Palace and politics in prewar Japan

Palace and Politics in Prewar Japan

Studies of the East Asian Institute Columbia University

Palace and Politics •

Ill

Prewar Japan David Anson Titus

1974 Columbia University Press

New York and London

This study, prepared under the Graduate Fac u l t ies of ( :ol ulllhia University , was selected by a committee of those fa(' u l t ies 1 o l'l'l'l'ivt· one of the Clarke F. Ansley awards given ann ua l l y hy ( :ol ulllhia University Press .

Library of Congress Cataloging in Publication Data Titus, David Anson, 1934Palace and politics in prewar Japan. Bibliography: p. I. Japan-Emperors. 2. japan. Kunaisho-History. 3. Japan-Constitutional history. 4. Japan-Politics. and government-1868- I . Title. JQ1641.T55 74-6109 354'.52'0312 ISBN 0-231-03622-1

Copyright © 1974 Columbia University Press Printed in the United States of America

For Anne B. and jesse E. Titus

THE

EAST

ASIAN

INSTITUTE

OF

COLUMBIA

UNIVERSITY

The East Asia n I nstitute of Columbia University was establ ished i n 1 949 t o prepare graduate students for careers deal ing with East Asia, and to aid research and publication on East Asia duri ng the modern period . The faculty of the I nstitute are grateful to the Ford Foundation and the Rockefeller Foundation for their fi nancial assista nee . The Stud ies of the East Asian I nstitute were i naugurated i n 1 962 to bring to a wider public the results of significant new research on modern a nd contemporary East Asia .

CONTENTS

Acknowledgments

1.

Introduction

2. 3.

Political Legitimacy and Palace Autonomy The Palace Bureaucracy: Gatekeepers of the

4.

The Offices and Officers of Palace Leadership

5.

Kido Koichi and the Matrix o f Palace Politics . Palace and Politics in Prewar Japan: An

lmperial Will

6.

Interpretation Bibliography Index Studies of the East Asian Institute

IX

13 51 97 1 93 31 1 335 343 359

VII

ACKNOWLEDGMENTS

RESEARCH FOR THIS STUDY was initiated in japan from

1 965

1 963

to

under the generous support of the Foreign Area Fellowship

Program.

Through the kind introduction of the late

Professor

Koizumi Shinzo, former President of Keio University and at that time tutor to the crown prince, I was given access to palace records and

an

opportunity

to

interview

palace

officials.

Without

his

cooperation, and that of many officials of the Imperial Household Agency, this study would not have been possible. I am also grateful to Professor Wada Hideo, Professor Yoshida Yoshiaki, and their colleagues in

the

Department

of

Law at

Meiji

University,

to

Professor Maruyama Masao, and to Professor Sato lsao for their im­ measurable assistance during my stay in .Japan. Portions of this study, fortunately, received the more than helpful criticisms of the University Seminar on Modern Japan at Columbia University and the Japan seminar at Harvard University. I am also indebted to Professor Fred

I.

Greenstein and my colleagues at

Wesleyan University, to Professor Shumpei Okamoto of Temple University, and to Professor James L. Payne of Texas A & M for their comments and encouragement during the last eight years of re­ search, writing and revising. The final revision of this manuscript, completed in Japan during

197 1 -72,

was made possible by a grant from the American Council

of Learned Societies and by Wesleyan University's generous leave and sabbatical arrangements. Professors Oka Yoshitake, lwanaga

X

,\t K:\ti\\'II·IH;MF:\TS

Kenkichiro, Sato Seizaburo, and Ito Takashi of Tokyo ll nivnsit y were most helpful not only in suggesting r evi sio n s l111t .dso i n Ktlid i n g m y research for a projected study o f politi c a l lc ad l 'r s h ip and po lit i('a l change in modern Japanese politics. Amy Heinrich, Cynthia Brodhead, and .Joa n l\ lr(�11a1·y provid l'd invaluable editorial suggestions during the l a st st ;tgl's of [ll"c·parottion for press.

I

am also grateful to Michael Harwood a nd Mary I )11r;1nt for

their encouragement and for their a s s i stanct' in styl.A.T.

Cromwell, Connecticut October 1973

Palace and Politics in Prewar Japan

Studies of the East Asian Institute Columbia University

CHAPTER ON E

INTRODUCTION

ORIGINALLY THIS WAS INTENDED to be a study of the imperial insti ­ tution i n Japan after World War I I . For three reasons it became a study of the prewar imperial i nstitution, going back to the theory of imperial prerogative evolved duri ng the Restoration settlement of 1 868-8 9 . First , the internal structure of the palace today is very close to what it had been from 1 885 to 1 945 , although the emperor's constitutional role has been drastically altered . Second , there is very little information in Japanese or English about the prewar palace , its leaders , and its concrete rel ation to politics , despite t he theoretical centrality of the imperial institution in politics . Third , the moderni­ zation of japan between 1 868 a nd 1 945 was centrally ma naged ; the articulated referent for modernization-political as well as economic and social-was the imperial i nstitution. The 1 947 Constitution of Japan produced profound changes i n the formal relationship between emperor and government . Accordi ng to the prewar constitution of 1 88 9 , the emperor had " combi ned i n his being the supreme rights of rule" and was conseq uently endowed with an immense range of constitutional prerogatives in military comma nd , civil admi nistration, and legislation. In the 1 947 consti­ tution, however, the people were made sovereign a nd the emperor was divested of all prerogatives . His constitutional position is now defined as "symbol of the State and of the unity of the people . " The Prime M i nister is no longer appoi nted by the emperor; he is "designated from among the members of the Diet by a resol ution of

2

I N T R O D U CT I O N

Diet ." Cabi net ministers are no longer appoi nted by the emperor but by the Pri me M i nister. The Office of Lord Keeper of the Privy Seal , an adj u nct palace office that had l i nked the emperor and the civil side of the prewar government , was abolished in 1 94 5 , as was the Office of Chief Aide-de-Camp, an adj unct palace office that had li nked the emperor to the military command . Palace personnel were reduced in number from approximately six thousand to one thou­ sand . I mperial lands , stocks a nd bonds , a nd other sources of fi­ nancial autonomy were elimi nated . I n 1 947 the imperial house be­ came dependent on funds provided in an annual budget passed by the legislature . The prewar aristocracy , including most bra nch houses of the imperial family, was abolished . M a naged by the Im­ perial Household Ministry , the prewar aristocracy had been the " bulwark" of t he throne . The I mperial H ousehold M i nistry , which had enj oyed a n autonomous position i n relation to the prewar government , w a s repl aced by the present I mperial H ousehold Agency under t he direction of the Prime M i nister. The i nternal organization of the palace , however, has remai ned all but uncha nged . There is still a palace office charged with the conduct of Shi nto rites , despite t he legal "denationalization" of Shinto as a state religion . There are i nformal offices i n the Imperial Household Agency which link the palace of today with the prewar aristocracy . There are a Poetry B ureau, a B oard of Chamberlai ns , a B oard of the Ceremonies-j ust as in prewar times . Although the formal relationship between emperor and government has been revised , therefore , many of the symbolic and actual roles played by t he emperor may have survived , overtly or covertly. Without an under­ standing of the prewar palace those roles could not be identified , much less eval uated . Second is the problem of information . The most I have found in a single work , for example, are lists of and partial descriptions for the men who held the four lead ing palace offices from 1 88 5 to 1 945 : the I mperial Household Minister, the Grand Chamberlain, the Chief Aide-de-Camp , and the Lord Keeper of the Privy Sea l . There is no systematic analysis of their roles, or of the institutional structure created to manage the emperor's relationship to politics and society : the I mperial Household Ministry . The lack of such information and the

I N T R O D U CT I O N

3

analysis is not without reason , however. Only recently has it been possible to·subj ect the prewar imperial institution to scholarly inquiry and objective analysis. The sanctity surround ing the prewar palace vanished in 1 94 5 ; since that time documents and private papers have become increasingly available, making it feasible to analyze the role of the palace in prewar pol itics . Postwar Japanese scholarship on the role of the imperial i nsti­ tution i n prewar politics has tended , however, to divide i nto two categories : the macrosocietal and t he micropolitical . In the macro­ societal category fall such scholars and writers as Maruyama Masao, I s hida Takes h i , I noue Kiyoshi , Yokota Kisaburo , M i yazawa Toshiyoshi , Kamiyama Shigeo, Yamazaki Tansho, a nd Yamakawa Hitoshi . Although there are vast differences in both approach and in­ terpretation among them, they all attempt to relate the imperial in­ stitution to one or more aspects of the total Japanese environment : to values and ideals in Japanese society, to the social structure, to the total legal order, or to norms a nd patterns of Japanese behavior. Marxist historians, for example , q uarrel over the class structure of the prewar "emperor system" (tennosei) . The debate centers on the nature of the class base of " absolutism" (zettai shugi) and "absolute monarchy" (zettai osei); the effort is to develop a model of prewar Japanese society. The founder of the absolutism school i n Japan, ac­ cording to Professor Toyama Shigeki , is Hattori Shiso: It was M r . Hattori Shiso's achievement (i n 1 928] to define the political regime produced by t he Meij i Restoration as "absolutism" and thus to locate it i n the developmental stages of w orld history . . . . Absolutism, or rule by an absolute monarch , is that state power w hich, resti ng on an equi­ librium of two contendi ng class forces (the feudal nobility-landlord class and the bourgeoisie) , carries out its policies by utilizing the opposition between the two while at the same time representing the i nterests of both. Absolutism is produced i n the course of the transformation from a feudal to a c a p i t a l i s t s y s t e m . . . . C o n s t r u c t i n g a state cons o l i d ated u nd e r centralized authority b y means o f a monarch's despotic and autocratic power, power t hat is protected both by an enormous sta nd i ng army and by an entrenched bureaucracy, absolutism simultaneously undertakes the ac­ celeration by force of primary accumulation, using its immense state power as a spur. 1 I. Toyama Shigeki , Meiji /shin Shi, "lwanami Zensho," vol . 1 28 , 1 7th ed . (Tokyo: Iwanami Shoten, 1 962) , p. 23 .

4

I NTRODUCTION

While they quite rightly poi nt to the bureaucracy as the engine of prewar Japanese politics , scholars and writers who adopt the absol u­ tism model go very little fart her i n their analysis of political i nstitu­ tions and behavior . They are content to argue whet her Japanese absolutism was more "feudal " than " bourgeois , " or the reverse . For example, Kamiyama S higeo, perhaps the leadi ng postwar exponent of the Hattori position and the feudal basis of Japanese absolutism , asserts the primacy of the bureaucracy i n the pol itical process . B ut the most one can fi nd about the operations of the bureaucracy in his writi ngs i s : The exercise o f the nation's powers coveri ng a l l domestic and foreign policy was formally united u nder t he i mperial prerogative; i n actuality , however, a flock of leaders i n the bureaucratic structures and the factions at court which enveloped t he emperor exercised [the emperor's preroga­ tives ] . 2

W h o were the bureaucratic leaders ? What were the court factions that " enveloped the emperor" ? How were bureaucrat ic leaders and court factions related ? The second category of J apa nese scholarship, not near as voluminous as the macrosocietal , I have termed the micropolitical . Scholars and writers i n this category focus on the emperor's indi­ vidual role in politics , viewi ngs events i n relation to his " personal " actions . This approach poi nts up contrad ictions i n the emperor's be­ havior, shows his i neffectiveness, or asserts his individual responsi­ bility for pol icy fail ures . Nez u Masashi , for example, argues that the emperor's actions during the Manchurian I ncident and its aftermat h in late 1 93 1 and early 1 932 were contradictory : the emperor condemned the Army's i ndependent acts of aggression in Manchuria and at the same time sent formal " i mperial words" of praise to that Army on October 1 8 , 1 93 1 and agai n at New Years in 1 932.3 Such apparent contradicti ons ca nnot be understood , however, wit hout consideri ng the emperor's opinions on pol icy , which he expressed to those "close to the throne , " in relation to his institutional role, 2 . Kamiyama Shigeo, Tenno Sei ni kansuru Riron-leki Sho Mandai (Tokyo: Ashi Shup­ panShinSha, !956),pp. l12-13. 3. Nezu Masashi, Ten no Sho wa Ki, 2 vols. (Tokyo: Shiseido, 1961) , vol. I, "Dai Nippon Teikoku no Hokai," pp. 110-14.

INTRO DUCTION

5

which was to legiti mate policies and decisions produced by ot hers for his ratification. The micropolitical approach tends to view the em­ peror as an autonomous individual i n an i nstitutional vacuum. The macrosocietal approach, in contrast , tends to see him in a social system without i nstitutional content or empirically verified patterns of political behavior. Third , the imperial i nstitution was central to Japa nese moderniza­ tion . A n effective national government was produced duri ng the Res­ toration settlement of 1 868-89. After the Restoration settlement the vast array of imperial prerogatives set down i n the 1 8 8 9 Constitution of the Great Empire of Japan gradually became i nstitutionalized , so that by the 1 920s Japanese politics was characterized by a high degree of confused competition among the i nstitutions of imperial prerogative to monopol ize t he Imperial Will in politics-that is, to make or control national policy making. The Diet , for example, insisted on its supremacy in politics by asserting the supremacy of the emperor's prerogative of legislation. The Army, on the other hand , i nsisted on its right to make policy on the basis of the em­ peror's prerogative of supreme command . All other i nstitutions of government were equally derived from imperial prerogative-and t herefore eq u a l l y j u s t i fi e d i n t h e i r p o l i c y - m a k i ng d e m a nd s . Pa ra l l e l i ng t he i n s t i t u t i o n a l i z a t i o n of i m p e r i a l p r e r og a t i ves , moreover, political leadership passed from an oligarchy of autono­ mous Restoration leaders to coalitions of institutional el ites who were creatures , not creators , of Japan's modern political institutions . Rapid and intense socioeconomic cha nge accompanied the i nstitu­ ti onalization of imperial prerogatives a nd political elites; by the 1930s, Japan was a formidable industrial and military power, her highly literate population enj oying universal adult male suffrage ( 192 5) and bei ng served by modern mass media . All these changes-political, economic, and social-took place a round a si ngle, fixed referent : the imperial i nstitution. As the pole st a r around which politics a nd society revolved , the imperial insti­ l ut ion w as looked to by most as the one fixed poi nt i n a vortex of ('hange. How did it mai ntai n its position i n the face of such drastic l'hanges? How could it survive the profound challenges of national d efeat in 1 945 and subseq uent occupation by a foreign power ? Only

6

I N T R O Dt: C T I O N

an exami nation of the theory and practice of palace politics and of the structure and leadership of the palace bureaucracy in prewar Japan could begi n to answer the first question and to suggest answers to the second . My concerns here are first with the theoretical position of the im­ perial institution i n the governi ng process, second with the evol ution of the palace as a formal structure and as personnel between 1 88 5 and 1 94 5 , and third with the relation o f the palace to the political process , primarily during the 1 930s . The theory of political legitimacy produced duri ng the Restoration settlement of 1 868-89 was based on the concept of imperial pre­ rogative to be discussed in chapter 2 . I n that concept were fused modified religious myths and court practices from Japan's ancient past , shogunal powers from Japan's i mmed iate past , and selected adaptations from Western constitutional monarchies . In theory , all political power derived from a " sacred and i nviolable"4 emperor who had " i n herited from Our A ncestors " "the rights of sovereignty of the State . " 5 If the imperial i nstitution was to be the eternal fount of political authority , however, measures had to be taken to ensure that political failures would not reach the emperor and that the em­ peror's human frailties would not enda nger the transcendental role of the throne in the Japanese polity . Duri ng the Restoration set­ tlement , therefore , court and government were separated i nstitu­ t i o n a l l y a nd e a c h w a s g i v e n i t s ow n w r i t t e n c o n s t i t ut i o n . Government officials were made responsible to the emperor for bot h policy maki ng and policy implementation. The emperor was to ratify policies resolved upon by the officials who "ad vised and assisted " him . The emperor did not ratify policies as an i ndividual monarch but as the representative of the I mperial Will-the beq ueathed i nstructions of his ancestors " of a lineal succession unbroken for ages eternal . "6 " I mperial Will" was thus to serve as a maj or political fic­ tion for legitimizing decisions of state while protecting the emperor from individual responsibility for the decisions he ratified . I n response to the theory of i mperial prerogative a nd the i nstitu4 . Constitution of the Great Empire of japan, Article 3. 5. Ibid . , Preamble.

6 . Ibid.

I N T R O DC C T I O N

7

tiona! separation of court and government , a palace bureaucracy developed between 1 868 and 1 889 w hich was designed not only to preserve the imperial institution's transcendental authority in politics but also to ma nage that institution's role as symbol of all t hat was true, good , and beautiful in the Japanese polity. The creation and evolution of the palace bureaucracy are the subjects of chapter 3 . Manipulating the transcendental I mperial Will i n politics and managing the symbolic centrality of the imperial i nstitution i n Japanese society had origi nally been seen a s complementary func­ tions . As will be seen in chapter 4, however, this was so only when the court was under strong cohesive leadership and coordi nated with equal ly strong and cohesive government leadership. S uch leadership existed in both court and government from 1 885 to 1 9 1 2 . But by the 1 930s the court had become a mosaic of bureaucratic representatives who responded in a variety of ways to contradictory external and i nternal pressures . These pressures came mainly from t he entrenched political i nstitutions that had grown to maturity under the 1 88 9 constitution. Pluralism and competition in the i nstitutions of government were reflected not only in the middle level of the palace bureaucracy but also in the palace leadership, the focus of discussion in chapter 4 . Between 1 885 and 1 9 1 0 the four leading palace officers were either dutiful Court Nobles or Restoration leaders who belonged to the charmed circle of statesmen directi ng the course of Japan's moderni­ zation. B ut by the 1 930s each of the four palace leaders was a distin­ guished burea ucrat drawn from a different component of the government bureaucracy-primarily the Foreign M i nistry , Home Mi nistry , Navy , and Army . As both palace persons and careerist representatives of bureaucratic i nterests, these palace leaders were to see to it that what the emperor ratified was the correct , appropriate , or desirable " national consensus , " despite competing demands from the multiple i nstituti ons of government . They were to protect the transcendence of the emperor as well as cope with the interests of their respective i nstitutional constituencies . Under such counter­ vailing pressures in the 1 930s , negotiating the national consensus taxed to the utmost the political capabil ities of the palace . No longer were palace leaders drawn from an i ngroup of men w ho were crea-

INTRODUCTION

tors , not creatures, of government institutions . No longer did they negot iate the imperial ratification of policies which were produced among themsel ves as Restoration comrades in arms and subj ect to personal , not i nstitutional , grievances , competitions , and policy preferences . Negotiati ng the national consensus for i mperial ratification was not , however, the responsibility of the four leadi ng palace officials alone . All or some of those who constitutionally or traditionally " advised and assisted " the emperor were i nvolved : t he four palace leaders , the Privy Councillors , the M i nisters of State , the C hiefs of the High Command, the Elder and Senior Statesmen . Between 1 88 5 and 1 9 1 2 the advisers to the throne were a small group o f like­ minded men who shifted offices freely among themselves , most notably during the period from 1 88 5 to 1 90 1 . Later, the number ad­ vising and assisti ng the emperor i ncreased , as did the institutions they represented . In the 1 93 0s the negotiation process leading to im­ perial ratification was complicated by the fact that partisan leaders i nitiated and implemented the policies which they themselves ne­ gotiated as national consensus . The outcome of the negotiation process was based on a balance of issue considerations, i nstitutional loyalties , a nd personal ambitions-all i nvolvi ng i nstitutionalized role perceptions on the part of the participants . By all criteria, however, the four palace leaders were to be the least involved i n policy i nitia­ tion a nd implementat i o n , the most conscious of t he national consensus , and the most experienced i n the arts of Japa nese domestic diplomacy . They and the reti red statesmen were to be the negotiators par excellence ; they were expected to discern the national consensus, regardless of their i nstitutional experiences , and to be able to ne­ gotiate that consensus i nto effect . Role behavior and the communication network l i nking court and government duri ng the 1 930s are the subj ects of chapter 5 . The focus of analysis is Marquis Kido Koichi , whose patterns of association re­ veal the nature of prewar Japa n's closed , or privatized , political process as well as the political roles req uired to sustai n that process . Between 1 930 and 1 945 Kido held three different palace offices : Chief Secretary to the Lord Keeper of the Privy Sea l , Director of Peerage Affairs , and Lord Keeper of the Privy Seal . Kido's behavior

I NTRODUCTION

9

i n each of these offices illustrates three roles that were basic to palace politics : the secretary , the bureaucrat , and the negotiator. The secretary was the functional eq uivalent of the feudal retai ner; it was via the loyal secretary that a political decision maker received in­ formation that was both reliable and indispensable in j udging the "trends of the times" on which he acted . The bureaucrat was an administrator w hose role i nvolved neutral or impartial adminis­ tration. The negotiator was responsible for eval uati ng the "trends of the times" and forging the consensus on which policy and personnel decisions were based . Kido's associates in his three pal ace offices also illustrate other political roles essential to privatized decision maki ng : the ratifier, who put the fi nal seal of approval on political decisions ; the policy advocate, who headed a faction within the government and whose partisan views and follower strength were key components of the "trends of the times " ; and the i nstrument of pressure-a person or group who attempted to i nfluence the political system "from below " or " from t he outs i d e " - i ncl ud i ng agitat ors , assassins , newsmen, i nterest groups , and political societies . C hapter 6 concludes with a discussion of privatized decision maki ng i n prewar Japan that attempts ( 1 ) to contrast the Japanese bureaucratic policy process and its privatization of conflict with the democratic policy process a nd its socialization of conflict , (2) to illustrate t he persistence of traditional patterns of decision making from Japan's i mmediate feudal past , and (3) to i ncorporate the role typology developed in chapter 5 i n an i ntelligible model of political decision making. Political controversy i n prewar Japan was not au­ thoritatively resolved in an open public forum but i n t he corridors of bureaucratic i nstitutions leadi ng into t he walled and moated imperial palace . Since decisions were to be presented to the public, not pub­ licly resolved , basic political communication was mai ntai ned by secretaries to leadi ng government and palace officials , negotiation took place verbally or "privately" among the principals involved , and the entire process was wrapped i n secrecy-as best it could be . Such privatization of conflict i n prewar Japan, which stands i n contrast to the socialization of conflict characteristic o f democracy i n the United States , was made both possible and desirable given the theoretical and actual role of the palace i n prewar Japanese politics .

10

INTRODUCTION

This study is bound exclusively to the structure and operations of palace politics in Japan between 1 868 a nd 1 945 . Si nce there have been no analytic descriptions of the worki ngs of the palace, much less the relation between the palace and the political process, I consider such a focus j ustified . Alt hough the concludi ng chapter suggests a model of prewar Japanese decision making that may ultimately prove useful for comparison with other political systems, the model is far from complete. The emphasis of this study is palace rather tha n government , a nd the model i s distorted b y the perspective taken. A great deal more needs to be done on faction building, factional coalitions, and political roles i n prewar Japanese politics before com­ parisons with other political systems and their development can be meani ngfully made. M ost critically needed in this respect is a study of the Japanese bureaucracy as it evolved under the theory and practice of imperial prerogative . Until very recently, Western and Japanese students of prewar Japanese politics have tended to ask w hy Japan was undemocratic a nd then to focus on the constitution, the legislature, pol itical parties , voting, and democratic values as we know them . Although the q uestion is certainly a proper one, the a nswers have been sought in the wrong sumo ring. It was the bu­ reaucracy , i nitiating and i mplementing policy , that shaped the course of successful political modernization in Japan, however undemocratic. If B ritish political development is any model of demo­ cratic development from a "traditional " society , moreover, one of the precond itions of democratic development i n national politics is the creation of a strong central bureaucracy . 7 Of course , the creation of an effective national bureaucracy is no guarantee that democracy will follow , but without it "the people" have nothing to fight agai nst­ a nd nothing to control . O nly in that society where the government did not govern and the land was " free" was there a different pattern of democratic development .8 7 . This is certainly the import of G. R. Elton's a nalysis of Tudor bureaucracy under Thomas C romwell i n t he 1 530s in his The Tudor Revolution in Government: Administrative Changes in the Reign of Hen ry VI/I (Cambridge : Cambridge University Press, 1 95 3 ) . 8 . James Young h a s analyzed with great perception a n d humor t he absence o f national government i n the United States during its foundling period i n The Washington Community, 1800-1828 (New York : Columbia University Press , 1 966) . Of course, the absence of government is no guarantee of democracy ; nor is a frontier.

