcoming legislative deadlock. Hence, existing explanations of the prewar Japanese political system cannot fully explain why under the intense pressure of potentially destabilizing political crises political actors who would have gained from doing so did not alter, suspend or replace the Meiji Constitution. - 6 - Barrington Moore’s Marxist Approach and its Theoretical Problems Perhaps one of the most recognized explanations of prewar Japanese politics comes from Barrington Moore, Jr. He discusses why Japan became a so-called fascist state before 1945 in his classic work, Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the Modern World. Moore‟s theoretical model is based upon the orthodox Marxist conception that economic development causes particular regimes to develop. He argues that there have been a number of distinct patterns of state development within the modern period: bottom-up peasant revolutionary development that produces „communist‟ societies (such as the Soviet Union and China); top-down revolutionary development that produce „fascist‟ societies (such as Imperial Japan and Nazi Germany); and middle class led development that produces democratic societies (such as France, Britain and the United States)2. According to Moore there are three key variables that determine a society‟s modernization or regime development: the degree to which a society adopts a commercialized market economy beginning with the commercialization of agriculture; whether or not a society‟s mode of agricultural production is „labour-repressive‟; and the „revolutionary potential‟ of the society‟s peasant classes3. Prewar Japan, like Germany, on Moore‟s account, experienced late industrialization and in the process maintained the traditional feudal class structure4. For this reason, Moore argues that Imperial Japanese politics were dominated by the landlord, capitalist and „militarist‟ classes5. Moore claims that prewar Japanese politics were „fascist‟ because the state combined internal repression, foreign expansion and nationalist ideology to check and manipulate the working, petty bourgeois 2 Moore, Barrington. Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the Modern World. Beacon Press, 1966. 418-420 3 Moore, 430-431, 433-438, 459-460, 467-468, 477-478 4 Moore, 254-275, 276-291 5 Moore, 291-313 - 7 - and peasant classes6 . Thus, according to Moore‟s theory the long term stability of the Meiji Constitutional system can be attributed to the development of a „labour-repressive‟ economic structure that prevented social revolution form occurring. Ultimately, Moore‟s analysis of political developments in prewar Japan is problematic because his model is too deterministic. Moore‟s theoretical framework relies on a Marxist understanding of economic structure and development but he fails to distinguish between what he refers to as „market‟ and „labour-repressive‟ commercial agriculture. In his discussion of agricultural production, Moore attempts to expand the Marxist notion of capitalist development by elaborating on the ways in which the pre-capitalist landlord class may adopt or fail to adopt commercial agriculture7. However, as Theda Skocpol states, Moore ultimately falls short of this task in that his distinction between „market‟ and „labour-repressive‟ commercial agricultural production „suffers from interrelated logical and empirical difficulties‟8. In addition, Moore‟s model fails to explain the criteria for evaluating the power of the middle class9. These two shortcomings combined with his vagueness on the question of „non-peasant allies‟ of the peasant class ultimately constrain his ability to fully explain important causal factors that distinguish his regime types10. In this context, Moore‟s obsession with neat regime categories tends to ignore significant empirical facts that might otherwise make his lumping of certain cases studies under one classification problematic11. Moore‟s structure-centered approach is insufficient for understanding the stability of the Meiji Constitutional system under pressure because it relies primarily on economic indicators to understand the behaviour and preferences of political actors. Guided by deterministic Marxist 6 Moore, 291-313 7 Skocpol, Theda. “A Critical Review of Barrington Moore‟s Social Origins of Dictatorship and Democracy,” Politics and Society, Vol. 4, No. 1 (1973), 1-34. 14-15 8 Skocpol, 14 9 Skocpol, 14-18 10 Skocpol, 14-18 11 Skocpol, 22-28 - 8 - assumptions which define politics as the sole product of economics, Moore‟s model does not address the institutional structure and dynamics of the Meiji Constitutional system let alone the established norms that stabilized them. The Imperial state was an oligarchy as Moore points out but his model does not address the complex and increasingly tenuous relationships that existed between actors from different institutions such as the Diet and the Genrō. Empirically speaking, Moore‟s argument has been called into question by many Japan specialists as there is little evidence that suggests agrarian landlords played a significant role in prewar national politics12. Hence, Moore‟s argument is unable to fully explain the stability of the Meiji constitutional system because it assumes that the Imperial state structure is stronger and more unified than it actually was and that structure alone can account for the trajectory of prewar politics. Richard J. Samuels’ Political Leadership Approach and its Theoretical Problems Using an actor-centered model, Richard J. Samuels argues that prewar Japanese politics can be explained by tracing certain policy choices of various leaders in the Meiji, Taishō and Shōwa periods. In his book, Machiavelli’s Children: Leaders and Their Legacies in Italy and Japan, Samuels offers a theoretical approach that in many respects promises to be the antidote to Moore‟s deterministic theory. Rejecting conventional approaches that emphasize structural constraints, Samuels argues that political outcomes are the result of choices made by political leaders 13 . He claims that leaders affect political outcomes more than most scholars have acknowledged as they tend to emphasize constraints on action14. However, Samuels states that what makes leaders casual factors is their unique ability to stretch structural constraints and that 12 Paxton, Robert O. The Anatomy of Fascism. Vinatge Books, 2004. 197-200 13 Samuels, Richard J. Machiavelli’s Children: Leaders and Their Legacies in Italy and Japan. Cornell University Press, 2003. See Introduction, 1-18 14 Samuels, 352-360 - 9 - creative actors have a wider range of choices than is commonly understood15. Hence, he seeks to demonstrate that the long term political trajectory of pre- and post-war Japan can be explained by analyzing the ways in which particular leaders in particular situations made choices that resulted in long term political outcomes16. Samuels argues that the trajectory of prewar Japanese politics can be explained by the Meiji oligarchs‟ choice of institutional design which created an authoritarian state with an independent military; Hara Kei‟s decision to delay expansion of the electorate, suppress the labour movement and not to expand party power by altering the institutional structure; and roles of Mutō Sanji, Kishi Nobusuke and Ayukawa Gisuke in the development of corporatist economic policies that fostered aggressive expansion on the Asian continent17. Hence, according to Samuels‟ theory the long term stability of the Meiji Constitutional system can be attributed to various policy choices made by different leaders in the prewar period that served to reinforce rather than undermine the prevailing institutions. While Samuels‟ actor-centered theoretical approach pays close attention to empirical facts and gives special attention to the agency of political actors, it is problematic because it does not clearly distinguish between established norms of elite behaviour and individual choices. Throughout Samuels‟ book he discusses how different prewar leaders starting with the Meiji oligarchs made particular policy choices that created and stabilized an authoritarian state-centric kind of politics. His theory does not discount political culture altogether, rather he claims that it is secondary to agency. Samuels states that culture “matters far more directly as a tool wielded by leaders . . . not as a great force but as a social mechanism.”18 He is correct to point out that actors 15 Ibid 16 Samuels, 352-360 17 Samuels, 48-63, 102-104, 114-123, 126-136, 140-151 18 Samuels, 352 - 10 - choose to affirm and perpetuate culture—culture is not a deterministic force that guides politics. However, Samuels‟ leader-centered model does not consider the ways in which established norms and beliefs might structure actors‟ incentives to pursue particular courses of action over others. This theoretical oversight serves to undermine his analysis of political institutions as they become merely static effects of individual policy choices. In this way, although not the extent that Moore‟s model does, Samuel‟s theory cannot account for complex institutional dynamics throughout the Meiji (1868-1912), Taishō (1912-1926) and early Shōwa (1926-1945) periods. Despite the fact that he takes an actor- rather than structure-centered approach, Samuels‟ theoretical model ends up making the same mistake that Moore‟s does as he also assumes that the Imperial state structure is stronger and more unified than it actually was. This is significant as Samuels‟ theory assumes that the institutional frame work of the Meiji Constitution will function as a seamless machine which moves from „illiberal‟ to „fascist‟ as result of no liberal policy choices by particular leaders. Samuel‟s theoretical explanation of Japan‟s prewar politics then cannot fully explain the stability of the Meiji Constitutional system because it assumes that agency dictates the degree to which culture and institutions can structure actors‟ behaviour. J. Mark Ramseyer and Frances M. Rosenbluth’s Rational Choice Approach and its Theoretical Problems Using cartel theory, J. Mark Ramseyer and Frances M. Rosenbluth argue that the Meiji Constitutional system and its long term trajectory can be explained through rational choice methodology in their book The Politics of Oligarchy: Institutional choice in Imperial Japan. According to Ramseyer and Rosenbluth, the developments of Japan‟s prewar politics can be explained by the institutional design crafted by rationally acting individuals within the Meiji oligarchy - 11 - and its evolution over time after their passing19. They point out that the policy choices made by the oligarchs constantly evolved out of internal struggles between its members to acquire, maintain or insulate their power from one another20. Hence, these scholars argue that the Meiji Constitution came out of a long term struggle between the oligarchs about how to institutionalize their rule21. Ramseyer and Rosenbluth articulate that the instability of the oligarchy explains why the Meiji Constitution was adopted while the need to identify and effectively co-opt political entrepreneurs explains the substance of it22. They account for the military‟s so-called takeover of the government in the 1930‟s by highlighting the fact that it had the ability to effectively veto cabinets and was accountable only to the imperial throne23. Ramseyer and Rosenbluth argue that the rise of political parties and the eventual „military dictatorship‟ were the products of an institutional design24. They state that the Meiji Constitution deliberately created a power vacuum resulting from the oligarchs‟ inability to reach a lasting consensus on institutional design that would directly institutionalize their power25. As rational actors, institutional actors sought to maximize their power within this system leading to first the ascendance of the House of Representatives and then the military26. Ramseyer and Rosenbluth‟s model is superior to both Samuels‟ and Moore‟s as it recognizes recognize the complex unity of the Imperial Japanese state. Yet, it falls short of adequately explaining the stability of the Meiji Constitutional system because it does not accurately explain how exactly the contradiction between clashing institutional actors was reconciled in the 19 Ramseyer, J. Mark and Rosenbluth, Frances M. The Politics of Oligarchy: Institutional Choice in Imperial Japan. New York: Cambridge University Press, 1995. See Ramseyer and Rosenbluth, Introduction, 39-40, 160-166 20 Ibid 21 Ramseyer and Rosenbluth, 30-32 22 Ramseyer and Rosenbluth, 9 23 Ramseyer and Rosenbluth, 160-166 24 Ibid 25 Ramseyer and Rosenbluth, 160-166 26 Ramseyer and Rosenbluth, 92-102, 160-166 - 12 - interwar years. In their discussion of the rise of military influence in politics during the 1930‟s, Ramseyer and Rosenbluth incorrectly assume that the military had control over the policy process. In this way, their paradigm obscures the power dynamics of institutions under the Meiji Constitutional system because it distorts the power of the military—that is assumes that it had more power vis-à-vis other state institutions that it actually had. Hence, Ramseyer and Rosenbluth‟s model fails to understand the role of established norms in prewar politics as it assumes that intra-institutional dynamics deterministically springs forth from institutional design. Cartel theory as wielded by Ramseyer and Rosenbluth seems to provide a strong argument when it is applied to the choices made in establishing the Meiji Constitution as an institutional framework. However, Ramseyer and Rosenbluth‟s model cannot explain the complex relationship between institutions and established norms that guided them. Hence, Ramseyer and Rosenbluth‟s rational choice approach cannot fully explain the stability of the Meiji Constitutional system during the interwar years of crisis. Shūichi Kato’s Institutionalist Approach and its Theoretical Problems In his article, “Taishō Democracy as the Pre-Stage for Japanese Militarism,” Shūichi Kato takes an institutionalist approach towards explaining the trajectory of prewar Japanese politics. Kato‟s main argument is that the institutional structure of the Meiji Constitutional system ultimately stayed the same but that its function changed over time to accommodate various different kinds of political actors as they came to dominate and control the system27. The reason behind this being that the institutional framework itself was oligarchic by nature28. Kato points out that Japan‟s prewar political trajectory was the result of changes in power dynamics between different actors who 27 Kato, Shūichi. “Taishō Democracy as the Pre-Stage for Japanese Militarism,” in Japan In Crisis: Essays on Taishō Democracy. Ed. Bernard S. Silberman and H. D. Harootunian. Princeton Universty, 1974. 217-236. 218 28 Kato, 226-236 - 13 - controlled different state institutions29. Hence, the prewar political system was durable because it was flexible enough to accommodate the „liberal‟ policies of political parties in the 1920‟s and the subsequent „illiberal‟ policies of the bureaucracy and the military in the Shōwa period30. Kato‟s institutionalist framework recognizes the complex unity of the Imperial state but does not explain how exactly the contradiction between clashing institutional actors was reconciled. Kato‟s analysis does a good job of highlighting the structural continuity of the Meiji Constitutional system. His work spends a lot of time discussing structural differences between interwar Japan and Germany which gives his analysis a great degree of clarity-he maximizes comparative leverage methodologically speaking. He is able to devise a good structural contrast between interwar Japan and Germany as Kato points out the ways in which Taishō liberalism fell critically short of challenging the authoritarian structure of the Meiji Constitutional system. He acknowledges that party politics in the Taishō „liberal‟ phase was in many ways a continuation of conservative, imperialist and repressive policies established in the Meiji period and continued into the Shōwa period31. Yet, Kato‟s argument stresses changes in function under a persistent structure. Kato‟s institutionalist argument is incapable of completely explaining the relationship between institutional continuity and varying degrees of functional continuity—that is the persistence of particular norms and values throughout the prewar period. Kato‟s institution-centric analysis is therefore problematic in terms of explaining the Meiji Constitutional system‟s long term stability because it conflates aspects of function with structure and ends up contradicting itself. 29 Kato, 226-236 30 Ibid 31 Kato, 220-226, 226-236 - 14 - II. OLIGARCHIC POLICY-MAKING THEORY AS AN APPROACH TO STUDYING PREWAR JAPANESE POLITICS This thesis makes two arguments about the Meiji Constitutional system‟s sustainability during the prewar years. First, it argues that the Meiji Constitutional system due to institutional design and elite political culture functioned in practice as an oligarchic state. Second, it argues that the reason the Meiji Constitution was never revised, suspended or discarded during the course of regime change was because political parties and high ranking military officers ended up using the same leadership strategies as the Meiji Genrō to in order to successfully maneuver the institutional structure of the policy-process. Hence, in the process of learning how to master the institutional dynamics of the political system, they eventually overcame legislative deadlock and in the process stabilized the oligarchic state without having to reform it in order to expand their power within it. Hence, this thesis argues that the institutional structure of the Imperial Japanese polity survived potentially destabilizing crises of the 1918 Rice Riots and the 1936 Incident because party- politicians and high-ranking military officers ultimately chose to perpetuate the norm of oligarchic policy-making. The two main arguments as expressed above are based upon R. P. G. Steven‟s argument that the Meiji Constitutional system functioned as a hybrid of the „separation of power‟ and „fusion of power‟ constitutional systems (see below). Steven‟s theoretical model, which explains the Meiji Constitution as a hybrid of „separation of power‟ and „fusion of power‟ in practice, will be used as a basis for explaining the policy-process and power dynamics of the prewar Japanese political system. Specifically, this thesis will use Steven‟s theoretical model as a basis for testing my theory that the Meiji Constitutional system remained stable throughout the prewar period. Despite the potentially destabilizing effects of the Rice Riots and the 1936 Incident because party- and military-leaders continued the established norm of oligarchic policy-making. Steven‟s theoretical model will be elaborated in two particular - 15 - ways. First, it will be expanded to explain the relative passivity of imperial rule during the prewar period. Second, it will be expanded to demonstrate how the norm of oligarchic policy-making created high incentives for party-politicians and military officers to work within the Meiji Constitutional system during political crises that could have provided them with a pretext to suspend, revise or even replace it. Methodologically speaking, this essay will use process-tracing to examine policy-making before, during and after the events of the Rice Riots and the 1936 Incident. Emphasis will be placed upon explaining the actions of political actors during these crises with specific attention paid to the ways in which party- and military leaders chose to use these crises to expand their political influence. R. P. G. Steven’s Hybrid Constitutional Theory: a Basis for Conceptualizing the Meiji Constitutional System in Practice In his article, “Hybrid Constitutionalism in Prewar Japan,” R. P. G. Steven argues that in their choosing of a constitutional model appropriate for Japan, the Meiji oligarchs rejected both the British and American constitutional models32. He claims that the Meiji oligarchs did not wish to adopt either of these constitutional models because they perceived the executive branch to be too weak under these systems33. Hence, according to Steven, the Meiji state-builders ultimately chose to base the Imperial Japanese constitution on the Imperial German model as they believed it would allow them to create a system with a powerful executive branch34. Ironically, Steven states, this constitutional choice led to a system which in practice had an executive that was actually weaker then both the American and British systems35. Though he does not explore in great detail the role 32 Steven, R. P. G. “Hybrid Constitutionalism in Prewar Japan,” in Journal of Japanese Studies, Vol. 3, No. 1 (Winter, 1977), 99-133. 100 33 Ibid 34 Steven, 100 35 Ibid - 16 - of the emperor in prewar politics, Steven argues that the cabinet under the Meiji Constitution was weaker than cabinets under the British or American constitutional systems because of passive imperial rule 36 . Steven‟s theory of hybrid constitutionalism posits that because the Japanese emperor was a passive actor in the prewar political system that it functioned in practice as a kind of hybrid of both the American „separation of power‟ and British „fusion of power‟ constitutional systems37. Specifically, he argues that the Meiji Constitutional system in practice possessed both the disadvantages of the „separation of power‟ and „fusion of power‟ systems while at the same time possessing the advantages of neither38. Steven‟s argument is particularly useful in its focus on the structural weakness of the cabinet under the Meiji Constitution. His method of conceptualizing the Meiji Constitutional system as a hybrid of „separation of power‟ and „fusion of power‟ systems allows one to gain a clear overview of how the Japanese political structure functioned during the prewar era. Before explaining the details of how he applies his theory of hybrid constitutionalism to prewar Japan, it is necessary to clarify how Steven‟s model conceptualizes „separation of power‟ and „fusion of power‟ systems. In defining all three typologies—„separation of power‟, „fusion of power‟ and hybrid constitutional systems—Steven focuses upon the relationship between the executive and legislative branches39. His model uses these typologies to explain how this relationship affects the strength and stability of the executive branch40. At this point, it is important to recognize the limits of his model‟s typologies as they are based purely on an understanding of the British and American political systems. Hence, Steven‟s model assumes that there will be a single member plurality electoral system in place to allow for the growth of stable and relatively large parties to develop in 36 Steven, 100 37 Steven, 100-109 38 Ibid 39 Steven, 100-109 40 Ibid - 17 - both the „fusion of power‟ and „separation of power‟ systems41. Another shortcoming is that he doesn‟t consider the possibility of proportional representative electoral systems under either constitutional system nor does he take into account that there might be several small parties. Despite these limitations, these typologies do allow one to gain an understanding of how the Meiji Constitutional system functioned in practice and how it can be distinguished from pre-existing democratic constitutional models. According to Steven‟s theory, the executive in „fusion of power‟ constitutional system produces strong and stable cabinets whereas the executive under a „separation of power‟ model creates stable but weak cabinets42 . Under the „fusion of power‟ system, there is likely to be harmony provided there is a majority party in the legislature because it will create the executive and only it can unseat the cabinet43. Steven argues that cabinets tend to strong under the „fusion of power‟ system because it is designed to create incentives for the legislature and the cabinet to work together and minimize legislative deadlock 44 . In contrast, the „separation of power‟ constitutional system, though it also creates stable cabinets, has a structurally weaker executive than the „fusion of power‟ system because cabinets are not created by the legislative branch45. The executive under a „separation of power‟ system is guaranteed stability because it is directly elected by the electorate46. However, due to the fact that the cabinet is not created and destroyed by the legislature there is less of an incentive for members of the legislature to support the cabinet on policy47. Hence, Steven‟s theory articulates that the executive in „separation of power‟ systems are 41 Steven, 100-109 42 Ibid 43 Steven, 100-109 44 Ibid 45 Steven, 100-109 46 Ibid 47 Steven, 100-109 - 18 - stable but weaker than those that exist under „fusion of power‟ systems because the former are designed to experience more legislative deadlock than the latter. Steven argues that a hybrid constitutional system is distinct from the „separation of power‟ and „fusion of power‟ systems because unlike the previous two, the hybrid system creates weak and unstable cabinets48. He contends that the executive under the hybrid system are therefore subject to high levels of legislative deadlock because they directly reflect the balance of power between state institutions represented in cabinet49. Like the „fusion of power‟ system, the hybrid constitutional order has an executive that is not separately elected as it is created by other state institutions50. Hence, the executive relies on other institutions including but not exclusive to the legislature for its existence—that is the formation and defeat of cabinets51. Like cabinets in „fusion of power‟ systems, the executive under the hybrid system cannot be created as a separate power by itself concentrated into one institution52. Yet, as with the executive branch in „separation of power‟ systems the cabinet in a hybrid system is more likely to face legislative deadlock because it is not the sole creation of the legislature53. In this way, cabinets under the hybrid system must rely on other institutions to create it and therefore necessarily reflects the balance of power within the state itself including elected and unelected bodies54. As a result, the executive branch in the hybrid model is essentially compromise between separate institutional bodies and he claims is therefore weaker than either the „fusion of power‟ or „separation of power‟ systems55. 