INTRODUCTION

11

This is not , therefore , a comparative study . Nor is it a study of the decision-maki ng process in terms of substantive issues, alternative courses of action, or reasons for ultimate policy decisions . These aspects of the political process are referred to only as illustrations of the i nstitutional role of the palace , the structure and style of palace politics , and the political roles and behavior that li nked palace and politics in prewar Japan. Finally, I am not much concerned here either with the per­ sonalities of the Japanese emperors or with the prewar social forces , unleashed by J apan's modernization, which pressured the political system " from below "-mass movements , political parties , societies , pressure groups. Unlike the B ritish monarch, or his portrait by B ritish historians , "personality" had very little to do with the political functioning of the Japanese emperor. Those responsible for advisi ng and assisting the emperor were responsible for making policies . This they did in view of the "trends of the times" outside the palace gates . As supreme ratifier in the Japanese state , the em­ peror was to put the seal of the I mperial Will on those policies . I n doing so, he was t o ensure as best he could that a national policy de­ cision presented to him for ratification was truly a consensus among all concerned leaders on the best way to cope with the trends of the times . Whatever policy preferences the emperor might have had were know n only to those close to the throne and of very limited ef­ fect . I n the prewar decision-making process, therefore , the emperor was an i nstitution, not an autonomous personality exercising an ar­ bitrary i ndividual will in politics. If the personality and the policy preferences of the Japanese em­ peror were not very relevant to prewar politics, social forces cer­ tainly were. There are two reasons for givi ng them only the most tangential treatment here . First , this study simply had to be con­ trolled in scope . Obviously not everything relevant to Japanese political development could be encompassed . Second , I do not think we have fully understood what these new social forces were pitted agai nst politicall y . I hope that by analyzi ng the core i nstitution i n the prewar political process " from above , " this study will aid our understanding of the forces " from below . "

CHAPTER TWO

POLITICAL LEGITIMACY AND PALACE AUTONOMY

IN 1 9 1 7 THE B I O G R A PH E R of Tanaka M itsuaki , Imperial Household Minister from 1 898 to 1 909, described the imperial palace as " a region of supreme scenic beauty, distant , far from the smoke of human habitations , in the verdant hills, above the blue expanse of sea . . . beyond the government , transcending, no one ever to i nvade its sanctity . " 1 B ut it was only after twenty years of gropi ng experi ­ mentat ion, from the restoration of t he emperor in 1 8 68 to the pro­ mulgation of the Constitution of the Great Empire of Japan in 1 88 9 , that such a transcendental position for the throne was agreed upon . The new imperial i nstitution was a creative fusion of t hree elements : ( 1 ) practices , rituals, a nd an organization unique to t he traditional imperial i nstitution on which nativist ideas about an idealized a ncient polity were based ; (2) formal powers of rule previously held by the shogun a nd " returned " to the emperor in 1 867 ; and , fi nally, (3) selected legal theories of constitutional monarchy prevalent in co nti nental Europe during the second half of the ni neteenth century . Alt hough the pressures of Western thought and power made the instituti onal and theoretical position of the restored emperor by no means certai n, historical precedents weighted the direction that the 1 . Tomita Ki">jiri">,

Tanaka Sezsan-haku

(Tokyo: Seisan Shoi n , 1 917), p . 2 9 3 .

13

14

POLI TI C A L

L E G I TIMA C Y

AND

PALACE

Al'TO N O M Y

new government was to take . This was especially true of the political transcendence of the emperor . The position of the imperial i nsti­ tution during the precedi ng Tokugawa period ( 1 603- 1 867) a nd earl ier had been tra nscendental as wel l . From the twelfth century , political power had been exercised by a shogun, and only in times of political turmoil and warfare had the emperor and his court enj oyed any latitude of political maneuverability . The shogun had i nitially acquired his position by mil itary power, and his de facto power was legitimized by the important fiction of imperial appoi ntment . Suc­ cessive shogu ns of a ruling house , whether they ruled or were manipulated by persons in the shogunal house or government, were all appointed by the emperor. Each appoi ntment , like that of the founding shogun , was a ritualistic ratification of a fait accompli. During the Tokugawa period the emperor and his court resided i n the imperial palace i n Kyoto, under the cl ose control of the shogunal government headquartered i n Edo-present-day Tokyo. The em­ peror was virtually a transcendental prisoner. He and his court , however, had performed unique religious rites in the Japanese pol ity from prehistoric times . These rites and the sociocultural status of the i mperial i nstitution had p roved useful to all de facto rulers throughout the course of J apanese history . As a transcendental prisoner, therefore , the emperor lent legiti macy to the Tokugawa shogunate , and for all shoguns this was a valuable mea ns toward stable government . After a series of political failures , foreign and domestic, the shogun returned his ma ndate of rule to the emperor . The stage was thus set for the leaders of the Restoration to unite the traditional religious aut hority of the imperial inst itution with the powers previously held by the shogun. Those powers had largely involved mil itary and pol ice functions , as attested by the shogu n's full title of " barbaria n­ subduing generalissimo. " Such a unification was concretely sym­ bolized when the emperor and court moved from Kyoto to Tokyo and took up residence in Edo castle, the fortress of the defu nct sho­ gunal government . By 1 889 the emperor's aut hority had been further enhanced . The Meij i Constitution, formally know n as the Constitution of the Great

POLIT I C A L

L E G I T I MA C Y

AND

PALACE

A U TO N O MY

15

Empire of Japan, made Japan a constitutional monarchy in which the emperor was to exercise far-reachi ng prerogatives . Traditional religious rites were implicit in this constitution, which termed the emperor " sacred and i nviolable . " Shogunal powers recast as im­ perial prerogatives , as well as constitutional powers modeled most notably after the Prussian constitution, were explicit . In direct com­ mand of the armed forces both in peacetime admi nistration and i n wartime operations, the emperor appoi nted a l l generals o f t h e Army and admirals of the Navy. He also appoi nted all Mi nisters of State , i ncludi ng the Prime M i nister. He exercised the legislative power with the consent of the I mperial Diet , declared war, made peace , and concl uded treaties . He had discreti onary powers to issue ordi nances in times of emergency . Paradoxically , however, the court and the government were separated structurally and legally during the Restoration settlement . The court acq uired its ow n autonomous bureaucracy , the I mperial Household M i nistry . The imperial i nstitution was also granted its own constitution, the I mperial House Law , concurrently with the enactment of the Meij i Constitution. Court affairs , furthermore , i ncreasi ngly involved ritual practices or rites , and t he public ap­ pearr.p ces of the emperor increasi ngly tended to be rigid , formalized affinnations of an imperial presence unique to Japan's sociopol itical order. Palace autonomy and the union of court and rites were paralleled by the union of government a nd rule . The emperor and court ritualized , the government decided and executed . This division of functions was also complementary : the emperor and court le­ gitimized government decisions. Far from bei ng his personal decision-maki ng powers , therefore , the emperor's prerogatives were the source of authority for rule by ot hers-at first , by the statesmen who had engi neered the Res­ toration and created the i nstitutions of prewar Japanese government ; later, by the leaders produced by the new i nstitutions that grew to maturity under the const itution of 1 88 9 . In short , the imperial prerogatives laid dow n in the constitution became institutionalized in a government structure over which the emperor reigned but did not rule : the civil and military bureaucracies , the imperial legislature ,

16

PO L I T I C A L

L E G I T I MA C Y

AND

P A LA C E

A U TO N O MY

and the courts . The emperor was once again a transcendental prisoner, but i n a new and different political system: a constitutional monarchy based on imperial prerogative . The separation o f court a nd government , t h e union o f court and rites , and the theory of imperial prerogative are the subjects of this chapter. These three elements constituted the theory of legitimacy on which Japanese government was based from 1 889 to 1 94 5 , despite changes in Japanese society and shifting power configurations among the leaders and component structures of government .

THE SEPA RATION OF COURT AND GOVERNM ENT : PALAC E AUTONOM Y

The " return to the kingly government of ancient times" (osei fukko), on the basis of which the emperor was restored physically to the center of the Japanese political order in 1 868 , had very little concrete content . "Direct imperial rule" (shinsei), a corollary slogan, was equally devoid of practical expression. It was the work of experi­ mental nation-building efforts and political compromises during the Restoration settlement of 1 868-89 that produced i nstitutional and legal arrangements relating the emperor to the new pol itical order. Between 1 868 and the creation of the cabi net system i n 1 88 5 , Res­ toration governments were based on a union of court and palace af­ fairs with government and state affairs under a literal i nterpretation of direct imperial rule . The first Restoration government of 1 86 8 m a d e n o d i s t i nct i o n b e twee n c o u r t a n d gove r n m e n t . 2 T h e g o ve r n m e nt orga n i z a t i o n e s t a b l i s hed b y t he I n s t r u m e n t o f Government (Seitaisho) o f 2 1 April 1 868 provided for an officer t o manage court affairs . He was an official withi n a unified government and had concurrent governmental functions as wel l . 3 2 . Nezu Masashi, p. 38 3 . Sorifu , Kancho Benra n , 20 vols. (Tokyo: Okura Sho l nsatsu, Kyoku, 1 958), vol . 2, " Sorifu I I , " pp. 3, 2 9 . The Gregorian calendar was not officially adopted in Japan until January I, 1 87 3 . Dates prior to this a re listed according to t he Japa nese calendar without conversion and in the Japa nese order: day, month, year.

POI.ITICAI.

I.EGITIMACY

AND

P A LA C E

A U TO N OM Y

17

B ut in early 1 869 a "court affairs admi nistrator" (naibenji), 4 later called a "court governor" (naitei chiji), 5 was created , only to be superseded by an " i mperial household secretary" (kunai kyo ) when the I mperial Household Mi nistry came i nto being on 8 July 1 869.6 All three of these 1 869 functionaries were to devote themselves exclu­ sively to court affairs . Although court affairs were no l onger handled by an official who had other concurrent admi nistrative duties , court admi nistration remai ned an i ntegral part of government . The Im­ perial Household M i nistry , l ike other mi nistries created by the 1 86 9 decree, was an administrative division of the Civil Government (Dajo kan). 7 The heads of the ministries created in 1 869 were not directly re­ sponsible for "advising and assisting" (hohitsu) the emperor, though they were to become so in 1 88 9 .8 Accordi ng to the 1 87 1 reorganiza­ tion of the ministries , for example , the privilege of assisting the throne directly was in theory limited to the Chancellor (Dajo Daijin), M i nister of the Left (Sa Daijin), M i nister of the Right (U Daijin), a nd the Councillors (Sangi). 9 The mi nistries , i ncluding the I mperial Household M inistry , were separate offices (bunkan) headed by secretaries . 10 The I mperial Household Secretary , however, was not i ncluded among those Secretaries of M inistries , Councillors , and M i nisters const i t u t i ng t he i n ner circle of eightee n , called t he "cabinet " (naikaku), w ho decided government policies . 1 1 By 1 87 1 evidence t hus emerges t hat court and government would eventually be i nstitutionally separated . In 1 869 the affairs of court had become the excl usive duty of one offica l ; two years later that official, t he l m4. Horei Zensho (Tokyo: Okura Shii I nsatsu Kyoku, 1 869-), p. 1 43 . Hereafter l isted by title, year, and volume number when appropriate. 5. Siirifu, Kancho Benran, 2 : 2 6. 6 . " Daj iikan Tasshi , " no . 622 , 8 July 1 869, in Horei Zensho, 1 869, pp. 249-64 . This di­ rect ive created the "eight ministry system" under a Chancellor, M i nister of the Left , M inister of t he Right , and Councillors . For t he officers in this system, see Toyama Shigeki and Adachi Yoshiko, Kindai Nihon Seiji Shi Hikkei (Tokyo: lwanami Shoten, 1 96 1 ) , pp. 26-27 . 7. Horei Zensho, 1 869, pp. 249-64 ; Kyutei Kishadan, Kunai Cho (Tokyo: Hiibun Sha , 1 957), p . 8 . 8 . " Dai Nippon Teikoku Kempii, " Article 5 5 , i n Horei Zensho, 1 889, vol . 2 , p . 6 . 9. " Daj iikan Tasshi , " n o . 400, 1 0 August 1 87 1 , in Horei Zensho, 1 87 1 , p. 3 1 7 . 1 0. Ibid. 1 1 . Tanaka Siigorii, Kindai Nihon Kanryo Shi (Tokyo: Tiiyii Keizai Shimpii Sha Shuppan Bu, 1 94 1 ) , p. 99.

18

POLITI C A L

L EG I T I M A C Y

AND

P A LA C E

AUTONOMY

perial Household Secretary, was not asked to sit on the highest governing body, the informal cabinet of leading government of­ ficials. 12 Paralleli ng the trend toward the structural separation of court and government between 1 868 and 1 87 1 was the displacement from central government offices of Court Nobles , the aristocrats who had accompanied the teenaged emperor from Kyoto to Tokyo. Between January and June 1 868 , forty-three of the 1 02 Conferees (Sanyo) ap­ poi nted were Court Nobles , i ndicati ng that the i nitial Restoration government was i n fact a fusion of court and government . Between June 1 868 and August 1 86 9 , however, only three of t he twenty-two appoi ntments were Court Nobles , and none of the twenty-six Councillors appointed between August 1 869 and December 1 88 5 was a Court Noble.13 Court Nobles d i d have a tenuous hold until 1 88 5 , however. The Chancellor, when that office was filled , was Court Noble Sanj o Sanetomi ( 1 837- 9 1 ) and the Mi nisters of the Right and Left were , with one exception, imperial pri nces or Court Nobles .14 B ut the union of court and government was becomi ng a fic­ tion, structurally and i n terms of personnel , between 1 868 and 1 87 1 . The crumbli ng union between court a nd government did not go unchallenged , however. The theory and practice of direct imperial rule found a staunch proponent within the court in the person of M otoda Eifu ( 1 8 1 8- 9 1 ) . At the recommendation of O kubo Toshi­ michi ( 1 8 3 1 -7 8) , one of the leading oligarchs of the day, Motoda had entered the I mperial Household M i nistry i n 1 87 1 as I mperial Reader to the emperor.15 In September 1 873 M otoda petitioned M i nister of the Right lwakura Tomomi to enlist the support of the government's ablest leaders for direct imperial rule . M otoda argued 1 2 . The inrormal cabi net or 1 87 1 is not to be conrused with t he rormal cabinet s ystem enacted in 1 885 . The "cabi nets" or 1 8 7 1 -8 5 , being coll ections or lead i ng government officials, were or cou rse the p recursors or this system. 1 3 . W. G. Beasley , " Councillors or Samurai O rigin i n the Early Meiji Government , 1 86 8 - 1 86 9 , " Bulletin of the School of Oriental and African Studies, University of London 20 (1 957) : 94 . 1 4 . Toyama and Adachi , p . 26. 1 5 . Kaigo Tokiomi, " M otoda Eiru , " Nippon Kyo iku Sentetsu Sos ho (Tokyo: Bukyo Shoi n, 1 942), 1 9 : 2 . Until his death i n 1 89 1 Motoda was a curtained i nflue nce at court . A ded icated Conrucian, he exercised considerable i nfl uence over the political development or t he Emperor Meij i as well as t he con tent or public education i n the new Japan's rormative years .

PO LI T I C A L

L EG I T I M A C Y

AND

PALACE

AUTONOM Y

19

that "at present the sovereign's wisdom is not yet extensive and his benevolence is not yet comprehensive. " 16 Cultivating the emperor's virtue must be the responsibility of the ablest men i n the country , and not left to second- and third-rate teachers as at present . The ablest men of the day have become Prime M i nister [ Chancellor] , M inisters of the Left and Right , and Councillors, but these men rarely consult the Emperor on political matters . 1 7

To this end , M otoda desired a union of court and government . M otoda's ultimate test came with the creation of a group of court advisers , the jiho (Advisers-in-Attendance) , on August 2 9 , 1 877 . Ac­ cording to Tsuda Shigemaro, biographer of jiho Sasaki Takayuki , the office of jiho " origi nated i n a si ncere i ntent to concentrate on cultivati ng the emperor's virtue . " When Sasaki Takayuki j oined the group in 1 878 there were ten jiho. 18 B oth educators of the emperor and potential implementers of direct imperial rule, the jiho, as court officers , came i nto competition with the M inisters and Councillors , collectively the government oligarchy . Like M otoda, Sasaki Takayuki wished to have a powerful and ca­ pable political leader head the jiho, given their "weighty office" of educati ng the emperor in political affairs . 19 He and the other jiho agreed that they were simply not prepared for the task of remedying the emperor's deficiencies as Japan's central political figure. Sasaki also complai ned of the personnel surrounding the emperor at court : they had left everythi ng as it was i n Meij i 's childhood . If the em­ peror refused advice on the most trifl i ng of matters the palace people let him have his way . As far as politics was concerned , the emperor left everything to the "two Mi nisters" (presumably , t he C hancel lor a nd t he M i nister of t he Left ) . Sasaki accused t he I mperial Household Secretary, Tokudaij i Sanenori , of weakness-allowing too much power to the two Ministers and fearing the emperor's temper. Head C hamberlain Higashikuze Michitomi , Sasaki com1 6. Quoted by Donald H. Shively, " M otoda Eifu : Confucian L ecturer to the Meij i Em­ peror," in Confucianism in Action, ed . David S. Nivison and Arthur F. Wright (Sta nford : Stanford University Press , 1 959), p. 3 1 5 . 1 7 . I bid ; Shively paraphrases M otoda's argument . 1 8 . Tsuda Shigemaro, Meiji Seijo to Shin Takayuki (Tokyo : Jishiikai, 1 928), pp. 401 , 403 . 1 9. Ibid . , " Preface" p. 3, and text p. 403 .

20

P O LI T I C A L

I.F.GJT I M A C Y

AND

PALACE

AUTO N O M Y

plai ned , lacked drive and could do nothi ng to assist the emperor's virtue. Conseq uently, Sasaki and the other jiho agreed to ask O kubo Toshimichi, possibly the most powerful member of the government at the time, to head the " B ureau of jiho. " O kubo agreed to turn over the Home M i nistry to Ito Hirobumi , with Ito's consent , and then to take over the post of I mperial H ousehold Secretary . B ut on the way to the palace on May 14, 1 87 8 , apparently to accept his new office, O kubo was struck down by assassins.20 The jiho once again found themselves powerless . The aftermath of Okubo's assassi nation reveals the difficulty the jiho faced i n "advising" a politically apathetic emperor as well as the power struggle implicit in their role of "cultivati ng the emperor's virtue . " The jiho agreed among themselves that the leadi ng assassin, Shimada lchiro, spoke no smal l measure of truth when he argued that Japanese politics origi nated neither in the will of the emperor nor in the discussions of the people but in the autocratic decision maki ng of a few officials. To meet such criticism from " public opinion , " Sasaki urged that the emperor be made a more active political figure; it was their duty as jiho to reproach the emperor soundly, without regard to their personal fortunes . When they ap­ peared before the emperor, Senior jiho Sasaki led the remonstration : if the emperor did not familiarize himself with foreign and domestic trends , t he whole work of the Restoration would collapse . All of t he jiho spoke to the same effect .21 Komeda Torao, jiho of the third rank, was the most blunt : " I f in the past [ Your Majesty] had show n as much care for politics as he had passion for horsema nship , no such criticism from the public as 'politics by two or three Mi nisters' would have occurred . " 22 If "cultivating the emperor's virtue" meant making the emperor a direct participant in politics, it would also make the jiho a for­ midable group should the emperor rule i n fact . The implications of such an interpretation of di rect imperial rule were not lost on the government . When the jiho reported to the Councillors following 20. Ibid . , 2 1 . Ibid . , 2 2 . Ibid . ,

403-07 . 408 - 1 0 . 410.

pp.

pp. p.

POLIT I C A L

LEG I T I M A C Y

AND

PALACE

A UTON O M Y

21

O kubo's assassi nation, they urged that the emperor be present at meeti ngs of the Councillors and M i nisters as well as attend cabinet meeti ngs daily, and that the jiho themselves take part in such sessions .23 Chancel lor Sanj o Sanetomi replied that arrangements for the emperor's presence at such sessions would be made . The jiho, however, would not be permitted to participate, si nce this would confuse the distinction between " i nternal " (court ?) a nd "external" (government ?) . Councillor and Home Secretary Ito H i robumi was much more explicit . It was the consensus of the cabi net , he said , that it would be improper for the jiho to have knowledge of the confi­ dential plans of the admi nistration; in present-day administrative systems , imperial house (teishitsu) and cabi net (naikaku) ought to be separate . Si nce the Restoration, Ito conti nued , it had not been the practice for the I mperial Household Secretary to serve concurrently as Councillor in the government . Ito concluded that the jiho were to concern themselves strictly with the emperor's ed ucation .24 By dearly acknowledging the separation of court and government , Ito fended off an attempt by court officers to assume pol itical power and 1 hereby break the monopoly of power held by the government ol igarchs in the cabi net . The jiho conti nued to be disgruntled after their bid for participa­ t ory power was denied . They complai ned that after their office was created in 1 877 the Chamberlains had remai ned i n charge of court ceremony and the Ministers and Councillors had retai ned control of political policy . As a result the jiho were neither fish nor fowl , having al most no power to carry out their functions . M otoda and Sasaki fi nally presented the government with an ultimatum : either increase the powers of the jiho or abolish the office . The government l'hose abolition .25 Although Sasaki 's biographer weakly claims that "t hey had absol utely no desire to violate the j urisdiction of the 21 . Professor Shively is much more direct on the i ntentions of the jiho. "Only four days [ t he jiho ] formally demanded that the Emperor be given the right to m;okc political decisions and that thejih u as his advisers, be permitted to participate in cabi net dc·l i herations. The oligarchs-lwakura and Ito, in particular-were indignant" (Shively, p . IU). 24 . Tsuda Shigemaro, pp. 411-12 . 2S. Ibid . , pp. 439-4 9.

.oft
32

PO L I T I C A L

LEGITIM ACY

AND

P A LA C E

A U TO N O M Y

Shi nto rites had always been a maj or function of the emperor and his court ; but to argue that rites had been a fundamental function of the imperial i nstitution historically is not to explain the particular place of rites in modern Japan. The experimentation with Shinto in­ stitutions between 1 868 and 1 87 7 suggests that the Restoration leaders were as eclectic a nd pragmatic i n adapting japan's heritage as they were in adopti ng Western forms and ideas to buttress the new na­ tional government . After 1 868 Shinto rites became an i nstrument for national izi ng public consciousness . No longer were the rites of court conducted in a secluded imperial palace for the knowledgeable members of a literate ruling elite. The Shinto reforms of the early Res­ toration years made court rites the property of the entire nation , linking those rites with local Shinto organizations i n an attempt to mobilize national sentiment around the emperor. The new organization and management of Shinto rites were i ntended to give the emperor a popular visibility that had never before existed . I noue Kiyoshi , a Marxist historian markedly unsympat hetic to the imperial i nstitution, has described the i ntent and actions of Res­ toration leaders as follows: None other t h a n the gods were to b e t h e fi nal base o f the emperor's , and his government 's, authority . W hen people have bee n submerged too long and too continuously in lives of poverty t hey somehow come to yearn for somet h i ng absolute to rel y o n . The Emperor System u t i l i zed t h i s sentiment . And , while severely suppressi ng t h e Christian faith, which argued that all men are equal before God , the Emperor System enj oi ned the people to believe in gods who were the retai ners of the Sun Goddess . To t hat end the emperor above all had to revere the gods himself. " Rites a nd rule are one" has its origi ns here . In terms of governmental organiza­ tion as wel l , a Division of Rites was created i n the first government system of J a nuary 1 868; then, in J uly of 1 869 , the Government of Rites was placed above t he Civil Government . Further, a shrine to the Eight Gods­ the G od of Creation a nd the others-was established in the Government of Rites and then transferred i nto the palace . (These gods were not derived from the ancient people's faith in nature but were conceived for the pur­ pose of apotheosiz i ng the a ncestors of the imperial clan around the sixth and seventh centuries , w hen the imperial clan at last became chief of the clans . "God of Creation" means no more than somethi ng miraculous that gives birth to things . ) On 3 January 1 870 the Rite of the Begi nning ( Gens hi Sai] was begu n i n order to celebrate the Gods of Heaven and

PO L I T I C A L

LE G I T I M A C Y

A l\i D

PALACE

AUTON O M Y

33

Earth, the Eight Gods, and t h e I mperial Spirit o f emperors gone before . Precedi ng the Commencement of Government Affairs [Se�.fi Hajime] on 4 January [every year] , this Rite was a nati onal holiday until the end of the War .68

I noue also states that fourteen shrines for imperial rites were es­ tablished i n 1 868 . Local shrines , hierarchically arranged and staffed by priest-officials , were created throughout Japan. Certain Shinto rites centering on the i mperial l i ne were also proclaimed national fes­ tivals. I n 1 87 1 , for example, the Rite of Emperor Jimmu was made a state holiday , as was the legendary Jimmu's legendary enthronement , later to be called Empire Day (Kigensetsu) , in 1 872 . As early as 1 874 the practice of prostration before the emperor's photograph was begu n .69 One need not accept I noue's value j udgments or his implications as to the effectiveness of such efforts to mobilize national consciousness in order to agree that the nationalization of Shinto was i ntended to secure popular loyalty to a transcendental emperor. The emperor and people were to be l i nked by national Shinto rites and a national orga­ nization for managi ng those rites that extended from the emperor and rourt i nto every village and hamlet . Of course, not all those who advocated a transcendental emperor a rgued that Shinto rites be made the sole basis of the emperor's exem­ plary role in society. We have seen, for example, that M otoda Eifu's concept of direct imperial rule i nvolved maki ng the emperor a paragon of Confucian virtues . This also meant linki ng sovereign and subject in terms of filial piety , the absolute loyalty due a father from his sons . Maki ng Shinto rites one function of an autonomous court in­ t erfered little with such Confucian concepts, however. Ethnocentric Shinto rites and universalist C onfucian moral force would merely n·i nforce one another in sustai ni ng the transcendental position of the i mperial house . The compromise that separated court and government u l timately proved acceptable to both Confucianists and Shi ntoists . ( :onfronting a rising storm of political ideas blow i ng i n from the West , 1 hey recei ved a sanctuary for t he i r conservat ism w hen ritual observances were delegated to an autonomous court . 6 R . Inoue Kiyoshi , pp. 63-64 . 69. Ibid . , pp. 64 , 66 , 67 .