48 Steven, 100-109 49 Ibid 50 Steven, 100-109 51 Also, cabinets under the hybrid system can be vetoed on policy directly by the legislature as well as other institutions. Steven, 100-109 52 Ibid 53 Steven, 100-109 54 Ibid 55 Steven, 100-109 - 19 - Steven‟s application of hybrid constitutional theory to the Meiji Constitutional order is useful as it provides an analytical framework for understanding the complex policy process and relatively high cabinet turnover that characterized prewar Japanese politics. He explains that as a hybrid constitutional system, the prewar Japanese executive was considerably weak particularly as the power of the Meiji Genrō began to wane in the Taishō and Shōwa periods56. Under the Meiji Constitution, the Japanese emperor was to be the final arbiter in executive and legislative matters with all other state institutions requiring imperial approval or sanction before finalizing policy57. While theoretically this meant that executive power was concentrated in the imperial throne, in practice the emperor was a relatively passive political actor during the entire prewar period58. As a result of this, executive prerogative, which in principle was wielded by a single monarch, was in fact dispersed across a variety of state institutions that were accountable solely to the imperial throne59. The Meiji Constitution had designated institutions whose function it was to check others, such as the Privy Council, both houses of the legislature and the emperor himself 60 . These institutions possessed the ability to veto policies without necessarily resulting in the defeat of a cabinet61. What complicated matters in practice was the fact that in addition to these, additional state institutions accountable only to the emperor could act on his behalf due his relatively passive role in politics62. Hence, though they performed separate functions passive imperial rule allowed some of these institutions to veto policies directly and all of them the ability to indirectly veto policies by forcing the collapse of cabinets63. Specifically, cabinets under the prewar Japanese political system required at least two policy-affecting institutions to defeat cabinets and two to three 56 Steven, 108-117 57 Ibid 58 Steven, 108-117 59 Ibid 60 Steven, 108-117 61 Ibid 62 Steven, 108-117 63 Ibid - 20 - to form them64. Hence, because prewar Japanese cabinets were neither a separate power nor were they fused to other institutions they were subject to having their policies blocked and/or being defeated by a variety of state institutions that enjoyed a degree of autonomy and had no clear incentive to support them65. Expanding Steven’s Hybrid Constitutional Theory to Explain the Long-Term Stability of the Meiji Constitutional System Steven‟s conception of the Meiji Constitutional system as a hybrid of the „fusion of power‟ and „separation of power‟ systems provides a greater clarity when it comes to understanding the structural weakness and instability of prewar Japan‟s cabinet system. Steven‟s hybrid constitutional model is able to explain the complex relationship between various institutional actors during the tenure of the Meiji Constitutional order. This contributes to the understanding of the balance of power between relevant institutional actors but it does not provide us with a full picture of the complex and changing relationship between these organs of the Imperial state. Despite this, Steven‟s theory is only able to provide a more refined knowledge of which actors were dominant over others in the cabinet formation process between 1890 and 1945. Hence, Steven‟s hybrid constitutional theory provides a very limited scope of analysis as it is focused solely on the rise and fall of cabinets under the Meiji Constitutional system. Hybrid constitutional theory is limited in its applicability to other aspects of the prewar political system in part because Steven takes for granted that the Meiji, Taishō and Shōwa 64 Steven, 108-117 65 Ibid - 21 - emperors were relatively passive political actors but does not explain why 66 . This is highly problematic as if one is to take a theoretical approach that emphasizes institutional dynamics as a causal mechanism then it would seem appropriate to address precisely why three Japanese emperors allowed institutions to usurp their executive prerogative. As my adaptation and expansion of Steven‟s theory will demonstrate, only political culture can explain why all three of Japan‟s prewar emperors were relatively passive political actors to such the degree that they allowed institutions to usurp their authority. Thus, the fact that Steven‟s theoretical model does not fully explain consistent passive imperial rule limits its ability to provide a fuller analysis of the political trajectory of the Meiji Constitutional system. While it provides great insight into the balance of power between institutional actors as reflected by the rise and fall of cabinets during the prewar period, Steven‟s model does not explain the stability of the Meiji Constitutional system in the face of potentially destabilizing crisis. Hybrid constitutional theory is limited in this way in part because of its scope but also because of its focus on structure. It relies on a purely institutional explanation of the trajectory of the prewar polity and does not directly address the relationship between involved political actors in the policy-process. Even if Steven‟s analysis did extend directly into the specifics of the policy-process, his model‟s purely institutionalist approach would not be able to explain why the emperor did not usually wield power directly with a degree of consistency. Nor would it be able to explain why party-politicians and military officers who would have had an incentive to circumvent the institutional structure to maximize their power in crisis situations, such as the Rice Riots and 1936 Incident, did not. For these reasons R. P. G. Steven‟s hybrid constitutional theory as an institutionalist approach cannot 66 Steven mentions that the Japanese monarch was passive briefly but does not explain why on page 108 and then proceeds to explain the affects of this in terms of gradual autonomy between institutions as a result of the waning Genrō influence in the 1920‟s and 1930‟s. - 22 - fully explain why the Imperial Japanese state did not collapse in the face of crisis during the interwar years. Ultimately the reason why existing scholarship has failed to completely explain the long term stability of the Meiji Constitutional system is due to the fact that it has overlooked the role of political culture. As Brian J. McVeigh points out, the use of culture as a variable in studying Japan has come to be looked upon with suspicion by social scientists of various fields67. He states that many researchers have come to be skeptical of culture-based arguments because they see them as: prioritizing so-called apolitical and therefore irrelevant phenomena (fine arts, folk and religious traditions, etc.) or as attempts to essentialize Japanese culture68. While these charges are not without merit, Robert J. Smith articulates in his work on Japanese political economy that culture is the context in which all decisions, including those on policy, are made69. This thesis bases its concept of culture as an independent variable on Lucien Pye‟s conception of political culture. Pye defines political culture as follows: …in any operating political system there is an ordered subjective realm of politics which gives meaning to the polity, discipline to institutions, and social relevance to individual acts. The concept of political culture thus suggests that the traditions of a society, the spirit of its public institutions, the passions and the collective reasoning of its citizenry, and the style and operating codes of its leaders are not just random products of historical experience but fit together as a part of a meaningful whole and constitute an intelligible web of relations. For the individual the political culture provides controlling guidelines for effective political behaviour, and for the collectivity it gives a systematic structure of values and rational considerations which ensures coherence in the performance of institutions and organizations. In essence, thus, political culture . . . “consists of the system of empirical beliefs, expressive symbols, and values which defines the situation in which political action takes place.” It encompasses both the political ideals and the operating norms of a polity.”70 67 McVeigh, Brian J. The Nature of the Japanese State: Rationality and rituality. Routledge, 1998. 28 68 Ibid 69 Smith, Robert J. “The Cultural Context of the Japanese Political Economy,” in The Political Economy of Japan: Cultural and Social Dynamics, Vol. 3 Ed. Shumpei Kumon and Henry Rosovsky. Stanford University Press, 1992. 13- 31. 15 70 Pye, Lucien W. “Introduction: Political Culture and Political Development,” in Political Culture and Political Development. Ed. Lucien W. Pye and Sydney Verba. Princeton University Press, 1965. 3-26. - 23 - As Pye makes clear, political culture consists of a comprehensive context that shapes and gives meaning to political action and is reflected in political values as well as norms of political behaviour. This thesis argues that political culture can constrain or expand an actor‟s behaviour as it sets identifiable limits on how politics are to be practiced. It structures individual and institutional actors‟ incentives and provides a context for the balance of power within and between institutions that make up the political system. Primarily, political culture normalizes and rationalizes political behaviour that defines the political system. The beliefs and norms that shape actors‟ behaviour may be enshrined in the institutional framework of the political system. Yet, in many cases the fact remains that institutional procedures and rules may be interpreted differently. Over time the practical application of politics within a particular institutional framework may change gradually or suddenly during crisis as actors who wish to make changes within or to the system consider, push for or accept alternative beliefs and norms. Political culture is in this sense is never deterministic or static rather it is constructed and evolves with political beliefs, behaviour and institutions. With regards to explaining the long term stability of the Meiji Constitutional system the role of political culture has been largely overlooked and thus, its causal effects have been obscured with other variables. As mentioned above, the main problem is that existing theories have failed to identify the complex relationship between both political culture and institutions that led to norms of oligarchic policy-making between elites from 1890 to 1945. This significant as even the most recent theoretical explanations of pre-war Japanese politics, though they are able to explain the complex unity of the Meiji state, are unable to explain how exactly the contradiction between clashing institutional actors in the policy-process was perpetually overcome. R. P. G. Steven‟s conception of the prewar Japanese political system is able to clarify in precise terms the intra-institutional - 24 - dynamics of the oligarchic Meiji state. This combined with a proper understanding of the norms that shaped the functioning of the policy-process under the Meiji Constitution will allow us to understand why this system was able to stabilize itself in the face of political crisis. This thesis will build a theoretical model that expands upon Steven‟s hybrid constitutional theory by taking that model‟s explanation of institutional dynamics and fusing it with analysis that looks to elite political culture and institutional framework as they pertain to the policy-process. Specifically, this thesis will develop a new theory that recognizes the symbiotic relationship between the cultural norms and institutional structures which guide the policy-process. It will be argued that these independent variables act together as a single causal mechanism. Specifically, this thesis is concerned with explaining the role of norms established and perpetuated by political elites that shaped institutional dynamics and created long term political stability. For the purposes of this thesis, political stability is reduced to the ability of actors who act through political institutions to play a role in the policy-process in such a way that it is sustainable and responsive to the needs of the system‟s actual policy-makers over time. Hence, this conception of political stability assumes that when systems are unable to sustain chronic legislative deadlock without altering the guiding norms and structures of the policy-process they politically unstable. According to this definition of political instability, the less stable the policy process is in the face of sustained legislative deadlock the more likely it will be circumvented and/or changed. In this way, regime change is characterized by a change in the policy-process that breaks with pre- existing norms and structures that guide the policy-process. Hence, this thesis conceptualizes politically stable regimes as those which can maintain the existing structure and norms of their policy-processes. - 25 - Blending Political Culture with Institutions: Oligarchic Policy-Making Theory This thesis articulates a theoretical model that incorporates both political culture and institutions when addressing the outcome of political stability, defined here as the absence of regime change (see above), in prewar Japan. It states that both political culture and institutional framework created a high propensity for elites to collectively rule as an oligarchy. This theory argues that during crises when one set of actors directly affiliated with a policy-effecting institution sought to establish a more prominent position in the balance of power within the oligarchy they had low incentives to circumvent the existing political structure. It is argued that even during dramatic shifts in power between policy-affecting institutions that the Meiji Constitutional system was able to withstand high degrees of legislative deadlock due to the perpetuation of oligarchic policy-making. Oligarchic policy-making had been a proven strategy used by the Genrō to manipulate the political structure with ease and to their advantage during the Meiji period. Despite the fact that party- politicians in the 1920‟s and military officers in the 1930‟s sought to dominate the political system, the very nature of the structure itself required them to revert to oligarchic policy-making in order to minimize legislative deadlock. This model, known henceforth as oligarchic policy-making theory, identifies two causal or independent variables: passive imperial leadership and the institutional framework of the Meiji Constitutional system. It maintains one intervening variable, the norm of oligarchic rule between state institutions with a high probability for stability of political institutions being the dependent variable. According to this theory, passive imperial leadership in executive and legislative affairs (IV 1) together with a constitutional order based upon imperial sovereignty (IV 2) cause norms of oligarchic rule between state institutions. This norm of oligarchic rule (Int. V) creates higher incentives for political actors to resolve political disputes and crises through existing institutional structures than by circumventing them. - 26 - Oligarchic policy-making theory claims that the prewar regime was never undermined during crisis by members of any policy-effecting institutions because oligarchic policy-making provides a mechanism for maximizing institutional autonomy. The norm of oligarchic rule allows political actors who work within different institutions a greater degree of autonomy than under a system in which the executive and legislative roles are more centralized. While certain political actors would gain from circumventing the norms and processes of the existing system to centralize power under their institutional base, doing so would necessarily undermine the relative autonomy and veto power possessed by other institutions. The fact that all policy-effecting institutional actors could and did veto policy by working to bring down the government through cabinet resignations— especially during crises—would have created perpetual legislative deadlock. Hence, during potentially destabilizing political crises, when one actor has the opportunity to gain more power over the system there is a high probability that they will expand their positive influence over the policy-process through compromise rather than confrontation. Compromise in this sense—that is in crisis situations—will involve bargaining between elite actors to achieve solutions to conflicts but in such a way that reshapes the balance of power between them. By taking more of a leadership role in the existing norms and processes through compromise with the other representatives of policy- effecting institutional actors, the ascending actor(s) gain more positive influence over the policy- process. Ultimately, this translates into political stability for the regime as a whole (DV). Oligarchic Policy-Making Theory Norm of Passive Individual Imperial Leadership (IV 1) + Institutional Framework (IV 2)  Norms of Oligarchic Policy-Making (Int. V)  Stable Political Regime (DV) - 27 - Observable Implications of Oligarchic Policy-Making Theory OLIGARCHIC POLICY-MAKING If this theory holds then we should observe imperial authority being wielded by policy- making/effecting leaders of policy-effecting state institutions rather than directly by the emperor on policy-decisions. This means that we should observe individuals/groups that control policy-affecting state institutions within the constitutional system being the source of all policy-decisions and initiatives. Specifically, we should observe actors who lead or possess strong influence over or within one or more policy-affecting state institutions initiating policies and coordinating their approval by other state institutions as well as playing a lead role in forming cabinets. We should observe elites from at least two or more policy-affecting institutions working together to initiate policy directly and attempts to gain tacit approval from other policy- effecting institutions to prevent their opposition and subsequent veto of the policy and/or the cabinet. This should be the core of the policy-making process in both times of political crisis and times of moderate to low conflict between political actors. Policies are therefore developed collectively between representatives and/or leaders of policy- effecting institutions who must be consulted before policy can be successfully enacted. This theory holds that the more developed alliances between two or more policy-affecting institutions the higher likelihood for compromise between them on policy. In these cases, there would be less incentive for allied political actors from different policy-affecting institutions to veto proposed policies. Conversely, this theory claims that the less developed alliances between two or more policy-effecting institutions the lower the likelihood for compromise between them on policy. In these cases, there would be more incentive for allied political actors from different policy-effecting institutions to veto proposed policies. - 28 - This theory defines alliance as the degree to which a political actor or group of political actors can establish cooperation between the policy-effecting institution that (s)he or they lead/represent and contacts that they have within other policy-effecting institutions. Effectively the strength of alliances is measured by the following criteria: a) The degree to which said policy-makers can gain access to resources and/or information that the allied institution(s) possess b) The degree to which said policy-makers can rely on members of the allied institution(s) to work towards pursuing the same policy goal(s) and executive actions held by and attempted by the leader in question The stronger these alliances are, the more power and influence these policy-makers should be in terms of being able block and put forth policy as well as have an influence on the formation of the cabinets. Specifically, in cases where policy-makers have strong alliances with members of other policy-effecting institutions we should observe all of the following:  high levels of influence on the policy-process being exerted by their policy-effecting state institution—their preferences should shape the initial drafts of policies and they should play a leading role in negotiation with other institutional actors on refining and finalizing proposed legislation; they should be able to successfully veto proposed legislation they oppose  the cabinet being led directly by them or indirectly through close cohorts directly affiliated with said leader‟s particular state institution(s) - 29 -  a high ratio of cabinet seats going to individuals picked by and/or affiliated with the leader in question In cases where policy-makers have weak alliances we should observe all of the following:  low levels of influence on the policy-process being exerted by their policy-effecting state institution—their preferences should not shape the initial drafts of policies and they should not play a leading role in negotiation with other institutional actors on refining and finalizing proposed legislation  the cabinet not being led directly by them or indirectly through close cohorts directly affiliated with said leader‟s particular state institution(s)  none all or a low ratio of cabinet seats going to individuals picked by and/or affiliated with the leader in question PASSIVE IMPERIAL RULE If this theory is correct then we should not observe the emperor actively formulating policy in an executive or legislative manner or regularly intervening in the policy process. The emperor should intervene and provide active leadership in the policy process rarely or never. The imperial throne should be an effective „rubber-stamping‟ agency for policies devised by groups of actors that control the key policy-directing state institution(s). Specifically this means we should see the emperor: a) not being the source of any policy decisions - 30 - b) having all of his decisions micro-managed and prepared by representatives of at least two policy-effecting institutions either directly or indirectly c) not engaging directly in leading any political organizations, debates or initiatives in the public or private sphere NEGOTIATED REGIME STABILIZATION BETWEEN OLIGARCHIC ELITES IN THE FACE OF POTENTIALLY DESTABILIZING CRISES If this theory holds then we should observe a particular sequence in terms of crisis resolution between elites from different policy-effecting institutions. If this theory is correct then we should observe leaders of disaffected predominantly cabinet-breaking institutions choosing to work within the existing institutional framework to increase their power during potentially destabilizing political crises. A potentially destabilizing crisis is defined here as a situation in which high levels of popular discontent overlap with a major dispute between political elites who operate different policy-effecting institutions. Responses to popular discontent by disgruntled elites in situations of potentially destabilizing crises (methods by which said elite actors could use mass movements to enhance their own power). Either they will use ties with mass movements to resolve crisis by forcing regime change either through negotiations with other elites or through revolution, or system destabilizing crises. Or actors will use ties with mass movements to resolve crisis by negotiating a power shift while preserving the regime, or balance of power crises. (For a fuller contrast of these two approaches see figures 2 and 3 on pages 29-30). Both balance of power and system destabilizing crises involve two elements (see figure 2 on page 29). First, they involve a relatively high degree of political unrest as expressed by a segment of the population against the government. Both types of crisis are initiated by high levels of political discontent on the part of the population expressed through popular protest mobilized by popular - 31 - movements71. Specifically, political unrest should be measured by the degree and frequency with which participants engage in peaceful protest and/or violence directed at state officials and government institutions. Second, balance of power and system destabilizing crises involve a political dispute over policy and/or on the way in which power is distributed between policy- effecting state institutions. This dispute must result in members from one policy-affecting institution threatening to or actually breaking cabinets formed in concert with or by other institutions (there must be disagreement between the former and members of the latter on a disputed issue or set of issues). In addition, this political dispute between elite political actors, while it is likely to originate before, must occur after the onset or during the onset of popular protest. What ultimately distinguishes balance of power crises from regime destabilization crises is the way in which actors who seek to use these crises to increase their power will attempt to resolve it. If leaders of disaffected state institutions enhance their power by use of existing norms and procedures for cabinet- and/or policy-making then they are able to enhance their power without altering the existing regime. In these cases, the potentially destabilizing crisis would be identified as a balance of power crisis as a negotiated change in the balance between policy-effecting state institutions by leaders/representatives of these served to resolve it at the elite level. In cases where leaders of disaffected state institutions attempt to circumvent the existing norms and procedures for 71 This theory defines popular movements as organizations that have a mass base of support among citizens at the local, regional or national level. Popular movements should be organized around a set of principles that its members claim to adhere to as well as specific policy goals. According to this definition, popular movements must have minimal or no qualifications for entry so that it is possible for members from different socio-economic backgrounds to possess membership. Though they may engage in alliances with the state and/or occupy positions of power within policy- effecting institutions and/or the cabinet, popular movements are separate from and organized independently from state institutions. By this definition political parties may qualify as popular movements. Specifically, popular movements with little or no connection to actors operating in policy-making/effecting state institutions should have no influence on policy-decisions. Movements with direct connections to policy-making/effecting state institutions should have an influence on said institutions‟ positions on policy in the following way: only if the faction/movement‟s members are in leadership positions within state institutions will these groups have an influence on policy-decisions. - 32 - cabinet- and/or policy-making then the regime collapse or change is inevitable as a result. Hence, these cases are known as system destabilizing crises. In destabilizing crises individual and/or institutional actors that have an interest in increasing their power and undermining that of others may seek to use high levels discontent as a pretext for power change (see figure 3 on page 30). What matters most about this process is whether those actors who seek to increase their power will attempt to work within the context of the existing institutional framework‟s rules and norms as they are defined and practiced. This could involve reinterpreting existing legal procedures. Such a course of action won‟t necessarily threaten the system with constitutional change or collapse but will result in revision of norms and procedures. It is when actors that seek to increase their power by circumventing any or all existing institutional rules and norms that their actions pose a serious threat to the survival and stability of that institutional framework. Actors might do this in a variety of ways: they might expand their power by utilizing exceptional provisions within the existing institutional framework, such as emergency executive powers for example. An alternative to this is to establish a new institutional framework with a new set of rules and norms. Failing these two, actors may simply operate in an ad hoc way that circumvents existing institutional restraints on their power and allows them to increase their influence over policy-making. Depending on the circumstances, any of these three methods for power maximization could result in either the indefinite suspension of, revision/replacement of or the collapse of existing institutional rules and norms as defined by and practiced under the constitution. - 33 - Table 1: Typologies for classifying potentially destabilizing political crises Balance of Power Crisis (Popular Dissatisfaction with Government) High degree of popular unrest translated into widespread political protest (Elite Dissatisfaction with Government) Political dispute between elites from different policy-affecting institutions (Resolution of Conflict) Leaders of disaffected state institutions enhance their power by use of existing norms and procedures for cabinet- and/or policy-making System Destabilizing Crisis High degree of popular unrest translated into widespread political protest Political dispute between elites from different policy-affecting institutions Leaders of disaffected state institutions enhance their power by attempting to circumvent the existing norms and procedures for cabinet- and/or policy- making If this theory is correct then we should observe all potentially destabilizing crises being resolved in such a way that they become balance of power rather than system destabilizing crises because no ascending political actor will circumvent existing norms of oligarchic policy-making. We should observe disaffected political actors using potentially destabilizing political crises attempting to increase their power and influence over the policy-making process during and the in the aftermath of these events. The method used by these actors should be through building strong alliances with members of other institutions and attempting to take on more of a leading role in the policy-process in the manner mentioned above (see first observable implication). Specifically we should observe potentially destabilizing crises develop in to balance of power crises and never into system destabilizing crises. - 34 - Table 2: Typologies for conceptualizing different responses by regime to potentially destabilizing crises Forced Regime Change Negotiated Regime Change Negotiated Regime Stabilization Regime Stabilization Level of Popular Dissatisfaction with Government High High High High Existence of a Major Political Dispute between Political Elites Yes Yes Yes No Strength of Ties between Elite Political Actors and Mass Movements Strong Weak Strong Weak Circumvention of Pre-Existing Norms for Cabinet and Policy-Making Yes Yes No No Outcome of Crisis Cabinet Collapse and Change in Elite Balance of Power creating Regime Destabilization and Change Cabinet Collapse and Change in Elite Balance of Power creating Regime Destabilization and Change Cabinet Collapse and Change in Elite Balance of Power, Regime Stabilization Cabinet Stability and Regime Stabilization - 35 - Methodological Approach A qualitative small-N case study will be conducted using John Stuart Mill‟s Method of Agreement. Mill defines the Method of Agreement in the following way: “If two or more instances of the phenomenon under investigation have one circumstance in common, the circumstance in which alone all the instances agree, is the cause (or effect) of the given phenomenon”72. This thesis will compare multiple cases with different historical situations and causes but that yielded the same effect on Japan‟s prewar political institutions—that is threaten the regime with the potential of collapse. The prevailing stability of the Meiji Constitutional system will be examined by examining two separate crises that threatened the stability of the Imperial Japanese state between 1890 and 1945. The first of these crises are the 1918 Rice Riots, a short period of severe economic crisis resulting in nation-wide riots and strikes and subsequently resulted in the expansion of political parties‟ power through the Imperial Diet. The second crisis to be examined is the 1936 Incident, the last and most threatening in a series of coup attempts by military officers in the 1930‟s that subsequently resulted in the expansion of the military‟s political influence through its cabinet position and connections with other institutions. Both cases will be examined in terms of context, how each situation might have destabilized the Imperial Japanese state and how oligarchic policy- making theory can fully explain politics before, during and after these crises. The 1918 Rice Riots and 1936 Incident are ideal for comparison through Mill‟s Method of Agreement as they constitute different circumstances under which the Meiji Constitutional system might have been directly challenged while resulting in regime stabilization. The 1918 Rice Riots was a popular response to rapid-inflation as a result of a Genrō-led cabinet‟s economic and military 72 Mill, John Stuart. System of Logic Ratiocinative and Inductive: Being a Connected View of the Principles of Evidence and the Methods of Scientific Investigation, Volume 1, Seventh Edition. Longmans, Green, Reader and Dyer, 1868. 