34

POLITIC A L

LEGITI M A C Y

AND

PALACE

A U TO N O M Y

POLITICAL AUTHORITY AND I M PERIAL PREROGATIVE

By the 1 930s Shinto rites conducted by the emperor and court were thought by some to be the very core of Japan's national polity . The political implications of the union of court and rites , so strongly sug­ gested by Professor I noue , were confirmed by Sakamaki Y oshio, an of­ ficial in the I mperial Household M i nistry from 1 9 1 8 to 1 93 6 , when he li nked Shinto rites, popular loyalty, imperial authority, and state power with fearsome logic : 1 . Man has a religious nature, w hich is to perceive the existence of gods above men and to resolve that all affairs should accord with the will of the gods. 2 . The power of the gods is operative in the common existence of a people or nation. 3. Therefore , t hat which governs a people or nation is the will of the gods. 4 . The i ncarnation of the will of the gods is the monarch. 5. The gods are eterna l , therefore the monarch-not one generation but the imperial l i ne-is also eterna l . 6 . Therefore , nation and imperial house are one a nd inseparable. 7. When strife occurs among groups or i ndividuals, all search for the j udgment of the gods. 8 . Therefore , the j udgment of the emperor, being the will of the gods, is obeyed .70

Elaborating, Sakamaki noted that " our nation has from ancient times believed i n the Way of the Gods, been proud of ours being the land of the gods, duly respected rites to the nation's ancestral gods, and revered the Shi nto shri nes . " 71 It is unquestionably the duty of all sub­ j ects to revere t he ancestral gods. The Shinto shri nes serve these deities ; they are public places and those who serve there are public of­ ficials. Supreme management of Shinto rites is vested i n the emperor , and his performance of rites is an example to all .72 I n taking the 70. Sakamaki Yoshio, Koshitsu Seido Gairon ( n . p . , n . d . ) . A copy of this work was kindly lent me by a former I mperial H ousehold M inist ry official w ho said it was one of t he few per­ sonal treasures he sal vaged from the fires during the war. The Gairon was evidently the basis for Sakamaki's Koshits u Seido Ko wa (To kyo: lwanami Shoten, 1 934), which is very s imilar in content . 7 1 . Ibid . , "Appendix , " p. 2 . 7 2 . Ibid .

PO LI T I C A L

LEGITI M A C Y

AND

P A LA C E

AUTONOMY

35

welfare of the nation as its primary principle , Shinto differs from other religions , which seek the salvation of all manki nd . To pray to the gods is to pray for the welfare of the nation. Therefore , "these Shinto services have a deep and special quality, in fact and ideally, i n that they are the divinely appoi nted work in the lifetime o f the em­ peror himself. " 73 Given the relation between rites and national welfare , " for our people to abandon reverence for the Gods of Heaven and Earth is completely impermissible. " 74 The ultimate result of uniti ng Shinto rites with the court was to make the emperor a transcendental figure embodying the will of the gods in the Japanese polity-at least for those who accepted and bel ieved in those rites . Assumi ng a more practical and less muddled pattern of thinking on the part of those who engi neered the Restoration settlement of 1 868-8 9 , how did the architects of the emperor's new role i n the Japanese polity relate rites and rule , the imperial institution and political power? Upon returning from Europe after his i nvestigation of Western constitutional monarchies in 1 883 , Ito Hirobumi i ndicated that the focus of popular loyalty would also be made the focus of state power : the imperial i nstitution. Havi ng discussed his views on a constitution for Japan with M otoda Eifu , Ito reported to the throne on September 2 3 , 1 883 in order to allay the many fears about his i ntentions re­ garding such a constitution. I n his discourse to the emperor, Ito declared that in Japan the emperor " of one line unbroken for ages eternal" exercised general control over all matters of state : "This is the national polity unparalleled throughout all nations . " 75 As the person primarily responsible for crea t i ng t he const itut i o n , I t o therefore i ntended to legitimize government power b y placi ng the em­ peror at the center of political authority . Si nce political power at this time had i n fact gravitated to the government oligarchs , court rites would thus be made to serve government rule. The relation of rites to rule eventually settled on was reveali ngly put by Ito in his explanation of the extensive imperial prerogatives granted by the draft Constitu tion of the Great Empire of Japan, 7 3 . Ibid . , p . 3 . 74. I bid . 7 5 . Shumpo Kiitsui Shiikai , 2 : 3 65 .

36

POLITICAL

LE G I T I M A C Y

AND

PALACE

A U TO N O M Y

popularly known as the Meij i Constitution .76 The draft was placed before the Privy Counci l , chaired not unnaturally by Ito hi mself, m June 1 88 8 . Ito argued : When we turn now to enacti ng a Constitution we must fi rst seek the pi­ votal axis of the nation, fi rmly establ ishing what that axis shall be . Without such an axis the Government will lose its discipl ine and the State will eventually col lapse when pol itics are entrusted to the reckless dis­ cussion of t he people . . . . In Europe , not only have the people become proficient in constitutional government si nce it first took seed ; there was also religion, and t his constituted the axis , deeply i nfusing t he popular mind . In this t he people 's hearts found unity . In our country , however, the power of t he religions is feeble ; not one can be deemed an axis of State . . . . I n our country there is only the I mperial House that can become such an axis. With this point in mi nd I placed great value on the Throne's prerogatives and endeavored as far as possible not to restrict t hem i n this draft Constitution .77

Ito thus saw i n the imperial house a native Japanese emotional equivalent to Western religion. With Shinto rites firmly lodged i n the court by 1 87 7 , the Meij i Constitution in effect grafted Western forms of monarchical prerogative onto the trunk of nationalized Shinto. I m­ plicit in Ito's reasoning was the proposition that rite makes might : rites antecedent to prerogative would unify Japanese loyalty around a Western form of state power. The Meij i Constitution was ratified by the Privy Council and pro­ claimed on February 1 1 , 1 88 9 , the anniversary of the day when the Emperor Jimmu was alleged to have taken the throne al most 2 , 5 5 0 years before . I n t hat document the emperor's moral authority was fused with pol itical authority in the form of vast imperial preroga76. Our concern here is with the implications or the Restoration settlement [or the imperial i nstitution's role in politics a nd the e mperor's personal relation to the political process, not the evolution or the Meij i Constitution . The theories and countertheories proposed up to 1 889 have been admirably covered elsewhere : see Pittau, Political Thought in Early Meiji Japan; and George Akita, Foundations of Constitutional Government in Modern japan, 7868- 7 900 (Cambridge, M ass . : Harvard University Press, 1 967) . It is sufficient to note here that the Meij i Constituti on, like palace autonomy and the nationalization or Shinto, was produced by experimentation a nd compromise-theoretical , institutiona l , and politica l . I have dwelt at length on the separation or court a nd government and t he union or court and rites simply be­ cause these critical aspects or the imperial institution's modern political development have not bee n adequately dealt with. 7 7 . Shimizu Noboru , Teikoku Kempa Seitei Kaigz, pp. 88-89, quoted in Maruyama M asao, Nihon no Shiso , 8th ed . (Tokyo: l wanami Shinsho, 1 964), pp. 29-30.

POLITICAL

LEGITI M A C Y

AND

PALACE

AUTONOM Y

37

tives . The emperor was declared "sacred and inviolable" (Article 3 ) , thereby maki ng explicit his transcendent position based on rites . He "combi ned in his being the supreme rights of rule" (Article 4) and " exercised supreme command over the Army and Navy" (Article 1 1 ) . From these two basic prerogatives flowed the many prerogatives of admi nistration , welfare , legislation, and emergency that were reserved to him by the constitution.78 This immense range of imperial prerogatives was new to the role of the emperor. With one exception in the fourteenth century , not si nce the ninth century had the emperor claimed or been argued to possess even closely equivalent powers . In 1 86 8 , however, the em­ peror had inherited the shogun's traditi onal political powers , most notably those involvi ng military command . These powers were recast as prerogatives similar to those held by authoritarian monarchs in the West . Other prerogatives , such as that of legislation, were derived solely from Western models. The emperor thus became the source of execut ive, legislative , and j udicial powers , a nd all government acts were issued i n his name . To argue that this fusion of rite and prerogative was a violation of the separation of church and state, as several scholars i n prewar Japan did , is somewhat misleading.79 Historically, there was no si ngle institution of church in pre- Restoration Japan to come in con­ flict with a si ngle i nstitution of state , in any way comparable to the struggles i n Europe . Nor was there any phil osophical issue of rendering unto Caesar that which is Caesar's and unto G od that which is God's. Lacki ng both institutional and philosophical bases for a church-state controversy , there was consequently no basis for the principle of separation of church and state . The emperor's cele­ bration of Shinto rites , moreover, resembled i n many respects the performance of high rituals of state by Western monarchs . M uch of the ceremony of the Western monarchies is of religious origi n, and some of the rituals of state still carry religious significa nce . The Queen of England remains the formal head of the C hurch of Eng7 8 . "Dai Nippon Tei koku Kemp o " , in Naikaku Kiroku Ka, 1942 , vol . 1 , pt . I, chap . 1 , sec . pp. 7 - 1 8 . 79. This argument w a s often adva nced b y t h e prewar Const itutional School o f Japanese legal scholars led by Mi nobe Tats ukichi . See Miyazawa Tos hiyoshi's review of M inobe's Chikuj1! Kemp!! Sezj.;z in Kokka Gakkai Zasshz 42 , no. 9 (September 1 92 8 ) , pp. 1 6 S -7 2 . 1,

38

POLITI C A L

LEGITIMACY

AND

PALACE

A U TO N O M Y

land . The i nvolvement of Western monarchs in such ceremonies and official positions has not generally been held to violate the distinction between church and state . Shinto was something less than a religion, something more than formal ceremony . As it was devel oped and applied to Japan's na­ tional life after 1 868, Shinto symbolized that the Japanese were Japanese , united in a national community that was peculiarly their own and somehow disti nct from and superior to other national entities . The weakness of the Japa nese religions , noted by Ito, meant that there was no i nstitutional chal lenge of any maj or import to the nationalization of Shinto. A nd the absence of a religious challenge to Shinto may have produced a more i ntense sentimental adherence to national ized Shi nto than might otherwise have occurred . Si nce the maj or figure in Shinto rites was the emperor, quasi­ religious aut hority was given to state power by Ito and the framers of the constituti on. In this respect , Sakuma Shozan's apt slogan for Japan's moderniz i ng aspirations, " Eastern ethics , Western tech­ niques , " may be applied with equal aptness to the government produced by the Restoration settlement . The basis of Japan's " im­ mutable and peerless " national polity (kokutai) was the imperial i n­ stitution, w hose spiritual and ethical authority ultimately derived from Shinto rites impl icit i n the constitution's provision that the em­ peror was " sacred and i nviolable . " 80 The constitution was explicit in stipulating a Western form of mutable state structure (seitai) : the 80. "National polity" (kokutaz) was the subject of academic and political controversy i n prewar J apan a s well as i n t h e immediate postwar period . There were at least three meanings, all of w hich were pred icated on the existence of the imperial i nstitution: ( 1 ) sovereignty residing in the imperia l institution, (2) the existence of an imperial l ine from w hich authority flowed , and (3) the presence of an imperial i n st itution symbolizing the spiritual unity of the Japa nese state and people. None of these definitions was clearly disti nct from the others and one or more co uld be combi ned or emp hasized depe nd i ng on w hat one wished to argue. M iyazawa Toshiyoshi, w ho inherited the mantle of l eadership of the Constitutional School from M i nobe Tatsukichi , has argued that imperial sovereignty, in the Western sense of being the fi nal l ocus of decision-making power, was the most widely accepted defi nition i n prewar Japa n . But because the emperor's spiritual role, based on Shi nto rites. was central to both sovereignty and ancestral authority, and because that role was the only consistent role played by the imperial i nstitution h istorically, I have taken a position that tends to support definitions (2) and (3). This ambiguous position is consonant with the fi nal and hopelessly am­ biguous discussion of t he national polity produced by t he Japanese government i n the 1 930s. "National polity " was literally the national system, "state structure" (seitai) was literally the government system. I n prewar Japan t he imperia l i nstitution was central to bot h .

P O LI T I C A L

LEGITIMA C Y

AN D

P A LA C E

A U TO N O M Y

39

Japanese monarchy . T o exaggerate , the existence of the imperial in­ stitution was to be eternal , the operation of the monarchy temporal . It is i n this context that Ito's i nsistence on the centrality of the im­ perial i nstitution to state power should be eval uated , not in the context of the separation of church and state . Ito and the Meij i oligarchs thus hoped to make the imperial i nsti­ tution an i mpregnable stronghold of political legitimacy, t he centripetal poi nt around which statesmen would gat her to govern. The theory of i mperial prerogative was the basis for political leadership that looked to the imperial i nstitution and not elsewhere for its j ustification . Religious and natural law , the " people," a social "class" -none had theoretical i ndependence from the emperor and none became sources of dist i nctive doctri nes that seriously challenged t he t heory of prerogati ve . 8 1 For example , M i nobe Tats ukichi , perhaps prewar Japan's leading constitutional theorist , argued that Japan was an organic state under law . His rationalization of consti­ tutional monarchy , widely accepted between 1 9 1 3 and 1 93 5 , did not , however, attack t he sovereign position of the emperor. He made the emperor the highest organ of state and attempted at most to make the Cabinet, appointed by the emperor, ultimately responsible to the lower house of the I mperial Diet-but on the basis of the emperor's legislative prerogative. M inobe i n effect made the elected house of the I mperial Diet the central of several governmental i nstitutions competing to declare the I mperial Will i n politics during the 1 920s and 1 930s.82 8 1 . See , for example, Robert A. Scalapino, Democracy and the Party Movement in Prewar Japan: The Failure of the First A ttempt (Berkeley and Los Angeles : University of California Press, 1962 ) ; a nd M aruyama M asao, Thought and Behaviour in Modem Japanese Politics, ed . Ivan Morris ( London : Oxford U niversity Press, 196 3 ) . 8 2 . M inobe i s the subject o f an outstanding st udy by Frank 0 . M iller, Minobe Tatsukichi: Interpreter of Constitutionalism in Japan (Berkeley and Los Angeles : University of California Press, 196 5 ) . Although M i nobe was forced to resign from t he House of Peers in 1935 for allegedly transgressing t he sacred national polity , it is difficult to conceive of a J apanese more loyal to the imperial i nstitution (ibid . , pp. 165-2 53). At the end of the war M i nobe resisted all revisions of the constitution that would demote the imperial institution or suggest popular sovereignty. He argued that t he Japanese have unbounded fai th in t he imperial house, a faith unparal leled in a ny other nation. The war was the wi llful doing of the military, which was completely oblivious to the general will of the people. Only from loyalty to the emperor d id the Japanese devote themselves t o the war effort. Should Japan lose the imperial house as its eentral i nstitution, there would be a ceaseless train of disturba nces ( M i nobe Tatsukichi , " M i nshu Shugi to Waga Gikai Seido," Sekai Uanuary 1 946] : 21-24 ) .

40

POLITI C A L

L EG I T I M A C Y

AND

PALACE

A U TO N O M Y

I f, however, the emperor's prerogatives were attributed to hi m personally or wielded by him i n fact , how could he avoid censure when the government failed to solve a maj or political crisis ? How could he wield prerogatives and yet be " i nviolable" ? In addition to palace autonomy and the union of court and rites , two measures were taken by the oligarchs . First the Ministers of State were to be i ndividually responsible to the emperor for the government and its policies . Article 55 of t he constitution made each M inister of State responsible for " advising and assisti ng" the emperor in the exercise of his prerogatives , and therefore the Mi nisters of State, not the em­ peror, were to be held accountable for governmental actions based on the emperor's prerogatives .83 Second , sovereignty i n the constitution was placed i n the imperial line, not in the emperor as an i ndividual ruler. As Ito explai ned i n his commentaries on the constitution: The sovereign power o f reigning over a n d o f governi ng the State , i s inherited by the Emperor from His A ncestors , and b y Him bequeathed t o His posterity . A l l the different legislative a s well a s executive powers o f the State, by means of which He reigns over the country and governs t he people, are united in this M ost Exalted Personage, who thus holds in His hands, as it were, all the ramifying threads of the political life of the country , . . 84 .

M iyazawa Toshiyoshi , a leading constitutional theorist i n postwar Japan, has termed the type of sovereignty that existed under the Meij i Constitution " oracle sovereignty . " 85 The emperor was the transmitter, not the i ndependent j udge, of the I mperial Will . The traditional role of the emperor for " ages eternal " as medium between the Japanese people and the gods was therefore adopted as the basis of sovereignty i n the Meij i Constitution. N oting t he vast difference that separated the Japanese emperor from Western divine-right monarchs , Maruyama Masao has argued that 83. " Dai Nippon Teikoku Kempo", in Naikaku Kiroku Ka, 1 942, vol . I , pt. I, chap. I, sec .

I, pp. 7 - 1 8.

84 . Ito H i robumi, Commentaries on the Constitution of the Empire of Japan, trans. Ito M iyoj i , 3d ed. (Chil-o Daigaku, 1 93 1 ) , p. 7 . 8 5 . M iyazawa Toshiyoshi, Kemp o, 4th ed . , re v . (Tokyo: Yuhikaku, 1 95 3 ) , pp. 2 5 , 1 57-58.

P O LI T I C A L

L E G I TI M A C Y

AND

P A LA C E

A U TO N O M Y

41

The amalgamat i o n of spiritual aut hority w i t h political power was regarded not as any new departure i n the concept of sovereignty , but simply as a return to 'the ancient days of the J i mmu Foundation. ' Though the Emperor was regarded as the embodiment of ultimate val ue, he was in­ fi nitely removed from t he possibility of creating val ues out of nothing. His M aj esty was heir to the I mperial l i ne unbroken for ages eternal and he ruled by virtue of the fi nal inj u nctions of his ancestors . The Imperial Constitution, granted to the people in 1 88 9 , was not regarded as having been created by t he Emperor himself; rather it was a document that 'transmitted t he i mmutable law accordi ng to which the land has been governed . ' Thus the Emperor too was saddled with a burden-i n his case a tradition that derived from the i nfinitely remote past . It was only because his existence was i nextricably i nvolved with t he ancestral tradition, in such a way that he and his I mperial Ancestors formed a single unit , that he was regarded as being the ultimate embodiment of i nternal values . 86

The emperor's prerogatives were prerogatives of the imperial l i ne , and the Meij i Constitution embodied the beq ueat hed i nstructions of the emperor's ancestors . His role as medium, through rites , between subj ects and ancestors placed political aut hority beyond his immediate person and fused the past , present , and future in an unbroken con­ ti nuum of eternal Right . This displacement i nto antiquity of political authority for present political rule was a del iberate and creative act of statesmanship by the drafters of the constitution. The M i nisters of State were responsible to an emperor who in turn was responsible to his ancestors . Si nce the emperor was not even in theory a free agent i n the process o f ruli ng but acted only in accorda nce with beq ueat hed i nstructions , the M i nisters of State were left potentially the sole free agents in governi ng the nation. The enactment of the I mperial House Law (Koshitsu Tempan) concurrently with the Meij i Constitution was the fi nal stone in the foundation of the Meij i state . Explicitly recognizing the pri nciple of palace autonomy , the Meij i Constitution reserved to the imperial house its self-governance . Ito's i ntent in the I mperial H ouse Law , as i ndeed in the constitution, was to preserve the dignity of the throne " coeval with heaven and earth. " 87 The I mperial House Law was not 86. Maruyama Masao, Thought and Behaviour, p . 20. 87 . Shumpo Kotsui Shokai , 2 : 57 4 ; Constitution of the Great Emp ire of Japan, A rticl e 74.

42

P O LI T I C A L

LE G I T I M A C Y

A :'I; D

P A LA C E

A t: TO N OM Y

even to be discussed , much less altered , b y the national legislature . I n his commentaries Ito states that the I mperial H ouse Law "will be regarded as the family law of the I mperial House . That these provi­ sions are not expressed i n the Constitution shows that no in­ terference of the subj ect shall ever be tolerated regardi ng them . " 88 Agai nst allowing the Diet a participatory role in determi ning when a regency should be i nstituted , for example, Ito argued that if "the de­ cision of a matter of great importance to the I mperial family is thus delegated to the will of the maj ority of the people there would be a tendency to bring about the degradation of the I mperial dignity . " 89 To make the Imperial House Law as sacred a nd i nviolable as the imperial i nstitution itself, therefore, Ito first placed it beyond the reach of "the people" and their representatives i n the Diet . At the same time, Ito attempted to buttress the immutability of the I mperial H ouse Law by the fiction that it merely set forth practices which had existed from time immemorial . The I mperial House Law was the ancestral law of the imperial li ne-provisions governing the imperial family t hat were a "clarification of bequeathed i nstruc­ tions" from the imperial forebears . 90 The I mperial House Law was not "enacted " but "clarified , " not created but already extant . It was not to be amended even at t he emperor ' s sole discret i o n ; 9 1 amendment , i f req uired , was to b e made by the I mperial Family Council and the Privy Council.92 The emperor was thus freed of per­ sonal responsibility even for his ow n house law . As with the consti­ tution, Ito and the government oligarchs made every attempt to remove from the emperor any personal initiative for which he might subsequently be held accountable. The I mperial House Law gave concrete expression to palace au88. Ito Hirobumi, p. 6. 89. Ibid . , pp. 33-34 . 90. Koshitsu Tempan, in Naikaku Kiroku Ka, 1 942, vol . 1 , pt. I , chap. 1 . , sec . 2, p. 2 1 . 91 . Shumpo Kiitsui Shiikai , 2 : 57 1 -7 5 . 9 2 . Naikaku Kiroku K a , 1 94 2 , vol . 1 , p t . I , chap. 1 , sec. 2 , p. 2 8 . Created b y the I mperial House Law of 1 889, t he I mperial Family Council as modified in 1 907 was composed of all male members of the imperial family of maj ority age w ho were eligible to become Regent . The Lord Keeper of t he Privy Seal , President of the Privy Council , I mperial Household M inister, M i nister of Justice, and President of the Supreme Court participated without vote. In addition to amending powers , th e I mperial Family Council and the Privy Council decided on modifica­ tions in the order of succession to the throne a nd on the creation of a regency . The I mperial Family Council also dealt with discipli nary matters involving members of the imperial family.

P O LI T I C A L

L EG I T I M A C Y

A !\O D

PALACE

43

A U TO N O M Y

tonomy .93 Its sixty-two articles dealt with succession, accession, and enthronement ceremonies , titles , regency , tutors and education, im­ perial family members , imperial house expenses , property litigation, the I mperial Family Counci l , amendment , and coming of age . As head of the imperial family, the emperor exercised supervision over that family (Article 3 5 ) . He sanctioned the marriages of the family members (Article 4 1 ) , appointed or approved tutors to their children (Article 37), approved travel abroad for family members (Article 43) , sanctioned the summons or arrest of a family member (Article 5 1 ) , and suspended or divested of rights a family member who humiliated or degraded the imperial house or was disobedient (Article 5 2 ) . I n the execution of all maj or duties the emperor was advised b y the I m­ perial Family Council and the Privy Council .94 Strangely enough, however, the I mperial House Law was not pro­ mulgated with the Meij i Constitution on February 1 1 , 1 88 9 . It is to be found neither in the Kampa (Official Gazette) for February 1 88 9 , in which t h e constitution appears , nor i n the Ho rei Zensho (Com­ plete Laws) for 1 889 or 1 890. When it was promulgated in later edi­ tions of the Ho rei Zensho, it was not countersigned by the M inisters of State, as was the usual practice with laws. Mi nobe Tatsukichi has surmised that the I mperial House Law was not promulgated in 1 889 because its framers considered it as applying only to the i nternal af­ fairs of the imperial house . M inobe claims that the drafters believed the I mperial House Law had no beari ng on the political life of the people and did not concern regulations of the state. I n support of his opinion, he cites Ito Hirobumi 's commentaries on the I mperial House Law , in w hich Ito declares that "the I mperial House Law is somethi ng the Imperial House has itself enacted as its house law; it is not something, therefore , to be promulgated to Uapanese] subjects by public ceremony . " 95 It would appear that Ito and those responsible for the I mperial House Law did not regard affairs of court as mat­ ters for either the government or t he public : the affairs of court should remain a mystery to ensure greater transcendence for the im­ perial i nstitution . 93. 94 . 95.

Siirifu, Kancho Ben ran, 2 : 3. Naikaku Kiroku Ka, 7942, vol. 7, pt. 7, chap. 7, sec. 2, pp. 2 7-28. M inobe Tatsukichi , Kempa Ko wa, 1 st ed . , rev . (Tokyo : Yuhikaku, 1 9 1 8) ,

p. 4 6 6 .

44

P O I . I T I C. A L

LEG I T I M A C Y

A !'; D

PALACE

AVTONOMY

M i nobe took issue with Ito's interpretation of the "private" nature of the Imperial House Law while agreei ng with him on the spiritual centrality of the imperial house . He argued that the Imperial H ouse Law was in fact a fundamental law of the land because it establis hed two rules of basic constitutional importance : provisions for succession to the throne and regulations for the creation of a regency . Far from concerning merely the private affairs of the imperial family, the I m­ perial House Law bore a critical relation to the state . As evidence , Minobe cited the 1 907 decree that made revisions and supplements to the Imperial House Law subj ect to countersignature by the Min­ isters of State and also to promulgation .96 The supplement to the I mperial House Law of February 2 1 , 1 907 was conseq uently coun­ tersigned by the Prime Minister and all Ministers of State. The only other prewar supplement or revision of the Imperial House Law-that of February 2 8 , 1 9 1 8-was similarly countersigned and promulgated . 97 At the same time, M i nobe argued that the I mperial House Law was of all state provisions the one " most based on national condi­ tions peculiar to Japa n . " From the standpoint of Japan's " u nique national polity , " matters pertai ni ng to the imperial house dare not be subject to participation by the people, and conseq uently the Diet 's power to consent to laws did not reach imperial house affairs i n any way whatever.98 Even as outspoken a constitutionalist as Mi nobe Tatsukichi willingly acknowledged the pri nciple of palace autonomy and its corollary : Japan's unique national polity centered on its im­ perial institution. Although Mi nobe may have disagreed with Ito on a host of issues regarding constitutional interpretat ion , the tran­ scendental position of the throne as the center of Japa nese life was not one of them . By the 1 930s the theoretical relationships between rites and rule and between the I mperial House Law a nd the Meij i Constitution had become blurred to a considerable extent . Some constitutional theorists , such as Uesugi Shi nkichi , argued that the "emperor is i n 96. Ibid . ; Koshiki Rei, "C hokurei" no . 6 , J anuary 3 1 , 1 907, Articles 4 and 5 , in Horei Zensho, I 907 , vol . I . 97 . Naikaku Kiroku Ka, 1 94 2 , vol . I , pt. I , chap. I , sec. 2 , pp. 31 -33, 37-3 8 . 98. M inobe Tatsukichi , Kempa Kowa, p . 468 .

POLITI C A L

LE G I T I M A C Y

A l\i D

P A LA C E

A U TO N O M Y

45

fact the state"-stressi ng the pri nciples of absol ute Western monar­ c hies adapted to J a p a n ' s unique national polit y . Ot hers , l i ke M i nobe , made the emperor the highest organ of a state under law­ stressi ng the constitutional aspects of Western monarchy . Yet others , notably Kakehi Katsuhiko, posited the centrality of Shinto and Shi nto rites in Japa n's prewar constitutional order. At the root of the confusion was the issue of sovereignty-its location and the con­ sequent distribution of powers among political entities such as the Diet . The Ort hodox School of Uesugi and others clearly mai ntai ned that the emperor was sovereign. The Constitutional School , led by Mi nobe , argued that the emperor was sovereign " u nder law" and that the "supreme rights of rule" were invested i n the state as a legal person. The Shi nto School dissolved the issue of sovereignty in a murky fusion of rite and rule i n the imperial l i ne , placi ng it in the eternal " becomi ng" of the imperial line in history , a Japanese version of the Hegelian unfolding of cultural progress concretely expressed by Japan's imperial i nstitution.99 Sakamaki Yoshio, the imperial household official whose tortured logic on the role of rites has already been noted , attempted to resolve the ambiguities that had arisen by integrating many aspects from many schools of i nterpretation. Following M i nobe's argument that the imperial house was a publ ic, not a private entity, Sakamaki asserted that since the imperial house had at its center the wielder of the supreme rights of rule in the state, "the state and the I mperial H ouse are one and i nseparable . " 1 00 "The vitality of the I mperial House is nothing less than the prosperity of the State . " 1 0 1 The 99. M iller, Mi nobe Tatsukichi , Kemp a Ka wa, Suzuki Yasuzo, Meiji Kempa to Shin Kempa (Tokyo: Sekai Shoi n , 1 94 7 ) , pp. 71 ff. ; Kokuritsu Kokkai Toshokan, Kempa Shirya Tenji Kai Mokuroku (Tokyo : 1 957 ) ; Uesugi Shi nkichi , Koku tai Kempa oyobi Kensei, 2d ed . (Tokyo: Yiihikaku, 1 91 7) . Kakehi Katsuhiko, Dai Nippon Teikoku Kempa no Kompon Gi (Tokyo: l wanami Shoten, 1 936). Among the plethora of sc hools of interpretation, the two most important were t he Ort hodox School (Uesugi Shi nkichi) and t he Const itutional School ( M i nobe Tatsuk irhi ) . The word in t he M eij i Const itution that has freq uently been translated as "sovereignty" is tachiken, which literally means "supreme rights of rule" or "prerogatives of government , " j ust as tasuiken means "prerogatives of supreme command . " The word used by M i nobe for sovereignty, or sovereign power, is shuken, the word used for sovereignty in the present Constitution of Japan. It is likely that a termi nological problem complicated the con­ t roversy over sovereignty and its relation to the " national polity" bot h in prewar Japan and i n t hr immed iate postwar period. 1 00 . Sakamaki Yoshio, Koshitsu Seido Ko wa, p. 1 0 . 1 0 1 . Ibid . , p . 9 .

46

P O L I TI C A L

LEG I T I M A C Y

A J\ D

P A LA C E

A U TO N O M Y

constituents of a monarchy , states Sakamaki , are a ruling monarch, a royal family from whom successors will come , and a general populace . It is the state's function , in accordance with these three constituents, to make the position of the monarch secure and tran­ quil , to make the monarch and the royal house "shine forth with increasing splendor," and to advance the people's welfare, tran­ q uil ity , and national consciousness . In a constitutional monarchy like Japan, argued Sakamaki , the people assist the monarch in ruling, the monarch seeks the advice of public leaders and the M i nisters of State, and the people participate in mai ntaini ng the public welfare . One aspect of the separation of court and government can be found i n the division of functions among those who advise the emperor . On court affairs the I mperial Household Mi nister, the Imperial Family Counci l , and the Privy Council advise . The separation of court and government , however, is merely a division of functions : the em­ peror's court and government prerogatives both derive from his position as Chief of State . Conseq uently, the Imperial House Law is as much a constitution as the constitution itself. Although the dis­ tinction between court and government embodied in these two basic laws makes Japan's a dual legal system, Sakamaki argued , both are eq ually state laws deriving from the emperor's i ndivisible supreme right of rule . 102 Separate but equal , the I mperial House Law and the constitution achieved reverential status by the 1 930s as the oracular pronounce­ ments of the apotheosized Meij i Emperor. As already noted , the I m ­ perial House Law w a s suppleme nted only twice between 1 889 and 1 94 5 ; the effect was to make the imperial house a more public entity and to elevate the I mperial House Law to equal status with the constitution. The Meij i Constitution remai ned unchanged to the let­ ter. Made structurally and legally independent by 1 88 9 , the imperial house was placed in a transcendental but symbiotic relation to political power. The pri nciple of palace autonomy was firmly united with the emperor's moral authority implied by the union of court and rites . The resultant moral transcendence was then made the 1 02 . Ibid . ,

pp.

1 3 -22 .

PO L I T I C A L

L EG I T I M A C Y

AND

P A LA C E

A U TO N O M Y

47

basis of political authority . These li nkages evolved as the result of ( 1 ) compromises between the theories and aspirations of traditionalists, such as M otoda Eifu , and those of the modernizing oligarchs, such as Ito Hirobumi ; (2) political conflict between the " i n " oligarchs a nd t he " ou t " leaders of " p ublic opi nion , " confl icts which threatened to embroil the emperor as well as u nseat the oligarchs ; and (3) creative nation-building efforts by the Meij i oligarchs , who eclectically applied both native Japanese and imported Western theories of state power in seeki ng to place their political power on an unassailable foundation of legitimacy . The implications of the theoretical and i nstitutional arrangements arrived at by 1 889 were twofold . First , the imperial institution was hopefully placed beyond the criticism of the "Japanese subject " . As high priest , shogun , and constitutional monarch, the emperor was to be the transcendental i nstitution of state power, " far from the smoke of human habitations , no one ever to invade its sa nctity . " This awesome dista nce was t o provide the basis of popular loyalty and obedience to the state . Second , the emperor was not to be a free agent in an open political process . His personal will , which was fallible, was not identical with the Imperial Will , which was by defi nition the eternal will of the imperial ancestors . This in turn meant restricting the emperor's public role to formal rituals, such as Shinto rites and the formal sanctioning of government decisions . By removi ng the emperor from open and direct participation i n the de­ cision-maki ng process , while issuing all government acts in the em­ peror's name , the oligarchs apparently hoped to ensure their ow n power to develop Japanese political i nstitutions around a core of bu­ reaucratic i nitiative supported by prerogative above and loyalty below . Surrounded by walls and moats i n the 2 7 5 -acre castle that had for­ merly been the fortress of shogunal power, the emperor found his per­ sonal freedom of movement , as well as his personal will i n politics , forfeited for shogunal powers recast i n Western monarchical form. An indication of the price the emperor paid by becomi ng the sacrosanct center of Japanese politics is the decreased number of imperial outi ngs from his Tokyo residence after 1 890. Table 1 lists the number of im­ perial tours from 1 868 to 1 9 1 0 , with a partial listing of the purposes

48 TABLE 1 .

POLITI C A L

L E G I TI M A C Y

AND

PALACE

A U TO N O M Y

IMPERIAL TOURS, 1 868-1 9 1 0 Purpose"'

rear

Total

1 868 1 869 1 870 1 87 1 1 872 1 873 1 874 1 875 1 876 1 877 1 878 1 879 1 880 1 88 1 1 882 1 883 1 884 1 885 1 886 1 887 1 888 1 889 1 890 1 89 1 1 892 1 893 1 894 1 895 1 896 1 897 1 898 1 899 1 900 1 90 1 1 902 1 903 1 904 1 905 1 906 1 907 1 908 1 909 1910

17 37 87 1 12 42 1 05 80 1 00 55 58 89 98 1 63 127 96 95 81 57 36 44

62 47 47 21 21 20 18 9 21 4 10 25 25 8 16 14 12 12 18 17 13 18 18

Within Tokyo

Outside Tokyo

9 36 86 111 38 101 77 97 51 56 86 96 161 1 18 91 91 75 55 33 42 60 46 44 17 19 18 14 8 18 3 9 23 21 7 15 II 12 10 17 15 12 17 16

8

4 4 3 3 4 2 3 2 2 9 5 4 6 2 3 2 2 1 3 4 2 2 4 1 3 1 1 2 4

3 0 2 1 2 1 2

M

R

s

I

p

0

4

3

0

0

80

50

9

0

39

1 35

13

0

8

0

7

29

9

0

9

24

9

0

7

0

2

0

G

4

9

11

11

0

2

4

3

2

2

0

5

9

0

2

0

7

0

2

P O LI T I C A L

LEGITIMA C Y

A !\ D

P A LA C E

49

A C TONOM Y

for which they were made . It is abundantly clear that the period of the Restoration settlement , from 1 868 to 1 88 9 , was also the period of the emperor's greatest mobility. From 1 87 1 to 1 880 there were an average of 90.2 imperial tours a year; from 1 88 1 to 1 890 the average was 6 9 . 2 per year. O nce the new state was formed , however, the emperor was increasi ngly restricted to his palace . Between 1 89 1 and 1 900 there were but 1 7 .4 imperial outi ngs a year; from 1 90 1 to 1 9 1 0 , only 1 4 . 6 . And , a s table 1 clearly shows, the emperor did not g o t o the government very often after 1 890; the government came to hi m. The decrease in imperial visits to the government does not necessarily mean, of course , that the emperor's political influence , if there was any , decreased . It does mean that whatever infl uence he had over government leaders was exercised withi n the palace sanctuary . His personal will was restricted to those close to the throne : leaders of the palace bureaucracy and those of the government , civil and mili­ tary , who were allowed access to his castled prerogatives . To exag­ gerate, the emperor's personal will ended at the palace gates where public scrutiny began. The keepers of the gates were the officials of the palace bureaucracy .

sou R C E FOR TABLE

I : Kunai Daijin Kambii Siimuka , Gyoko Hyo (Tokyo: 1 933). The code u nder Purpose , based on places visited by the emperor, is as follows : G government ; M military, including maneuvers and academies ; R religious institutions, such as shrines ; S social, including prefectural tours, attendance at cultural events, and visits to factories ; I international, such as banquets with royal visitors held outside the palace ; P personal, such as summer visits to the imperial villas. •

=

=

=

=

=

=

CHAPTER THREE

THE PALACE B UREAUCRACY : GATEKEEPERS OF THE IM PERIAL WILL

THE OLD SHOGU NAL FORTR ESS was ideally suited to be the new im­ perial palace . With its spacious grounds , securely walled and moated , it gave physical meani ng to palace autonomy and imperial transcendence . It also provided ample space to house the palace bu­ reaucracy which managed the affairs of the imperial i nstitution and isolated the emperor as a person from the public I mperial Will ; by 1 900 t he number of palace officials had grow n to approximately four thousand . What was the structure and composition of the palace bureau­ cracy ? What specifically did it manage and how ? We may begi n by attempti ng a schematization of the roles created for the emperor as a l ogical result of the theory of legitimacy and the i nstitutional ar­ rangements evolved by the Restoration settlement of 1 868-89 (fig . 1 ) . Although the imperial i nstitution tended to be undifferentiated to its more dedicated adherents , the analysis of the separation of court and government in chapter 2 suggests that there were two funda­ mental imperial roles : Emperor-in-Court and Emperor-i n-State . B oth roles had two aspects : Emperor-i n-Public and Emperor-in­ Chambers . The Emperor-i n-Public represented the Imperial Will in politics and society; the Emperor-in-C hambers was the emperor as a person who expressed his ow n views and ideas to those "close to the throne . " 51

52

TH E

P A LA C E

BUREAVCRACY

Ideal Polity (I nstitutions of Court) ( Rites) EMPEROR-I N-COURT

I

visible

I

\

invisible

\

E M P E R O R -I N - PU B LI C

EM PER OR - I N - C H A M B E R S

( I mperial Will)

(Personal Will)

\

visible

\

I

invisible

I

EM P E R O R -I N - STATE

( Rule) ( I nstitutions of Government) Real Polity FIGURE

1.

IMPERIA L ROLES I N PREWAR

J A PA N

The Emperor-in- Court. B agehot i nforms u s that the B ritish monarch i n the late nineteenth century performed the "dignified " function i n government and that the monarchy 's dignity derived from its " mystery . " The dignified mystery of the monarchy , however, was directed not so much toward the "efficient " political leaders as toward the general public. Its mystery served to make the monarchy the focus of popular loyalty and identification, deflecting public at­ tention away from the tough busi ness of actually ruling the nation and how that was accomplished . To exaggerate , the dignified mystery of the monarchy served the purposes of efficient rule by eliciting popular obedience to political decisions produced by an al­ most i nvisible political process . 1 Mystery , by definition, meant that the act ual worki ngs of the monarchy would be screened from popular scrutiny so that the monarch would appear to be above the ordi nary lot of men. In Japan the need for such transcendence was apparently i ntensified by the vast changes in Japanese society that were planned or a nticipated by the moderniz i ng oligarchs after 1 86 8 . Fosteri ng an almost rel igi ous at1 . Walter B agehot , The English Constitution ( London: Kega n Paul , Trench, Triibner, 1 904) .

TH E

P A LA C E

BUREAUCRACY

53

tachment to the imperial institution, Restoration leaders and their heirs attempted to make the Imperial Will a diffuse mystery around which all ki nds of i ndividuals and social groups could be mobilized . The mystery of the Japanese imperial institution was, of course , al­ ready there to be adapted by the Restoration leaders . Throughout Japan's recorded history the emperor had been the most exalted figure in Japa nese society. By virtue of his hereditary position, he had exercised an excl usive monopoly over certain religious rites that many considered basic to the well-being of the entire society . The court had also been the repository of ancient Japanese culture and its accretions from C hina. I ndeed , from the twelfth century onward , rites and cultural practices were the only consistently autonomous functions performed by the imperial institution. Finally, the emperor and his court had been isolated from society at large and from the seat of de facto government for over 2 5 0 years when the Restoration occurred i n 1 86 8 . The restored emperor was very much a transcendental mystery to the vast maj orty of J apanese . Given this backgrou nd and the fact that the emperor had not ruled for centuries , the Japanese imperial i nstitution had perhaps more potential than any other modern monarchy to become the symbol of the nation as a people, as disti nct from the nation as a state . This di­ stinction may very well have been the true basis for the separation of court and government and for the functional union of court and rites on the one hand and of government and rule on the other, as discussed in chapter 2 . The new imperial i nstitution, bri nging its distinctive character to bear in the building of a new Japan, could be made to symbolize the unity and continuity of Japanese society as it underwent the disruptive effects of extensive modernization. Thus , when the emperor made a tour of the provi nces , presided over an im­ perial poetry contest , or performed rites at one of the national Shinto shri nes , he symbolized the unity of the Japanese people, their shared cult ural heritage and skills, and their unique rel igious communality , of which he was chief priest . He was Emperor-i n-Court acting as Emperor-i n-Publ ic, presumably reminding all .Japanese that , by virtue of his maj esty and mystery , they were a unique and un ited people. The Emperor-in-Court also symbol ized somethi ng new i n his

54

TH E

PALACE

BUREAUCRACY

public functions : the desirable course of Japan's modernization. The emperor and imperial family were among the first to don Western dress and to use the latest modes of transportation. The imperial house was fi rst i n line to invest in new manufacturi ng enterprises . Modern educational facil ities received imperial support and the em­ peror distributed rewards t o those who successfu l l y i nt roduced "Western techniques" i nto Japa nese society. The emperor and his court were thus made to stand for the moderniz i ng aspirations of the Japanese people as wel l . The first duty o f the palace bureaucracy was the ma nagement o f the Emperor-i n-Court as Emperor-in-Publ ic. I n this function the palace bureaucrats were gatekeepers of the I mperial Will in Japa nese so­ ciety, seali ng off the emperor's personal will so that the Imperial Will might be made the focal point of unquest ioning popular esteem and reverence . To do so, the palace bureaucracy was divided i nto two components : an outer side and an inner side. The outer side contai ned t he bureaucratic machi nery for mai nt a i n i ng t he t ra nsce ndental dignity of the emperor i n society. It arra nged formal appearances of the emperor and imperial family members , wrote and regulated publicity , managed the imperial estate and fi nances of the imperial house , and watched over its ow n bureaucratic concerns . The i nner side ma naged the Emperor-in-Court as Emperor-in-C hambers . The officials of the i nner side were chamberlains, valets , aides , tutors , ladies-i n-waiting. They provided for the personal well-bei ng and private education of the emperor and imperial family. As immediate compa nions and servants to the emperor, they were the innermost guardians of his privacy . They were conseq uently responsible for keepi ng the emperor's personal will and his human i ndiscretions from escapi ng i nto the public arena . The palace bureaucracy , both its outer and inner sides , was to see that the emperor as a person was made in­ visible . Only the emperor as a social paragon , as an ideal , was to be visible . Fi nally, the imperial i nstitution was to be the expression of all that was true, good , and beautiful in the Japanese polity. In its structure , therefore , the court was .to b e a model o f the ideal polity, o f the ideal social arrangements that ought to prevail i n Japan. The second basic duty of the palace bureaucracy was to maintai n itself as social paragon .

TH E

PALACE

BUREAUCRACY

55

The Emperor-in-Sta te . The emperor was also t o be the transcendental source of state power. For centuries , of course, the emperor had been the source of political legitimacy . It was his theoretical and physical proximity to politics that was new . As Em­ peror-i n-State the emperor was to express the I mperial Will i n politics , and the I mperial Will was to be the referent for a l l acts of government . J ust as the court was responsible to the Emperor-in­ Court for mai ntaining the " ideal polity , " the government was responsible to the Emperor-i n-State for actuating the " real polity . " Agai n , the emperor was not t o be i nvolved as a person but as a n in­ stitution. The Emperor-i n-State , as Emperor-i n-Public , was to ratify decisions reached by responsible government leaders . As the fi nal act in the prewar decision-maki ng process, imperial ratification meant that a given policy was a legitimate decision of state , having been sanctioned by the I mperial Will . The Imperial Will i n politics was not t he emperor's personal will but the will of the " imperial line unbroken for ages eternal . " Far more than was the case with the emperor's personal will as Emperor-in-Court , it was imperative that the emperor's personal will as Emperor-i n-State be hidden from public view . If the personal preferences of the emperor were publicly known on a sharply disputed issue prior to its resolution by the government , the emperor might well be charged with political partisanship and his i ndividual role challenged . And if those preferences were widely credited as the source of a disastrous political policy , the imperial i nstitution itself might be q uestioned , undermining the foundation of the state. Because the mysterious dignity of the imperial institution was the basis of state power, however, the gatekeepers of the I mperial Will in politics were not only palace bureaucrats but virtually all political leaders of account : M inisters of State, generals and admirals of the Supreme Command , Privy Councillors , Elder Statesmen, and all others who by law or custom " advised a nd assisted" the emperor . Hut it was primarily the responsibility of four palace officials to ma nage the Emperor-i n-State so that the palace remai ned the in­ vi olable sanctuary of the emperor's personal will i n politics . These four were the I mperial Household M i nister, the Grand Cham­ berlain, the Chief Aide-de-Camp , and the Lord Keeper of the Privy Seal . To ensure the transcendental position of the throne in politics

56

TH E

PALACE

BUR EAUCRACY

as well as i n society, they were primarily concerned that the Em­ peror-i n-State as Emperor-i n-Public appear as the representative of the I mperial Will i n politics , and therefore that the emperor's per­ sonal will be made invisible. The emperor was to be neither a political partisan nor a policy maker. He was to ratify decisions produced by government leaders with the I mperial Seal . With these considerations in mind , the four palace leaders cooperated with the government leaders of the day in arranging the public display and i nvolvement of the Emperor-i n-State. In practice , of course, the distinction between Emperor-i n-Court and E mperor-i n-State , as well as t hat between Emperor- i n ­ Chambers and Emperor-i n-Public, proved ambiguous and very flexible, depending on the discretion of court and government leaders of the day , the nature of political pressures , and the state of public opi nion. The Emperor-i n-State and the Emperor-i n-C ourt were one and the same person. A nd the gatekeepers between Emperor-in­ Chambers and Emperor-i n-Public for both these roles were the same palace officials. Ultimately it proved as difficult to preserve a fixed distinction between Emperor-in-Public and Emperor-i n-C hambers as it was to maintai n the distinction between Emperor-i n-State and Emperor-in-Court . During the 1 930s, for example, the emperor pursued his i nterest in marine biology within the pal ace sanctuary and collected specimens near the imperial villa at Hayama . Many found this activity beneath the sublime Japanese emperor. Was this avocation public or private ? Public criticism of the emperor's pas­ time in the 1 930s illustrates the prevailing state of public opi nion as well as a maj or breach in the human fortifications around the Em­ peror-i n-Chambers . At this time i n Japan's modern history not even the emperor's private life was free from public scrutiny. By way of contrast , there was l ittle danger prior to 1 9 1 0 that w hat the emperor said or did in chambers as Emperor-in-State or Em­ peror-i n-Court would leak out i ndiscreetly to the public. Those who surrounded the emperor in the early half of Japan's modern history formed a tight knot of political leaders , the oligarchs , w ho may have disagreed among themselves over policies but not on the privileged status of remarks made by t he Emperor-i n-C hambers . There is no i ndication, for example, that t he criticism of the emperor's passion for horsemanship made by the jiho i n 1 87 8 reached the public or be-