428 - 36 - policy (see below). This economic crisis triggered anti-government protests throughout the county and resulted in nation-wide riots and attacks on police stations for several days across Japan. As an economic crisis that triggered unprecedented protests and violence directed at the state and those believed to have been responsible for profiteering by hoarding rice the Rice Riots might have led to the collapse of the Imperial state if political parties had pushed for it73. In the 1936 Incident, almost 1500 officers from the Imperial Army seized government buildings and assassinated cabinet members and other high-ranking government officials in an attempt to bring a military dictatorship. High ranking army officers sympathized with this attempt and though this manifested itself in their delayed suppression of the uprising, they did not openly support it from their end by arbitrarily installing a new cabinet (see below). Hence, the Rice Riots and the failed coup of 1936 were both different instances in which the Meiji Constitutional system was under severe pressure yet both proved the robustness of Japan‟s political institutions as neither case led to regime change. 73 Lewis, Michael. Rioters and Citizens: Mass Protest in Imperial Japan. University of California Press, 1990. 100-115 - 37 - III. THE DEVELOPMENT OF OLIGARCHIC POLICY-MAKING IN THE MEIJI PERIOD The Legacy of Passive Imperial Leadership Throughout the medieval and early modern periods, the Japanese emperor‟s role in politics had evolved from supreme political leader to a non-partisan sanctifier of political legitimacy74. Centuries prior to the Meiji Isshin, the Imperial throne had been a symbol of political legitimacy and national unity and was an intrinsic part of Japanese political culture since the Nara period when Japan adopted the Chinese imperial state model. Japan‟s first imperial government was closely modeled after that of T‟ang dynasty China with emperor serving as the chief executive and legislative officer who presided over a small state council75. While China‟s pre-modern imperial state was able to last several hundred years, Japan‟s collapsed during the middle ages as the rise of incessant civil war facilitated the descent into feudalism76. As military officers became more involved in political affairs, the emperor‟s role began to shift to more of a symbolic role—rather than actively determine policy, the emperor now appointed executive officers to rule in his place 77 . This seems to have been a symbiotic relationship developed between the imperial house and power generals as the former was usually provided with some material support and maintained prestige whereas the latter was afforded political legitimacy78. By the time of Japan‟s unification under Tokugawa Ieyasu at the beginning of the seventeenth century, the imperial house had come to depend on the resources and protection of 74 Titus, David Anson. Palace and Politics in Prewar Japan. Columbia University Press, 1974. 52-57 75 Duus, Peter. Feudalism in Japan in Studies in World Civilization Systems Design and Implementation, Third Edition (1993). McGraw-Hill, 1976. See Chapters 2 and 3. 76 A major reason for this institutional collapse was the gradual erosion of the ritsuryo taxation system which the imperial government had relied upon heavily for revenue. Yamamura, Kozo. “The Decline of the Ritsuryo System: Hypotheses on Economic and Institutional Change,” in Journal of Japanese Studies, Vol. 1, No. 1 (Autumn, 1974), 3- 37. 26-29 77 Duus {1976}, see chapters 2 and 3 78 Ibid - 38 - various different feudal warlords79. All three of Japan‟s great unifiers sought to secure a position at the imperial court as the officially designated shōgun and this role of the emperor as symbolic head of state was continued throughout the seventeenth and eighteenth centuries80. The Re-Emergence of Passive Imperial Rule during the Meiji Period This trend continued into the nineteenth century but was interrupted by the expansion of western trade and colonialism into East Asia under the active reign of emperor Kōmei. The tradition of passive imperial leadership was interrupted as the Meiji Emperor‟s father, the last to reign over the Tokugawa polity, encouraged resistance to foreign encroachment81. Kōmei advocated samurai to fight Europeans and Americans who demanded that Japan relax its highly restrictive trade policy82. He sought to halt the negotiations between Japan and western powers and in this way was active in politics83. The extraordinary circumstances seem to be the cause of this deviation from the norm as Japan was perceived to be on the verge of colonization and subjugation by western powers84. In the wake of the Tokugawa regime‟s collapse, the Restorationist samurai and the imperial court were in a unique position as both needed the other in order to build a stable regime which could strengthen Japan in the face of western imperial advances into Asia. Though they had planned to restore the emperor‟s power as supreme executive and legislative officer, the samurai 79 Duus {1976}, see chapters 2 and 3 80 Duus {1976}, see chapter 4 81 Shillony, Ben-Ami. “Conservative Dissatisfaction with the Modern Emperors,” in The Emperors of Modern Japan. Ed. Ben-Ami Shillony. in the Handbook of Oriental Studies series, Vol. 14. Koninklijke Brill, 2008. 137-162. 137-138 82 Ibid 83 Shillony {2008}, 137-138 84 Ibid - 39 - who overthrew the Tokugawa bakufu (lit. “tent government”) did not have an exact programme85. The imperial throne was to be the locus of power, that much the Restorationists had agreed on but in practice the imperial court proved to be too inexperienced in political affairs86. The emperor was a teenager and the court officials were neither prepared to advise the young emperor nor rule on his behalf as the imperial house had long been removed from the policy-making process87. By contrast, the Restorationists had up to this point been samurai officials under the Tokugawa regime88. These figures were not only trained military officers they also possessed bureaucratic training and experience and thus, they were more equipped to govern than the imperial court89. Yet, it would be difficult for certain cliques of Restorationist samurai from the south of Honshu to establish a centralized state over the rest of the country. Throughout the Tokugawa period and before, Japan had been a loose collection of feudal territories, in which caste and local clan association had determined the course of politics90. In essence, the Restorationists were not merely building a new regime but building a different kind of regime in which the pre-existing feudal authority structure would be replaced with a central state91. Hence, they needed some kind of symbol or figure that could serve as basis of legitimacy for a government which would essentially transcend and abolish the pre-existing status-based and highly decentralized socio-political 85 Banno, Junji. Democracy in Pre-War Japan Concepts of Government, 1871-1937: collected essays. Trans. Andrew Fraser. Routledge, 2001. 1-3; Jansen, Marius B. The Making of Modern Japan. Belknap/Harvard University Press, 2000. 331-343 86 Titus, 13-21 87 Silberman, Bernard S. “Bureaucratic Development and the Structure of Decision-Making in the Meiji Period: The Case of the Genro,” in The Journal of Asian Studies, Vol. 27, No. 1 (Nov. 1967), 81-94. 87-89; Silberman, Bernard S. Cages of Reason: the Rise of the Rational State in France, Japan, The United States and Great Britain. University of Chicago Press, 1993. 160-168 88 Silberman {1967}, 87-89; Silberman {1993}, 160-168; Titus, 13-21 89 Silberman {1967}, 87-89; Silberman {1993}, 160-168; Titus, 13-21 90 Eisenstadt, S.N. Japanese Civilization: A Comparative View. Chicago & London: University of Chicago Press, 1996. 185-189; Ikegami, Eiko. The Taming of the Samurai: Honorific Individualism and the Making of Modern Japan. Harvard University Press, 1995. 158-159; Jansen, 123-127 91 Silberman {1993}, 159-160 - 40 - system92. In this way, the imperial court needed the Restorationist samurai to perform the practical tasks of government while the former would serve the new regime as a symbol of political legitimacy. While formally the head of state, the Meiji emperor was in practice a passive and malleable figurehead whose role in politics was largely symbolic. Though the Meiji emperor was to be an important symbol of national political unity, structural separation between court and state impeded the development direct imperial rule in practice93. Initially there was no distinction between the emperor‟s role in government and sacred Shinto rites94. Court officials and imperial advisors had administrative duties in addition to their roles at court95. By 1871, the role of court officials had changed they were removed of their administrative duties and were distinct from the Civil Government96. Hence, though there was a separation between the administration of the imperial court and the state the emperor was still the center of government97. From the early days of the Meiji state, the imperial court played a largely symbolic role in the policy-process. Before 1869 only three out of twenty-two court nobles were appointed as Councillors and from 1869 to 1885 none were98. Court nobles did hold, in most occasions, the offices of Chancellor as well as Ministers of the Right and Left99. Court advisor Motoda Eifu (1818- 1891) attempted to reverse the trend towards structural separation between the imperial court and the government proper in 1877 with the creation of Jiho (Advisors-in-Attendance) 100 . These advisors were intended to educate the emperor on political affairs as well as directly carry out the 92 Silberman {1993}, 178-180 93 Titus, 16-17 94 Ibid. 95 Titus, 16-17 96 Titus, 16-17 97 Ibid 98 Titus, 18 99 Ibid 100 Titus, 18-21 - 41 - imperial prerogative in executive, legislative and judicial decisions 101 . Yet, the role of Jiho necessarily threatened to undermine the oligarchic government that had developed by directly competing with the Ministers and Councillors of the Dajōkan and for this reason it was abolished soon after its establishment102. The young emperor had come to rely on the Chancellor and the Minister of the Left to finalize executive, legislative and judicial decisions103. There were exceptions to the passive rule of the Meiji emperor. He did intervene directly in politics during the 1890‟s as he refused the resignation of the Home Minister in 1892 and facilitated a compromise on the ship- building budget in 1893104. For the most part however, though he formally gave his consent and approved policy-decisions presented to him by high-ranking government and military officials105. Thus, the Meiji emperor was never an independent or active player in government before or after the establishment of the Meiji Constitution106. Oligarchic Policy-Making in Early Meiji Japan: The Product of Passive Imperial Rule and Oligarchic Institution-Building The emergence of an oligarchic policy-process in the early Meiji period was ultimately a by-product of the passive imperial rule and the nature of institution-building before 1889. In considering the development of oligarchic policy-making in the early Meiji period it is important to understand the situation of the Restorationist samurai who took over the state in 1868. Though they had collectively worked to overthrow and replace the Tokugawa government, those Restorationist samurai who took over the practical administrative duties of their predecessors did 101 Ibid 102 Titus, 18-21 103 Ibid 104 Hara, Takeshi. “The „Great Emperor‟ Meiji,” in The Emperors of Modern Japan. Ed. Ben-Ami Shillony. in the Handbook of Oriental Studies series, Vol. 14. Koninklijke Brill, 2008. 213-226. 220 105 Titus, 55-56 106 Steven, 128-129 Titus, 55-56 - 42 - not centralize political leadership into one formal institution (see below). These figures agreed on general policy goals but few of the Restorationist samurai who took over the state could actually agree on the means through which it should be achieved107. Coming from particular regions in what is now southern Honshu, the Restorationist samurai had the same socio-economic background and their group had a collegial dynamic to it108. Most of these individuals didn‟t want to open up the polity to include the wider populace nor did most of them want politics to revert to the institutions of the Tokugawa regime109 . In this way, the Restorationist samurai wanted to keep the political system closed to capable and like-minded individuals but had no institutional trapping, nor a clear strategy for building modern state110. Over time the lack of institutional criteria for participation in the policy-process combined with a passive imperial court fostered the norm of oligarchic policy-making. The Meiji emperor, as mentioned above, was to be the center of the new regime as it declared itself a restoration of direct imperial rule and promised to modernize the country and ward off foreign subjugation111. Officially, the emperor was Japan‟s legal sovereign and the locus of supreme executive, legislative and judicial authority. While the emperor ruled in theory, Japan‟s new government of a large group of former samurai officials ran the state in practice112. The ways in which this group built institutions inevitably led to its transformation into a small oligarchy by the 1880‟s113. 107 Banno, 1-3, 29-30; Ramseyer and Rosenbluth, 15-28 108 Moody, Peter. “Genro Rule in China and Japan: A Comparative Perspective,” in Journal of Chinese Political Science, Vol. 12, No. 1, 2007. 29-49. 30-31; Silberman {1993}, 178-187 109 Banno, 1-3, 29-30; Ramseyer and Rosenbluth, 15-28 110 Silberman {1967}, 90-94; Silberman {1993}, 178-187 111 Totman, Conrad. “The Meiji Restoration: From Obsolete Order To Effective Regime,” in Japan Examined: Perspectives on Japanese History. Ed. Harry Wray and Hilary Conroy. Honolulu : University of Hawaii Press, 1983. 72- 78. 71-74 112 Silberman {1967}, 90-94; Silberman {1993}, 178-187; Titus, 16-21; Umegaki, Michio. After the Restoration: The Beginning of Japan’s Modern State. New York University Press, 1988. 81-96 113 Silberman {1967}, 90-94; Silberman {1993}, 178-187 - 43 - Wielding imperial authority to legislate and use executive power, the early Meiji rulers did not immediately build an institutional framework for guiding the policy-process before 1889. Prior to the establishment of the Meiji Constitutional system, the oligarchs had used the Dajōkan system centered on the imperial throne to build a modern Japanese state114. Yet, there was a great deal of overlap between government and bureaucratic decisions. Before the Meiji Constitution was enacted, the oligarchs used imperial decrees issued in the name of a passive emperor to make, adjust or change laws115. As pragmatic revolutionaries, the Meiji oligarchs focused on combining state-building with building a powerbase within the state itself 116 . This seems to have been motivated by a desire on the part of many influential oligarchs to protect and maintain their position within the oligarchy117. Over time, the most influential oligarchs came to derive their power from maintaining their power and influence over certain state institutions118. Despite their role as Japan‟s modern state-builders, few members of the oligarchy seem to have been interested in establishing a basis of power outside the state119. When this did finally begin to happen it was motivated by a desire to recreate a position of influence with the oligarchy itself120. Hence, core state institutions such as the imperial army and navy as well as the bureaucracy were consolidated institutions before the political system itself was completely institutionalized121. 114 Umegaki, 81-96 115 Silberman {1967}, 90-94; Silberman {1993}, 178-187 116 Ramseyer and Rosenbluth, 15-28 117 Ibid 118 Ramseyer and Rosenbluth, 15-28 119 Banno, chapter 1; Ramseyer and Rosenbluth, 15-28 120 This seems to have happened at various stages during the oligarchy. In the 1870‟s Saigō Takamori established a powerbase outside the state by uniting many samurai in the Satusma area; Itagaki Taiskue established a powerbase among organizations that made up the Popular Rights and Freedom Movement in the 1870s and 1880s; Itō Hirobumi and later Yamagata‟s protégé Katsura Tarō would establish powerbases in the House of Representatives by creating political parties. Banno, chapter 1; Duus, Peter. Party Rival and Political Change in Taishō Japan. Harvard University Press, 1968. 39-44; Scalapino, Robert A. Democracy and the Party Movement in Prewar Japan: The Failure of the First Attempt. University of California Press, 1962. 43-46, 57-61, 95-96, 179-181, 193-194 121 Ramseyer and Rosenbluth, 15-28 - 44 - The Restorationist samurai were transformed into an oligarchy also because they did not institutionalize their membership in the polity—that is they did not create formal institutions to regulate their rule or select successors122. As a result of this, decision-making became increasingly centralized in the hands of a progressively smaller group which had no criteria for defining its membership123. As a group of Restorationist samurai from select regions in southern Japan, the Meiji oligarchs were united by common professional training geographic and socio-economic background124. Junji Banno notes, there was much squabbling between various personalities within the oligarchy during the 1870‟s as the most prominent figures attempted to gain consensus on and support for different policies125. Had the imperial court been more experienced in policy affairs it might have served as a genuine agent of coordination as there was no other traditional authority which might have been able to act as a centripetal force in the legislative process126. Influence within the oligarchy was often-times dictated by the popularity of one‟s ideas as well as the ability to create loyal networks of personal support within new state institutions127. For this reason, the Meiji oligarchy as a governing body gradually evolved from a loose body of a few hundred former samurai officials into a small clique by the 1890‟s 128 . By this time, the Meiji oligarchs had established networks to coordinate otherwise what would have otherwise been autonomously operating institutions and choose their successors from them129. By the 1890‟s, the Meiji state had come to be dominated a smaller groups of men than had originally been involved in the state-building process during the late 1860‟s130 . Ultimately the 122 Silberman {1967}, 90-94; Silberman {1993}, 178-190 123 Silberman {1967}, 90-94; Silberman {1993}, 178-190 124 Silberman {1967}, 90-94; Silberman {1993}, 178-190 125 Banno, chapter 1, especially pages 1-3, 29-30 126 Eisenstadt, 266-277 127 Ramseyer and Rosenbluth, 15-28 128 Silberman {1967}, 90-94 129 Ibid 130 Silberman {1967}, 90-94 - 45 - adoption of a constitutional system was intended to ward off full-scale democratization and ease domestic pressure for a more representative political system while at the same time appeasing norms of western state-craft. In many ways the establishment of a constitutional system was meant to facilitate Japan‟s movement towards parity with western nations that had dictated extraterritoriality and favorable trade terms in the 1860‟s131. Since the Iwakura Mission, the Meiji oligarchs had recognized that modernization would require Japan to adopt western institutional designs and standards of law before it could successfully make the case for removing the extra- territoriality clauses in treaties with Europe and the U.S.132 . For Japan‟s political leaders, the adoption of a constitutional system was supposed to be part of a larger process that would make Japan into a modern state 133 . In addition, the establishment of a constitutional system was intended to formally institutionalize the policy-process to appease mounting domestic unrest that had been building since the 1870‟s134. By the 1880‟s, the rise of an autonomous press and political parties had proven that the oligarchs could not stop the politicization of the wider public by issuing 131 Jansen, 355-361; Kashiwazaki, Chikako. “Jus Sanguinis in Japan : The Origin of Citizenship in Comparative Perspective,” International Journal of Comparative Sociology, 1998. 278-300 132 Ibid 133 Beasley, W. G. “Meiji political institutions,” in The Cambridge History of Japan, Volume 5: The Nineteenth Century. Ed. Marius B. Jansen. Cambridge University press, 1989. 620-673. 651-652 134 Although a good proportion of the Meiji leadership did not prioritize the establishment of deliberative institutions, by the mid to late 1870‟s and early 1880‟s pressure from below combined with shifts in popular culture led to a demand for a more democratic order. Many in the Meiji oligarchy were initially opposed to instituting democratic institutions on the grounds that it would either lead to undermining the process of building a modern state and capitalist economy and/or lead to political destabilization. Along these lines, the Meiji state pursued a policy of suppressing political activities and debates in the public sphere that seemed to question or attack the government-this included tight control over the press, detaining and prosecuting political dissidents as well as using the military to suppress protests. Yet, the feasibility of maintaining this position was eroded considerably by two occurrences. The first of these was the establishment of legislative assemblies in regional governments. In the early years the oligarchy had allowed peasants and local notables, under the authority of an appointed regional government to attend to their own affairs as the national government was more concerned with developing federal structures and economic infrastructure. By 1878, the government gave into the demand for local assemblies by formally instituting regional legislatures throughout the country with a limited franchise based upon property qualifications. Banno, 31-34, 56-68; Halliday, Jon. A Political History of Japanese Capitalism. Pantheon Books, 1975. 23-31; Scalapino 32-38, 40-57; Sims, Richard. Japanese Political History since the Meiji Renovation 1866-2000. Palgrave, 2001. 5-12, 33-40, 42 - 46 - repressive administrative ordinances135. Hence, due to domestic pressure to democratize and the oligarchs‟ goal of putting Japan on par with the west in terms of trade and diplomatic relations, a constitutional system was adopted. The Institutional Framework of the Meiji Constitution In building a constitutional system, the oligarchs wanted to preserve their power as much as possible but they faced a dilemma: how maintain control of the system in the long term. If they adopted a system which vested sovereignty in the populace, they would easily and quickly lose control over the political system. The Japanese monarchy had been the figurehead of the Meiji state for several years at this point and the key to the oligarchs‟ political power and legitimacy136. These factors combined led the Meiji oligarchs to design a system that would vest supreme political power in the emperor137. The Meiji oligarchs envisioned that the Meiji Constitutional system would be a more elaborate and popularly sanctioned continuation of behind-the-scenes oligarchic rule 138 . Hence, establishing a constitutional system based upon imperial rather than popular sovereignty guaranteed that the oligarchs would maintain control over politics despite the fact that they lacked a single institution which could maintain a closed polity and guarantee succession. The Meiji Constitution of 1889 is comparable to the Imperial German and Russian constitutions as all three were based upon imperial sovereignty. As in Germany and later Russia, Japan‟s first modern constitutional system was strategically designed by authoritarian political elites to maintain their power in the face of domestic pressure for democratization. Like the Prussian Constitution of 1871 and the Russian Basic Laws of 1906, the Meiji Constitution vested 135 Banno, 31-34, 56-68; Halliday, 23-31; Scalapino 32-38, 40-57; Sims, 5-12, 33-40, 42 136 Pyle, Kenneth B. “Meiji conservatism,” in The Cambridge History of Japan, Volume 5: The Nineteenth Century. Ed. Marius B. Jansen. Cambridge University press, 1989. 674-720. 700-701 137 Beasley, 651-665 138 Ibid - 47 - sovereignty in the imperial throne139. In theory, the constitution concentrated supreme, executive, legislative and judicial power in the emperor140. He was commander-in-chief of the armed forces; held executive veto power and approved all laws141; appointed cabinet ministers and members of the upper house of the legislature142; could issue executive orders by decree; and could dissolve the legislature at will143; and in times of national emergency wielded special legislative powers144. In addition, the Meiji constitution established a holistic nationalist interpretation of the imperial throne. The Preamble paid homage to conservative and Shinto nationalist ideas about the Emperor being a descendent of Japan‟s first so-called „divine‟ emperor145. This section as well as Articles I to IV established a legal definition of the sovereign as a sacred deity146. The fact that the document as it refers to its origins and provisions as stemming from the benevolence of the Japanese emperor demonstrates the degree to which the oligarchs become dependent upon using imperial 139 Articles I to IV taken together confine the right of sovereignty to the imperial line to be execercised by the emperor in accordance with the legal framework established by the Meiji Constitution (Siemes). This was based upon the Prussian Constitution of 1871 and bares a striking similarity to the Russian Constitution of 1905. Berghahn, Volker R. Imperial Germany, 1871-1918: Economy, Society, Culture and Politics. (2005 Edition) Begrhahn Books, 1994. 176-184. ;Lee, Stephen J. Russia and the USSR, 1855-1991: Autocracy and Dictatorship. Routledge, 2006. 42; Siemes, Johannes. Herman Roesler and the making of the Meiji State: An Examination of His Influence and the Founders of Modern Japan & The Complete Text of the Meiji Constitution Accompanied By His Personal Commentaries and Notes. Sophia University and Charles E. Tuttle Company, 1968. 55-66 140 Article IV states that the “Emperor combines in Himself all government power.” Article V specifies that the emperor wields supreme legislative power with the consent of the Imperial Diet. Siemes, 64-66 141 Articles XI to XIV defined the emperor as commander-in-chief of the army and navy; the leadership over the army‟s and navy‟s organization in peace time; the right to declare war, make peace and conclude treaties; and the right to proclaim marshal law. Articles VI, IX, and X state that he gives sanction to laws and promulgates their implementation; the authority to organize government administration and the right to issue ordinances within the confines of the Constitution. Siemes, 69-73, 83-92, 93-103 142 Article XV states that the emperor “confers titles of nobility, rank, orders and other marks of honour.” Article XXXVI states that the “House of Peers shall, in accordance with the Ordinance concerning the House of Peers, be composed of members of the Imperial Family, of nobles, and of Deputies who have been nominated by the Emperor.” Siemes, 104-106, 156-160 143 Articles VII, XLII-XLIII state that the emperor opens, closes, prorogues, and dissolves the Imperial Diet and can call special sessions of any length determined by imperial ordinance. Siemes, 74-77, 172-174 144 Article VIII states that in cases of emergency when the Diet is not in session that the emperor has the power to legislate by issuing imperial ordinances which are subject to post-humous review by the Diet during its next session and can be retro-actively vetoed. 145 Siemes, 53-54 146 Siemes, 55-66 - 48 - nationalism to justify their rule147. Defined in these terms, the Meiji oligarchs believed that they had built a strong and lasting legal bulwark to mitigate and contain any popular involvement in government that this document guaranteed. The Early Years of the Meiji Constitutional System and the Persistence of Oligarchic Policy-Making the in Imperial Japanese Polity Despite the fact that the Meiji Constitution vested sovereignty in the imperial throne, this institutional framework crystallized the pre-existing system of oligarchic rule. In practice the pattern of oligarchic rule through control of state institutions performing executive and legislative functions on the emperor‟s behalf continued as the established legislative process148. As stated above, theoretically the executive was unified and concentrated with legislative and judicial power in the emperor. In practice, however, this system was highly susceptible to internal conflict and competition between institutions as the imperial prerogative could be used without limit in multiple jurisdictions149. In this sense, the Meiji constitution‟s institutional framework, whether power was wielded by the emperor himself or on his behalf, required a minimal degree of coordination and unification of state institutions to prevent high turnover of cabinets, deadlock and contradiction on policy 150 . Given the fact that the Meiji emperor was in most cases a passive monarch, the propensity for deadlock between state institutions that fell under the imperial prerogative was increased and therefore a high degree of coordination was required to effectively pass legislation151. In their design of the single elected body under the Meiji Constitution, the House of Representatives (the lower House of the Imperial Diet), it is clear that they intended to manipulate 147 Siemes, 55-66 148 Hall, John Witney. “A Modern Monarch for Modern Japan,” in Political Development in Modern Japan. Ed. Robert E. Ward. Princeton University press, 1968. 11-64. 56-57 149 Ibid. 150 Steven, 108-114, 128-129 151 Ibid - 49 - this institution and in so doing continue to maintain a relatively closed oligarchic polity. Initially the electoral laws allowed only the richest 1% of male subjects to elect representatives to the House of Representatives152. Thus, only a small minority of the population, which was in many ways from the same socio-economic background as the imperial court and the oligarchy, was politically enfranchised while the entire population was subject to conditional civil rights 153 . Additionally, Imperial sanction could be used to prorogue or dissolve the lower elected House of the Imperial Diet, the House of Peers—the appointed upper House could veto the House of Representatives and in emergency situations laws could be passed by imperial decree154. The oligarchs knew that they could use their control over through appointments the Privy Council, Civil Service, the Imperial Army and Navy as well as the House of Peers to gain cohesion within cabinets155. This combined with the marginally representative and weak lower house of the Diet would seem to allow the oligarchs to easily control the legislative process as they could continue to manipulate the imperial prerogative to consolidate and exercise real executive power. Much to the oligarchs‟ dismay, the first ten years under the Meiji Constitution would prove not only to be a continuation of struggle with political parties but also a serious test of their power. Elected officials would find a way to utilize the Diet to veto policies of Genrō-controlled cabinets. Though the oligarchs had placed countless checks on the House of Representatives in their constitutional design, they underestimated the ability of this institution to approve cabinet budget 152 Ramseyer and Rosenbluth, 33-44 153 That the establishment of propertied and gendered suffrage was meant to minimize popular involvement in politics is corroborated by the establishment of similar qualifications for regional parliamentary bodies and legal precedent. Ibid 154 Ramseyer and Rosenbluth, 33-44 155 Ramseyer and Rosenbluth, 33-44. Articles XXII to XXX contain legal loopholes for the state to undermine guarantees to rights of privacy, speech, petition, unlawful search and seizure of property, religious belief and assembly, 127-144; In addition, Article XXXII denies military staff full rights as guaranteed to other subjects in the above- mentioned sections, 148-149 - 50 - proposals156. Ultimately, it was the lower House‟s ability to reject the cabinet‟s annual budget that allowed the country‟s fledgling political parties to outmaneuver the oligarchy as this power could be used to defeat Genrō-controlled cabinets157. Despite the fact a cabinet could defer to the previous year‟s budget, and in some cases did, the demands of imperial expansion, war and industrialization during these years required increased revenue for the state 158 . Hence, the House of Representatives‟ power to approve budgets in practice was a potentially powerful veto on policies but also on the government itself. By the early 1900‟s the oligarchs would find themselves gradually developing alliance with political parties in order to avoid legislative deadlock in the lower House159. The early years of the Imperial Diet demonstrate the lengths that the oligarchs were willing to go to in order to keep political parties from influencing policy. In 1892, acting Prime Minister Itō Hirobumi attempted to raise taxes but met with stiff opposition in the lower House as representatives refused to pass the bill160. Rather than try to compromise with this body he immediately dissolved it and called new elections in which he used the Home Ministry used police and hired toughs (sōshi) to influence the outcome in the government‟s favor by stuffing ballot boxes and harassing party candidates161. After the election when deadlock persisted, the state used police and sōshi to attack Diet members while in session and forcibly close lower House and was followed by the collapse of the cabinet162. While multiple oligarchs had tried to coerce the Diet into accepting its policies, within four to five years the 156In early drafts of the Meiji Constitution it is clear that imperial sovereignty and a weak parliament were core principles that guided the oligarchs in their drafting of that document. All drafts of the Meiji Constitution outlined a parliament that was intended to be a superficial part of the legislative process with minimal influence. Beasley, 656-658 157 Steven, 114-118 158 Banno {1992}, chapters 2-3; Gordon, 91-92; Ohara, Masayo. Democratization and Expansion: Historical Lessons, Conetmporary Chellenges. Praeger, 2001. 72-73 159 Scalapino, 152-168 160Akita, George. Foundations of Constitutional Government in Modern Japan, 1868-1900. Harvard University Press, 1967. 101-105; Mitchell, Richard H. Censorship in Imperial Japan. Princeton University Press, 1983. 116-119; Scalapino, 160-164 161 Akita, 101-105; Mitchell {1983}, 116-119; Scalapino, 160-164 162 Akita, 101-105; Mitchell {1983}, 116-119; Scalapino, 160-164 - 51 - party-politicians in the lower House had clashed with the Genrō in six separate Diet sessions163. Yet, none of these figures was willing to appoint a party member to the cabinet let alone negotiate on policy with leaders in the House of Representatives to prevent deadlock164. Eventually, the defeat of several purely Genrō cabinets and one failed attempt at an all-party cabinet led to the beginning of a compromise between the two in which parties were given cabinet seats, money and some say on policy matters in exchange for their support of government bills165. 163 Scalapino, 152-168 164 Ibid 165 Scalapino, 152-168 - 52 - IV. TEST CASE ONE: ASSESSING THE POLICY-PROCESS IN THE ERA OF THE RICE RIOTS Understanding the Political Context of the Rice Riots Though the Meiji oligarchs had faced political crises and mass unrest before 1918, the Rice Riots were unprecedented in terms of the degree to which Genrō-dominated politics were threatened. The Rice Riots of 1918 was a crisis like none the Meiji state had faced before as it was the culmination of two trends that had gradually emerged in the late Meiji period: the rising influence of political parties and the increase in popular protest. It was the convergence of two long-term trends had made this crisis potentially explosive to such the degree that it might have resulted in the revision, suspension or even replacement of the Meiji Constitutional system. Increased party power along with steady increases in popular protest had created problems for the Meiji oligarchs before as these two elements had overlapped considerably in the years leading up to the establishment of the Meiji Constitution166. The Meiji oligarchs had hoped that said document would allow them to contain and separate these two elements indefinitely by co-opting party politicians and tightly restricting popular participation in politics167. This strategy was effective in helping the oligarchs maintain power with varying degrees of legitimacy into the twentieth century but it relied upon maintaining a balance between restricting party influence while still providing them with a role in government168. The oligarchs‟ strategy for maintaining the reigns of power also sought to create a relatively passive civil society with low levels of political discord169. Ultimately, the continued modernization of Japanese society would feed into the sustenance of these trends170. 166 Duus {1968}, 107-112; Gordon, Andrew. Labour and imperial Democracy in Prewar Japan. University of California Press, 1991. See chapter 2.; Ishida, Takeshi. “Movements to Protect Constitutional Government—A Structural- Functional Analysis,” in Democracy in Prewar Japan: Groundwork or Façade? Problems in Asian Civilizations Series. Ed. George O. Totten. D.C. Health and Company, 1965. 75-89. 75-79 167 Ramseyer and Rosenbluth, 30-31 168 Ibid 169 The Meiji oligarchs did not seek to establish an ideologically charged and highly politicized polity. Prior to 1890, they adopted a three pronged strategy of dissemination of propaganda, censorship and strong discouragement of political activities. This included tight censorship of the press, public meetings as well as restricting state employees from - 53 - By 1918, political parties were established partners of the Genrō in governing but sought to enhance their power by establishing party-dominated cabinets. Since before 1890, the Meiji oligarchs had struggled with political parties to minimize the latter‟s influence and maintain a relatively closed-political system supported by an authoritarian state171. However, parties had used the lower house of the Imperial Diet to gradually increase their influence in policy-making and the formation of cabinets172. In various struggles with the Diet to approve legislation, the Genrō had largely moved away from to the tactic of excluding party members from cabinet posts as this had often led to stiff deadlock created by party members in the lower House173. By the late Meiji and early Taishō periods Seiyūkai in particular had established itself as an ally of the Genrō in the lower House of the Imperial Diet (see below). Seiyūkai, supported by other parties and popular opinion, had played a crucial role in bringing down the third Katsura cabinet in 1913—a clear defeat for the Genrō as Katsura had been attacked for holding multiple offices simultaneously (see below). On the eve of the Rice Riots, parties in general but particularly the Seiyūkai were determined to establish the precedent of having all cabinet posts filled by members or allies of a single political party. This was to be a major factor in the Rice Riots as the unprecedented strength of political parties vis-à-vis the Genrō would give the former considerable leverage over the latter. As a series of popular protests, riots and strikes nation-wide for several days, the Rice Riots were the largest popular disturbances that the Meiji state had ever encountered. The participation in political parties, meetings and demonstrations. These policies were continued after 1890 as the Meiji Constitution allowed for the limitation of rights to speech, assembly, etc. In addition, the oligarchs had used Article 34 of the 1896 Uniform Civil Code, designed to keep the attention of citizens focused on fulfilling „fukoku kyoohei‟, and prevent the rise of strong NGOs Gluck, Carol. Japan’s Modern Myths: Ideology in the Late Meiji Period. Princeton University Press, 1985. 53-55; Mitchell {1983}, 64-65, 74-80, 80-88, 104-105, 111, 120-121, 124-125, 130-131; Pekkanen,Robert. Japan’s Dual Civil Society: Members Without Advocates. Stanford University Press, 2006. 51-53; Scalapino, 63-65 170 Gordon {1991}, see chapter 2 171 Mitchell {1983}, 91, 104-105, 111, 120-121, 124-125, 130-13; Pekkanen, 51-53 172 Duus {1968}, 83-86; Steven, 118-119 173 Duus {1968}, 83-86; Steven, 118-119 - 54 - establishment of parliaments at the local and national level combined with the integration political parties into those institutions as well as the rise of print media contributed to an unprecedented opening up of politics to the wider populace174. Yet, the fact that suffrage was restricted by gender and property qualifications separated a large part of the population from direct involvement in the political system175. This combined with the Peace Preservation Ordinance and the Uniform Civil Code imposed heavy restriction on political participation176. While most imperial subjects were legally restricted from political participation, by the early twentieth century the rise of a modern socio-economic structure and modern print-press led to the development of a more diverse and increasing politicized population177. In the two decades leading up to the Rice Riots, public protests over policy-issues increased dramatically and on several occasions erupted into clashes with police and full-scale riots178. The rapid transformation of Imperial Japan‟s economy in the years leading up to the Rice Riots also posed new challenges for the state. By 1918, Japan had a rapidly expanding industrial economy and a modern working class and in general wages lagged behind the cost of living for this group as well as increasing numbers of the middle class179. In this period, labour disputes reached unprecedented levels—particularly in 1917 and 1918--as the demand for collective bargaining rights and social welfare became more pronounced180. The Japanese state took on a new role as strike-breaker and policy-makers engaged in repressive campaigns directed against the proliferation of radical leftist literature and organizations181. 174 Gordon {1991}, chapter 2; Gordon {2009}, 125-135 175 Gordon {1991}, chapter 2; Gordon {2009}, 125-135 176 Mitchell {1983}, 91, 104-105, 111, 120-121, 124-125, 130-13; Pekkanen, 51-53 177 Gordon {1991}, chapter 2; Gordon {2009}, 125-135 178 Gordon {1991}, chapter 2; Gordon {2009}, 125-135; Lewis, Michael. Rioters and Citizens: Mass Protest in Imperial Japan. University of California Press, 1990.2-3; Ohara, 37-57 179 Gordon {1991}, chapter 2; Gordon {2009}, 125-135; Lewis, 2-3; Ohara, 37-57 180 Gordon {1991}, chapter 2; Gordon {2009}, 125-135; Lewis, 2-3; Ohara, 37-57 181 Gordon {1991}, chapter 2; Gordon {2009}, 125-135; Lewis, 2-3; Ohara, 37-57 - 55 - Despite the oligarchs‟ best efforts, political parties would eventually grow strong enough to challenge the Genrō‟s hegemonic position. This combined with the rising tide of popular discontent could have created a situation in which parties used popular discontent to directly attack the regime to win more power for themselves. In the years leading up to the Rice Riots, political parties became more powerful and ambitious while at the same time the wider populace became more involved in protests directed against the economic and political status quo. It was the way in which these two elements began to converge in the Rice Riots that made this event threatening. Characterized by unprecedented outbreaks of streaks and nationwide riots at a time when ambitious political parties sought to remove and replace the oligarchs, this event could have resulted in regime change had these two elements colluded completely. In this way, the Rice Riots threatened the stability of the Meiji Constitutional system. The Rice Riots and the Long-term Political Consequences The Rice Riots of 1918 were a series of protests involving strikes and riots in many parts of Japan from the 13th to the 16th of August182. This crisis originated from the government‟s decision to send 70,000 troops to Siberia in order to prop up the White Russian forces and allied supplies in Vladivostok183. Though Prime Minister Terauchi Masatake‟s cabinet was determined to support an anti-communist resistance in Russia, this move was received with great reluctance by the Japanese population that was forced to pay for it184. The crisis began when the government seized large amounts of domestic rice reserves to supply troops in the field as this move resulted in a 182 Lewis, 1-11, 15-26 183 Gordon, Andrew. A Modern History of Japan: From Tokugawa Times to the Present, Second Edition. Oxford University Press, Inc. 2009. 174-175 184 Lewis, 1-11, 15-26 - 56 - shortage of rice and rapid inflation 185 . With the government doing nothing about the sudden increase in the cost of this staple foodstuff in matter of hyper inflation had spread to many other consumer goods and an economic crisis186. On August 14th protests and some strikes against the rise in consumer prices and the government led to riots in localities throughout the countryside187. The riots lasted for four days and saw unprecedented destruction of property including the widespread destruction of police boxes and bombing of police stations188. In the midst of the crisis, party leaders met and openly denounced the government‟s economic policy189. By the end of the crisis Seiyūkai leader Hara Kei met with Yamagata Aritomo and threatened to lead a public demonstration against the government together with the other parties190. Within a month of this incident Yamagata had appointed Kei prime minister and the Terauchi cabinet had resigned191. The Rice Riots, though they did not result in a full-scale political revolution, were a turning point for the prewar political system and did lead to important power shifts between political actors and significant changes in domestic policy. It is important to point out there were a lot of continuities between the Meiji and Taishō periods as many aspects of foreign and domestic policy remained unaltered or even intensified during the latter192. Yet, the parties‟ enhanced power during the 1920‟s was reflected in their expansion of the electorate, adoption of social welfare programs and movement towards a foreign policy characterized by increased diplomacy193. The Rice Riots are often seen by political historians as the turning point at which political parties were finally able to seize control over the policy-process and the beginnings of the so-called Taishō democratic 185 Lewis, 1-11, 15-26 186 Ibid 187 Lewis, 1-11, 15-26 188 Gordon {2009}, 130-134; Lewis, 1-11, 15-26 189 Ibid 190 Gordon {2009}, 130-134; Lewis, 1-11, 15-26 191 Ibid 192 Kato, 220-236 193 Sims, 129-141 - 57 - period194. While it is clear as demonstrated below that party-politicians did not cease complete control over the policy-process, the Rice Riots lead to a shift in the balance of power between the Genrō and the House of Representatives (see below). The consequences of this were an increase in positive influence over the policy-process of political parties at the expense of the Genrō without affecting the Meiji Constitution. The Marxist Interpretation of the Rice Riots and its Theoretical Problems According to Barrington Moore‟s Marxist interpretation, the Rice Riots were nothing more than „scattered resistance to the transition from the premodern agrarian system to a new one.‟195 He articulates that the landlord class together with „militarists‟ and industrial capitalists dominated prewar politics and used the state to suppress the weak and divided lower classes196. For this reason the Rice Riots, though they were revolts by the peasants against the system because feudal social relations were preserved these did not result in revolution197. Moore is right to point out the politics under the Meiji Constitution were manipulated and dominated by a small group of elites. The Meiji political system by this point was an oligarchic state and popular movements had at best a partial influence on the policy-process. The electorate was limited largely to a wealthy minority and other policy-affecting institutions were accountable only to the imperial throne198. These combined with the fact that on average members of institutions other then the House of Representatives made up on average roughly 70-75% of the cabinets throughout 1890-1918199. While Moore‟s Marxist interpretation of the Rice Riots does highlight the elitist nature of prewar politics, his explanation is factually inaccurate because it assumes that the ruling classes 194 Scalapino makes this argument. See Scalapino, 210-211 195 Moore, 258 196 Moore, 289-299 197 Ibid 198 Ramseyer and Rosenbluth, 41-48 199 Steven, 112-113 - 58 - were more unified than they actually were. The problem with Moore‟s theory as it pertains to the Rice Riots is that it doesn‟t recognize the fact that this incident was a serious political crisis as it exacerbated divisions between certain political actors in the so-called ruling classes. Political parties by this point had become legitimate governing partners and rivals of the Genrō simultaneously. Moore‟s theory assumes that because parties were aligned with propertied classes who made up the industrial and agrarian sectors and because suffrage was limited that parties, along with the Genrō, were merely the tools of economic interests. Yet, the fact that the Rice Riots allowed parties the leverage to increase their power at the expense of the Genrō by having more control over the formation of subsequent cabinets disproves Moore‟s theory. When one explores the full depth of this crisis it is clearly evident that the Rice Riots exacerbated a mounting dispute between these two types of political elites. Had the parties chosen to, they could have easily aligned themselves with the public as they both had a common interest in seeing the fall of the cabinet and the general decline of the Genrō‟s influence in politics. The Political Leadership Theory Interpretation of the Rice Riots and its Theoretical Problems Richard J. Samuels‟ work does not explain the crisis of the Rice Riots, rather it deals with the choices made before it that put Hara Kei in a position of alliance with Yamagata Aritomo as well subsequent decisions made by the former that solidified Seiyūkai‟s power200. If one extrapolates from his leader-centered model, that leaders stretch structural constraints through utilizing different resources in the choices they make, then we might assume that choices of individual leaders were causal variables. If applied to the Rice Riots, Samuels‟ theory would have us believe that the reason this crisis did not destabilize the Meiji Constitutional system because of the choices made by key political actors involved. The most appropriate actor to focus on would seem to be Hara Kei 200 Samuels, 102-105, 114-123 - 59 - as this fits with Samuels‟ argument that Hara Kei‟s choices in the Taishō period resulted in the stabilization of pre-existing regime without fully democratizing it. This interpretation would explain the resolution of the Rice Riots as the product of Hara‟s decision to take over the system from within by not involving Seiyūkai in public protests and using his alliance with Yamagata to ascend to power through appointment to the cabinet as prime minister. Thus, it is implied by Samuels‟ model that the reason the Rice Riots did not destabilize the Meiji Constitutional system was because Hara chose to align himself and his party with the Genrō so that they would be allowed to form a government to help defuse the crisis. On first glance it appears as though Samuels‟ theoretical model can explain the stabilization of the Meiji Constitutional system in the wake of the Rice Riots but ultimately, it falls short as it renders the actions of other actors involved less important. Essentially, it overlooks the fact that none of the party leaders were willing to take to the risk of directly leading protests against the government, though they benefited from its undermining of the Terauchi cabinet 201 . The significance of this must not be underestimated as it demonstrates that there was an established consensus among all mainstream party leaders that the overthrow of the government by illegal means was out of the question202. These actors together with Seiyūkai had the most to gain by subverting the system at this time because they would have stood a good chance of being able to oversee constitutional revision or even replace it to expand their power. This assumption that Hara‟s choice to work within the system rather than work against through marshalling popular opinion was critical but it was not the deciding factor nor was he the only party leader willing to compromise with Yamagata. For this reason, the logic of Samuels‟ theory cannot fully explain the Rice Riots as it only considers the choices made by Hara Kei. In this way, Samuel‟s leader- 201 Lewis, 15-26 202 Ibid - 60 - centered model looks backward to trace Hara‟s political ascension rather than look forward and understand the context of those choices as they might have been influenced by prevailing beliefs and norms. The Rational Choice Theory Interpretation of the Rice Riots and its Theoretical Problems According to Ramseyer and Rosenbluth‟s argument, for prewar state organs with the exception of the military, institutional equilibrium was inefficient203. In terms of how this affected the stability of the Meiji Constitutional system, their theory argues that this institutional framework was too difficult to change through legal means204. It can be inferred that the failure of political parties to collude created structure induced equilibrium205. Like Samuels‟, Ramseyer and Rosenbluth do not directly address the crisis of the Rice Riots but their rational actor model but as with the former the latter provide enough information to infer how they might explain why this event did not destabilize the prewar political system. According to Ramseyer and Rosenbluth, political parties and voters had higher incentives to stave off the expansion of the electorate206. For voters the expansion of the electorate would dilute Diet members‟ loyalty to representing their interests as the electorate up to this point had been restricted to the wealthiest 1 and then later, 5 percent of the population207. For political parties the expansion of the electorate would mean higher costs for election campaigns as they would have to engage a larger and more diverse electorate208. Ramseyer