THE

PALACE

BUREAUCRACY

57

came a source of public criticism of the emperor at that time (see above , p. 20) . Differences among the Mei j i oligarchs, however great , were fought out person-to-person without visibly i nvolving the Em­ peror-in-Chambers . The dearth of information about the M eij i em­ pero r ' s p e r s o n a l v i e w s on p o l i t i c s a nd his p e r s o n a l p o l i c y preferences , i f h e had any, is a tribute to t he cohesiveness o f those close to the throne in preservi ng the privileged status of the Em­ peror-in-C hambers in both his roles as Emperor-in-State and Em­ peror-i n-Court . By t he 1 930s this knot of political leaders had been untied . Those close to the throne i ncluded institutional spokesmen, whose primary loyalties were to their respective i nstitutions of imperial prerogative­ Army , Navy , Home M i nistry , Foreign M i nistry . If the Emperor-in­ Chambers favored one policy over another (personally or at the prompting of advisers), one of these i nstitutional representatives would invariably be slighted . When the slighted representative conferred with his i nstitutional cohorts, leaks to the outside often followed . Omi nous remarks about " evil advisers" close to the throne might also filter through the inj ured institution and seep out to the press as wel l . The privileged nature of the emperor's remarks i n chambers was thus destroyed b y the encroachment o f institutional pluralism i nto the very heart of the palace . Our concern i n this chapter is the i nstitutional setti ng of palace politics : the structure and functions of the I mperial H ousehold Ministry , which managed the day-to-day activities of the Emperor­ in-Court and coordi nated the routine affairs of the Emperor-in-State . We are concerned , therefore , with the palace bureaucracy , as struc­ t ure a nd personnel , between 1 885 and 1 945 . C hapter 4 describes the nature and function of palace leadership, w hich set t he policies for ma nagi ng the public Emperor-i n-Court and coordinated the process of imperial ratification by the Emperor-i n-State . THE OFFICES AND PE RSONN EL OF THE PALAC E BU R EAUC RACY

M any theorists and commentators have argued that the prewar . J a panese state was a family writ large . Japanese subj ects were not

58

TH E

P A LA C E

B U R EA U C R A C Y

i ndividual citizens but members of families which i n turn were units of the national family headed by the imperial house . The emperor was the fat her of his people and all his subj ects were bound to him by the absolute loyalty owed a father by his children.2 If the prewar Japanese state was a family writ large , however, the imperial insti­ tution was a state writ smal l . The imperial institution had its ow n constitution, t he I mperial H ouse Law , and its own counci l , the Im­ perial Family Council, w hich j oi ntly with the Privy Council advised the throne on t hat constitution. The I mperial Household M inister was responsible for advising and assisting the emperor i n all matters of court , much as the Prime M i nister and other Ministers of State were responsible in all matters of government . The I mperial Household M inistry housed i n miniature all that in the large was encompassed by the government 's bureaucracy : fiscal and property management , capital i nvestment , foreign affairs , police functions , m i l i t a ry o p e r a t i o n s , e d u c a t i o n , a nd w e l fa r e . T h e I m p e r i a l H ousehold M i nistry was t hus a concrete symbol o f modern bureau­ cratic government . Not only did the Restoration .settlement produce a separation of court and government and a unification of court and rites ; it also resulted i n a new administration for court affairs . The creation of modern bureaucratic structures i n government was paralleled i n court , especially duri ng I t o H i robu m i ' s t e nure as I mperia l Household Secretary and I mperial Household M inister from 1 884 to 1 887 . When he left the I mperial Household M i nistry the bureau­ cratic structure for managi ng court affairs had been largely consoli­ dated . From the end of the Restoration settlement in 1 889 to the end of 1 945 , w hen the prewar government ceased to operate as it had , only minor revisions were made i n the functional domai n and structure of t he mi nistry . A list of its offices i n 1 945 i ndicates the wide range of functions under its j urisdiction (see table 2). The earliest date for each office , or its cognate predecessor, also indicates that the struc2 . For example, see Maruyama M asao, Thought and Behaviour, pp. 36-37 and passim, for a critique, a nd H ibino Yukata, Nippon Shindo Ron, or the National /deals of the Japanese People (Cambridge : Cambridge University Press, 1 928}, for a glorification of family-state ideals.

TH E

PALACE

TABLE 2.

59

BUREAUCRACY

OFFICES OF THE IMPERIAL HOUSEHOLD MINISTRY, 1 945

Title of Office

Minister's Secretariat Board of Chamberlains Board of the Ceremonies Office of Peerage Affairs Office of Imperial M ausolea Office of Archives Office of Court Physicians Office of Imperial Cuisine Office of the Privy Purse Office of Imperial Works Office of Imperial Stables General Affairs Bureau Bureau of Imperial Guards Board of Rituals Board of the Empress's Affairs Board of the Crown Prince's Affairs Board of the Empress Dowager's Affairs Imperial House Audit Bureau Imperial Poetry Bureau Imperial House Museum Sh6s6in Curator's Office Bureau of Imperial Lands and Forests Peers School Peers School for Women Board of Prince Ri's Affairs Kyoto Office Office of Lord Keeper of the Privy Seal Office of Aides-de-Camp

Date Created 1 908 1 884 1 877 1877 1 878 1 883 1 886 1 886 1 885 1 885 1 886 1 94 1 1 886 1 939 1 869 1 889 1 869 1 889 1 888 1 886 1 884 1 885 1 877 1 885 191 1 1 883 1 885 1 896

souRCEs : " Sho Sanko Hy6," handwritten document provided by the Imperia l l lousehold Agency ; Sorifu, Kancho Benran , 2 : 26-32 ; Matsushita Yoshio, Meiji Gunsei S/li Ron , 2 vols, (Tokyo : Yiihikaku, 1 956) , 2 : 206. N OTE : The English names of the offices have been taken mostly from prewar Japanese yearbooks and biographical dictionaries in English. Where I considered the name clumsy or ambiguous I took the liberty of changing it. The dates listed are as accurate as possible, given the frequent changes in office names from 1 869 through 1 945.

l ure of the palace bureaucracy remained almost unchanged after I H 8 9 . Of t he twenty-eight offices in 1 945 , only one was entirely new l o the twentieth century : the Board of Prince Ri's Affairs created after the annexation of Korea in 1 9 1 0 . The M i nister's Secretariat a nd the General Affairs B ureau, created in 1 908 and 1 94 1 respec-

60

THE

PALACE

BUREAUCRACY

tively, were organizational refi nements of two or three secretarial di­ visions of 1 886-8 9 . As noted i n chapter 2, the Board of Rituals origi nated as a subdivision of the Board of the Ceremonies around 1 880. Other offices , such as the B oard of the Empress Dowager's Af­ fairs, were created and abolished according to the composition of the imperial family at any given time . In addition, ad hoc offices were established and then termi nated when their work was completed : palace construction bureaus, i nvestigatory bodies, and t he like . One o f t he offices listed i n table 2 w a s not formally i ncl uded i n the I mperial H ousehold M i nistry's organization chart : the Office of A i d e s - d e - C a m p . O rg a n i z at i o n a l l y , t he A i d e s - d e - C a m p w e r e considered members o f the General Staff Office , a n d i t w a s the military , not the palace bureaucracy , that i nitiated t heir appoint­ ments . But because the Aides were i n fact regular attendants on the emperor and members of t he imperial family, they were opera­ tionally very much a part of t he palace bureaucracy , if not the I m­ perial Household M i nistry strictly speaki ng. It also appears that a n office for the Ladies-i n-Wait i ng was never created a s a formal part of the I mperial H ousehold Mi nistry . Whether or not they were ac­ knowledged as " officials" of t he ministry , however, there were al­ ways Ladies-i n-Waiting in the inner circle surrounding the Em­ peror-i n-C hambers . Finally, there were a number of Court Advisers (Kyuchu Kom an) and I mperial Household M i nistry Consultants (Kunai Sho Goyogakari) w ho advised the court on such matters as fi nances , dipl omatic etiq uette, Shinto ceremonies , and t he dis­ tribution of honors . Some of them were extra court appoi ntees while others served concurrently i n formal court offices . As t he formal and i nformal offices of the I mperial H ousehold M inistry reveal , management of the court was an extensive and many-faceted operation. A considerable bureaucracy was req uired to mai ntain the imperial i nstitution as a state within a state . Table 3 lists at three-year intervals the total number of I mperial Household M i nistry officials from 1 924 to 1 942 , the numbers of persons in the two highest civil service categories of direct imperial appoi ntee (shin­ nin) and i mperial appoi ntee (chokunin) for those years , and the number of officials in the Tokyo palace and in t he provi ncial offices

TH E

PALACE

TABLE 3 .

61

BUREAUCRACY

IMPERIAL HOUSEHOLD MINISTRY OFFICIALS, 1 92 4-1 942

rear•

Total

Direct Imperial Appointeesb

Imperial Appointeesb

Other

Tokyo

Provinces

1 924 1 927 1 930 1 933 1 936 1 939 1 942

4 , 749 4 , 58 1 4 , 922 4 , 899 5 , 1 45 5 , 342 5 , 677

8 7 8 8 6 6 8

101 1 09 Ill 1 15 1 34 1 40 1 33

4 , 640 4 , 465 4 , 803 4 , 776 5 , 005 5 , 1 96 5 , 536

3 ' 161 3 , 209

2 , 181 2 , 468

SOURCE : Kunai Sho Shokuin Roku, 1 924-42 . Although this source goes back to 1 92 1 , figures o n total personnel and civil service allotments by rank are not listed until 1 924. Three-year intervals have been selected because two of the annual Kunai Sho Shokuin Roku, those for 1 926 and 1 932, are missing from the Imperial Household Agency files and because there was neither a very great increase in officials over the twenty­ year period nor much variation on a yearly basis. b The number of direct imperial appointees from 1 935 on includes both those offi­ cially of that rank and those treated as such. In 1 942, for example, 6 of the 8 direct imperial appointees were officially of that rank. The same applies to those of imperial appointee rank : of the 1 33 imperial appointees in 1 942, 75 were imperial appointees strictly speaking. •

for years in which these figures were available. The personnel break­ down of the mi nistry's officials in 1 943 was as follows : 3 Di rect imperial appointee or treated a s such I mperial appoi ntee or treated as such Executive appoi ntee or treated as such Official or treated as such Attached staff Staff Employees Total

9 1 32 356 2,631 203 I, 1 6 1 1 ,325 5 ,8 1 7

With ni ne officials o f direct imperial appointee rank i n 1 94 3 , and from six to nine of that rank from the 1 920s on, the I mperial l l ousehold M i nistry ranked among the top bureaucratic structures i n prewar Japan. Only the Foreign M i nistry , whose mi nister and 3. Kunai Sho Shokuin Roku, 1 94 3 , p . 1 9 .

62

THE

P A LA C E

BUREAUCRACY

ambassadors plenipotentiary were direct imperial appoi ntees , and the armed forces , whose generals and admirals were also of that rank, outranked the Imperial H ousehold M i nistry i n this respect . And only the Foreign M i nistry outranked the I mperial Household M i nistry in its ratio of direct imperial appoi ntees to total personnel . Only the Prime Minister and the Speakers of the two houses of the I mperial Diet , moreover, received higher official salaries than the I mperial Household Mi nister. In 1 93 9 , for example, the maximum official salaries (yen per annum) of the leading officials of the government and the court were : 4 Government Prime M inister Speaker of t he House of Peers Speaker of the House of Representatives M inisters of State President of the Privy Cou nci l Ambassadors Plenipotentiary Admirals and generals Vice M i nisters of State

9,600 ¥ 7 , 500 7 ,500 6,800 6,600 6,600 6,600 5 , 800

Court I mperial Household M inister Lord Keeper of t he Privy Seal Chief Aide-de-Camp Grand C hamberlain

6,800 ¥ 6,800 6,600 6,200

Finally, in the years from 1 924 to 1 943 there were approximately fifty officials at court w ho held concurrent posts i n the government but were not carried on the I mperial Household M i nistry's official roster. Of t hese, about twelve were direct imperial appoi ntees i n rank. As measured by the ranks and salaries of its official s , therefore , the I mperial H ousehold M inistry was a very powerful and pres­ tigious bureaucratic organization . The transcendental dignity of the imperial i nstitution, of course , made it necessary for the I mperial Household M i nistry to be prestigious and dignified as wel l . Why the mi nistry should have been led by such high-ranking bureaucrats, however, is a q uestion that cannot be answered merely i n terms of status, as we shall see in the following chapters . Here we turn to the 4 . Shokuin Roku, I 939, pt. 2, pp. 1 - 1 57 .

TH E

PALACE

B U R EA U C R A C Y

63

institutional and personnel structure around the emperor for the routi ne management of the imperial dignity .

TH E OUTER SIDE : BUREAUC RATIC OPERATIVES OF THE EM PEROR-IN-COURT AS EM PEROR-IN-PUBLIC

Functionally, the I mperial Household M i nistry had two sides : an " outer" side and an " i nner" side .5 The outer side comprised the modern bureaucratic aspects of court admi nistration-arrangement of imperial tours, management of imperial lands and forests , ac­ counting, records , public relations-and was located in provincial of­ fices or in office buildings outside and around the inner palace , the Fukiage park, where the emperor resided . The inner side comprised those in regular attendance on the emperor and his immediate family : Chamberlains , Ladies-i n-Waiting, Aides-de-Camp , tutors to the im­ perial family and so on. As "family" members of the court , these persons had access to the emperor and imperial family in the inner palace . As will become apparent , the line separating the outer and inner sides was not clear-cut or absolute ; there was a constant flow of communication between the two. As a rule, however, the work of the outer side was office work, that of the inner side, attendance . The former were bureaucrats, the latter, imperial emissaries and ele­ gant valets. A glance at the 1 945 offices listed in table 2 will reveal generally which of the offices were outer offices : the M i nister's Secretariat , Of­ fice of Imperial Mausolea, Office of Archives , Office of the Privy Purse , Office of I mperial Works , General Affairs B ureau, etc . Ap­ proximately eighteen of the twenty-eight offices were clearly involved with court administration and estate management . The Minister's Secretariat , for example, was responsible for " secret " matters , the promotion, status, and retirement of mi nistry officials, pensions , cor­ respondence , statistics, and reports for the Official Gazette, coordi­ nation of mi nistry offices and bureaus , investigations , t he drafting of S. Kyiitei Kishadan, pp. 3-4 . Although this source applies the " outer" (ornate) and " i n­ owr" (oku) distinction to the postwar court bureaucracy , it is equally valid, if not more so, for 1 he prewar court admi nistration.

64

THE

PALACE

B U R EA U C R A C Y

papers , and matters not assigned to other offices . The General Af­ fairs B ureau handled arrangements for imperial tours , public an­ nouncements and photographs , and miscellaneous work for other ministry offices . The I mperial Guards Bureau was responsible for palace security, escorting imperial processions and other police func­ tions , sanitation, air defense , and fire-fighting. Reception of foreign dignitaries and protocol matters fell u nder the Foreign Affairs Section of the B oard of the Ceremonies .6 The outer side was thus responsible for the Emperor-i n-Court acting as Emperor-i n-Public, a charge they guarded j ealously. I f the living emperor's public i mage, as well as that of the im­ mediate imperial family, was the first duty of the outer side , the dignity of t he imperial li ne-the imperial presence in the land from antiquity-was also in its charge . I t was the responsibility of the Of­ fice of I mperial Mausolea to maintain the 734 tombs and shrines of past emperors and imperial family members that were located i n twenty-seven o f t h e forty-seven l ocal divisions o f Japan.7 These were physical reminders to the people that theirs was a divine land which had existed from " time i mmemorial . " Supporting the physical presence of the imperial l i ne among the populace were the rites conducted by the B oard of Rituals on national festival days . To em­ phasize the cultural centrality of the imperial l i ne, the I mperial Household M i nistry also managed the Shosoi n , repository of Japan's cultural treasures and court relics dating from the eighth century on . A ncient court music was preserved by the music division of the B oard of the Ceremonies .8 The management of court fina nces. Nothing better illustrates the essentially administrative nature of the outer side, and , incidentally , the separation of court and government , than the management of court fi nances . M anaging the fi nancial operations of t he palace was a maj or function of the outer side, and approximately half of the M i nistry's total personnel were engaged in imperial estate and fi­ nance management . 6. " Kunai Shii Kansei , " Article 1 0, Koshitsu Rei, no. 7 , October 7 , 1 92 1 , as amended t hrough 1 94 1 ; in Naikaku Kiroku Ka, 1 942, vol . I, pt. 3, chap. I. sec. 3, pp. 1 6, 1 7 . 7 . " Ryiibo S u Ichiran Hyii," handwritten document provided b y the I mperial Household Agency . In the 1 930s Japan comprised forty-three prefectures, three municipalities, and one circuit . 8. Siirifu, Kancho Benran, 2 : 20-2 1 ; Kyutei Kishadan, p. 50.

TH E

PALACE

BUREAUCRACY

65

As part of their preparations to create a transcendental emperor during the Restoration settlement , the Meij i oligarchs strengthened the i ndependent base of court organization by transferring vast amounts of state land to the imperial house and by placing the court 's finances under sole j urisdiction of the palace bureaucracy . I n 1 882 lwakura Tomomi pressed upon the cabi net a strongly worded argument in favor of imperial house properties which would serve as the " material foundation" of the constitution soon to be d rafted and implemented . He argued that all state lands as well as railways and a number of industrial plants should be placed under autonomous control of the imperial house .9 I mperial house property would give the i mperial i nstitution fi nancial autonomy commensurate with its all-embracing political authority . The oligarchs were agreed that the creation of an autonomous source of income for the imperial house was an essential step to be taken before the constitution came into ef­ fect . 1 0 The legal a nd admi nist rat ive separat i o n of court and government was therefore to be bolstered by granti ng financial i nde­ pendence to the court . The court 's financial autonomy was consolidated during Ito Hirobumi 's tenure as I mperial Household M inister from 1 885 to 1 887 . First , lands and capital i nvestments were turned over to the Imperial Household M i nistry i n such amounts that by 1 9 1 5 the court was able to defray the maj or part of its expenses from income derived from its own i ndependent sources . Second , budgets and ac­ counts, not only for its ow n revenues from capital and lands but also for its allotment from the national treasury to meet court expenses , were completely under court (I mperial H ousehold M inistry) control . The government had no formal j urisdiction over court fi na nces ex­ cept over i ncreases in the national treasury allotment . I n 1 885 a B ureau of I mperial Lands was created to admi nister the vast public lands transferred or to be transferred to the management of the Imperial Household Ministry . Between 1 885 and 1 890 the imperial estate increased from 54,7 1 7 acres to some 8 . 7 million acres . Article 45 of the I mperial House Law of 1 889 also provided for hereditary imperial property; approximately 2 . 5 million of the 8 . 7 9 . Kuroda Hisata, Tenno Ke n o Zaisan (Tokyo: Sanichi Shinsho, 1 966), pp. 22-24 ; I noue Kiyoshi, pp. 88-89. 1 0. Kuroda Hisata, pp . 24 , 26; I noue Kiyoshi, p . 89.

TH E

P A LA C E

B U R EAUCRACY

acres were consequently designated as hereditary , "forever" bel ong to the imperial family.U Between 1 890 a nd 1 945 , however, 1 l w imperial estate decreased considerably. Disparities between rq�istered area and actual surveys, the sale of unnecessary lands and la nds difficult to administer, and t he return of lands , origi nally ac­ quired for development purposes , to prefectural authorities largely accounted for the decrease . In 1 895 , for example, 3 . 3 million acres of imperial land in H okkaido were returned to local admi nistration . L a n d s i n Y a m a n a s h i p r e fect u re , a p p a re n t l y acq u i re d fo r development , were largely returned to local prefectural adminis­ tration by 1 9 1 6 . Surveys of i mperial lands proved the registered area of imperial land to be as much as 1 8 percent greater than the actual area . 12 Even the hereditary lands were greatly reduced . Consequently, the imperial estate in 1 940 totaled 3 ,226,388 acres , not i ncludi ng " unnecessary lands" yet to be disposed of, of which 5 1 2 ,204 acres were hereditary imperial property. Forest lands constituted approximately 97 percent of the imperial estate , while palaces , villas, and imperial tombs accounted for 0.5 percent . The remai nder was agricultural land . I mperial lands were held i n forty of the forty-seven l ocal divisions of Japan. 13 At least 1 ,300 of the Im­ perial Household M inistry 's 5 ,400 registered officials were engaged in the management of the imperial lands a nd forests in 1 940. 14 In ad­ dition to maintai ning an imperial presence throughout the land , the B ureau of I mperial Lands and Forests also provided a substantial in­ come to the court from 1 897 on. In 1 9 1 5 it is estimated that three million of the imperial house's total income of sixteen million yen derived from the B ureau of I mperial Lands. 15 The period from 1 885 to 1 890, years of enormous expansion of the imperial estate , were also years of great i ncrease in imperial capital investments, most notably i n the field of banking. I n 1 884 Matsukata Masayoshi, leadi ng government oligarch i n the world of finance, urged that government stocks i n the Bank of Japan and the Yokohama Specie B ank be turned over to the i mperial house. The 1 1 1 illion

lo

11. 12. 13. 14. 1 5.

Kuroda Hisata, pp. 2 6 , 3 1 -32 , 3 7 . I bid . , pp. 1 08-09 passim, 46-48, 5 1 - 5 2 . Ibid . , pp. 1 08-09, 1 1 1 . Kunai Sho Shokuin Roku, 1 940. Kuroda Hisata, pp. 1 32 , 248-4 9 , 1 00 .

TH E

PALACE

B U R EAUCRACY

67

government subsequently transferred shares valued at five million yen in the former and one million yen in the latter to the Chief of the Privy Purse, Imperial Household M i nistry , in February 1 885 . These stocks were to remain the nucleus of imperial house i nvestments until 1 947 , when all imperial house properties were dissolved . Addi­ tional stocks were acquired between 1 88 5 and 1 890 i n the N ippon Yusen Kaisha (a shippi ng company), Japan Railways , Sapporo Sugar, Tokyo H otel , and industrial railways in H okkaido. 16 And al­ t hough the imperial estate dwindled after 1 890, imperial house in­ vestments in banking and industry conti nued to grow until 1 945 . Each venture in foreign expansion netted monies for the Privy Purse . I n 1 898 part of the i ndemnity garnered from the Sino-Japanese War, in the amount of twenty million yen , was granted to the im­ perial house by resolution of both houses of the I mperial Diet . This money was then invested by the court i n government securities . The imperial house also invested i n the B ank of Taiwan ( 1 899), Taiwan Sugar ( 1 90 1 ) , the Bank of Korea ( 1 909) , the Chosen I ndustrial Bank ( 1 9 1 8) , t he Manchurian Railway ( 1 924-2 5 ) , and banks belonging to private combines , such as the Sumitomo Bank. 17 M oreover, imperial house capital i nvestments grew steadily i n each of these enterprises as the result of stock i ncreases and divisions . For example , the modest holding of 4 , 000 M a nchurian Railway stocks i n 1 92 5 grew to 84 , 3 7 5 stocks b y 1 945 . 18 Profits from lands, forests, and capital i nvestments were supple­ mented by sums allotted from the national treasury to meet the " im­ perial house expenditure" (koshitsu hi) . As early as 1 869 the Civil Government had allotted a regular sum for court expenses , but the expenses of t he i mperi al family a nd t he sal aries of I mperia l Household M i nistry personnel were paid b y the M inistry o f Fi­ nance . From 1 886 onward , however, a l ump sum to cover all court expe nses was gra nted to the I mperial H ousehold M i nistry . Moreover, the I mperial H ousehold M inistry was i n no way ac­ countable to the government for the use of the government allotment ; the court was to have fi nancial autonomy even with respect to 1 6. Ibid . , pp. 27-29, 63-64 . 1 7 . Ibid . , pp. 6 1 -62 , 66, 92-93 . 1 8 . Ibid . , p. 93 .