and Rosenbluth imply that the Rice Riots played a role in convincing party politicians that universal male suffrage was inevitable209. Hence, it can be inferred that there is one reason why this crisis 203 Ramseyer and Rosenbluth, 165-166 204 Ibid 205 Ramseyer and Rosenbluth, 165-166 206 Ramseyer and Rosenbluth, 48 207 Ramseyer and Rosenbluth, 41-48 208 Ibid 209 Ramseyer and Rosenbluth, 48 - 61 - did not destabilize the Meiji Constitutional system: the leaders of political parties did not prioritize expansion of the electorate210. The logic of this explanation would assume that if the political parties had directly used these protests in order to destabilize the regime they would have had to expand the electorate before they were willing to do so. Party leaders had become accustomed to the existing institutional framework as it shaped their policy preferences for maintaining a small propertied electorate and were not willing to change or scrap the existing system. The problem with Ramseyer and Rosenbluth‟s logic as it is applied to understanding the Rice Riots is that it assumes that parties were more complicit with the Genrō during the crisis then they actually were. Their model‟s assumptions about the role of political parties in the prewar system imply that the political parties had little incentive to marshal popular support because they were satisfied with the status quo. Despite the fact that they did not directly involve themselves in protests during the Rice Riots, they definitely had an incentive to undermine the credibility of the Terauchi government and maximize their power by increasing the influence of the lower House. Indeed, various party leaders had experimented in the early days of the crisis with participation and coordination of anti-government protests211. While this did not last long the parties, especially Seiyūkai, knew that protesters would be powerful allies if Yamagata had chosen not to compromise with them on the formation of the next cabinet212. Hence Ramseyer and Rosenbluth‟s logic is insufficient as it cannot fully explain why parties did not adopt a more aggressive stance against the Terauchi government during the Rice Riots when it would have been in their interests to do so. 210 Ramseyer and Rosenbluth, 48 211 Lewis, 15-26 212 Ibid - 62 - The Institutionalist Interpretation of the Rice Riots and its Theoretical Problems According to Katō‟s theory, the Rice Riots did not destabilize the regime because state institutions were stronger than political parties213. This political crisis, while it led to the ascension of „party cabinets‟ it did not result in any institutional change because parties were too dependent upon the existing institutions and too disconnected from the masses214. Due to the institutional framework, the scope of political parties was limited as they were never mass movements and largely disconnected with the population215. This theory implies that the political parties never chose to align themselves with the anti-government protests and take on leadership roles within them because the former was removed from the latter. It is assumed that as political actors operating within the same institutional structure, party politicians had more incentive to preserve it than they did to attempt to abolish or drastically change it because they were too weak to do so. While Katō is right to point out the huge disconnect between the participants of the Rice Riots and the mainstream parties, his explanation assumes too much complicity on the part of political parties vis-à-vis the Genrō. In this way, his interpretation assumes that the interplay between different institutional elite actors was simpler than it actually was. The Rice Riots offered the real possibility for regime change as parties were determined to remove the Genrō from power and take on a hegemonic position with regards policy-making. The fact that Yamagata and Hara were able to reach a compromise and agreed on many policies should not obscure the real tension between parties and the Genrō had turned the Rice Riots from an outbreak of widespread protest into a political crisis. As mentioned above, it is clear that party leaders were using popular protest to undermine the Terauchi cabinet and depending on how Yamagata responded would have 213 Katō, 230-231 214 Ibid 215 Katō, 230-231 - 63 - considered more radical actions. That they did not go this route but did benefit from the Rice Riots reflects their complex relationship within the Meiji Constitutional system at the time. Oligarchic Policy-Making Theory’s Interpretation of the Rice Riots As demonstrated above, the existing theories that might explain the reason why the Rice Riots did not destabilize the Meiji Constitutional system all fail to capture the complex relationship between the Genrō and the mainstream political parties. It will be the purpose of this section to use Oligarchic policy-making theory to explain why political parties did not use widespread anti- government protests that occurred during the Rice Riots in order push for regime change. Oligarchic policy-making before the Rice Riots As during the late Meiji period, effective political leadership required actors to maintain strong alliances with figures in relevant policy-affecting institutions in order to minimize the threat of political deadlock which in many cases might have resulted in the collapse of the cabinet. In the two decades before the Rice Riots, we see an increasing role being played by Diet Members in general as they steadily gained more cabinet posts during this period. The average number of cabinet seats held by Diet Members was 3.4 per cabinet (out of 12 in total), whereas the average number held by Genrō in this period was just over double that at 6.9 seats per cabinet216. From 1900 to 1911, the average number of seats held by Diet Members was 3 seats per cabinet, yet it should be noted that of the four different cabinets that existed during this period two excluded Diet Members altogether217. From 1911 to 1916 we see an increase in the number of seats held by Diet 216 Steven, 112-113 217 These are the first Saionji and second Katsura cabinets. Ibid - 64 - Members in the cabinet with an average of 6.5 going to these figures218. This coincides with a decline in the number of seats held by the Genrō as the average of 7.3 from 1900 to 1911 drops to 4 per cabinet between 1911 and 1916219.The general trend towards greater involvement of Diet Members was reversed by the Terauchi cabinet which only included 1 elected official from the lower House but maintained the average number of 4 seats going to Genrō220. Hence, there was a steady rise in the number of Diet Members holding cabinet posts in the years leading up to the Rice Riots. Of the political parties in the House of Representatives, it is clear that the Seiyūkai or Association of Political Friends led the way towards party-led cabinets that would begin to emerge immediately after the Rice Riots. They had led protests against the government during the Constitutional Crisis of 1912-13 and the Rice Riots in 1918221. Yet, it is clear that political parties had a complicated symbiotic relationship with the Genrō before and after 1918. While there was debate about the degree to which parties should cooperate with the Genrō amongst party leaders, as the largest party, it is clear that the Seiyūkai ultimately benefited from compromise with said group and was not willing to sacrifice this relationship easily 222. In the years leading up to the Rice Riots the Seiyūkai led by Hara Kei had developed a close relationship with the Genrō223. Like many of his contemporaries, Hara Kei principally opposed the practice of transcendent cabinets 224 . Inspired by the British parliamentary system, Hara sought to open politics up to such a degree that 218 Steven, 112-113 219 Ibid 220 Steven, 112-113 221 Hara, Takeshi. “Taishō: An Enigmatic Emperor and his Influential Wife,” in The Emperors of Modern Japan. Ed. Ben-Ami Shillony. in the Handbook of Oriental Studies series, Vol. 14. Koninklijke Brill, 2008. 227-240. 76-79 222 Najita, Tetsuo. Hara Kei in the Politics of Compromise, 1905-1915. Harvard University Press, 1967. 74-76 223 Connors, Lesley. The Emperor’s Advisor: Saionji Kinmochi and pre-war Japanese politics. Croom Helm, 1987. See chapters 2-3; Najita {1967}, 57-52, 194-204 224 Duus {1968}, 71-77 - 65 - governments would be made up of and accountable primarily to elected officials225. Despite his ideals, Hara worked methodically to build his party into the most influential organization in the lower House of the Diet through establishing strong ties with members of the Civil Service and Genrō (see below). In the two decades preceding the Rice Riots, Hara Kei had offered his party‟s support on government policies in exchange for influence in the cabinet226. The brainchild of Itō Hirobumi, Seiyūkai was established in 1900 to serve as a means to organize support for government policy in the Imperial Diet227. The party was used partially as a tool to coordinate and utilize support from members of Itō‟s faction as well as to expand his influence into the lower House by co-opting elected officials228. In this way, Hara was in the Diet—as Katsura was in the Civil Service—a second generation political leader in the Meiji Constitutional system as he relied on Genrō connections to ascend to a high level position within a policy-affecting institution229. It was Hara‟s relationship with imperial advisor and Genrō Saionji Kinmochi that would help him lead Seiyūkai to be the most influential party in the House of Representatives. As nobility, Saionji had connections to the imperial court and by 1908 had held the positions of President of the Privy Council, President of Seiyūkai, Minister of Education and Prime Minister 230 . Hara‟s relationship with Saionji provided Seiyūkai with two unique advantages that other parties did not have at the time: control over the Home Ministry and diplomatic inroads to the Yamagata Aritomo‟s faction, the most influential Genrō at the time231. During the 1890‟s, Hara had been a bureaucrat who joined Itō‟s Seiyūkai and replaced Communications Minister Hoshi Toru when that figure 225 Duus {1968}, 71-77 226 Connors, 53-58 227 Scalapino, 179-181 228 Connors, 48-52; Duus {1968}, 10-13 229 Connors, 48-52; Duus {1968}, 10-13 230 Initially, serving as Minister of Education in one of Itō‟s third cabinet (January to June 1898), Saionji quickly became an important liaison between different policy-affecting institutions on behalf of the former. Connors, 6-10, 11-13, 24-30 231 Najita {1967}, 57-52, 194-204 - 66 - stepped down after a corruption scandal232. Hence, Hara had developed some political experience before taking on a leadership position within the party233. Yet, Saionji proved to be a valuable ally for Hara as the former secured the latter the post of Home Minister in the first Saionji cabinet in 1906234. This position, held several times by Yamagata, allowed Hara to extend the influence of Seiyūkai into the bureaucracy—through administrative reforms Hara was able to control the Police Department235 and appoint loyal followers as regional Governors236. Hara, who was to hold the position of Home Minister from 1906-1911 and 1913-1914, used this position to establish elaborate networks of support for Seiyūkai through material rewards and control over various bureaucratic positions. These bureaucrats coordinated party-approved policies of central government, dispensed contracts with state funds for building local infrastructure to party-friendly businesses and cultivated party loyalty in areas where it was weak including rigging elections for the Seiyūkai237. In addition to Hara‟s establishment of networks in the Home Ministry, Saionji‟s friendship allowed him to make considerable inroads to the Yamagata faction. This was to prove crucial to Seiyuukai‟s success as by 1912 Yamagata was simultaneously the most influential Genrō and the 232 Duus {1968}, 77-79 233 Hara Kei was not a protégé of Itō Hirobumi in the way that Katsura Tarō was of Yamagata Aritomo—in contrast to the relationship between Katsura and Yamagata, Hara was neither close to Itō nor directly supported by him. Despite this, Hara mimicked Genrō strategies in his leadership of Seiyūkai as he maintained that party‟s pre-existing relationships with figures outside the Diet such as Saionji Kinmochi and through this figure built a relationship with Yamagata. Connors, 23-30, 45-50; Duus {1968}, 77-79 234 Connors, 24-29 235 Hara took control of Japan‟s police by first replacing Ōura Kanetake, a member of the Yamagata faction who had harassed parties in election campaigns, as metropolitan police chief and purging his men from the police. Second, Hara revised the rules pertaining to chain of command so that the head of police was directly responsible to the Home Minister and selected Anraku Kendō on the condition that he would serve Hara‟s interests and only work under him. Najita {1967}, 36-37 236 Hara rewarded Seiyūkai loyal governors and fired disloyal ones. He sent strong party-advocates to places with weak party representation. He made elaborate reshufflings of governors beginning in July 1906, then in December 1906, January 1907, March and April 1908 just before the general election of May 1908. Hara also began to suspend governors if they demonstrated opposition to cabinet policies. It is clear that he was determined to check or remove Genrō influence in general and Yamagata‟s in particular on Home Ministry affairs. Najita {1967}, 37-45 237 It is well-established that Hara used these networks to secure electoral and financial support for Seiyūkai candidates as well as harass candidates of other parties during election campaigns. Najita {1967}, 63-69; Samuels, 114-115 - 67 - renowned for his distrust of party politics238. As one of the few remaining oligarchs and given his experience with institution-building, Yamagata had spent years manipulating the Civil Service, the Army, Navy, House of Peers and Privy Council239. These institutions were stacked with many individuals in top-level positions whose careers they owed to Yamagata and were often working on his behalf to help him maintain a presence in the policy-process240. Although not a Genrō himself until 1913, Saionji was a powerful asset for Hara and Seiyūkai as the former was able to use his connections with the Genrō to facilitate or personally negotiate deals between these two groups241. Through maintaining a close alliance with Saionji, Hara was able to position his party as the most favored party as by 1918 Yamagata and the Genrō found Seiyūkai to be a reliable and influential governing partner (see below). Since the early 1900‟s, Seiyūkai had been winning over a third of the seats in the House of Representatives and Hara allied with Saionji had used this electoral strength of the party to secure concessions on government policies and cabinet posts. One example of this was the party‟s support for the signing of the Portsmouth Treaty which would conclude hostilities between Japan and Russia242. Prime Minister and prominent Yamagata faction member Katsura Tarō knew that such a bill would be unpopular because the public had no knowledge that the historic victory over Russia had exhausted the Japan‟s finances and military in the process243. For this reason, Katsura and other cabinet members sought to push the drafted Portsmouth Treaty through the Diet quickly with as little opposition as possible244. Katsura‟s government was not in a weak position due to the fact that it had the support of Yamagata who in turn and through other Genrō had strong ties with 238 Najita {1967}, 7-14, 48-56 239 Banno {1992}, 100-121; Duus, 84-85; Hackett {1965}, 261-271 240 Banno {1992}, 100-121; Duus, 84-85; Hackett {1965}, 261-271 241 Connors, 11-23, 43-52 242 Najita {1967}, 27-30 243 Ibid 244 Najita {1967}, 27-30 - 68 - multiple factions and wielded strong influence over several policy-affecting institutions245. This was crucial as it provided the government with support from the House of Peers, Privy Council, Imperial Army and Navy as well as the Imperial throne on the proposed treaty246. Despite the backing of these institutions through Yamagata and other Genrō, Katsura also needed the support of the House of Representatives247. The fact that Seiyūkai had 133 seats out of 379 in the lower House required that Katsura gain their support on the bill in order to pass a majority vote248. Hara was able to use his party‟s position to gain a larger say in the policy-process by dictating the composition of the next cabinet and securing a degree of independence from Genrō- controlled policy-effecting institutions249. Yet, his attempt to reform the gun system of administration demonstrates that in order to successfully pass legislation coordination of multiple policy-affecting institutions through negotiating consensus on policy was necessary. In exchange for Seiyūkai‟s support if the Portsmouth Treaty in the House of Representatives, Hara demanded that Katsura allow for the party to take a lead role in the next cabinet with Saionji serving as Prime Minister250. He claimed that if Katsura failed to allow Seiyūkai to choose the next cabinet with a guarantee that the Genrō would restrain the Privy Council, House of Peers and themselves from interfering in the day to day affairs of cabinet that his party would lead popular demonstrations denouncing the treaty251. After consulting Itō, Katsura agreed and following popular riots in response to the signing of the Portsmouth Treaty he resigned allowing Saionji to form the next cabinet in 1906252. Serving as the new Home Minister, Hara sought to reform the administration system of regional localities 245 Najita {1967}, 7-13, 27-30, 48-56 246 Ibid 247 Najita {1967}, 7-13, 27-30, 48-56 248 Najita {1967}, 66 249 Najita {1967}, 27-30 250 Najita {1967}, 27-30, 46-55 251 Ibid 252 Najita {1967}, 27-30, 46-55 - 69 - established by Yamagata in and known as the gun system253. Hara had sought to reform the system to centralize and rationalize administration in such a way that would grant the Home Minister more direct power over local bureaucratic affairs254. While his draft of the bill passed the House of Representatives it was defeated in the House of Peers under the direction of Yamagata255. Hara would attempt to draft a version of this bill in 1921 and would get it passed successfully as by this point he would have a strong alliance with Yamagata whose support he could count on256. In the years leading up to the Rice Riots, there was an increase in reliance on Seiyūkai to secure support for policies put forth by Genrō-led cabinets257 . On the eve of Rice Riots this relationship of was as strong as ever between the Genrō and the Seiyūkai. During the so-called Taishō Political Crisis, the party along with others had led popular demonstrations against the government demanding cabinets entirely free of Genrō influence throughout the country258. Yet, the Seiyūkai gladly abandoned its hardline position once Hara was able to negotiate with the Genrō through Saionji and play a leading role in forming a cabinet led by Admiral Yamamoto Gonnohyōe259. After the Siemens Scandal which brought the Yamamoto cabinet down, it went from 209 to 105 seats in the lower House in the general election of 1915260. Though the election was tainted with widespread interference by police acting on behalf of the pro-government Dōshikai, the scandal and the party‟s actions at the end of the Taishō Political Crisis made it unpopular261. 253 Najita {1967}, 27-30, 46-55 254 Ibid 255 Najita {1967}, 27-30, 46-55 256 Ibid 257 Connors, 53-76; Najita {1967}, 27-30, 46-55 258 Najita {1967}, 167-175 259 Connors, 53-54; Najita {1967}, 167-175 260 Najita {1967}, 66, 199-204 261 Ibid - 70 - From 1916, the Seiyūkai had struck bargains with the Terauchi cabinet to support its policies in exchange for election support and a small number of cabinet seats262. This included its decision to send troops to fight in the Russian civil war and to seize domestic rice reserves for military purposes263. This party could have easily vetoed proposed by this cabinet especially that government‟s disastrous economic policies but did not264. Seiyūkai because of its majority in the lower House could have also vetoed the Teruachi cabinet‟s budget but supported it (given that Japan had sent troops to Russia the 1918 budget was a larger one as it was directly increased by these military expenditures)265. Hence, oligarchic policy-making structured the friendly relationship Seiyūkai had with the Terauchi and Genrō in the years leading up to the Rice Riots which became quite strong just before it with the party having a significant impact on policy. Oligarchic policy-making after the Rice Riots Oligarchic policy-making continued after the Rice Riots as Hara Kei increased his influence over the policy-process by becoming Prime Minister in 1918 and establishing the first cabinet which excluded Genrō. Despite the unprecedented strength of political parties at this point, the norm of oligarchic policy-making remained as Genrō figures such as Saionji and Yamagata still had networks within and power over various policy-effecting institutions including the Privy Council, the House of Peers, the Civil Service and the military266. These alliances were still required to secure the stability of one‟s cabinet as well as the passage of legislation267. An example of how Hara‟s power expanded after the Rice Riots as a result of his ties to Yamagata is his attempt to reform the electoral system. As Home Minister in the second Saionji 262 Connors 57-58; Lewis, 11-15; Najita {1967}, 203-204 263 Connors 57-58; Lewis, 11-15; Najita {1967}, 203-204 264 Connors 57-58; Lewis, 11-15; Najita {1967}, 203-204 265 Connors 57-58; Lewis, 11-15; Najita {1967}, 203-204 266 Connors, see chapters 4-5 267Connors, see chapters 4-5; Duus {1968}, 84-85 - 71 - cabinet, Hara had established a working group to examine reform of the electoral system268. Seiyūkai like other parties had high incentives to replace Japan‟s multi-member district single non- transferable electoral system because of the high costs it imposed 269 . Hara‟s study group recommended a return to single member districts established under the 1889 Election Law and he put a bill to do just that to the Diet in 1911270. It passed the lower House but was defeated in the upper. Yamagata had meticulously designed the shift to an MMD SNTV system in 1900 to weaken political parties, and his strong influence over various policy-effecting institutions271. Hence, the initial attempt to reform had failed because of a failure on the part of Hara to secure the prior support of other policy-affecting institutions, namely the House of Peers272. Between 1911 and 1918, there had been a growing pressure amongst some parties and the larger populace for universal suffrage273. After he became Prime Minister, Hara re-attempted to reform the electoral system but in such a way that would maximize Seiyūkai‟s electoral advances without enfranchising all subjects274 . Before tabling a bill that would create an electoral system based upon single member districts he discussed the matter with Yamagata and convinced him to endorse it275. In 1919 the bill to revise Japan‟s electoral system passed both the lower and upper houses of the Diet and Hara‟s cabinet remained intact—that is no attempt was made by any other institutions to break it as a veto276. It has sometimes been claimed by scholars that the period from 1922 to 1924 was a brief return to transcendental cabinets—cabinets manipulated and dominated by the Genrō. On closer 268 Ramseyer and Rosenbluth, 46-47 269 Ramseyer and Rosenbluth, 46-50 270 Ramseyer and Rosenbluth, 46-47 271 Connors, see chapters 4-5; Duus {1968}, 84-85; Ramseyer and Rosenbluth, 46-47 272 Ramseyer and Rosenbluth, 46-47 273 Ramseyer and Rosenbluth, 47-49 274 Ramseyer and Rosenbluth, 46-49 275 Ibid 276 Ramseyer and Rosenbluth, 46-49 - 72 - inspection this seems to be inaccurate due to the fact that only one cabinet of the four that existed during this period were composed of and led by non-party affiliated individuals277. Nor do we see any Genrō holding seats in any of them let alone having one of them or their followers serve as Prime Minister278. The fact that it took no fewer than three policy-affecting institutions to raise these cabinets with the aid of the House of Representatives demonstrates the strong influence that Seiyūkai had developed through said body by this point279. As the most powerful Genrō in this period, Saionji had a direct role in the negotiation of all cabinets in this period and he had become dependent upon singling out Seiyūkai as an acceptable governing partner280. While it had been involved in creating the two cabinets following Hara‟s assassination, as a governing partner the party had lost its ability to secure majority support for policy in the lower House due to internal party fracturing281. Saionji and the other Genrō feared that leaders of the other major parties were too inexperienced and less willing to compromise with other policy-effecting institutions282. In 1924 after the fall of the Kiyoura cabinet, the Kenseikai proved itself to Saionji as being capable enough to form a stable government as it secured a majority of seats in the Diet and was subsequently asked to form a government283. Hence, the so-called hiccup in party-cabinets from 1922 to 1924 can be explained by a temporary break down in relations between major parties and other policy- affecting institutions due to the temporary implosion of the Seiyūkai. 277 Connors, 77-97 278 Steven, 112-113 279 Ibid 280 Connors, 77-97 281 Ibid 282 Connors, 77-97 283 Ibid - 73 - Passive imperial leadership before and after the Rice Riots The Taishō emperor has often been singled out for his passive rule and this seems to have been the norm. From the beginning, the Taishō emperor played a passive role in national politics as legal sovereign. Initially, Yamagata‟s Genrō protégé, Katsura Tarō, appointed imperial advisor under the recommendation of the former, had manipulated and directed all imperial decisions before the Taishō Political Crisis of 1912-1913 (less than a year)284. In fact, it was Katsura‟s role as imperial advisor which had served to undermine his third and final cabinet in said crisis285. Katsura had come under a lot of criticism from both the public and political parties over his holding the positions of Lord Keeper of the Privy Seal, Grand Chamberlain and Prime Minister simultaneously because it was believed he was manipulating imperial power286. While no Prime Minister after 1913 was to simultaneously hold the position of imperial advisor, the emperor‟s failing health greatly minimized the likelihood that he would actively intervene in issuing orders, selection of appointees and policy-decisions287. From 1913 until 1921 (he was „retired‟ in 1921 by Crown Prince Hirohito), the Taishō emperor‟s seal of approval on policies seems to have issued on his behalf by his senior advisor and recognized Genrō, Saionji Kinmochi288. Bix argues that the Taishō emperor was so incapable of political and military decisions that the Genrō actually came to play a stronger role in imperial affairs than they had during the Meiji period289. Decisions made in the name of the Taishō emperor such as approving policies were always subject to the deliberation of members of the Genrō as well as members of other policy-effecting institutions they controlled like the Privy Council, Civil 284 Connors, 37-42 285 Ibid 286 Connors, 37-42 287 Connors, 37-42; Hara, Takeshi. “Taishō: An Enigmatic Emperor and his Influential Wife,” in The Emperors of Modern Japan. Ed. Ben-Ami Shillony. in the Handbook of Oriental Studies series, Vol. 14. Koninklijke Brill, 2008. 227- 240. 231-235 288 Bix, Herbert P. Hirohito and the Making of Modern Japan. HarperCollins, 2000. 