68

THE

PALACE

BUREAUCRACY

monies granted from the national treasury . From 1 88 9 to 1 909 the sum disbursed from the national treasury was fixed at 3 million yen . I n 1 909 the sum was increased t o 4 . 5 million, where i t remai ned until the end of World War II. 19 The only accurate figures on imperial house fi nances from 1 897 are this disbursement from the national treasury and the fi nancial records of the B ureau of I mperial Lands and Forests. I mperial in­ vestments in public bonds and in both public and private enterprises were cloaked in secrecy, with only occasional figures available on certain investments.2° Kuroda Hisata, however, estimates the annual income of the imperial house in 1 9 1 6 as fol lows : 21 National treasury Bureau of I mperial La nds and Forests Stocks Public and other securities Total

4 , 5 00,000 ¥ 3,000,000 4 , 5 00,000 4 ,000,000 1 6 ,000,000 ¥

By 1 9 1 6 at the latest it is clear that the imperial house fi nanced the maj or portion of its operations from its own resources . From 1 9 1 4 it even had its own bank, the J ugo Cinko. At the end of the war im­ perial capital (stocks , public securities , cash on hand) was valued by the Occupation authorities at 3 3 6 , 1 5 8 ,890 yen, yielding annual in­ come of possibly as much as 2 0 million yen . I ncome derived from lands and forests plus sales of land in 1 937 was approximately 8. 7 million yen .22 A very rough estimate of annual imperial house in­ come around 1 940 would thus be : 23 National t reasury Bureau of I mperia l La nds and Forests Capital investments Total

4,500,000 ¥ 9,000,000 1 5 ,000,000 28,500,000 ¥

1 9 . I bid . , pp. 1 2 - 1 3 , 1 4 , 36, 7 8 . 20. I bid . , p . 7 2 . 2 1 . I bid . , p. 1 00 . Accord ing to Japanese yearbooks, t he average exchange rate of U . S . dollars to y e n i n 1 9 1 6 w a s 5 50.50 to 1 00 ¥ . The i ncome o f t h e imperial house w a s t herefore over 58,000,000 per a nnum in 1 9 1 6 U . S . d ollars. 22. I bid . , pp. 1 20, 1 22 , 249. 23. The biographers of lshiwatari Sotaro, the last I mperial Household M i nister, claim that t here had never been a complete estimate of imperial house properties until the Occupation au­ t horities dema nded such a valuation after the war. The value fi nally arrived at by I shiwatari was 1 . 6 billion yen (lshiwatari So taro, [Tokyo: lshiwatari Sotaro Denki Hensan Kai, 1 954] ,

TH E

PALACE

BUREAUCRACY

69

All but autonomous in its fi nancial resources , the imperial house was accountable only to itself for both revenues and expenditures . Only increases in the national treasury allotment req uired Diet ap­ proval , and by 1 940 the national treasury allotment was a mere fift h of the total court i ncome . The court audit, as well as its budget , was completely disti nct from the government's audit and budget and ex­ clusively under the control of the Imperial Household Ministry . Even the court 's fiscal year (January-December) after 1 89 1 was dif­ ferent from that of the government (April-March) .24 The chief advisory body on court fi nancial operations was the Im­ perial House Economic Counci l , created in 1 89 1 . According to the 1 89 1 regulations governing the council , it was to be composed of spe­ cially appoi nted imperial house economic advisers , the I mperial H ousehold Mi nister, the Vice M i nister, the Chief of the Privy Purse , and the Chief of the Bureau of I mperial Lands. The emperor was to be present at council meeti ngs , which decided o n regul ations governing the management of the hereditary estate, designation of hereditary properties , general provisions concerning the imperial estate, capital i nvestments, the amount and use of the current ac­ count , budget increases and decreases in the imperial estate and in capital investments, and regulations governing the court audit . Three accounts were created for auditing purposes by the I mperial H ouse Finance Law of March 1 89 1 : the imperial estate account , the capital account , and the current account . The C hief of the B ureau of I mperial Lands was responsible for the estate accou nt , the Chief of the Privy Purse for the capital account , and division and bureau chiefs of the mi nistry for the current account .25 Three accounts were retai ned after the revision of imperial house auditing procedures i n 1 9 1 2 a s well : current , special , and capital . The special account after p . 478). It is also claimed that t he imperial house had origi nally acqui red stocks to give the public confidence in this new form of economic activity : if the emperor invested , others would follow suit (ibid . , p . 47 9). However accurate these claims-and bot h are subject to serious doubts-it remains t hat imperial property a nd i nvestment contributed greatly to the political pur poses of palace autonomy. The excha nge rate at the Yokohama Specie Bank i n 1 940 was U . S . $23 .40 to 1 00 ¥ . The annual i ncome of t he imperial house in 1 940 was therefore around $6.7 million in 1 940 U . S . dollars. 24. Kuroda Hisata, pp. 36, 37-38, 44 . 2 5 . Ibid . , pp. 39-44 .

70

THE

P A LA C E

B U R EAUCRACY

1 92 1 applied exclusively to the imperial estate and was the responsi ­ bility of the C hief of the B ureau of I mperial Lands and Forests. The Chief of the Privy Purse looked after both current and capital ac­ counts and was responsible to the I mperial Household Mi nister for drafting the court budget . These accounts were i nvestigated by the Imperial H ouse Audit Bureau, which did not , however, examine the inner court and " secret " expenditures . Apparently, the budgeting and accounti ng practices of t he court were considerably more flexible than those of the government .26 I mperial lands and forests, imperial house capital, and an all but u nt o u c h a b l e g o v e r n m e n t d i s b u rs e m e n t gave t h e c o u r t t h e wherewithal for legal and administrative autonomy . How these resources were employed , however, depended on w ho controlled the I mperial H ousehold M inistry . With great wealth at its disposal , the palace bureaucracy could conceivably use court funds for political purposes . This would appear to be all the more likely if the Imperial Household M i nister had pol itical aspirations or owed obl igations to gover nment leaders , or if his tenure i n office depe nded o n cooperation w i t h the government leadership of the day . Contribu­ tions to individual government leaders and politicians , and even to polit ical parties , l i kewise depended on t he nat ure of palace leadership . The secrecy that enshrouded the fi nancial operations of the court is circumstantial evidence that court funds were not used strictly for court purposes . The very autonomy of the court , para­ doxically, constituted a potential for political abuse . As will be show n in chapter 4, there is some evidence that this potential was indeed exploited , bl urring in operation the distinction between court and government . Managing the aristocracy. If in fact the imperial institution were to be an operati ng symbol of all that was true, good and beautiful i n the Japanese polity, the I mperial Household M i nistry logically would also manage the affairs of social status . This was concretel y implemented by giving the M inistry control over the aristocracy . After several unsuccessful attempts to settle on a system of ranks following the Restoration, a modern aristocracy fi nally came into being in 1 884. The new aristocracy was to be open to ex-samurai 2 6 . Ibid . ,

pp.

80-83 .

TH E

P A LA C E

B U R EA U C R A C Y

71

and commoners who had made outstandi ng contributions t o the new Japanese polity as well as to ex-feudal lords and ex-Court Nobles . By 1 87 8 the C ouncillors had agreed to create five ranks , and these were formalized in 1 884 as follows : prince , marquis, count , viscount , and baron . Centering on the imperial house , the new aristocracy was to be the statused bulwark of the throne . On the day following the establishment of the aristocracy the emperor reported the i nvestment of titles to the i nner shri ne (kashik odokoro ) , to the shri ne of the im­ perial spirits (koreiden) , and to the shri ne of the gods (shinden) . The newly invested aristocrats then paid their respects to the inner shrine, dedicati ng themselves to t he service of the imperial house .27 As early as 1 877 a bureau withi n the I mperial Household M i nistry had been establis hed to ma nage t he affairs of the aristocracy . 28 I n 1 884 the Mi nistry was given custody of the family registers of the aristocracy . Marriages and adoptions req uired its approval; 29 in 1 886 the Mi nistry was charged with managing the hereditary property of the aristocracy as well . 30 After several changes in name and clarifications of j urisdiction, the I mperial Household M inistry office in charge of the aristocracy was finally styled the Of­ fice of Peerage Affairs (Sochitsu Ryo ) . As such, it handled ( 1 ) matters regarding the aristocracy , the imperial family, pri ncely families, and such families as the Korean royal house, i ncluding conferences of princes-of-the-blood and of princes as well as meetings of the I mperial Family Counci l ; (2) supervision of personnel assigned to the imperial family, pri nces-of-the-blood , and princes ; (3) announcements of con­ ferrals of court rank and peerage titles ; ( 4) regulations governi ng the Peers School and the Peers School for Women; and ( 5 ) preparation of the electoral list of counts, viscounts, and barons for the H ouse of Peers . 31 The director of the Office of Peerage Affairs was either a direct i mperial appoi ntee or an imperial appointee . 32 Ito H irobumi , architect of the new peerage system of 1 884, not 27. Shumpo Kotsui Shokai , 2 : 384-88; I noue Kiyoshi, pp. 94-95 . 2 8 . "Kunai Sho Tasshi Koka, " no. ! , January 1 0, 1 87 7 ; in Horei Zensho, 1 87 7 , p. 950; Kyiitei Kishadan, p . I I . 29. Shumpo Kotsui Shokai , 2 : 386. 30. Ho rei Zensho, 1 886, vol . 1 , pp. 1 78-8 1 . 3 1 . Naikaku Kiroku Ka, 1 942, pt. 3, chap. I, sec. 3, p. 1 6 ; Shokuin Roku, 1 943, pt. I , p . 60 . 3 2 . Naikaku Kiroku K a , 1 94 2 , pt . 3, chap. I , sec. 3, p. 1 9 .

72

TH E

P A LA C E

BUREACCRACY

only managed the affairs of this new class as I mperial Household M inister but was i nstrumental in clarifying the relationship of the new aristocracy to the social order. First , the Peers School was to train the sons of the aristocracy in service to the nation. Second , Ito i nsisted that the new aristocracy reward achievement as well as family heritage i n its composition. 33 One of Ito's first acts after becoming Imperial Household Secre­ tary in March 1 884 was to resolve the long-standing issue of j uris­ diction over the Peers School . Created in 1 887 , the Peers School was i nitially supported by contributions from the court and the " peers club , " the latter i ncluding the entire aristocracy at t he time. After considerable difficulty , Ito had the Peers School made a government school under the management of t he I mperial Household Ministry on April 1 7 , 1 884. In so doing, Ito argued that the aristocracy existed to support the social order and to preserve the prosperity of the imperial house . For the aristocracy to operate merely for its ow n group i nterest was " d isl oyal . " 34 Refl ect i ng t he public service orientation of the aristocracy as i ntended by Ito, the curriculum of the Peers School emphasized military subjects and ethics as well as the natural sciences and literature .35 The new aristocrats were to embody the cultural values of the new Japanese nation, combi ning in their education both Eastern ethics and Western techniques . An em­ phasis on military subj ects reinforced the role of the aristocracy as bulwark of the throne and hence , of national defense . Almost in­ variably the headmaster was a promi nent military figure; among them were such famous war horses as General Nogi Kiten and General Tani Kanj o . Except those given special exemption, princes­ of-the-blood all became military officers . 36 Moreover, the Peers School gave concrete express i o n to t he identificati o n of t he aristocracy with imperial rites . Peers School holidays, for example , included the anniversary of t he death of Emperor Komei ( 1 866) , the Rite of Emperor J immu , Empire Day, the emperor's birthday , and Shi nto festivals of the autumnal and vernal equinoxes , new rice , and 33. 34. 35. 36.

S humpo Kotsui Shokai , 2 : 384 . Shumpo Kotsui Shokai , 2 : 378-80. Horei Zensho, 1 884 , p. 1 2 . Goseitoku Fukyii Kai , p . 1 03 .

TH E

P A LA C E

73

BUREAUCRACY

harvest .37 The desti ny of the aristocracy as a public class was thus made i nstitutionally coeval with the fortu nes of the state and the im­ perial house . I n the formative years of the new aristocracy ( 1 884-87) , the aristocracy of merit i ncreased from 33 to 80 families38 B y 1 92 8 the total number of aristocratic families was 956, excl udi ng pri nces-of­ the-blood . As i ndicated in table 4 , a bare maj ority of aristocratic families ( 487 of 956, or 50.8 percent) were of " miscellaneous" family origi ns . This category was mixed in composition and i ncluded highly stat used hered itary priest s , high-ranki ng feudal branch families , leading retainers of the greater feudal l ords (including the shogun) , and families of high status i n pre-Restoration japan-some of the old Osaka business families, for example-as well as those families whose titles were acquired solely as reward for service to the state and society.39 In this latter category-the merit aristocracyTABLE 4. COMPOSITION OF THE JAPANESE ARISTOCRACY IN 1 928, EXCLUDING MEMBERS OF THE IMPERIAL FAMILIES (PRINCES-OF-THE-BLOOD) Family Origins

Title

Prince M arquis Count Viscount Baron Total souRCE :

Imperial House

Court Noble

Feudal Lord

Miscellaneous

Total

0 5 4 0 0

9 10 29 85 43

3 16 33 220 12

6 9 42 74 356

18 40 1 08 379 41 1

9

1 76

284

487

956

Gendai Ka;:;oku Fuyo (Tokyo : Nihon Shiseki Ky6kai, 1 929), p. 720.

3 7 . Horei Zensho, 1 884, p. I , 298. 38. I noue Kiyoshi, p. 96. 39. I am i ndebted to Herschel Webb for call i ng my attention to t he diversity i n the back­ grounds of those aristocrats in t he miscellaneous category . In overall composition, therefore, t he pre-Restoration ascribed aristocracy outnumbered the post-Restoration merit aristocracy , even though the merit aristocracy domi nated in the miscellaneous category .

74

TH E

P A LA C E

BUR EAUCRACY

were Ito Hirobumi and Yamagata Aritomo, both of whom were eventually created princes . Of low rank within the warrior class , both acquired their titles on the basis of merit alone. A nd like the titles of all prewar aristocrats, those granted to Ito a nd Yamagata were made hereditary . The conferral of titles on " servants of the throne" conveyed fiscal reward as well as status . Around 1 884 , for example, those granted the title of count , such as Ito Hirobumi and Yamagata Aritomo, were each given lump sums of 3 5 ,000 yen. In 1 887 each count was given 3 5 0 shares of stock, valued at 1 7 ,500 yen, i n the Nippon Yusen Kaisha . 40 B oth the 1 884 and the 1 887 awards were disbursed from the emperor's Privy Purse . S i nce the I mperial H ousehold M i ni nstry managed both t he affairs of t he aristocracy and an autonomous im­ perial estate, the financial resources of the imperial house were used to sustain not only the dignity of the imperial house but also that of its bulwark . The Imperial H ousehold M i nistry , therefore, was the bureaucratic means of fusing the hierarchy of social status with that of wealth, or at least of ensuri ng that social status and wealth were commensurate. The aristocracy discharged its public service function in several ways . One of the maj or areas of service was the I mperial Household M i nistry itself. Of the fifty-two men serving in the six leading palace offices between 1 885 and 1 945 , forty-five were or became titled . There were nine princes ( i ncluding one pri nce-of-the-blood), six marquis, ten counts, seven viscounts, and thirteen barons . All twelve men who served as Directors of Peerage Affairs were titled : two pri nces, five marquis, three viscounts , and only two barons , i ndi­ cating that the aristocracy played a considerable role i n defining its bulwark functions as well as i n managing the affairs of the I mperial Household M i nistry . THE INNER S I D E : SANCTUA RY OF THE EM PEROR-IN-CHA M B E R S

While some aristocrats held leadi ng palace posts, others served in lesser roles , primarily on the inner side of the I mperial H ousehold 40. I noue Kiyoshi, p . 96.

Til E

P A LA C E

BUREAUCRACY

75

M inistry as C hamberlains and Stewards. The inner side of the palace bureaucracy and certain positions, such as Master of the Ceremonies , on the outer side , constituted refuges for aristocrats con­ sonant with their inherited dignity and status while not demanding from them the tough-mi nded abil ities required outside the palace to achieve social and political success-abilities that i n many cases had been possessed by t hose who first acquired titles . Among those who served a lifetime in the palace , for example, was no less a figure t h a n t he adopted son of I t o H i robumi , Pri nce I t o H i rokuni ( 1 870- 1 93 1 ) . After graduati ng from the Peers School and studying briefly in Germany , Hirokuni entered the I mperial Household M inistry , where he remai ned for thirty-nine years . Eventually achievi ng the post of Grand Master of the Ceremonies ( 1 926-29) , he became know n as the "lord of the great i nner mountain."41 If Ito H irokuni is a classic instance of an aristocratic scion taking refuge in the I mperial Household M i nistry , he was by no means the only one . I n 1 93 1 , for example, we find the grandson of oligarch Yamagata Aritomo, Prince Yamagata Arimichi , among the C hamberlains and concurrently among the Masters of the Ceremonies .42 The Cham berlains. It was among the C hamberlains i n particular that many an aristocrat found sanctuary . In 1 942 t he Imperial Household M i nistry's regulations provided for a Grand Cham­ berlain, two Deputy Grand C hamberlains, and ten C hamberlains.43 I n 1 94 3 , however, there were twelve C hamberlains in addition to the Grand and Deputy Grand C hamberlains . 44 Of the fifteen, eight were titled or in line to succeed and at least three were younger brothers of titled aristocrats. In addition to their predominantly aristocratic background , the C hamberlains also found a measure of group co­ hesion in educational experience . Twelve of the fifteen C hamberlains in 1 943 had graduated from prewar Japan's foremost university, Tokyo Imperial University .45 Forming the inner ring around the em4 1 . Daz Jimmei jiten, ed . Shimonaka Yusaburo, 10 vols. (Tokyo: Heibon Sha, 1 937-4 1 ) , 1 : 26 1 . 42. Shokuin Roku, 1 93 1 , pt. 3 , p . 1 0. 4 3 . Naikaku Kiroku Ka, 1 942, vol . 1 , pt. 3, chap. 1 , sec. 3, p. 1 . 44. Kunai Sho Shokuin Roku, 1 94 3 . 4 5 . Data on t he personal backgrounds a n d careers o f t he Chamberlains, as w e l l a s all other officials discussed hereafter, have been obtained from published rosters of officials, biographical dict i onaries, biographies , diaries, interviews, and the I mperial Household Agency's personnel files.

76

THE

P A LA C E

BUREAUCRACY

peror and his immediate family, the Chamberlains , usually of high educational attainment and impeccable li neage , were elegant men­ servants a nd i mperial messengers-paragons of the I mperial H ousehold M i nistry's i nner side . I n contrast to the bureaucrats of the outer side, whose duties were clearly outli ned in the organic law of the I mperial Household M inistry, the Chamberlains had no prescribed duties other than to serve at the emperor's side. Even after the palace admi nistration had become highly bureaucratized , the basic regulation governing the I mperial H ousehold M inistry merely stated that the Board of Cham­ berlains "takes charge of matters at the side of the emperor. "46 A characteristic of the ministry's inner side , as demonstrated by the Chamberlains, was its lack of formal structure a nd officially stipu­ lated duties . This was , of course , natural to offices that called for constant companionship with the emperor and imperial family. I rie Sukemasa , Chamberlai n to the emperor since 1 934 and formerly a viscount , has articulated by hyperbole what that companionship en­ tailed : Outside t here may be persons whom the emperor trusts deeply. Outside there may be persons w hom the emperor respects. But those he considers j ust like t he atmosphere, like t he breezes, paying no heed that they are there-being j ust the same w hether there or not-those are Cham­ berlains . 47

I rie notes that the role of companion and messenger played by the Chamberlains might invol ve helping the emperor collect mari ne life specimens or goi ng to Taiwan to express the emperor's condolences on such occasions as the great earthquake of 1 93 4 .48 l rie also recalls that considerable freedom prevailed at the side of the Emperor-i n-C hambers . I n 1 93 5 , for example, the Chamberlains kept a book of fines for improprieties i n speech and the sums collected were used for a year-end party. Outside the palace such improprieties , claims I rie, would have been dealt with as disrespect , causing offenders to lose their positions or suffer other serious penalties . Even during the Pacific War, conti nues lrie, such freedom 46. Naikaku Kiroku Ka, 1 942 , vol . 1 , pt. 3 , chap. 1 , sec. 3. 4 7 . I rie Sukemasa , Tenno-sama no Kanreki (Tokyo: Asahi Shimbun Sha, 1 962), p . 242 . 48. Ibid . , pp. 232-3 3 , 238-39.

TH E

P A LA C E

BUREAUCRACY

77

was protected in the palace . Yoshida Shigeru ( 1 878-1 967 ) , a prewar Foreign M i nistry bureaucrat a nd one of J a p a n ' s outst a ndi n g postwar leaders , w as arrested during the w a r for mentioning the possibility of Japanese defeat. Yet "we" at the palace talked openly about the i nevitability of defeat . 49 "At the emperor's side freedom of speech has been guarded at every turn; there is absol utely no more democratic a place than this . "50 Such "freedom" at the emperor's side also meant the pranksteri ng freedom that prevails among those of an intimate ingroup . Count Kanroj i Osanaga , classmate of the Taisho Emperor (r. 1 9 1 2-26) at the Peers School and a lifetime official of the I mperial Household M i nistry's inner side , states that both the Meij i and Taisho Em­ perors were surrounded by a band of stalwarts fond of practical j okes . Kanroj i 's uncle, Kadenokoj i Sugekoto ( 1 860- 1 92 5 ) , was one of them, claims Kanroj i . When " Kade" was a page to the Meij i Em­ peror he , like the other pages , slept in the young emperor's room . On one occasion Kade, unable to sleep , decided to kick the bed i n w hich the emperor was sleeping soundly. The emperor bolted out of bed-it was a time of earthquakes-and looked around hi m. Kade was of course very fast asleep by then. This same Kade , according to Kanroj i again, was know n to yank the sleepi ng Taisho Emperor's blankets off when his ow n had fallen off in the dead of the night . Kanroj i also relates that when he himself and the Taisho Emperor were schoolboys together, they were taught to swim by being throw n willy-nilly into the water by a Colonel Tachibana .5 1 Such pranks , according to I rie, continued as a regular part of inner court life i n the Show a period ( 1 926-) a s wel l . 52 Even acknowledgi ng the at49. I rie Sukemasa , jiju to Paipu (Tokyo: Mai nichi Shimbun Sha, 1 957), pp. 97-98 . 50. l rie Sukemasa , "Okajiki," Matsudaira Tsuneo Tsuiso Roku, ed . Tokyo PR Tsiishin Sha (Tokyo: M atsudai ra Tsuneo-shi Tsuioku Kai, 1 96 1 ) , p . 83. l rie's essays on court life have all been written in the postwar period . Given the t heoretical role played by the emperor in prewar Japan and the protective shell imposed by the palace bureaucracy , such anecdotes could not have been publ ished then. With t he postwar "democratization" of the imperial insti­ tution, however, l rie and other court persons have published essays, anecdotes, and memoirs about the inner court during the prewar period . I rie in particular has gained a reputation for bei ng the emperor's foremost public relations man. Although his views may be exaggerated and one-sided in an effort to create a more humanistic image for the emperor and court , his essays capture at least some of the atmosphere and activity of court life in prewar Japan. 5 1 . Kanroj i Osanaga, Sebiro no Tenno (Tokyo : Tozai Bummei Sha, 1 957), pp. 42-43 , 4 5 . 5 2 . l rie Sukemasa, "Okajiki . "

78

THE

P A LA C E

B U R EA U C R A C Y

tempts to " humanize" and " democratize" the emperor and court that are so read ily apparent in these postwar accounts of prewar in­ ner court life , there was clearly a very great gap between the rigid formality and cold sanctity surrounding the Emperor-in-Public and the freewheeling i nformality and intimate fraternity surrounding the Emperor-in-Chambers. Also characteristic of the C hamberlains was the practice of holding concurrent posts in both the inner and outer sides of the Im­ perial Household M i nistry . Of the fifteen Chamberlains in 1 94 3 , for example, one was concurrently Grand Steward to the Empress , one headed a section of the Office of Court Physicians , and three were on the tutorial board for the children of the imperial family . Not un­ naturally, a Chamberlain headed each of the three sections of the B oard of Chamberlains, the outer office through which communi­ cation between the inner court and outer mi nistry offices was chan­ nelled .53 Since audiences granted to foreign dignitaries as well as visits of worship to the inner shri ne were scheduled by the Board of the Ceremonies ,54 it is not surprisi ng that three C hamberlains served concurrently as Masters of the Ceremonies . One Chamberlain also served concurrently as a secretary i n the General Affairs B ureau,55 an outer office concerned with imperial tours and public relations . The holding of concurrent posts was evidently deemed necessary be­ cause it was the C hamberlains who had primary responsibility for managing access to the Emperor-in-Chambers , as well as providing companionship to the emperor and members of the imperial family. The primacy of the Chamberlai ns in mai ntaining the privacy of the imperial family by controll i ng access to the i nner court , at least during the 1 930s and 1 940s , is also clear from the official duties of the B oard of Chamberlains. Although the I mperial Household M inistry 's organic law did not specify the functions of the board , the ministry's regulations governing divisions and sections did . The fi rst section of the Board of Chamberlains, the Miscellaneous Section, handled ( 1 ) imperial lectures , (2 ) audiences at court , (3) court allowances and imperial gifts, (4) imperial family records, (5) cus53. Kunai Sho Shokuin Roku, 1 94 3 . 54. " Kunai S h o Bunka Kitei , " in Shokuin Roku, 1 943, pt . 1 , p . 60. 5 5 . Kunai Sho Shokuin Roku, 1 943.

TH E

PALACE

B U R EA U C R A C Y

79

tody of the board 's seal , and (6) matters not assigned to the other two sections. In coordination with the Office of Aides-de-Camp, that section also scheduled audiences for military personnel . The I nner Court Section of the board managed such matters as butleri ng. The t hird and last section, the Management Section, was in charge of ac­ counts and possessions of the imperial family .56 Thus the Board of C hamberlains managed the i nner court a nd liaison between the Em­ peror-in-Chambers and t he Emperor-in-Public. The term of service for a Chamberlain was generally his lifetime . The fifteen Chamberlains i n 1 943 averaged nine years i n office , rangi ng from three to twenty-nine years service . Six of the fifteen were still C hamberlains in 1 962 .57 Obviously, the post of C hamberlain was not a rung on the ladder of advancement to higher office; it was a termi nal appoi ntment . It may be surmised that their seniority , as well as their aristocratic station and educational background , gave the C hamberlains considerable authority in deali ng with the outer side bureaucrats , enhanci ng their ability to preserve t he privacy of imperial family life from outside scrutiny. The Ladies-in- Waiting and A ides-de- Camp . The C hamberlains did not , however, have a complete monopoly over the i nner court cir­ cuit . There were two other groups w ho shared i n t he function of regular attendance : the Ladies-i n-Waiti ng and the Aides-de-Camp . Until the early twentieth century t he Ladies-in-Waiting had had quite a different function from others in regular attendance : to ensure that there would be an heir to the t hrone. In fact , a court lady of the second rank (gon-tenji) , Yanagihara Naruko ( 1 855-1 943) , was the mother o f the Taisho Emperor. Meij i 's empress had no offspring .58 The Ladies-in-Waiting had been members of the inner court t hroughout t he recorded hist ory of J a p a n , a nd Kanroj i Osanaga estimates that some three hundred of them accompanied the young emperor to Tokyo w hen he was restored as the center of the Japanese polity in 1 86 8 . 59 Only w hen the present emperor became regent i n November 1 92 1 , 5 6 . " Kunai Sho Bunka Kitei , " in Shokuin Roku, 1 94 3 , pt . 1 , pp. 59-60. 57. Kunai Sho Shokuin Roku, 1 943; Kunai Cho C hokan Kambo Hisho Ka , Kunai Cho Shokuin Roku (Tokyo: 1 962) . 5 8 . Kyiitei Kishada n, pp. 1 7 , 1 5 1 . 5 9 . Kanroj i Osanaga, p. 1 42 .

80

TH E

P A LA C E

B U R EA U C R A C Y

however, was a successful reform of the Ladies' court life begun, and it was not until 1 92 6 , when the present emperor began his reign , that monogamy was established .60 Until 1 92 1 , and especially in the early Meij i period , the Ladies-i n-Waiti ng were a power to be reckoned with. Like the C hamberlains, they acted as i ntermediaries between the emperor and the outside world . It is even rumored that some leading Restoration figures were turned away by the Ladies when they appeared for audience at the emperor's inner residence . Such Restoration leaders as Sanj o Sanetomi, Iwakura Tomomi, and Saigo Takamori were victims of their perversity .61 In 1 87 1 an at­ tempt was evidently made to put a stop to the political maneuveri ngs of the Ladies by havi ng prominent civilian and military officials ap­ pointed to court posts . Among the appoi ntees were the future jiho Yoshii Tomozane and Komeda Torao, two " modernizers" devoted to the reform of t he imperial institution.62 B ut even with maj or ef­ forts at housecleaning, the traditions of life in the " back i nterior" of the emperor's residence continued up to the end of the Taisho pe­ riod .63 Until the Showa period , the lives of the one hundred Ladies and their attendants , isolated from the outside world , were replete with jealousies, viciousness , and feuds .64 Even such a life-long member of t he court's inner side as Kanroj i Osanaga states that their existence was " i nhuman, i nappropriate to the new times . "65 When the present emperor married i n 1 92 4 , the old system a nd practices of court Ladies-in-Waiting were fi nally abolished . The " machi nes for produci ng t he emperor's heirs" were replaced by salaried women at­ tendants . 66 The attempted reform of such a ncient court practices and i nstitu­ tions indicates the i ntent of the Meij i oligarchs to make the imperial i nstitution a modern structure around w hich popular l oyalty could 60. Nezu M asashi , p. 3 9 ; Kyiitei Kishadan , p. 1 56. 6 1 . Kyiitei Kishada n, pp. 1 52-53 ; Kanroj i Osanaga , p. 1 43 . 62 . Tokutomi l ichiro, Koshaku Yamagata A ritomo Den, 3 vols. (Tokyo: Yamagata Ari­ tomo-ko Kinen Jigyo Kai, I 933), 2 : 288-89 . 63. Kyiitei Kishadan , p. 1 49; Watanabe lkuj i ro, Meiji Tenno, 2 vols. (Tokyo: Meij i Tenno Shotoku Kai , I 958), I : 1 2 5 -3 1 . 64 . Yanagihara Byakure n , " Y a nagihara l c h i i Tsubone no Kai n i n , " Bu ngei Shu njii (Tokushii) , no. 10 ( 1 956) : 44; Kyiitei Kishadan, pp. 1 50-5 3 . 6 5 . Kanroji Osanaga, p . 1 4 2 . 6 6 . Nezu Masashi , p . 3 9 ; Kyiitei Kishadan, p . 1 45 .

TH E

PALACE

BUR E A UCR A CY

81

b e mobil ized i n support of the " new Japa n . " The emperor was not to be the captive of i n ner court anachronisms but a modern symbol of old virtues , and the difficulties posed by as seemi ngly innocuous an institution as the Ladies-i n-Waiti ng were typical of the whole range of problems confronti ng the politicia ns of modernization. Quite a different component of the court 's i n ner circle was the Of­ fice of Aides-de-Camp . Like the Chamberlains , the Aides-de-Camp served as imperial messengers and regular attendants . Whereas the C hamberlai ns in t heir capacity as messengers linked emperor and government , however, the Aides as messengers li nked emperor and military . The Office of Aides was i nstitutionally unique in the Im­ perial Household Ministry : provisions governing the organization and duties of the Aides were established by I mperial Ordi nance, not by I mperial H ousehold M inistry decree . Personnel assigned to the palace as Aides were listed i n the Army and Navy rosters as well as in the official roster of the I mperial H ousehold M i nistry , i ndicating that the Aides were less court officials than representatives of the military at court .67 The organization of the Office of Aides-de-Camp i n 1 943 was es­ tablished by I mperial Ordinance Number 3 1 9 of December 2 8 , 1 908 , as ame nded i n 1 9 1 9 a nd 1 92 0 . That ord i na nce was countersigned by the Prime M i nister, M i nister of the Army , and M inister of the Navy-but not by the I mperial Household M inister.68 Its provisions reveal that control over the Aides rested only minimally with the Imperial Household M i nistry : 1 . The C hief Aide shall be a General or Lieutenant General of the Army or an Admiral or Vice Admiral of t he Navy , by direct i mperial ap­ pointment [shimp o ] . 2 . Aides shall be officers ranking from Army Captain or Navy Lieutenant 67 . There had been Aides to the emperor from at l east 1 87 5 . In 1 873 a dispute occurred over whether Aides should be officers of the Imperial Guard or of the armed forces. Yamagata Aritomo's argument for Aides appoi nted from the military services won the day . As early as 1 87 9 a Chief Aide-de-Camp was named directly. In 1 885 General Katsura Taro proposed the creation of an Office of Aides-de-Camp after i nvestigati ng European military organizations; his proposals apparently were the basis for the establishment of the Office of Aides-de-Camp which finally occurred in 1 896 (Matsushita Yoshio, 2: 205-06; Tokutomi I ichiro, Koshaku Yamagata, 2: 2 88-92 ; Nezu Masashi , p . 1 1 5) . 6 8 . "J ij ii Bukan Sei" , i n Kunai Sho, ed . , Genko Kunai Sho Hoki Shu, 2 vols. (Tokyo: Dai Nippon Horei Shuppa n Kabushiki Kaisha , 1 92 7 ) , with changes entered through 1 938.

82

3.

4. 5. 6.

TH E

P A LA C E

B U R EAUCRACY

to Army Lieutenant General or Navy Vice Admiral; there shall be five from the Army and three from t he Navy . The functions of t he Chief Aide and Aides are : a . ordinary services t o the emperor b. reporting a nd replying to the throne on military matters c. conveying military orders and commands d. accompanying the emperor at mil itary parades , maneuvers , tours, festivals, rites, banquets, audiences . The Chief Aide and Aides shall be dispatched to mil itary exercises and mil itary in­ spections . The Chief Aide supervises the work of the Aides . The C hief Aide and Aides are to be treated as members of the General Staff [ Samba ] . When t he C hief Aide a nd Aides are at court they shall observe the regu­ lations of the I mperial H ousehold M inistry .69

Only the last provision gave the I mperial H ousehold M i nistry any measure of control over the actions of the Aides . The Office of Aides-de-Camp was thus an extra-court institution imposed on the very center of court life . In addition to the eight Aides there was also a staff of seven military officials attached to the office .7° Furthermore, military Aides from the Army and the Navy were attached to every male of maj ority age in the imperial families . The regulations governi ng the duties of Aides to imperial family members were almost identical to those governing the Aides to the emperor. Like the ordinance creati ng the Office of Aides-de-Camp , these regulations were countersigned only by the Prime Minister, Army Minister, and Navy M inister.71 As table 5 indicates , the total number of mil itary appoi ntees at court varied from twenty-eight to thirty-eight during the period from 1 92 1 to 1 943 . Si nce there were only eight Aides to the emperor , they and their staff were only one of the military's wi ndows into court life. Moreover, the maj ority of Aides to the emperor as well as Aides to imperial family members were Army officers . After 1 93 1 the predominance of Army over 69. Ibid, paraphrased . 70. I bid . 7 1 . " Kozoku-tsuki Rikugun B ukan Ka nsei , " I mperial Ordinance no. 2 8 1 , August 1 0 , 1 896, as amended in 1 900, 1 91 5 , and 1 93 2 ; " Kozoku-tsuki Kaigun B ukan Kansei ," I mperial Ord i nance no. 3, October 1 4 , 1 897 , as amended in 1 899 and 1 9 1 6 (ibid . ) . I nterestingly, the rules for Aides to imperial family members were made by each of the two services separately and not by cosponsored I mperial Ordinance.

TH E

PALACE

83

B U R EA U C R A C Y

TABLE 5. AIDES TO THE EMPEROR AND IMPERIAL FAMILIES : ALTERNATE YEARS, 1 92 1 -1 943 Office of Aides (Aides and Staff) rear 1 92 1 1 923 1 925 1 927 1 929 1931 1 933 1 935 1 937 1 939 1 94 1 • 1 943•

Army

Total 38 35 33 30 28 28 30 31 31 29 29 29

24 21 20 19 17 17 20 21 21 20 20 22

(63 % ) (60% ) (6 1 % ) (63 % ) (6 1 % ) (6 1 % ) ( 67 % ) (68 % ) (68 % ) (69 % ) (69% ) ( 76 % )

Navy 14 14 13 II II II 10 10 10 9 9 7

( 37 % ) (40% ) ( 39 % ) (37 % ) ( 39 % ) ( 39 % ) (33 % ) (32 % ) (32 % ) (31 %) (3 1 % ) ( 24 % )

Total Army 15 19 18 15 14 13 13 13 13 13 13 13

10 12 II 10 9 8 8 8 8 8 8 8

Navy 5 7 7 5 5 5 5 5 5 5 5 5

Imperial Family Aides Total Army 23 16 15 15 14 15 17 18 18 16 16 16

14 9 9 9 8 9 12 13 13 12 12 14

Navy 9 7 6 6 6 6 5 5 5 4 4 2

SOURCE : Kunai Sho Shokuin Roku, 1 92 1-43. Aides attached to the Korean royal house have been excluded. It is interesting to note that the roster of officials for the govern­ ment in toto (Shokuin Roku) does not list the Aides under the Imperial Household Ministry but under the Army and Navy. The Imperial Household Ministry's official roster (Kunai Sho Shokuin Roku) lists the Aides under " attached office" and under each of the imperial princes. • In 1 94 1 and 1 943 one civilian was attached to the Office of Aides ; he is excluded from the tabulation.

Navy at court was even more marked , especially among Aides to im­ perial family members. That the i nitiative in appointi ng Aides to the emperor lay with the armed forces is strongly suggested by the fact that t he Aides were members of their respective General Staffs . General Honj o Shigeru , Chief Aide-de-Camp from 1 933 to 1 93 6 , has verified that during the 1 930s the military proposed and the emperor a nd court disposed on appointments to the Office of Aides . I n his diary entry of May 25 , 1 93 5 H onj o states that after considerable difficulty in arrivi ng at its selection of t he candidate to be its ranking Aide-de-Camp , the Navy finally proposed two candidates to the throne with clear preference indicated for the first nominee . The emperor, however, selected the

84

THE

P A LA C E

B U R EA U C R A C Y

second candidate, Rear Admiral Hirata N oboru . Honj o immediately notified the Navy of the emperor's decision. The Navy M i nister told Honj o that it was almost without precedent for the emperor to select the second nominee . Admiral Fushimi H iroyasu , Chief of the Navy General Staff, bel ieved that the emperor had q uestioned the " i n­ tegrity" of the Navy M i nister, or that someone in the palace was en­ gaging in disruptive activities . The Navy M i nister, however, told Honj o that the matter was closed because the emperor had acted .72 If i n fact this was an unprecedented case , it reveals that the military had more than the power to initiate the appoi ntments of Aides-de-Camp : a recommendation by the mil itary was cl oser to an order that , if not followed by the emperor and court , could create a difficult situation. The emperor did review cand idates for the post of Aide and had some voice in the matter, as the above case indicates . General Yoshihashi Kaiz o, Aide-de-Camp from December 1 944 to November 1 945 , has stated that the emperor checked the q ual ifica­ tions of each Aide prior to his appointment . In addition to " ability , " qualifications such a s " honesty" and the capacity t o perform liaison work reliably were required .73 Although the appointment of Aides was subject to a measure of negotiation between the military and the court , however, the military had the upper hand . Strengtheni ng the independence of the Aides as palace personnel was the transient nature of their service at court . The Aides were not " permanent " servants of the palace , as were the Chamberlains , but served a tour of duty at the side of the emperor. Chief Aide Honj o Shigeru , for example, obtai ned permission from t h e emperor in 1 934 to have Aides serve three-year tours to duty .74 I n 1 943 the eight Aides , including the Chief Aide , had been in office an average of sl ightly over two years , ranging from nine months to three years eleven months (the Chief Aide, General Hasunuma Shigeru) . Only the Chief Aide resembled anythi ng like a permanent official close to the throne . By contrast , we have seen that the fifteen Chamberlains had been i n office an average of nine years as of August 1 943 . 7 2 . Honjo Shigeru , Honjo Nikki (Tokyo: Hara Shobo, 1 967), p. 2 1 2 . 7 3 . I nterview with General Yoshihashi Kaizo, March 24 , 1 96 5 . A t the time of the in­ terview , General Yoshihashi was President of the Self-Defense Force Staff College. 74. Honjo Shigeru , p. 1 99 .

TH E

PALACE

BUREAUCRACY

85

M oreover, a tour of duty as Aide-de-Camp was usually followed by promotion in rank and advancement in the military hierarchy .75 Of the forty-three Aides w ho served from December 1 92 6 , when the era name changed to Showa , to November 1 94 5 , when the Office of Aides-de-Camp was abolished , seven were i n office w hen the office was abolished and ten were listed as havi ng been transferred without any further notation. Of the remaining twenty-six, twelve were transferred to commands.76 A tour of duty as Aide-de-Camp , therefore , was a stepping-stone to higher office within the military bureaucracy . To take an extreme example, General Anami Kore­ chika became Army M i nister under Prime M i nister Suzuki Kantaro in April 1 945 ; Anami had been an Aide-de-Camp from August 1 92 9 t o August 1 93 3 during Suzuki's tenure a s Grand C hamberlai n . Appointment to the post of Aide-de-Camp sometimes involved status as well as strictly military q ualifications . I n 1 943 , for example, at least one of the Aides was an aristocrat : Baron Lieutenant Colonel Yamagata Ariakira , a grandson of Pri nce Yama­ gata Aritomo. A glance at the roster of forty-three Aides between 1 92 6 and 1 943 reveals a few more aristocrats. Viscount Machij i ri Kazumoto, who eventually became a genera l , was an Aide from May 1 930 to March 1 93 5 and agai n from March 1 93 7 to October 1 93 7 . A career Army officer, Machij iri was married to the eldest sister of Prince-of-the-B lood Kaya Tsunenori and was considered i nfl uential because of his marriage . Marquis Rear Admiral Daigo Tadashige , Aide from December 1 93 8 to October 1 94 1 , was of Court Noble origi n : his wife was the eldest daughter of Pri nce Mori M otoaki . I n addition t o those of aristocratic background , there were also Aides related to officials who had been close to the throne . Hirata Noboru , a naval officer who served as Aide from 1 93 5 to 1 93 9 , for example , was the son of former Lord Keeper of the Privy Seal Hirata Tosuke.77 Given the emphasis on military careers for aristocrats , however, the number of aristocratic Aides was not extraord i nary . It is quite possible that more military officers of aristocratic origi n 7 5 . Yoshihashi i nterview ; i nterview with Chamberlain Irie Sukemasa , 1 964 . 7 6 . Photocopy of a document written by General Yoshihashi Kaizo on I mperial Household Agency stationery [ 1 965 J . 7 7 . I bid . ; Yoshihashi interview .

86

TH E

P A LA C E

B U R EA U C R A CY

served tours of duty as Aides-de-Camp than did those of common origin . As I rie Sukemasa remarked , only "the best" were allowed to become Aides .78 I n general, however, the Aides-de-Camp, who managed the i nner court 's liaison with the military commands , were very different from t he C hamberl a i n s , w ho ma naged access t o t he Emperor- i n ­ C hambers and were primarily responsible for liaison between the in­ ner court and civilian personnel , both court and government . The Aides were career military officers whose service at court was tempo­ rary . The C hamberlai ns were overwhelmingly civilian and their positions were all but permanent . A tour of duty as an Aide was in most cases a step up in the mil itary , not the palace bureaucracy . There was virtually no upward mobility among the Chamberlains , at court or elsewhere . Although some Aides were of aristocratic origi n, they were far less aristocratic than the Chamberlains . Fi­ nally, all the Aides were graduates of t he service academies. Only two of the fifteen C hamberlai ns in 1 943 were products of the aca­ demies . INSTITUTIONAL PLURALISM AT COURT

The Office of Aides-de-Camp exemplifies the presence at court of institutional representatives from the " outside" -personnel from the numerous i nstitutions of imperial prerogative beyond the palace gates . Since the emperor was the supreme commander of the armed forces , the General Staff officers of the Office Aides, w ho linked the emperor to the operational commands of the military establishment , could be fully j ustified on constitutional grounds . B ut the emperor also possessed extensive prerogatives in government admi nistration , foreign affairs, and legislation. By the same logic that led to the presence of the military at court , one would expect that the other basic i nstitutions of imperial prerogative-the government mi nistries and the I mperial Diet-would also be represented there . By the 1 930s the court was in fact a mosaic of such institutional representatives . This was especially true of palace bureaucrats i n 7 8 . I rie i nterview .

TH E

PALACE

BUREAUCRACY

87

positions of responsibility. M oreover, the higher the position of a palace official , the more likely it was that he had had extensive bu­ reaucratic experience outside the palace . As we shall see in the follow i ng chapter, for example, the four top palace officers after 1 927 were invariably brought into the palace directly from high positions in the government and military bureaucracies . B ut the " outside" i nstitutional representation at court was se­ lective. Not all institutions of imperial prerogative were represented , nor were they represented equally. Table 6 illustrates the nature of institutional representation at court by giving a statistical summary of the careers and backgrounds of eighty-three palace officials i n responsible positions a t court during 1 943 , from the top four palace officers down to and including the officials who managed the various sect ions of w hich the I mperial H ousehold M i nistry ' s boards , bureaus , and offices were composed . Ladies-in-Waiti ng and some thirty palace advisers have been excluded from the tabulation. Of the eighty-three persons i ncluded , fifty-seven were section chiefs or above , fifteen were Chamberlains , eight were Aides-de-Camp, and three were court physicians . Thirty-six of the eighty-three ( 43 .4 percent) had had at least one year's experience i n the civil bureau­ cracy prior to entering the I mperial Household M inistry . Twelve ( 1 4 . 5 percent) had come from the military establishment , and ni ne (1 0 . 8 percent) had some degree of other professional experience, such as teaching in private universities . Thus , 6 8 . 7 percent of the Im­ perial Household M i nistry's leading officials in 1 943 came from the "outside . " Those whose careers were apparently confi ned exclusively to the palace numbered only twenty-six (3 1 .3 percent) . The personnel mosaic at court was essentially made up of the civi l , court , and military bureaucracies , these three clusters ac­ counting for 89.2 percent of the leading palace officers in 1 943 . Most numerous were representatives from the civil bureaucracy (43 . 4 percent) , w h o outnumbered their " outside" colleagues from the military bureaucracy by three to one. At least seven of the eleven ministries of t he central civi l i a n bureaucracy in 1 940 were represented . One of those ministries , the Home Mi nistry , was the most strongly represented of all the outside institutions of imperial prerogative, civilian or mil itary : 1 8 . 1 percent of the leading palace

88

TH E

P A LA C E

BUREAUCRACY

TABLE 6. CAREERS, EDUCATION, AND STATUS OF LEADING IMPERIAL HOUSEHOLD MINISTRY OFFICIALS, 1 943 (83 PERSONS) Number of persons

Percentage

CAREER BACKGROUN D :

Civil Bureaucracy Home Ministry Ministry of Agriculture and Commercea Finance Ministry Foreign Ministry Ministry of Justice Ministry of Communications Ministry of Education Court Bureaucracy Military Bureaucracy Army Navy Otherb Business and banking Medical Education Mass media

Total

36

43 . 4 15 6 4 3 3 3 2

26 12

18. 1 7 .2 4.8 3 .6 3.6 3.6 2 .4 31 .3 14 . 5

7 5 9

8 .4 6.0 10.8

3 3 3 0 83

3 .6 3 .6 3.6 0 1 00 . 0%

LEGISLATIVE EXPERIENCE :

House of Peers House of Representatives None Total

5 0 78 83

6.0 0 94 . 0 1 00 . 0%

COLLEGE EDUCATION :

Tokyo Imperial University Other state universities Military academies Private universities None Total

49 15 12 4 3 83

59 . 0 18 . 1 14 . 5 4.8 3 .6 1 00 . 0%

TH E

PALACE

89

BUREAUCRACY

TABLE 6.-Continued Number of persons

Percentage

STATUS:

Titled or in line to succeed but not taxpayers Titled taxpayers Titled subtotal Nontitled taxpayers Taxpayer subtotal Neither titled nor taxpayer Total

12

14 . 5 15 . 7

13 25

25

30 . 1 30 . 1

38

33 83

45 . 8 39 . 8 1 00 . 1 %

souRCE : Kunai ShO Shokuin Roku, 1 943, and terse data available in Japanese bio­ graphical dictionaries and palace files . Given the limited information available, the information presented in table 6 may not be very complete. It is highly likely, for ex­ ample, that many more than 13 of the 25 aristocrats had tax obligations. N OTE : 1 943 has been chosen because one might expect that in the midst of Japan's greatest war military representation at the side of the emperor would be greater than at any other time. This was obviously not the case. From my perusal of personnel lists and official rosters after 1 920, it would appear that 1 943 was not unrepresentative of the personnel complexion of the Imperial Household Ministry after 1 920, when the institutionalization of imperial prerogatives was largely completed. • I ncludes the Ministry of Agriculture and Forestry and the Ministry of Commerce and Industry, which were created from the Ministry of Agriculture and Commerce in 1 925. b Other is not quite synonymous with " private sector," since private educational institutions were not entirely free from regulation by the Ministry of Education and certain business and banking activities were closely involved with the state bureaucracy.