40; Connors, chapter 3 289 Bix, 40 - 74 - Service and military 290 . In most cases, policies seem to have been approved among representatives of the Diet and the Genrō on the emperor‟s behalf 291 . Hence, the imperial prerogative, although officially wielded by the emperor in terms of formally approving policies and executive orders as they were documented, was in fact wielded by influential leaders and members of policy-effecting state institutions. The Taishō emperor, similar to but much more so than the Meiji sovereign, was also quite passive when it came to the rise and fall of cabinets. Unlike his predecessor, the Taishō emperor never intervened in disputes between cabinet and other institutions292. Cabinet members during this Taishō emperor‟s reign were neither appointed nor dismissed on the initiative of that figure but by Genrō in consultation with other policy-effecting institutions, primarily the Diet293. The most influential actors in during this era were the Genrō—whose influence was still decisive but increasingly undermined and checked by the increasing power of party politicians through the Diet294. By 1912, political parties had come to play a crucial role in the formation of policy and were regularly involved in cabinet making and breaking295. The Taishō political crisis had confirmed the fact that parties were too powerful to be ignored296. Following the Rice Riots, the House of Representatives and Genrō were still the chief cabinet-making and -breaking institutions with a higher propensity for other policy-effecting institutions to be involved than during the Meiji period297 290 Connors, chapter 3 291 Ibid 292 Connors, chapter 3 293 Ibid 294 Connors, chapter 3; Steven 112-113 295 Connors, chapter 3;Duus {1968}, 83-86 296 Najita {1967}, 143-160 297 Steven 112-113, 118-119 - 75 - The Rice Riots as a Balance of Power Crisis As the explosion of unprecedented nation-wide riots and popular protest at time when Seiyūkai and other political parties sought to establish all-party cabinets, the Rice Riots qualifies as a potentially destabilizing crisis for the Meiji Constitutional system. The sequence of this crisis also falls easily under the category of balance of power crisis as it never developed into a system destabilizing crisis. The crisis began with widespread popular reaction to the economic crisis resulting from the Terauchi government‟s grain policy. These riots were never directly orchestrated by any centralized organizations structures that could coordinate activities across localities298. Yet, it is clear that the riots did originate from protests that were either occurred spontaneously and were organized by local community leaders after the fact or organized loosely prior to the event by local community organizations299. Groups that were primarily involved in organizing the protests that spawned the Rice Riots were labour unions, newspapers, community-based organizations spontaneously formed prior to the event or service-based ones such as the fire department300. Initially, local branches of political parties were also involved in organizing the first wave of protests at the beginning of the crisis though they seem to have cut ties with or disguised their involvement in the actual riots301. Hence, popular movements were involved at the local level in organizing protests at least in the initial stages of the Rice Riots. Though it had its roots in a longstanding ambition of party leaders to replace the Genrō as the cabinet-making institution and play a leading role in setting policy agendas manifested itself after the onset of the economic crisis that sparked the Rice Riots. As mentioned above, the party leaders including Hara Kei collectively deliberated on their position regarding the crisis and issued 298 Lewis, 45-64, 100-113, 158-164, 219-227 299 Ibid 300 Lewis, 45-64, 100-113, 158-164, 219-227 301 Ibid - 76 - public statements that criticized the Terauchi government‟s grain policy302. Despite the fact that the party leaders had not opposed the cabinet‟s economic policies that had triggered the crisis hitherto this point, they were clearly intent on profiting from this failed policy303. As mentioned before, the strongest party which could have prevented the crisis if it had resolved to veto the policies that provoked it, either because of ignorance or shrewdness chose not to304. Yet, Seiyūkai clearly played a significant role in sealing the Terauchi cabinet‟s fate by taking a stance against it305. The parties‟ criticism of the Terauchi cabinet during the crisis was ultimately motivated by a deeper dispute that they had with the regime: namely the fact that they sought to shift the balance of power between them and the remaining Genrō. Conventional historical approaches when explaining the Rice Riots have implied or explicitly argued that it was a negotiated regime change as parties „seized‟ control of the policy making process and therefore attempted to build a new regime from within. These approaches are right to point out the convergence of elite dispute with popular discontent with the government that resulted in the appointment of the first party-created cabinet and -led cabinet under Hara Kei. Yet, they would be wrong to identify this crisis as resulting in negotiated regime change from above. The reason for this being that political parties increased their power as a result of their role in the crisis but did not actually seize complete control over the policy process. This would have entailed removing the influence of the Privy Council, Imperial Army and Navy, and Civil Service and especially the Genrō (policy-affecting institutions outside the Imperial Diet) as they could still veto policy and break cabinets306. It is clear that Hara Kei used the crisis to increase his party‟s position vis-à-vis the Genrō to such the degree that he successfully shifted the balance of power between 302 Lewis, 11-15 303 Ibid 304 Lewis, 11-15 305 Ibid 306 Steven, 118-119 - 77 - the Genrō and the Diet. Yet, this was a case of negotiated regime stabilization because the oligarchic regime itself was unaltered only the balance of power between institutional actors was changed. This change in the balance of power was significant and the product of several years of compromise with between the two sets of actors rather than of a circumvention of institutions or established norms. The fact that there was a temporary return to non-party led cabinets after Hara‟s assassination demonstrates the degree to which this power shift was dependent upon alliances between other elites rather than the marshalling of popular opinion. - 78 - V. TEST CASE TWO: ASSESSING THE POLICY-PROCESS IN THE ERA OF THE 1936 INCIDENT Understanding the Political Context of the 1936 Incident Like the Rice Riots, the 1936 Incident occurred at a time when a once dominant policy- affecting institution was gradually losing power it once had over the policy process. Unlike the Rice Riots, the 1936 Incident was not an eruption of popular sentiment against the government rather it was a failed attempt by military officers to establish a military dictatorship. Despite the differences between this event and the Rice Riots, the 1936 Incident was also significant as it also might have resulted in the revision, suspension or even replacement of the Meiji Constitutional system. As was the case with the Rice Riots, the attempted coup d‟état of 1936 was preceded by several years build up of political discontent among elites in policy-effecting institutions and the populace. Since the Meiji era, politicians in the Imperial Diet had periodically clashed with the Imperial Army and Navy regarding the allocation of state funds for military expenditure307. During the Taishō period, party politicians while generally supporting Japan‟s imperial foreign policy had sought to scale back military expenditure and place limits on the country‟s arsenal308. Military officers of ranks high- and low- who were concerned about geo-strategic threats in an age of mechanized total war had seen this as a betrayal of Japan‟s national defenses and especially the military‟s relative autonomy (see below). By the 1930‟s, the mismanagement of the domestic economy in the face of economic crisis by party-led cabinets and the Manchurian Incident had led to many segments of the population to side with the military and the bureaucracy over party politicians on policy issues309. The attempt by young military officers to overthrow the government was the convergence of these two trends and posed a serious threat to the Imperial state. 307 During the 1890‟s, following the Russo-Japanese War and during the Taishō Political Crisis. See Chapter 3 308 Ohara, 88-96 309 Berger, Gordon Mark. Parties Out of Power in Japan, 1931-1941. Princeton University Press, 1977. 80-86 - 79 - In the years leading up to the 1936 Incident, there was a broad consensus among the various bodies that comprised the Imperial Japanese Army high command that Japan should not only maintain but expand its presence on the Asian continent 310 . The roots of this idea had emerged out of several factors including but were not limited to the proliferation of total war planning propagated by figures like General Nagata Tetsuzan311. Since the 1920‟s, Nagata claimed that because future wars would depend on cooperation entire populations to effectively wage, that the military required more direct intervention into societal affairs 312 . This kind of political arrangement was similar to but distinct from the fascist conception of state, as it was designed to instill radical national solidarity but for purely military purposes313. By the early 1930‟s, Nagata‟s ideas combined with the belief that Japan should use its presence in Manchuria to strengthen its position geo-strategically against China and the USSR as well as to solve domestic economic crisis were widespread in and outside the military (see below)314. At this time the majority of high-ranking officials in the Imperial Army and Navy sought larger military budgets and less interference in military affairs by the civilian government315. Combined with the military elite‟s severe alienation from the policies of the party-cabinets of the 1920‟s, were radical nationalist movements comprised of both civilians and young military officers which pursued violent pressure on the state316. It is important to recognize that these groups were diverse and were, with the exception of some imperial nationalist parties in the Diet, 310 Brown, Delmer M. Nationalism in Japan: an introductory historical analysis. Russell & Russell, 1955. 187-199; Ohara, 97-110 311 Berger, 86-93 312 Nagata had been an advocate of introducing reforms that would mechanize the Imperial Army by using funds to develop modern aircraft, tank and advanced artillery (such as anti-aircraft units) divisions (Fn). He argued that the Japanese military, if it was to maintain on the cutting edge of modern warfare, would have to be prepared to fight in total war situations. According to Nagata, Japan needed to enact programs that would expand military budgets to modernize weaponry. Berger, 86-93 313 Ibid 314 Berger, 86-93 315 Ibid 316 Brown, 187-199 - 80 - loosely organized factions with a core membership numbering in several hundred at most317. Though the radical nationalist movements of the Shōwa period never developed into mass political parties such as the Italian Fascist Party or the National Socialist German Workers‟ Party, these groups began to exert a great deal of pressure on Japan‟s government and military establishments318. Many of the young officers who participated in these groups were graduates of the Not War College and had been indoctrinated with kokutai nationalist ideas319. Many of these officers also came from peasant families and experienced first hand the economic crises that had plagued the countryside during the interwar period320. Although the Imperial Rescript for Soldiers and Sailors and the law of Public Meetings had forbade the participation of military officers in partisan politics, military officers had started a trend of political terrorism directed against party and government officials 321 . By the early 1930‟s, young military officers had conducted several assassinations of government officials and planned four coup d‟états in an attempt to allow the military to gain more control over the Imperial Japanese state 322 . In this sense, there was a developing radical right in the form of military and civilian groups that used violence to put pressure on the government during this period. Throughout the prewar period there had been periodical clashes between the Diet and the military—either directly or indirectly through various Genrō—regarding the amount of revenue allocated to the latter. In the early Shōwa period, the Imperial Navy and Army had come to be relatively autonomous as had other state organs with gradual wane of the Genrō‟s power and two 317 Shillony, Ben-Ami. Revolt in Japan: The Young Officers and the February 26, 1936 Incident. Princeton University Press, 1973. 13-24 318 Brown, 187-199; Shillony, 13-24 319 Brown, 187-199, 200-205; Shillony, 56-80 320 Shillony, 3-13 321 See chapter 3 322 Shillony, 25-36 - 81 - factions of army officers began to clash with the government on foreign policy issues323. Both factions wanted to pursue imperial expansion in China and reorganize domestic administration but they differed in their long-term goals324. The Kōdōha or Imperial Way faction, wanted to build a new kind of state based upon direct imperial leadership and establishing a national socialist state325. The Tōseiha or Control Faction sought to focus on reorganizing the state less along ideological lines but rather to facilitate the modernization of the Imperial Army in the context mentioned above (see above). In addition to these two factions was a group of low- and mid-ranking officers numbering in the thousands known as Shōwa Restorationist nationalists, that sought achieve Kōdōha objectives by force326. By the early 1930‟s these three groups would coalesce against a common enemy that is party-politicians whose attempts at pursuing „liberal‟ policies were perceived by all three to be detrimental to national and army goals327. For this reason, military officers began to ignore the government‟s policy and pursue their own agendas328. The convergence of these two groups: reformist high and radical low ranking army officers could have led to establishment of a military dictatorship either from within or by force. The public was sympathetic to the rising influence of the military and due to the pre-existing legal structure329. The Japanese state‟s authoritarianism could have been easily utilized by the military to crack down on resistance had it chosen to support the radical nationalist officers‟ coup or if it had staged an internal coup. In fact, it is often claimed by literature on the 1936 Incident that it was case of the Tōseiha pursuing the latter course of action330. This coup attempt, had more members of the military high command chosen to support it, could have resulted in the circumvention of the existing 323 Berger, 86-93; Shillony, 37-55 324 Berger, 86-93; Shillony, 37-55 325 Berger, 86-93; Shillony, 37-55 326 Berger, 86-93; Shillony, 37-55 327 Berger, 86-93; Shillony, 37-55 328 Berger, 86-93; Shillony, 37-55 329 Berger, 86-93; Shillony, 25-36 330 Shillony makes this argument. See pages 209-219 - 82 - policy-process and formed or dictated the formation of a cabinet331. Even if the emperor was opposed to such actions (as he was) it is possible that he could have been silenced, coerced or even replaced by the army and his brother Chichibu332. Hence, the 1936 Incident could have resulted in regime change. The 1936 Incident and its Long Term Political Consequences On February 26th, 1936 roughly 1500 soldiers from the Shōwa Restorationist movement with connections to the Kōdōha faction of the army marched on central Tokyo and took over government buildings including the Imperial Diet and held them for three days 333 . They assassinated court officials and many of the cabinet ministers excluding the Prime Minister (who by stroke of luck was able to escape but believed to be struck down during the event)334. These officers sought to rouse the emperor to lead them in establishing a new cabinet that would enact radical reforms to reform the country along the lines of famous radical nationalist Kita Ikki‟s ideas about establishing a socialist Imperial Japan guided directly by imperial rule with a removal of the existing oligarchy335. The emperor and his advisors (those who survived) resisted the demands of the plotters and martial law declared336. Members of factions in the Imperial Army attempted to negotiate on behalf of the rebel troops and the government337. Yet, on the second day when it is clear the imperial court won‟t budge on keeping the existing cabinet in place until the crisis is over they turn against the rebel troops338. On the third day an amnesty was issued for the rebel troops 331 Shillony 120-134 332 Shillony, 95-109 333 Shillony, 135-152 334 Ibid 335 Shillony, 64-80, 110-134 336 Shillony, 95-109, 167-197 337 Shillony, 152-166 338 Shillony, 184-198 - 83 - and they were ordered back to barracks, those who resisted were struck down by other divisions of the army339. Many have argued that this event set the pretext for the Toseiha faction of the Army high command, led by Colonel Ishiwara Kanji (the man who had engineered the Manchurian Incident) to take over the government from the inside and establishing a military dictatorship340. The military‟s influence in policy-making was on the rise before this incident and did increase in the aftermath of this event as once the crisis was over the cabinet fell and Kanji himself handpicked members of subsequent cabinets 341 . Yet, the military‟s power was checked by members of other policy- affecting institutions, specifically the Civil Service and the Diet as the two cabinets following this event fell as a result of deadlock on the issue of passing proposed Army reforms342. In the years following this event the ground war with China intensified and the military did eventually succeed in passing its proposed administrative and economic reforms that were designed to turn Japan into a „national defense state‟ and later General Tōjō was able to legally ban political parties343. However, these actions were only possible because of successful bargaining with the Civil Service and the Diet to which important concessions were made before the Army could secure these policies (see below). While the balance of power shifted towards the military in the early Shōwa period and away from the House of Representatives, the former‟s attempts to positively influence the policy-process were unsuccessful until it submitted to the norm of oligarchic policy-making. 339 Ibid 340 Shillony, 209-219; Sims, 197-204 341 Sims, 197-201 342 Steven, 112-113 343 Shillony, Ben-Ami. Politics and Culture in Wartime Japan. Clarendon Press, 1981. 29-36 - 84 - The Marxist Interpretation of the 1936 Incident and its Theoretical Problems Moore argues that although Japan was never a fascist dictatorship—that is a one-party state under an absolute dictator—in the way the Italy and Germany were, it was none-the-less fascist regime following the 1936 Incident344. Moore emphasizes that National General Mobilization Bill of 1938, the dissolving of trade unions and the banning of political parties in 1940 as well as the establishment of the Imperial Rule Assistance Association (designed to coordinate domestic civil affairs during the war) attest to the development of a „fascist‟ regime345. Moore does not argue that regime change occurred as a result of this event, rather he seems to claim that the 1936 Incident led to the final evolution of Imperial Japan into a „fascist‟ state346. As with his assessment of the Rice Riots, Moore oversimplifies inaccurately depicts the political structure and assumes that it is necessarily wedded in an unholy alliance with the economic structure. The problem with Moore‟s assessment of the 1936 Incident is that he does not recognize the fact that despite the further shift in the balance of power towards the Army following this event that this institution did not completely take over the policy-process. Reformist military officers of the Imperial Way faction did rise to prominent cabinet positions after 1936 and the military was able secure larger budgets for itself and reform domestic administration (see below). Despite these successes, Moore‟s interpretation of the 1936 Incident and its aftermath overlooks the struggle between the House of Representatives and the Imperial Army during the latter‟s attempt to pass initial drafts of the National General Mobilization Law (see below). Nor does Moore recognize the fact that Imperial Japan falls outside the definition of a fascist regime as it was always an oligarchic and never became a one-party state. 344 Moore, 300-305 345 Moore, 300-305 346 Moore, 289-299, 300-305 - 85 - The Political Leadership Interpretation of the 1936 Incident and its Theoretical Problems Ironically Samuels‟ book, which highlights the role of leaders, fails to fully explain role of political leaders the so-called rise of fascism in Japan during the 1930‟s as he does not address the 1936 Incident. He explains that “Japanese fascism was animated by the institutional constructions of authoritarian oligarchs as well as by the intimidation of liberal elites and by the failure of party politicians . . . Japan‟s fascism came not from a break with the past but from a refusal to break with it.”347 Hence, according to Samuels‟ Japan‟s political structure combined with the rise of corporatist economic planning at first in Manchuria and then in Japan led to the rise of fascism348. As with his hypothetical interpretation of the Rice Riots, Samuels‟ theory assumes that Imperial Japanese state was more cohesive and unitary than it actually was. While Samuels‟ outlines the details of the authoritarian framework of the Meiji Constitutional system established by Itō Hirobumi and Yamagata Aritomo, his analysis starts to unravel in his discussion of leaders during the Taishō and Shōwa periods. Samuels is right to highlight the importance of agency and his argument that the Meiji political system paved the way for the increased military influence in politics in the 1930‟s is correct. His argument seems clear enough on first glance, that Japanese politics was conducive to fascism. Yet, at no point does he explicitly define fascism nor does he discuss the differences between the authoritarianism of the Meiji regimes (pre-constitutional and post-constitutional) as distinguished from the military/bureaucratic dictatorship that emerged in the 1930‟s. The problem here is that if the Japanese political system could so easily slide into fascism without leaders to push it in that direction then what is to say that it wasn‟t always fascist? Overall, the problem with leaders as an explanatory variable is that Samuels‟ cannot clearly identify political 347 Samuels, 122-123 348 Samuels, 140-153 - 86 - leaders who led Japan into the so-called „fascist‟ period of government349. Vaguer than Moore‟s assessment, Samuels‟ also manages to overlook the tension between the House of Representatives and the Imperial Army during the late 1930‟s. The Rational Choice Interpretation of the 1936 Incident and its Theoretical Problems Ramseyer and Rosenbluth do not specifically address the 1936 Incident but based on the logic of their argument that the oligarchs collectively created an institutional framework that allowed for military autonomy it is logical to assume that they would interpret this event as resulting from flawed institutional design. Hence, these scholars would likely argue that the reason that the 1936 Incident did not destabilize the Imperial Japanese state was because the military didn‟t need to circumvent the system to take control of it350. They would state that the military‟s power vis-à-vis the Diet gave it the power to undermine that institution‟s power and usurp it351. Thus, Ramseyer and Rosenbluth would explain the 1936 Incident as a step towards the consolidation of a military controlled state that was inevitable given the institutional framework of the Meiji Constitution. The problem with this hypothetical interpretation of the 1936 Incident is that it assumes that the military had more control over the policy-process than it did even at the height of its power (see below). In this way Ramseyer and Rosenbluth‟s analysis of the military‟s so-called gradual takeover of politics cannot fully explain how and why this entity‟s increased meddling in politics didn‟t provoke a system-destabilizing constitutional crisis. Specifically, these scholars ignore the fact that the military—which had a history of meddling in politics as they pertained to military affairs 349 In contrast to Italy and Germany, which had fascist movements with clearly identifiable leaders who paved successful paths to power in those countries, Japan had no successful fascist movement. Nor was there any one particular leader who pushed Japan away from party-led governments. 350 Ramseyer and Rosenbluth, 160-166 351 Ibid - 87 - before the 1930‟s—did not fully take hold of the political system without the help and consent of other actors. The Institutionalist Interpretation of the 1936 Incident and its Theoretical Problems Like Ramseyer and Rosenbluth, Kato proposes a structuralist argument that implies that institutional design allowed the military to gradually establish control over the policy-process though he does provide an explanation for why the 1936 Incident did not destabilize the government. Kato does not directly address the significance of the 1936 Incident but argues that in the 1930‟s and 1940‟s the political structure remained the same but that it went through a phase of „bureaucratization and militarization‟352. In this sense, Kato attributes regime stability despite the pressure of the 1936 Incident as attributable to the continuity of an institutional structure that is authoritarian at its core and could be radicalized. This seems to be an accurate assessment but the problem with Kato‟s interpretation is that he emphasizes institutional continuity but he fails to recognize the role of oligarchic policy-making that operated on top of the institutional structure. The problem with Kato‟s theory is that it cannot fully explain the complex relationship that the military had with the other institutional actors after 1936. Kato draws a distinction between interwar Germany and Japan in terms of how their political structures and trajectories were different. Despite this he fails to precisely define the difference between fascism and an oligarchic authoritarian system in which one institutional actor becomes dominant in the elite balance of power. After 1936 Japan adopted many policies that bore striking similarity to fascist policies in Italy and Germany. Yet, if one examines the ways in which these policies were enacted and their substance it is clear that the military‟s heightened influence over 352 Kato, 234-236 - 88 - policies in the late 1930‟s and early 1940‟s did not break with pre-existing norms of oligarchic policy-making that had existing since the late Meiji period. Oligarchic Policy-Making Theory’s Interpretation of the 1936 Incident Oligarchic Policy-Making before the 1936 Incident The Shōwa period is distinct from the Meiji and Taishō periods as there seems to be a higher degree of autonomy and conflict between policy-affecting institutions during compared with the previous two eras of the prewar political structure. However, as will be demonstrated below, oligarchic policy-making seems to have been the norm from 1932 until 1945. Unlike the rise of the House of Representatives in the Taishō period, the advance of one institution‟s power cannot be easily designated vis-à-vis the others in the Shōwa. But one thing is clear, bureaucrats not military officers seem to have held the most cabinet seats in this period. On average the most cabinet seats in this period went to the Civil Service followed by the House of Peers353. The proportion of cabinet seats going to these institutions does not seem to have been affected by the 1936 Incident. Before that crisis the average number of seats in a cabinet held by the House of Peers was 4.5, while the Civil Service held on average 8.5354. After the crisis the House of Peers held on average 4.4 cabinet seats whereas the Civil Service retained 7.3355. When reviewing the data on the composition of cabinets from 1932 to 1940, there appears to be more of a decline in the number of Diet Members holding cabinet posts than a dramatic rise in the number of seats going to military officers. The Imperial Army and Navy scored about the same number of seats as the House of Representatives (marginally more or less) throughout this 353 Steven, 112-113 354 Ibid 355 Steven, 112-113 - 89 - period356. Before the 1936 Incident there were two cabinets, with the average number of seats held by members of the House of Representatives at 6.5, while the average number of seats held by military officers was 4.5 ( 2 for the Imperial Army, and 2.5 for the Imperial Navy)357. After the crisis, the average number of cabinet posts going to members of the lower House declined sharply to 3.1, while the average number of seats held by military officers increased marginally to 4.6 (2.5 being held by the Imperial Army, 2 being held by the Imperial Navy)358. This is a substantial increase for the number seats going to military officers which was 2.8 in the Taishō period (1.5 going to the Imperial Army, 1.4 going to the Imperial Navy)359. Hence, the number of cabinet posts going to members of the House of Representatives vis-à-vis military officers before and after the crisis demonstrates a marked decline in the former‟s influence but less of a change in the latter‟s. It is evident from Steven‟s data that there is a greater diffusion of power between policy- affecting institutions during the Shōwa period as opposed to the Meiji or Taishō periods given the significant rise in the average number of institutions involved in raising and defeating cabinets. In the Meiji period, the average number of institutions involved in raising cabinets was 1.6 with the Genrō being involved in creating 9 out of the 11 cabinets that existed during this period360. The average number of institutions involved in breaking cabinets was 2361. The chief cabinet-breaker was the House of Representatives which broke out of the 11 cabinets that existed during this period362. The average number of policy-effecting institutions involved in building cabinets during the Taishō period was 2.8 363 . The chief cabinet maker during this period was the House of 356 Ibid 357 Steven, 112-113 358 Ibid 359 Steven, 112-113 360 Ibid 361 Steven, 112-113 362 Ibid 363 Steven, 112-113 - 90 - Representatives which played a role in raising all 16 cabinets of the Taishō period364. The average number of institutions involved in breaking cabinets during this era was 2.2, with the Genrō (broke 10 out of 16) and the House of Representatives (broke 9 out of 16) being the chief cabinet breakers365. During the Shōwa period the average number of policy-effecting institutions involved in building cabinets was 4, where as the average number of cabinet-breaking institutions was 2366. The cabinet-making institutions of this era include Imperial Army (9 times), Genrō/Jushin (8 times), the House of Representatives (7 times), the Civil Service (5 times), the Imperial Navy (3 times), the Privy Council (2 times) and the House of Peers (1 time)367. The chief cabinet-breakers during the Shōwa period were House of Representatives (8 times), the Imperial Army (5 times), the Civil Service (2 times) and Imperial Navy (2 times)368. Hence, there was more conflict between policy- affecting state institutions in the Shōwa period resulting from increased levels of institutional autonomy which correlates positively with the gradual wane of the Genrō‟s power369. This data reveals that the military‟s influence in government did increase since the Taishō period, but that it did not have the largest number of seats in cabinets from 1932 to 1940370. Therefore the rising influence of the military in the policy-process during the 1930‟s must be seen in a larger context of a trend towards greater conflict resulting from increased autonomy of institutions. While much of the conventional literature on the politics of the Shōwa period emphasize a gradual diminishing of party influence coupled with a rise in military influence, careful scrutiny of the policy-process reveals a more complex relationship between political actors. The trend towards 364 Ibid 365 Steven, 112-113 366 Ibid 367 Steven, 112-113 368 Ibid 369 Steven, 112-113 370 Ibid - 91 - institutional autonomy reflects the steady decline of the Genrō influence on the policy-process from the 1920‟s and into the 1930‟s. Although Saionji was able to hang on until the late 1930‟s, most of his colleagues had died by the Shōwa period371. As a result, policy-affecting institutions became more autonomous over time and legislative deadlock became more likely with preferences of various institutional actors becoming more acute 372 . It is clear that the military did gain an unprecedented influence over the policy-process in this period as it consistently vetoed policies by defeating cabinets beginning in 1930 with the second Wakatsuki cabinet 373 . Yet, a closer examination of the policy-process reveals that no policy-effecting institution was shut out of the policy-process and intra-state compromise was still crucial for avoiding legislative deadlock. Before the military‟s so-called takeover of the policy-process in the mid 1930‟s, compromise among multiple policy-effecting institutions was crucial for the formulation of state policies. One example of this is the decision by the Hamaguchi cabinet to agree to the limiting of Japan‟s naval armaments in the Pacific at the London Naval Conference in 1930. This was a highly contentious issue in Japanese domestic politics as it was seen by many military officers and nationalists as a violation of the military‟s prerogative or the right of Supreme Command374. In the lead up to the London Naval Conference, the U.S. and Great Britain had negotiated with Japan on joint limitations on their naval arsenals to avoid an arms race between the powers in the Pacific region as all three possessed territories there 375. Initially, it was agreed that all three powers would be able to guarantee stability and security with ratio of 10 (U.K.):10 (U.S.):7 (Japan) for naval cruisers and destroyers376. However, in the midst of the conference, the American delegates had 371 Connor, 99-100 372 Titus, 150 373 Steven, 112-113 374 Connors, 117-126; Sims, 151-153; Titus, 150-153 375 Ibid 376 Sims, 149-153 - 92 - insisted on a 10: 6 ratio vis-à-vis the U.S. to Japan377. Although a new compromise was eventually reached that was closer to the initial 10: 10: 7 ratio which the Japanese delegates preferred, the acceptance of a ratio that was below said agreement was received by hardliners in the Navy high command with resentment378. Often cited as an example of political parties‟ exercising unrestrained control over the policy-process during the first half of the interwar years, the ratification of the London Naval Treaty was the result of oligarchic policy-making. At the end of the Taishō period and on the eve of the Manchurian Incident, the decision to accept the terms of the London Naval Conference symbolized one of the last major policy decisions made by party-led cabinets. While liberal preferences of the Hamaguchi cabinet played a role in the ratification of the London Naval Treaty of 1930, if we examine the policy-process that led to this outcome it is clear that oligarchic policy-making was crucial. As mentioned above, this was a highly divisive policy stance. Members of the Navy high command, the House of Peers and Representatives as well as the Army and Privy Council were decidedly against the approval of this policy379. In this sense, the motion to approve the signing of the London Naval Treaty could have been defeated by a number of institutions. The Army minister could have resigned causing the defeat of the cabinet; though Hamaguchi‟s party enjoyed a majority in the lower house the upper house could have defeated it; the Privy Council could have rejected it on the grounds that it violated the right of Supreme Command; the Supreme War Council—a body that deliberated on behalf of the Army and Navy on all major military policies—could have rejected it; or Naval officers could have appealed directly to the throne to have the decision overturned380. 377 It was perceived by American military intelligence that only such a ratio could secure U.S. supremacy over the Japanese Imperial Navy in case of war. Sims, 149-153 378 Sims, 149-153 379 Connors, 117-126; Sims, 151-153; Titus, 150-153 380 Due to the fact that the acting Navy Minister had approved the decision the defacto veto of the cabinet through resignation was not a possibility. Connors, 117-126; Sims, 151-153; Titus, 150-153 - 93 - These threats were real and could ultimately be overcome only through the aid of Saionji who could as a senior imperial advisor and prominent Genrō figure built support for the London Naval Treaty across policy-affecting institutions381. Saionji played a crucial role in securing the approval of the London Naval Treaty as he negotiated on behalf of Hamaguchi to secure support from Konoe Fumimaro, an influential faction leader in the House of Peers and persuaded the Army minister not to resign his post382. In addition, Saionji was able to use imperial prerogative to by- pass the Supreme War Council and colluded with the pro-treaty leadership of the Navy Ministry on the issue383. Saionji and his followers at the Imperial Court blocked attempts by anti-treaty naval officers to make direct appeals to the emperor to halt the treaty ratification by refusing to allow access to the sovereign384. With the solid majorities in both Houses of the Diet, the Navy Minister, Army Minister and backed by official imperial sanction, Hamaguchi and Saionji were able to overcome resistance and delays by the Privy Council‟s sub-committee as this body attempted to reject the bill385 . Hence, the decision to ratify the London Naval Treaty of 1930 reflects the oligarchic nature of policy-making in at the end of the Taishō period and just prior to the rise of military influence in cabinets. Oligarchic policy-making after the 1936 Incident In the years leading up to and after the 1936 Incident the military, particularly the Imperial Army, was to exert an unprecedented influence in the policy-process as it was to clash with the lower House of the Imperial Diet in the struggle to direct the policy-process. Despite the fact that the military did begin to achieve more of its policy goals after the 1936 Incident—specifically the 381 Connors, 117-126; Sims, 151-153; Titus, 150-153 382 Connors, 117-126; Sims, 151-153; Titus, 150-153 383 Connors, 117-126; Sims, 151-153; Titus, 150-153 384 Connors, 117-126; Sims, 151-153; Titus, 150-153 385 Connors, 117-126; Sims, 151-153; Titus, 150-153 - 94 - building of a national defense state infrastructure—oligarchic policy-making lasted until 1945. The passage of the National General Mobilization Law legally expanded the power of the cabinet to by- pass the Diet in designated times of war and use imperial decree to coordinate the domestic economy and implement martial law386. The Imperial Army had pushed for this type of legislation for several years by the time it was implemented and it did strengthen the military‟s influence in the policy-process387. However, as the process through which and the final details of the National General Mobilization Law demonstrate, the military did not establish a dictatorship because this bill altered the Diet‟s role in but did not remove it from the policy-process (see below). Nor did this law minimize or remove the influence of other policy-affecting institutions from the policy-process (Ibid). It was the Manchurian group of the Imperial Army led by Colonel Ishiwara Kanji that initiated and drafted the first proposal of what was to be the National General Mobilization Law388. In the wake of the 1936 Incident, this group drafted long-term plans organizational strategies for Japan to be able to effectively strengthen its economy and military in preparation for a possible war with the Soviet Union and maximize its naval power in the Pacific389. Formed in conjunction with the Genrō, Civil Service and the House of Representatives, the military-led cabinet under General Kōki Hirota introduced the first draft of this reform bill390. This draft was composed by Major Sato Kenryo of the recently established Policy Unit within the Military Affairs Bureau under the supervision of Kanji391 . In order to achieve the goal of developing a comprehensive national defense state infrastructure, Kanji‟s faction sought to create extensive administrative reforms to concentrate power in the cabinet: 386 Berger, 156-157 387 Ibid 388 Berger, 93-99 389 Ibid 390 Berger, 93-99 391 Berger, 93-99; Sims 201 - 95 - The implications of these sweeping proposals were…in no way contrary to the Constitution, they implied the elimination of a large number of official posts, and the removal of administrative jurisdiction over several areas from one ministry to another. They also contained a recommendation for the creation of a new and powerful budget and policy research organ, which would supersede the individual ministries in power and prestige392 Prime Minister Hirota in conjunction with the Army and Navy Ministers presented these policy reforms to the cabinet in two separate conferences on the affects of reforms related to the central and local administrative systems393. In submitting this reform package to the non-military cabinet ministers, Hirota was prepared to accept amendments and revision on the part of the ministries that they represented394. While the Army had increasing leverage over the political process at this time, the initial draft of the National General Mobilization Law was received negatively by both the Civil Service and the Diet to such a degree that both groups demanded that it be scraped or drastically altered. Both the bureaucracy and the Diet recognized the proposed administrative changes would obviously allow the Army to circumvent the bureaucracy and the Diet through expanding the powers of the Cabinet 395 . It was this aspect that caused dissention among many Ministers, particularly those who had held seats in the Diet396 . It is unclear whether or not one of the Ministries would have pushed their cabinet representative to break the cabinet by resigning but given the radical nature of these proposed reforms they very likely could have397. Before it could get to this point a clash between Army Minister Terauchi and Diet Member Hamada Kunimatsu over this reform and the increasing influence that the Army was exerting on politics eventually led 392 Berger, 94-95 393 Berger, 93-99 394 Ibid 395 Berger, 94-99, 106-120 396 Ibid 397 Berger, 96 - 96 - to irreconcilable division of the cabinet and Hirota‟s resignation398.Though Hirota and the Army had been willing to negotiate the specifics of the proposal the radical reform proposal led to the cabinet‟s collapse in January 1937399. The Diet was to prove as decisive in bringing down the government in 1937 as it was the following year over the proposed National General Mobilization Law because the Army was unable to effectively compromise with it. The cabinet formed after the fall of the Hirota cabinet which was largely put together by the Army with the aid of the Civil Service, Genrō and House of Peers was headed by General Yūzō Hayashi400. This cabinet initially showed promise as it was able to build bridges with members of the bureaucracy in such a way that administrative reforms were toned down and a less imposing tax increase was introduced to reach out to the Diet 401 . Despite Hayashi‟s movement away from the initial reform proposal, he was too forceful in his dealings with the members of the House of Representatives to secure their support. Although there were almost no Diet members in the cabinet, General Hayashi was able to push Army‟s reform proposal through the Diet402. At the end of March the cabinet‟s budget and reform proposals had passed the Diet but this victory was undermined by Hayashi‟s unprovoked and sudden decision to prorogue it on the 31st of that month403. Hayashi‟s announcement explaining this action further alienated the Diet as he stated that he intended to hold new elections in order to “purify the Diet”404. This led to 398 Berger, 106-115 399 Ibid 400 Berger, 115-120 401 Berger, 115-120 402 When choosing cabinet ministers, Hayashi offered portfolios to members of fringe rather than larger parties and only on the condition that these individuals leave their respective party affiliations. This was done to help realize Kanji‟s goal of building a new national socialist party in the Diet to replace the existing parties and secure majority support in said institution for the government‟s reform proposals. Out of the three asked, only one of these Diet members joined Hayashi cabinet. Berger, 115-120 403 Ibid 404 Berger, 115-120 - 97 - the Hayashi government‟s collapse shortly thereafter and further delayed the implementation of the Army‟s reform agenda405. Despite the fact that Imperial Army had increased its influence on the policy-process in the past decade or so, two Tōseiha picked cabinets failed to the effectively deal with the members of the Imperial Diet and were unable to get two separate drafts of the reforms they wanted. Ultimately, the military had to appease the Imperial Diet by further compromising with it and by relying on a Diet politician to re-introduce the bill. After the defeat of two army-led cabinets, it was the cabinet of Konoe Fumimaro that was able to pass the final version of the National General Mobilization Law in 1938. In the spring of 1937 the Army had regrouped and hammered out another version of its proposed administrative reforms406. A respected and influential aristocrat with connections to the Imperial court and in the House of Peers yet at the same time sympathetic to the Army‟s national defense state reforms, Konoe made passing this set of policies a core priority407. Konoe‟s first cabinet had paved the way for the successful passage of this legislation as it was able to pass a lot of legislation that supported state war-planning policies during the intensification of the China war in 1937408. As respected statesman and charismatic leader, Konoe was able to ease the lower House of Diet into a more cooperative dynamic as he worked for army policy goals but understood the concerns of elected representatives409. In early 1938 Konoe was able to re-table the bill and reach a compromise with the House of Representatives and the House of Peers410. Konoe was able to pass the National General Mobilization Law in exchange for the addition of two clauses411. First, the addition of a clause that 405 Ibid 406 Berger, 122 407 Berger, 122 408 Berger, 136-142 409 Ibid 410 Berger, 149-161 411 Ibid - 98 - specifies certain preconditions (wartime conditions) before the law can be invoked412. Second, the addition of a clause which states that sixty per cent of the cabinet planning council which is to recommend which laws to issue by imperial decree—by-passing the Diet—be made up of sitting members of the House of Representatives413 Thus, by compromising with the Diet and allying itself with a key figure with experience in said institution, the military relied upon compromise within the context of oligarchic policy-making to enact its policy goals. Passive Imperial Leadership before and after the 1936 Incident Unlike his predecessors, Emperor Hirohito was perhaps the most active Japanese sovereign since his great grandfather , who actively instigated the sonnō joi movement in the years leading up to the Meiji Isshin. Despite the fact that Hirohito was more politically engaged than his father or grandfather in that on several occasions he clearly sought to influence policy-decisions, he was for the most part a relatively passive political actor in the prewar political system. The main reason for this was Saionji and later Lord Keep of the Privy Seal Kido Koichi‟s restriction of the emperor‟s influence on major policy negotiations414. From the early 1920‟s when Hirohito reigned for his sick father as regent, the issuing of imperial will was—as it had been with his predecessors—coordinated by court advisors and the Genrō415. In the 1920‟s, Saionji had used his influence over court and Privy Council appointments to stack top level positions with allies in these institutions416. Saionji and Koichi believed in the established norm of passive imperial rule as they 412 Berger, 149-161 413 Ibid 414 Connors, 101-102, 105, 121-122, 126, 131-132, 196, 201-208; Large, Stephen S. Emperor Hirohito and Shōwa Japan: A political biography. Routledge, 1992., 209-210; Titus, 261-272 415 Connors, 101-102, 105, 121-122, 126, 131-132, 196, 201-208; Large, 10-11, 209-210; Titus, 261-272 416 Connors, 87-89;Titus, 108-112 - 99 - continuously acted in such a way that would keep the emperor from directly influencing the policy- process417. Before and after the 1936 Incident emperor Hirohito, though historical records show that he was opposed to the increasing influence of military in politics during the 1930‟s and 40‟s. Yet, Hirohito never endorsed, condemned or worked to support or undermined policies in the public sphere. Historical evidence demonstrates that Emperor Hirohito and his advisors Makino and Saionji were consistently opposed to the rising tide of military involvement in political affairs and their long term policy goals. The emperor had opposed the Kwantung Army‟s actions in northern China from the beginning and was increasingly concerned about the negative affects that it would have upon Japan‟s diplomatic relationship with western countries418. He was in favour of the London Naval Treaty, and was opposed to the public attack on Professor Minobe Tatsukichi‟s constitutional theory by anti-liberal pundits as well as the Shōwa Restorationists who perpetrated the 1936 Incident419. Despite his moderate political convictions, Hirohito never made any public statements endorsing the London Naval Treaty or Minobe‟s constitutional theory 420 . Nor did Hirohito publicly condemn the military officers involved in the Manchurian Incident, or any of the assassinations or attempted coup d‟état before or after the 1936 Incident under pressure from his advisors to stay out of politics421. Although he was increasingly troubled by the course of Japanese military expansion and its diplomatic movement towards Nazi Germany, Hirohito was to keep his doubts and anxieties mostly to himself and out of the public sphere422. 417 Connors, 101-102, 105, 121-122, 126, 131-132, 196, 201-208; Large, 10-11, 209-210; Titus, 261-272 418 Large, 34-40 419 Large, 40-45, 57-65 420 Large, 34-40, 40-45, 57-65 421 Large, 46-55, 65-75 422 After the crisis, Hirohito was troubled by the course of military expansion on the Asian continent and seems to have been opposed to the continued escalation of conflict after the Marco Polo Incident in 1937. From 1938-1939, he had deep reservations about Japan‟s movement away from the U.S. and Great Britain towards an alliance with Nazi Germany. Large, 88-89 - 100 - Hirohito was not entirely passive during the 1936 Incident and it has been argued that his steadfast opposition and resolve to have the rebels put down helped maintain political stability in a time of crisis423. Yet, despite his resolve in this crisis, the norm of passive imperial leadership was to constrain his ability to restrain and punish the rebellious officers in particular and the higher ranking military officers who were more sympathetic to the radicals424. With his chief advisor Saionji away from Tokyo at the time, Hirohito took a firm stand against the rebels as soon as he was informed of the situation425. The emperor was opposed to the rebellion and actively used his right of supreme command to issue military orders to have the rebellion suppressed. It also appears that Hirohito refused to accept the resignation of the cabinet before the crisis was over426. It is unclear whether Kido advised him on this issue or he simply followed Hirohito‟s orders to maintain cabinet stability427. The Emperor did play a decisive role in the 1936 Incident by refusing to side with or even entertain the idea of establishing a new cabinet before the suppression of the rebels but it is unclear whether or not chief advisors had also taken this position and advised him accordingly. Yet, it is also clear that high ranking officers of the Tōseiha and Kōdōha delayed the actual movement against the rebels by a few days and that during or in the aftermath of this crisis the imperial throne refused to confront or denounce the army or the rebellious soldiers428. Indeed the very man who would take the lead role in forming the next few cabinets, Ishiwara Kanji met with contacts he had close to the rebel forces to facilitate sharing knowledge about the situation on both sides429. Kanji as well as other figures in the military were hesitant to attack their comrades as they 423 Bix, 297-305; Large, 65-75 424 Large, 65-75; Shillony {1973}, 198-207 425 Bix, 297-305; Large, 65-75; Shillony {1973}, 95-109, 111, 113, 126-128 426 Bix, 297-305; Large, 65-75; Shillony {1973}, 95-109, 111, 113, 126-128 427 None of the literature I reviewed was clear on this matter. 428 Shillony {1973}, 150-166 429 Ibid - 101 - hoped that a new cabinet could be negotiated with the emperor430. By the second day it was clear to all sides that Hirohito was not willing to compromise and so the Imperial Army offered amnesty for those soldiers who would return to barracks431. This is significant as most officers implicated in this rebellion did just that and were never punished for their actions432. In the aftermath of the 1936 Incident, Hirohito might have used his authority to denounce the military‟s increasing interference in politics but chose not to under pressure from Saionji who thought it unwise for the emperor to do such a thing433. Hence, the emperor did play an usually active role in during the 1936 Incident as he maintained political stability during this crisis and firmly opposed the coup attempt itself. However, the norm of passive imperial leadership did have an influence on the way in which the Imperial Army dealt with the rebels and in turn influenced Hirohito‟s failure to publicly address the military‟s challenge of his authority or punish army officers for it. The Shōwa Emperor had attempted on several occasions to direct the course of policy- decisions but was usually restrained by his advisors. From the beginning of his reign, Saionji had worked closely with the Genrō, both Houses of the Diet, the Privy Council and other policy-effecting institutions to varying degrees to select cabinets and negotiate the terms of major legislation434. While technically the emperor had the final veto on political decisions his prerogative was micro- managed and his assent to executive and legislative decisions were in practice prepared for him by Saionji435. This was a practice that was continued by Koichi into the war years and up until 1945436. While some scholars have made much out of Hirohito‟s role in the prewar political system and in terms of his role in military decisions, it is important to remember understand active leadership was 430 Shillony {1973}, 150-166 431 Shillony {1973}, 167-198 432 Shillony {1973}, 198-220 433 Connors, 201-208; Large, 65-75 434 Titus, 261-272 435 Connors, 101-102, 105, 121-122, 126, 131-132, 196, 201-208 436 Large 209-210; Titus, 261-272 - 102 - the exception with him not the norm 437 . As Stephen Large points out, Hirohito‟s famous confrontation of Prime Minister Tanaka in 1929 was prepared and debated rigorously and approved by his advisors before hand and was meant to warn not force the resignation of that figure438. That this was an aberration in Hirohito‟s leadership style is born out by the expressed dismay of Saionji and the emperor himself with the subsequent fall of the Tanaka cabinet and Hirohito‟s expressed intent to unconditionally support cabinet decisions439. In 1934, 1937 and 1941—various points in time when Japan‟s foreign aggression resulted in or was about to create further escalation of armed conflict—the emperor attempted to play a direct role in shaping the policy-process440. Yet, on all of these occasions Hirohito, though his advisors took stock of his preferences, was restrained by his chief advisors441. As with his predecessors, Hirohito usually approved decisions already made by his advisors. The 1936 Incident as a Balance of Power Crisis As with the Rice Riots, the 1936 Incident qualifies as a balance of power crisis as it was a convergence of high levels of popular discontent with a dispute between political elites from different policy-effecting institutions. The populace seems to have been at best indifferent to the attempted coup d‟état although it is unclear how they would have reacted had the rest of the country known what was going on or if it would have lasted more than a few days (see Shillony). Though it numbered only some 1400 members, the Young Officers‟ movement by our oligarchic policy-making theory is a popular movement. Demanding the radical restructuring of Japan, these soldiers seized government buildings and killed several prominent government officials and 437 Bix makes this argument 438 Large, 36-40 439 Ibid 440 Large 87-88, 107-108; Titus, 263-264 441 Large 87-88, 107-108; Titus, 263-264 - 103 - attempted establish a military dictatorship guided by direct imperial rule. The Young Officers had organized previous coup attempts and were a well-organized faction within the Imperial Army with direct links to mid- and high-ranking officers from the Kōdōha faction and civilian nationalist groups. Like the Tōseiha and Kōdōha army factions, like the Fleet faction of the Imperial Navy, were opposed to the liberal-leaning foreign policies of party-politicians during the 1920‟s. Members from both these groups had been involved in various activities that were designed to circumvent the Tanaka and then the Hamaguchi cabinets‟ attempt to interfere in military affairs442. Like many revisionist bureaucrats, these two factions sought to undermine what was perceived as incompetent rule by political parties in the face of financial crisis443. By the eve of the 1936 Incident the military had been playing an increasing role in cabinet-making and breaking- and officers from the Tōseiha were intent on passing sweeping reforms of the domestic economic and administrative systems444 . Officers from the Kōdōha and Tōseiha became cabinet members and started to become active in radical nationalist activities445. Yet, Kōdōha faction was most directly implicated in the coup because it was known that the Young Officers had direct links to high ranking officers from that group. On the one hand, the Tōseiha officers wanted to further expand their influence into the policy-process by having an army run cabinet but on the other hand remove all influential Kōdōha officers to consolidate their influence in the army high command. These officers under the direction of Ishiwara Kanji and others the Imperial Army sought to delay imperial orders to put down the rebellion and even attempted to negotiate the creation of an all military cabinet once the coup was underway. 442 Shillony {1973}, 37-55 443 Berger, 80-99 444 Berger, 80-99; Shillony {1973}, 37-55 445 Berger, 80-99; Shillony {1973}, 37-55 - 104 - As with the Rice Riots, many scholars who have analyzed the 1936 Incident have claimed that it resulted in negotiated regime change as the Tōseiha officers directly benefited from the crisis as they were able to use it as a pretext to purge their rivals and have a larger role in the cabinet. While the expansion of the Tōseiha faction‟s power within the Imperial Army and within the policy- process were consequences of the 1936 Incident, at no point did these officers circumvent institutional structure or prevailing norms in their ascent to power. Ishiwara Kanji and his cohorts used the crisis to establish more direct ties with Saionji and other imperial advisors as the balance of power shifted further toward the Imperial Army and away from the Imperial Diet. At no point did the Imperial Army ever seize control of the policy-process, rather they enhanced their role in it gradually and the same way that political parties had done before them, through establishing alliances with other policy-affecting institutions. Like the political parties in the 1920‟s, the military saw the most influence in politics when it was able to gain a position of dominance in the oligarchic policy-making process through building alliances with other elites based upon compromise. - 105 - CONCLUSION Prewar Japan, like many European countries during the interwar years, experienced profound political crises that potentially threatened the stability of the existing regime. In early postwar years some European countries experienced revolution in large part because of severe economic depression resulting from involvement in the First World War, specifically Russia and Germany. Unlike these two countries, Imperial Japanese government never faced the consequences of losing a major war, either in the form of dictated surrender terms stipulating regime change or in the form of widespread criticism by the populace for having lost what became an unpopular war. Despite this, the Imperial Japanese state did face unprecedented countrywide upheaval in the 1918 Rice Riots. While these riots did not lead to revolution involving the breakdown or revision of the Meiji constitution, it is entirely possible that they could have if mainstream party politicians had chosen to align themselves with popular opinion and aggressively lead it against the regime. During the 1930‟s, in a fashion strikingly similar to but distinct from Italy and Germany during the interwar years, the Meiji Constitutional system again faced the possibility of revision, suspension or replacement as military officers began to attack state officials. While there were several of these attacks and a few attempted coup d‟états prior to, the 1936 Incident was the most serious in terms of its threat to the existing political structure. Mid- and low-ranking military officers stationed in Tokyo overtook the capital and held it for a couple of days demanding that the emperor dissolve the existing cabinet and lead the way in creating a military-led state. Like the party-leaders during the Rice Riots, high ranking military officers during the 1936 Incident could have used the impending crisis to serve as a pretext for expanding their power by changing and/or subverting existing political institutions but in the end chose not to. - 106 - Assessing Oligarchic Policy-Making Theory‟s Interpretation of the Rice Riots and the 1936 Incident and the Limits of Oligarchic Policy-Making Theory Both case studies have revealed that oligarchic policy-making was the norm before and after both the Rice Riots and the 1936 Incident. Party-politicians and military officers, experienced difficulty initially with the policy-process as they seem to have underestimated the veto power of other policy-affecting institutions. Before and after the Rice Riots we see that Hara Kei was working to expand the power of the Seiyūkai within the Imperial state as he used his position as Home Minister to use the bureaucracy to facilitate the building of strong support for his party. Likewise Ishiwara Kanji sought to expand the Imperial Army‟s influence over government administration to facilitate the building of a „national defense state‟ in the 1930‟s. Both figures had attempted to push bills that they drafted through the Imperial Diet and secure its passage as official law but in both cases, their initial proposals were defeated. The reason for this in both cases seems to be the fact that they possessed at best weak relationships with the leaders of influential policy-affecting institutions who had high stakes in the proposed reform. In Kanji‟s and Hara‟s initial attempts to gain the passage of their proposed reforms neither had directly consulted with the institution that vetoed their bill---the Genrō and the House of Representatives. Hence, the first drafts of the reform of local administration and electoral rules in 1906 and 1911 proposed by Hara, and the National General Mobilization bill proposed by Kanji were vetoed. In both cases, only when actors wishing to secure a major policy reform that would strengthen their power were able to negotiate closely with influential representatives of potentially opposing policy-affecting institutions were they able to get these bills passed. It also seems that neither the Seiyūkai nor the Control Faction of the Army were willing to unconditionally risk challenging the institutional structure of the Meiji political system. Rather, both - 107 - crises were used to facilitate power shifts between various different policy-effecting institutions. Hara‟s experience and Kanji‟s also reveal the degree to which the Genrō monopoly over the Meiji Constitutional system was breaking down by the 1920‟s and 30‟s. Hara‟s use of the Rice Riots to further develop his party‟s control over the political system was calculated and was possible in large part because he had used the state to build the Seiyūkai into an electoral machine that could secure majorities in the lower House. He had positioned the party in such a way that when the time was right Hara could force more concessions from the Genrō and yet present the Seiyūkai as the government‟s only reliable governing partner in the House of Representatives. Similarly, Kanji and other Control Faction officers used the 1936 Incident as a pretext to consolidate their control over the Imperial Army and purge it of their rivals the Imperial Way Faction. Kanji and his faction, in their delayed suppression of the coup, had deliberately tested the strength of the cabinet and the imperial throne. These military officers would be able to use the attempted coup as leverage to gain more influence in the cabinet and over policy to prevent future attacks. The decline of the Genrō had rendered many policy-affecting institutions more independent than ever before and cabinets considerably less stable. Kanji and his fellow officers could indirectly threaten continued deadlock if the Army did not gain some concessions in the wake of the crisis. Hence, after the 1936 Incident other policy-affecting institutions conceded Army-led cabinets. It seems that the greatest contrast between the two cases studied is the response and activeness of the imperial throne. In the period leading up to, during and after the Rice Riots we see a consistently inactive emperor. The Taishō emperor seems to have been immobilized because of health just as much as prevailing norms and beliefs about passive imperial leadership. One might argue that his weak health exaggerates to some degree to the norm of passive imperial leadership. In contrast, Hirohito was considerably more active in his reign and especially during the 1936 Incident. Unlike his father, it is clear that Hirohito took an active interest in political affairs and - 108 - was deeply concerned about the Kwantung Army‟s plans for and illegal invasion of Manchuria. At the beginning of the Manchurian Incident Hirohito inadvertently dismissed Prime Minister Tanaka for because Hirohito was displeased with the cabinet‟s inability to reign in the Army. This had significant political consequences as it resulted in the collapse of the cabinet which only strengthened the position of the Army vis-à-vis the government. During the 1936 Incident Hirohito was also decisive in halting the coup even though the Army hesitated to put it down and high ranking officers encouraged the emperor to break the cabinet and form a new Army-led one as per the demands of the rebels. Later during the war it also appears as if Hirohito was considerably involved in the military high command in making military decisions. Hence, while Yoshihito‟s reign seems to offer a record of extreme passivity, Hirohito‟s is marked by relative passivity in political affairs with numerous exceptions and is therefore a tougher case. Ultimately oligarchic policy-making theory is limited in its explanation of the nature of prewar Japanese politics because it assumes that passive imperial leadership results solely from cultural norms. theory assumes that all passive emperors are consistent in their passivity because of cultural norms that signify the imperial throne as political symbol and source of legitimacy in national politics. It overlooks possibility that varying degrees of imperial passivity can be attributed to the relationship that the monarch has with his advisors and their interests. Factors involved in calculating the way in which the relationship between an emperor and his advisors may or may not reinforce the norm of passive political leadership is the level of training and education. For example, the Meiji emperor ascended the throne with no political training and his advisors were for the most part were also ill-equipped to make political decisions (see above). The relationship that the monarch might have with his advisors is also a potential contributing factor as mentioned above the Taishō emperor suffered from sever health problems throughout his reign to such the degree that his advisors seem to have had a paternalistic relationship with him. - 109 - A second blind spot of oligarchic leadership theory is its failure to distinguish between military and political decisions. Specifically, this involves the degree to which the emperor is involved in the military decision-making process Does a monarch qualify as a passive emperor if he is to some degree involved in the military chain of command in terms of being briefed on decisions, having the ability to issue orders and hold subordinates accountable for theirs? In the case of Imperial Japan, there are issues which might fall under the jurisdiction of the Imperial Army or Navy such as the treatment of non-combatant civilians in conquered territory but have political implications given the discretion afforded military officers in the field. During times of war if military leadership coincides with political leadership in a domestic context (or in a foreign one as the situation mentioned above would classify) it would seem that drawing clear distinctions between imperial involvement in military as opposed to political decisions could be difficult to ascertain. Given that country‟s imperial expansion during the prewar era and the atrocities that the Japanese military committed during the Second World War, the distinction between passive imperial leadership in military as opposed to political affairs is an important question. Why Imperial Japan Never Fully Democratized The issue of whether or not Imperial Japan during the 1920‟s was a democratic or democratizing regime has been debated at length by many Japan specialists. Many scholars recognize that there was „liberal‟ trend in Japan at this time with regards to aspects of foreign policy and the rise of a mass electorate and party-led cabinets as well as the development of many social welfare programs (see above). Yet, the fact remains that throughout this period sovereignty remained in the hands of elite actors operating under the rubric of imperial sovereignty. Hara Kei, the first party-leader to become Prime Minister could have passed but worked to defeat proposed - 110 - legislation for establishing universal suffrage in 1919446. When the bill was finally approved some six years later it excluded women and was coupled with a renewal of the repressive Peace Preservation Ordinance now referred to as the Peace Preservation Law and was clearly directed at suppressing radical leftist movements447. Mainstream political parties had been slowly co-opted by the Genrō since the late Meiji period and high election costs combined with years of limited suffrage laws had provided parties with little to establish close ties with the wider populace. Another major contributing factor in the weak linkage between parties and the electorate was the high level of interference in elections by state and party officials. The state and party leaders are known to have systematically manipulated the electoral system during the Meiji, Taishō and Shōwa periods. Thus, political elites consistently undermined the electoral process during the prewar period. Due to the fact that the mainstream political parties in the lower House of the Imperial Diet were able to benefit from and seize power within the imperial state through establishing alliances with other political elites they never had a high incentive to alter it. Why Imperial Japan Never Became a Genuinely Fascist State Many scholars have described Imperial Japan during the 1930‟s and 1940‟s as a fascist regime. This seems to stem from comparisons with Germany and Italy in this period. Though these countries bare a striking similarity to Imperial Japan in many respects, these two countries possessed fundamentally different institutional designs. Both Italy and Germany were full-scale democracies by 1918 which were manipulated by populist radical nationalist movements. The Italian Fascist and Nazi Parties were first able to use the low threshold proportional representative 446 Duus {1968}, 154, 156-157 447 Mitchell, Richard H. Thought Control in Prewar Japan. Cornell University press, 1976. See chapter 2 - 111 - electoral systems to gain seats in their countries‟ national legislatures. Second, these parties were able to secure institutional control by gaining control of the executive office and then consolidate that power by using constitutional provisions that granted the executive emergency powers to build a legal path to dictatorship. This process did not occur in Imperial Japan because the institutional framework under the Meiji Constitutional system did not allow for the rise of a populist leader or movement to gain control over the legislature or executive without the support of at least one other policy-affecting state institution. That is not to say that popular movements and leaders in interwar Italy and Germany did not have to gain allies within the establishment before consolidating their power, they did448. But the German and Italian political systems during the interwar years were based upon popular sovereignty and therefore guaranteed the supremacy of the elected legislature (and executive in Germany)449. In Japan, the executive was not only beholden to the legislature but also several other non-elected policy-effecting institutions that could effectively break cabinets at will. In Germany and Italy, the use of emergency powers paved the way toward fascist dictatorship450 but in Japan the cabinet was too weak to be a basis for this kind of institutional manipulation. With exception of the emperor, no one actor or group of actors could guarantee the stability of the executive. Thus, the entrenched norm of weak imperial leadership coupled with Imperial Japan‟s institutional framework made it impossible for the establishment of a dictatorship without revision, replacement or suspension of the Meiji Constitution. As conflict on the Asian continent and in the Pacific intensified there was an increasing blur between military and political decisions as the state began to micromanage various aspects of domestic affairs more closely but at no point did Japan become a military dictatorship. While 448 Paxton, 87-106 449 Paxton, 42-44, 55-67 450 Paxton, 87-106 - 112 - military officers did come to hijack foreign policy in that they engaged in continental expansion in such a way that was controlled solely by military institutions and not answerable to the central government, they did not establish full control over the policy-process. Officers within the Kwantung Army did circumvent the existing institutional framework—that is they were afforded more autonomy in their actions on the Asian continent through a semi-decentralized military command due in part to imperial sovereignty451. In this sense, a military structure that afforded commanders in the field to act first and report later combined with the fact that the military was theoretically answerable to the emperor alone did allow military officers to circumvent the central government on issues of foreign policy452. Yet, these institutions operated in a decentralized manner in practice only because passive imperial leadership was the norm. Imperial Japan‟s Stability Attributed to Norm of Oligarchic Policy-Making What is fundamentally problematic about most conventional explanations of the political trajectory of the prewar period is that there is a tendency to confuse the prominence of particular institutional actors within the system with the dominance of institutional actors over the system. Like Interwar Germany and Italy, Imperial Japan is commonly understood by many scholars to have gone through periods of democratic and dictatorial rule. Yet, this claim is unwarranted as this thesis has demonstrated that the Meiji Constitutional system in practice was an oligarchy that, though it changed over time, contained a weak executive that was easily manipulated by several policy-affecting institutions. For this reason, Imperial Japan‟s polity functioned as an oligarchy that was able to sustain potentially destabilizing political crises through compromise between representatives of various policy-affecting institutional actors. The institutional framework combined 451 Ohara, 142-150 452 Ibid - 113 - with passive imperial rule necessitated the establishment of oligarchic policy-making on major issues between policy-affecting institutions. It was this norm which allowed the Meiji state to survive well into the 20th century as it could accommodate various shifts in the intrastate balance of power as well as guarantee that no actor would radically alter the system to their advantage. Imperial Japan though it contained political groups and crises that could have transformed that polity into a fully-fledged democracy or centralized dictatorship, never ceased functioning as an oligarchic polity from 1890 to 1945. - 114 - WORKS CITED Akita, George. Foundations of Constitutional Government in Modern Japan, 1868-1900. Harvard University Press, 1967. Banno, Junji. The establishment of the Japanese constitutional system. Trans. J.A.A. Stockwin. Routledge, 1992. Banno, Junji. Democracy in Pre-War Japan Concepts of Government, 1871-1937: collected essays. Trans. Andrew Fraser. Routledge, 2001. Beasley, W. G. “Meiji political institutions,” in The Cambridge History of Japan, Volume 5: The Nineteenth Century. Ed. Marius B. Jansen. Cambridge University press, 1989. 620-673. Berghahn, Volker R. Imperial Germany, 1871-1918: Economy, Society, Culture and Politics. (2005 Edition) Begrhahn Books, 1994. Berger, Gordon Mark. Parties Out of Power in Japan, 1931-1941. Princeton University Press, 1977. Bix, Herbert P. Hirohito and the Making of Modern Japan. HarperCollins, 2000. Brown, Delmer M. Nationalism in Japan: an introductory historical analysis. Russell & Russell, 1955. Buruma, Ian. Inventing Japan: 1853-1964 in Weidenfeld & Nicolson Universal History Series. London: Weidenfeld and Nicolson, 2003. Connors, Lesley. The Emperor’s Advisor: Saionji Kinmochi and pre-war Japanese politics. Croom Helm, 1987. Duus, Peter. Party Rival and Political Change in Taishō Japan. Harvard University Press, 1968. Duus, Peter. Feudalism in Japan in Studies in World Civilization Systems Design and Implementation, Third Edition (1993). McGraw-Hill, 1976. Eisenstadt, S.N. Japanese Civilization: A Comparative View. Chicago & London: University of Chicago Press, 1996. Gluck, Carol. Japan’s Modern Myths: Ideology in the Late Meiji Period. Princeton University Press, 1985. Gordon, Andrew. Labour and imperial Democracy in Prewar Japan. University of California Press, 1991. Gordon, Andrew. A Modern History of Japan: From Tokugawa Times to the Present, Second Edition. Oxford University Press, Inc. 2009. - 115 - Hackett, Roger F. “The Meiji Leaders and Modernization: The Case of Yamagata Aritomo,” in Changing Japanese Attitudes Toward Modernization. Ed. Marius B. Jansen. Princeton University Press, 1965. 243-282. Hackett, Roger F. “Political Modernization and the Meiji Genrō,” in Political Development in Modern Japan. Ed. Robert E. Ward. Princeton University press, 1968. 65-98. Halliday, Jon. A Political History of Japanese Capitalism. Pantheon Books, 1975 Hall, John Witney. “A Modern Monarch for Modern Japan,” in Political Development in Modern Japan. Ed. Robert E. Ward. Princeton University press, 1968. 11-64. Hara, Takeshi. “Taishō: An Enigmatic Emperor and his Influential Wife,” in The Emperors of Modern Japan. Ed. Ben-Ami Shillony. in the Handbook of Oriental Studies series, Vol. 14. Koninklijke Brill, 2008. 227-240. Hara, Takeshi. “The „Great Emperor‟ Meiji,” in The Emperors of Modern Japan. Ed. Ben-Ami Shillony. in the Handbook of Oriental Studies series, Vol. 14. Koninklijke Brill, 2008. 213-226. Analysis,” in Democracy in Prewar Japan: Groundwork or Façade? Problems in Asian Civilizations Series. Ed. George O. Totten. D.C. Health and Company, 1965. 75-89. Jansen, Marius B. The Making of Modern Japan. Belknap/Harvard University Press, 2000. Kashiwazaki, Chikako. “Jus Sanguinis in Japan : The Origin of Citizenship in Comparative Perspective,” International Journal of Comparative Sociology, 1998. 278-300 Kato, Shūichi. “Taishō Democracy as the Pre-Stage for Japanese Militarism,” in Japan In Crisis: Essays on Taishō Democracy. Ed. Bernard S. Silberman and H. D. Harootunian. Princeton Universty, 1974. 217-236. Large, Stephen S. Emperor Hirohito and Shōwa Japan: A political biography. Routledge, 1992. Lee, Stephen J. Russia and the USSR, 1855-1991: Autocracy and Dictatorship. Routledge, 2006. Lewis, Michael. Rioters and Citizens: Mass Protest in Imperial Japan. University of California Press, 1990. McVeigh, Brian J. The Nature of the Japanese State: Rationality and rituality. Routledge, 1998. Mitchell, Richard H. Thought Control in Prewar Japan. Cornell University press, 1976. Mitchell, Richard H. Censorship in Imperial Japan. Princeton University Press, 1983. Mill, John Stuart. System of Logic Ratiocinative and Inductive: Being a Connected View of the Principles of Evidence and the Methods of Scientific Investigation, Volume 1, Seventh Edition. Longmans, Green, Reader and Dyer, 1868. - 116 - Moody, Peter. “Genro Rule in China and Japan: A Comparative Perspective,” in Journal of Chinese Political Science, Vol. 12, No. 1, 2007. 29-49. Moore, Barrington. Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the Modern World. Beacon Press, 1966. Najita, Tetsuo. Hara Kei in the Politics of Compromise, 1905-1915. Harvard University Press, 1967. Ohara, Masayo. Democratization and Expansion: Historical Lessons, Conetmporary Chellenges. Praeger, 2001. Paxton, Robert O. The Anatomy of Fascism. Vinatge Books, 2004. Pekkanen, Robert. Japan’s Dual Civil Society: Members Without Advocates. Stanford University Press, 2006. Pye, Lucien W. “Introduction: Political Culture and Political Development,” in Political Culture and Political Development. Ed. Lucien W. Pye and Sydney Verba. Princeton University Press, 1965. 3- 26. Pyle, Kenneth B. “Meiji conservatism,” in The Cambridge History of Japan, Volume 5: The Nineteenth Century. Ed. Marius B. Jansen. Cambridge University press, 1989. 674-720. Ramseyer, J. Mark and Rosenbluth, Frances M. The Politics of Oligarchy: Institutional Choice in Imperial Japan. New York: Cambridge University Press, 1995. Richardson, Bradley M. and Flanagan, Scott C. Politics in Japan. Little, Brown and Company, 1984. Samuels, Richard J. Machiavelli’s Children: Leaders and Their Legacies in Italy and Japan. Cornell University Press, 2003. Scalapino, Robert A. Democracy and the Party Movement in Prewar Japan: The Failure of the First Attempt. University of California Press, 1962. Shillony, Ben-Ami. “The February 26 Affair: Politics of a Military Insurrection,” in Crisis Politics in Prewar Japan: Institutional and Ideological Problems of the 1930’s. Ed. George M. Wilson. Sophia University, 1970. Shillony, Ben-Ami. Revolt in Japan: The Young Officers and the February 26, 1936 Incident. Princeton University Press, 1973. Shillony, Ben-Ami. Politics and Culture in Wartime Japan. Clarendon Press, 1981. Shillony, Ben-Ami. “Conservative Dissatisfaction with the Modern Emperors,” in The Emperors of Modern Japan. Ed. Ben-Ami Shillony. in the Handbook of Oriental Studies series, Vol. 14. Koninklijke Brill, 2008. 137-162. - 117 - Siemes, Johannes. Herman Roesler and the making of the Meiji State: An Examination of His Influence and the Founders of Modern Japan & The Complete Text of the Meiji Constitution Accompanied By His Personal Commentaries and Notes. Sophia University and Charles E. Tuttle Company, 1968. Silberman, Bernard S. “Bureaucratic Development and the Structure of Decision-Making in the Meiji Period: The Case of the Genro,” in The Journal of Asian Studies, Vol. 27, No. 1 (Nov. 1967), 81-94. Silberman, Bernard S. Cages of Reason: the Rise of the Rational State in France, Japan, The United States and Great Britain. University of Chicago Press, 1993. Sims, Richard. Japanese Political History since the Meiji Renovation 1866-2000. Palgrave, 2001. Skocpol, Theda. “A Critical Review of Barrington Moore‟s Social Origins of Dictatorship and Democracy,” Politics and Society, Vol. 4, No. 1 (1973), 1-34. Smith, Robert J. “The Cultural Context of the Japanese Political Economy,” in The Political Economy of Japan: Cultural and Social Dynamics, Vol. 3 Ed. Shumpei Kumon and Henry Rosovsky. Stanford University Press, 1992. 13-31. Spaulding, Jr., Robert M. “Japan‟s „New Bureaucrats‟, 1932-45” in Crisis Politics in Prewar Japan: Institutional and Ideological Problems of the 1930’s. Ed. George M. Wilson. Sophia University, 1970. Steven, R. P. G. “Hybrid Constitutionalism in Prewar Japan,” in Journal of Japanese Studies, Vol. 3, No. 1 (Winter, 1977), 99-133. Szpilman, Christopher. “Fascist and Quasi-Fascist Ideas in Interwar Japan, 1918-1941,” in Japan in the Fascist Era. Ed. E. Bruce Reynolds. Polygrave-MacMillan, 2004. Titus, David Anson. Palace and Politics in Prewar Japan. Columbia University Press, 1974. Umegaki, Michio. After the Restoration: The Beginning of Japan’s Modern State. New York University Press, 1988. Wilson, George M. Radical Nationalist in Japan: Kita Ikki, 1883-1937. Harvard East Asian Series 37. Harvard University press, 1969. Yamamura, Kozo. “The Decline of the Ritsuryo System: Hypotheses on Economic and Institutional Change,” in Journal of Japanese Studies, Vol. 1, No. 1 (Autumn, 1974), 3-37.