officials in 1 94 3 had served i n the H ome M inistry prior to entering the I mperial Household M i nistry . Notably underrepresented was the Imperial Diet . None of the eighty-three officials had ever been elected to t he House of Representatives . O nly five were from the House of Peers, and those five, two princes a nd three marquis, served exclusively by right of hereditary title. None of the eighty­ three ranki ng officials of the palace in 1 943 had ever been elected by any kind of electorate to any kind of political office. Finally, representatives from the " private sector" of Japanese so­ ciety-mass media (0 percent ) , private education ( 3 . 6 percent) ,

' Il l

THE

P A LA C E

B U R EA U C R A C Y

a nd banki ng (3 . 6 percent) , and other professions (3 . 6 -were also underrepresented . The palace bureaucracy was I " · , t . 1 1td foremost a mosaic of representatives of the bureaucratic in­ ·.t tl 1 1 1 i o ns of imperial prerogative . Underlini ng the bureaucratic ' l t. t ral'lcr of these palace officials was the fact that in 1 94 3 , 5 9 . 0 p•·rn·nt had graduated from prewar Japan's most prestigious state u n iversity, Tokyo I mperial University , w hile only 4 . 8 percent were grad uates of private universities . Si nce the imperial i nstitution was the state writ small , it is not un­ natural that offices resembli ng those of the state existed at court . B ut it does not necessarily follow that officials from the state bureaucracy should serve in the Imperial Household M inistry , however closely state and court offices resembled each other. In other words, an explanation is required for the relatively low percentage (3 1 . 3 percent) of purely court bureaucrats in the mi nistry and for the se­ lective representation of outside i nstitutions . I n the first place, the palace was the hub of the negotiation process in prewar Japanese politics . The theory of legitimacy set forth in the Meij i Constitution, if practiced , would make the imperial i nstitution the ultimate locus of political resolution because it was at court that policies were to receive fi nal ratification. To have representatives at court from the maj or political components i nvolved in policy maki ng might conceivably facilitate t hat process . The existence of the Office of Aides·de-Camp a nd the B oard of C hamberlains attests to the i nteraction of the military and civil bureaucracies with the court . All of the Aides were career officers i n the Army or Navy and four of the fifteen Chamberlains in 1 943 had prior careers i n the civil bureau­ cracy . It was the emperor's role as Emperor-i n-State that made such offices and outside representation necessary . The absence of officials with any experience in the elected H ouse of Representatives and the small number and hereditary nature of those from the H ouse of Peers suggests that at least by 1 943 the legislature simply was not important to the process of political ne­ gotiation a nd ratification that centered on the Emperor-i n-State. And the 1 943 composition of the palace bureaucracy was not , i n fact , atypical of the I mperial Household Mi llistry throughout the prewar period , especially after imperial prerogatives had become thoroughly l o l l 'o l nt·ss I " ' ' ,. , , , )

TH E

PALACE

B U R EA U C R A C Y

91

i nstitutionalized in the 1 920s. As we shall see in the following chapter, for example, none of the four top palace officers from 1 885 to 1 945 had ever been elected to the H ouse of Representatives . Even though the H ouse of Representatives had acquired considerable power by 1 92 0 , it never ma naged to p lace its i nstitutional representatives i n positions of power at the side of t he emperor. The selectivity of i nstitutional representation at court substantiates the profoundly bureaucratic nature of the political process in prewar Japan . Although a law , to be a law , required passage by the Im­ perial Diet , the role of law per se was very limited , and all laws re­ quired imperial sanction as well . Such vital decisions as declarations of war and appointments of Prime M i nisters were always made by the responsible M inisters of State, military chiefs , and imperial advisers and then ratified at the palace , not i n the I mperial Diet . To exaggerate , it was t he i mperial seal t hat u l t i mately made a government policy t he legitimate policy of t he state , not its passage by t he I mperial Diet . The imperial i nstitution was intricately i nterwoven not only with the government but also with society at large . The arrangements for imperial tours by the Emperor-in-Court , for example, c � uld" best be made by officials familiar with l ocal operations; hence the impressive number of ex-Home M inistry officials among the leading court of­ fici a l s . T he same applies to ma nagement of imperial house properties scattered throughout Japan and the number of officials at court from t he M inistry of Agriculture a nd Commerce. Finally, the imperial i nstitution was the ideal polity writ smal l . In addition to the bureaucracy , representatives of status a nd wealth were prominent at court . At least 30. 1 percent of the leadi ng palace of­ ficials in 1 943 were titled or in l i ne to succeed, and at least 4 5 . 8 percent were wealthy enough to pay taxes . The I mperial H ousehold M i nistry was t herefore an i nt ricate mosaic of burea ucracy , aristocracy , and pi utocracy-the elements i n prewar Japanese so­ ciety most highly valued as components of t he " ideal . " B ut i f t hese court officials symbolized t he " ideal" structure of the prewar Japanese polity , they also operationalized that symbol both in government and i n society. Guardians of the emperor's personal will , they were also responsible for translating the imperial symbol

92

THE

PALACE

B U R EA U C R A C Y

into the I mperial Will. By the 1 930s the human mosaic at court no longer reflected a unified and stabilized agreement concerning the privileged status of the Emperor-i n-Chambers . With the gradual disappearance of a unified elite in government after t he turn of the century , the imperial i nstitution itself lost its internal unity. By the 1 930s struggles for supremacy among t he institutions of imperial prerogative outside the palace had i nfected even the lower levels of the palace bureaucracy. The following incident reveals the extent to which the walls around the Emperor-in-Chambers had been breached . In October 1 93 1 a Maj or Endo Saburo called on his section chief in the General Staff Of­ fice, Colonel Imamura Hitoshi . 79 Had Col onel I mamura ever met Sekiya Teisaburo, the Vice Mi nister of the I mperial Household M inistry , Maj or Endo asked . I mamura repl ied that he had met Sekiya but once , and only formal ly. Endo then told Imamura that after he had returned from Europe with Sekiya 's son he had been a regular visitor to the Sekiya house . Suddenly, conti nued Endo, Sekiya had cal led him to his official residence and asked him about the October I ncident , a plot by young Army officers to overthrow the government in October 1 93 1 . Sekiya inquired if it were true that one objective of the plot was to " rectify" the I mperial Household M inistry . Si nce Sekiya had heard from the police that Endo's supe­ rior, Colonel Imamura , was investigating the matter, he asked Endo to make an appointment for him to see I mamura . Commenting, Sekiya said that he was not worried about the fate of the I mperial Household M i nistry but was concerned over "causing a dist urba nce at the emperor's side . " After delayi ng two or three days because Ima­ mura was so " busy , " Endo had fi nally come to Imamura wit h Sekiya 's req uest . I n response , I mamura told E ndo that he had heard that Sekiya was a practici ng Christian, but that he would meet him after checking with the Military Affairs Bureau in the Army M inistry . This precaution was necessary , claimed I mamura , because to discuss 7 9 . The following account is from I mamura Hitoshi, Kozoku to Kashikan (Tokyo: Jiyii Ajia Sha, 1 960), pp. 234-44 . Although M aj or Endii is listed in the 1 93 1 and 1 932 Shokuin Roku as a member of t he General Staff Office, Colonel I mamura is not . Accordi ng to I ma­ mura's account, he became Chief of the Strategy Section, General Staff Office, on August 1 , 1 93 1 but was transferred to t he Shanghai front in February 1 932 . This would explain his absence from both official rosters, w hich are dated July ( I mamura H itoshi, pp. 1 92 , 2 5 8 ) .

TH E

PALACE

B U R EA U C R A C Y

93

the October I ncident with " outsiders" required permission from the bureau, which had general j urisdiction over such i ncidents . The bu­ reau consented to I mamura's meeti ng with Sekiya. I mamura then had E ndo arrange a meeti ng with Sekiya for 8 : 30 p . m . at Sekiya's official residence, t he time and place havi ng been selected so that few would see them meeting. W hen they met , Vice M i nister Sekiya told I mamura that persons in t he Metropolitan Police Office had told him that Colonel I ma­ mura had tracked down the plan of the October I ncident written by its organizers , and that , after divulging its contents only to the top Army leaders , the C hief of the Military Affairs Bureau, General Koiso Kuniaki, had burned the document . Sekiya said it had bee n discovered that the radicals had i ntended to discipline the I mperial H ousehold M inistry thoroughly. Colonel I mamura was dismayed that there might be persons withi n the Army who had set such ru­ mors flying and that t he Metropol itan Police Office should have spied out the plan of the radicals . He then denied that the organizers were in any way disloyal , asserting to Sekiya that those of the twelve arrested whom he knew were loyal and far above the ordi nary in i n­ tegrity , however improper their met hods. B ut it was true, I mamura conti nued , that the plotters had intended to rectify the I mperial H ousehold M i nistry . He guessed that t he I ncident 's organizers harked back to the early Meij i period , when the emperor was surrounded by talented men " of the people" ; at present those at court were mostly aristocrats , and " persons truly able to render assistance to His M ajesty are not to be found . " Praising their " l oyal hearts , " I mamura then warned Sekiya that many officers of the I mperial Guards had been involved in the blood­ sealed pact among the organizers . Knowing full well w hat transpired at the palace , Imamura stated , the officers of the Imperial Guards detested the officials of the I mperial Household M i nistry . Asked to give the reasons for this hatred , I mamura replied that ac­ cording to what he had heard , the officers of the I mperial Guards felt that I mperial H ousehold M inistry officials treated the emperor as "their emperor, " separati ng him from the eyes and hearts of the people . With police and officials surrounding him every mi nute , the tl t"l : 0

r

C"J trl

>

r

>

'tl

...,

0

1:'1 " en

ri

..., ...,

0

> z 0

C"J 1:'1 en

;!

...,

0

Ishiwatari Setaro

Fujita Hisanori

Hata Shunroku Hasunuma Shigeru

Position

Kido Koichi

Yuasa Kurahei

IHMD ; Shokuin Roku ; Ijiri Tsunekichi, Rekidai Kenkan Roku (Tokyo : Choyokai, 1 925) ; and Toyama and Adachi. N OTE : Ito Hirobumi had headed the Imperial Household Ministry since 1 884 and Tokudaiji had been Grand Chamberlain since 1 87 I . In 1 936 Ichiki Kitokuro was appointed Privy Seal for one day as a procedural step to have the murdered Saito replaced. The list of Chief Aides is as accurate as possible, given the lack of information on this office available from the Imperial Household Agency before 1 926.

souRCE :

1 940 1 94 1 1 942 1 943 1 944 1 945

1 937 1 938 1 939

rear

TABLE 7.-Continued

:r ,

:;!l til

t!l )o t:l t!l

r

)o n t!l

r

, )o

'"rl

0

:;!l til

� n t!l

'"rl

0

)o z t:l

til

� n t!l

'"rl

0

0

.....

.....

OFFICES

AND

OFFICERS

OF

PALACE

L E A D E R SH I P

111

rently Grand C hamberlain, leadership at the palace like that in the government was firmly in the hands of the oligarchs , who either held Cabi net offices ( 1 8 8 5 - 1 901 ) or exercised control i ndirectly as Elder Statesmen and via proteges in the Cabi net ( 1 90 1 - 1 2 ) . Palace leaders were securely under the oligarchs in terms of both status and alle­ giance throughout the Meij i period , despite the fact that , with the one exception of Ito Hirobumi from Choshu , they were either Court No­ bles or restorationist samurai from other than Choshu or Satsuma , the two domai ns that had spearheaded the Restoration. That a Court Noble, Prince Tokudaij i Sanenori , was able to serve concurrently as both Grand Chamberlain and Privy Seal for over twenty years , from 1 89 1 to 1 9 1 2 , attests to the solidity of the Restoration settlement as far as palace leadership was concerned . The first ten years of the Taisho period , from 1 9 1 2 to 1 92 1 , were , however, years of transition and confusion in the nature of palace leadership . This is indicated by the high rate of turnover in the leading palace offices (a change in one of the four officers every 3 . 4 years) and the diversity i n origi ns and status of t hose holdi ng leading palace posts during that period . A political protege of oligarch Yama­ gata Aritomo (Katsura Taro) , oligarchs as Elder Statesmen ( O yama lwao and Matsukata Masayoshi) , a pri nce-of-the-blood (Fushimi Sa­ danaru) , former Court Nobles ( O gimachi Sanemasa and Takatsukasa Hiromichi ) , successful civil bureaucrats (Watanabe Chiaki and Ha­ tano Takanao) , and creatures of the military bureaucracy (Uchiyama Koj iro and Nakamura Yujiro) all found their way into the four leading palace offices . There seems to be no pattern in their appoi nt­ ments-no effort , for example, to balance institutional forces at the side of the emperor-although Yamagata and his allies apparently domi nated at court . Judged by the marked increase i n personnel changes and the diversity of palace leadership between 1 9 1 2 and 1 92 1 , the oligarchic unity surroundi ng the throne up to 1 9 1 2 was clearly lost . Those close to the throne were no l onger exclusively the band of Restoration " stalwarts" praised by Kanroj i Osanaga .9 With the appoi ntment of Count Maki no N obuaki as I mperial Household M i nister in February 1 92 1 , palace leadership entered a new era . The leaders who domi nated the palace from Makino's ap9 . See above,

p. 7 7 .

1 12

OFFICES

AND

O FFICERS

OF

PALACE

L EA D E R S H I P

pointment until 1 936 were constitutional monarchists-proponents of M inobe Tatsukichi 's theory of constitutional monarchy or a more conservative variant-led outside the palace by Pri nce Saionj i Kim­ mochi and i nside the palace by , most notably, Count Maki no, B aron lchiki Kitokuro, and Admiral Suz uki Kantaro . After the death of Prince Matsukata Masayoshi in 1 924, moreover, the powers pre­ viously held collectively by the Elder Statesmen devolved solely on Prince Saionj i as the " last of the Elder Statesmen . " From 1 92 1 to 1 93 6 , therefore, palace leadership was primarily i n the hands of a small group of partisan bureaucrats under Pri nce Saionj i ' s tutelage who were the products of Japan's new bureaucratic and educational system. With one exception they were political leaders brought i nto the palace from successful careers " on the outside . " A measure of stability at the side of the emperor was also restored by this like­ minded group of palace officers . Between 1 92 1 a nd 1 936 a change i n one o f the four palace officers occurred o n an average o f once every five years . The young officers' unsuccessful but bloody coup of February 2 6 , 1 93 6 , followed b y t he only complete shift i n palace leaders during the entire period from 1 885 to 1 945 , marks the start of the last phase i n the evolution o f palace leadership. The new palace group emerging duri ng the 1 930s, the traditionalists, achieved preemi nence i n court councils with the appoi ntment of M arquis Kido Koichi as Privy Seal in 1 940 and the death of Prince Saionj i that same year. Led by Pri nce Konoe Fumimaro outside the palace and by Marquis Kido inside, the traditionalists disavowed the partisan stance of the constitutional monarchists . In effect , they restored the " neutrality" of the throne in politics by recommendi ng personnel and policies " appropriate to the new times . " As such, the traditional ists were concerned more with na­ tional unity than with substantive policies , more with discovering and havi ng the emperor ratify a true " national consensus" than with pursuing a specific set of domestic or foreign policies . The constitutional monarchists persisted in their policy and person­ nel preferences until the assassinations a nd threats of assassi nation from 1 930 to 1 93 6 forced them to desist . One reason why the tradi­ tionalists , Kido Koichi in particular, took a " neutral" stance was precisely to preserve t he transcendental position of the emperor : to

OFFICES

AND

OFFICERS

OF

PALACE

L E A D E R S HI P

113

persist i n recommendi ng to the throne "liberal " domestic policies and " pro-Anglo-American" foreign policies was to invite assassi nation not only of government leaders but also of the leading palace officers , thereby jeopardizing the emperor and the imperial institution . The traditionalists , like the constituti onal monarchists at court , were all creatures of Japa n's modern bureaucracy . They also main­ tained the same balance of institutional forces at the emperor's side that had been established by the constitutional monarchists. Whereas the constitutional monarchists had manipulated this institutional balance as best they could to realize their partisan purposes , however, t he traditionalists used it to ensure the accuracy of the " national consensus" on which they based their recommendations to the throne . Since the traditionalists were " neutral , " moreover, they could allow new policies and men to lead the nation i n accordance with the "trends of the times" revealed by the national consensus , regardless of the direction of those trends. Thus Privy Seal Kido, as the emperor's chief adviser from 1 940 to 1 945 , could preside over both the com­ mencement of war agai nst the United States in 1 94 1 and the sur­ render of Japan in August 1 945 . In short , he was able to recommend the appointment of a Prime Mi nister who would be able to lead the nation i nto war " i n accord with the times" in 1 94 1 and one who would be able to surrender the nation " i n accord with the times" in 1 945 , before the nation was obliterated . That meant General Toj o in 1 94 1 and retired Admiral Suzuki Kantaro i n 1 94 5 . Finally, a s i n 1 9 1 2 -2 1 , the rise of the traditionalists was marked by great i nstability in the four palace offices , chiefly as a result of the Feb­ ruary 2 6 I ncident of 1 93 6 . Between 1 93 6 and 1 940 a change in one of the four officers occurred every 2 . 3 years . With Kido's appoi ntment as Privy Seal i n 1 940 and the Japanese government under the near hege­ mony of the military , stability was once again restored at the side of the emperor. Accepti ng this summary statement about the evol ution of palace leadership from 1 88 5 to 1 945 for the moment , what general charac­ teristics did all four palace leaders have in common throughout that period ? First , to poi nt out what palace leaders were not , it is most conspicuous that none had ever held elective public office (excluding those elected to the House of Peers from among their aristocratic peer

1 14

OFFICES

AND

OFFICERS

OF

PALACE

LEAD E R SH I P

group) . Second , only five of the thirty-seven men ( 1 3 . 5 percent) who held the top four positions at court from 1 88 5 to 1 945 could possibly be identified as career court figures : Tokudaij i Sanenori , Iwakura To­ m os a d a , T a k at s u k a s a H i r o m i c h i , T o k u g a w a S ad a t a k a , a n d O gimachi Sanemasa. All o f the others were brought into positions of leadership at the emperor's side from successful careers outside the imperial household. After 1 927 not a si ngle appointment was made from among career court bureaucrats, although some of those ap­ pointed served for a time in the palace prior to assuming one of the four top positions . Table 8 presents career data on the thirty-seven men who held one or more of the top four offices at court from 1 88 5 to 1 945 . Quite ob­ viously, palace leadership was composed predominantly of members of the House of Peers with considerable bureaucratic experience and achievement : 5 9 . 5 percent were at one time or other members of the House of Peers , 4 5 . 9 perce nt Privy Counci l l ors , 3 7 . 8 percent Ministers of State , and 2 7 . 0 percent Supreme War Councillors. Si nce the number of times a person held high official position is also in­ dicative of his political success or importance , column 5 has been in­ cluded in table 8 . Most revealing i n this respect is the fact that the fourteen who had served as M i nisters of State served thirty-seven times in that capacity, suggesting a high degree of political success and competence prior to enteri ng the palace . That the thirty-seven palace leaders should have among them six Prime Ministers , servi ng in that capacity twelve t i mes , is also a measure of t he high political achievement of the group i n general . Table 8 (columns 3 and 4) also reveals that appoi ntment to one of the four leading palace offices generally termi nated the given person's official public career. Of the 1 2 6 times the thirty-seven palace leaders served in high government office, 1 04 times (82 .5 percent) occurred prior to or concurrent with holding top palace office; only 22 times ( 1 7 .5 percent) occurred after leavi ng palace office . Si nce the average age of the palace leaders upon assumi ng office was 6 1 .7 years , there was , of course, not much time left for future political aspirations . I n fact , thirteen o f the thirty-seven (35 . 1 percent) died within two years of leavi ng office . The career backgrounds of the palace leaders also suggest that the

OFFICES

AND

OFFICERS

OF

P A LA C E

115

L EA D E R SH I P

TABLE 8 . HIGH GOVERNMENT OFFICES OF THE FOUR LEADING PALACE OFFICIALS, 1 885-1 945

(37 PERSONS) Number of Times in Position

Number• of Persons (2)

Before or Concurrent with Palace Office (3)

6 ( 16 . 2%) 1 4 (37 . 8 % )

7 36

5

12 (9 . 5 % ) 3 7 ( 29 . 4 % )

( 10 . 8% ) (45 . 9 % ) (27 . 0 % ) (2 1 . 6 % ) (59 . 5 % ) (8 . 1 % ) (0 % )

2 14 8 ll 22 4 0

5 6 3

0 0

7 20 ll 12 23 4 0

1 04

22

Position (l)

Prime Minister Minister of State President of the Privy Council Privy Councillor Supreme War Councillor Top military postsh House of Peers• Ambassadors House of Representatives Total

4 17 lO 8 22 3 0

After Palace Office (4)

Total Number of Times (5)

(5 . 6% ) ( 1 5 . 9% ) (8 . 7 % ) (9 . 5 % ) ( 18 . 3% ) (3 . 2% ) (0 % )

1 26 ( 1 00 % )

soURCEs : Miwa Kai and Philip B. Yampolsky, Political Chronology of Japan, 1885-1957 (New York : East Asian Institute of Columbia University, 1 957) ; Toyama and Adachi ; Gikai Seido Shichijii Nen Shi (Ki:r.okuin-Sangiin Giin Meikan) , ed. Shiigiin-Sangiin (Tokyo : Okura Sho Insatsu Kyoku, 1 960) ; biographical dictionaries in Japanese and English. • Since almost all of the 37 palace officers served in more than one important govern­ ment position, the totals for column 2 add up to more than 37 and 100 percent. b For the purposes of tabulation, "top military posts" were limited to the following : Chief of the Army General Staff, Chief of the Navy General Staff, Chief of the Military Affairs Bureau (Army Ministry) , Inspector General of Military Education, and Chief of the Naval Affairs Bureau (Navy Ministry). • Only one person, Baron General Nakamura Yiijiro, was elected or appointed to the House of Peers twice. Of the 23 times the 37 leading palace officers served in the House of Peers, 9 were by imperial appointment, 10 by right of title, and 4 by election (among counts, viscounts, and barons).

palace leadership was even more a mosaic of the valued i nstitutional forces in prewar Japanese political society than was t he second echelon of palace officialdom described in chapter 3 . In terms of achievement , it was a far more prestigious group of military and ci­ vilian leaders ; fully twenty-eight of the thirty-seven (75 .7 percent)

116

OFFICES

AN D

O FFICERS

OF

P A LA C E

LEA D E R SH I P

held a t one time o r another one o r more of the highest military o r ci­ vilian offices , excl uding ambassadorships and memberships i n the H ouse of Peers . Palace leaders were also of high status . Of the thirty-seven, at least thirty ( 8 1 . 1 percent) achieved or inherited titles . Of these thirty , one was a prince-of-the-blood , eight were pri nces , one a marquis, eight counts , four viscounts , and eight barons . I n proportion to the total aristocracy (table 4) , palace leaders were drawn more often from the upper than from the lower artistocracy in terms of titles ulti mately achieved . Palace leaders were also draw n more from the merit than the hereditary aristocracy : twenty-six of the thirty acquired their titles as rewards for service to the state . Top palace office was not generally a refuge for the scions of aristocrats . Palace leaders were , however, linked by family ties not only among themsel ves but also with political and social leaders outside the palace . I n the Meij i period , Pri nce Tokudaij i Sa nenori epitomized the i ntri­ cate web of family that connected Japan's sociopolitical elites by mar­ riage and adoption. One of Tokudaij i 's younger brothers was Pri nce Saionj i Kimmochi ( 1 849- 1 940) , who was closely aligned with Ito Hirobumi 's side of the oligarchy . 10 The political opponent of Yama­ gata Aritomo's willful protege , Katsura Taro, Saionj i alternated as Prime M inister with Katsura between 1 90 1 and 1 9 1 2 . As mentioned previously, Saionj i became the leading spirit of the constitutional monarchists at court and leading adviser to the throne after the last of the oligarchs, M atsukata M asayoshi , died i n 1 92 4 . Another of T o k ud a ij i ' s b r ot hers was B a r o n S u m i t o m o K i c h i z a e m o n ( 1 864- 1 92 6), a leadi ng figure i n one of prewar Japan's four leading i ndustrial and fi nancial combi nes . Tokudaij i 's eldest daughter mar­ ried Pri nce Takatsukasa Hiromichi ( 1 8 5 5 - 1 9 1 8 ) , Grand Chamber­ lain from 1 9 1 2 to 1 9 1 8 . Tokudaij i 's second daughter married a mar­ quis sci on of the feudal domain of Akita ; his third , a M itsui of the M itsui combine; his fourt h, a viscount scion of a nother feudal do­ main; and his fifth, a prince of the Satsuma domai n . One of Tokudaij i 's sons became a consultant (goyogakari) to the Household 1 0. Saionj i served in three of Ito's four Cabinets and Matsukata Masayoshi's second Cabi net ; he never served under Yamagata or other u ncomprom i s i ng ad vocates of "transcendental Cabi nets . "

OFFICES

AND

OFFICERS

OF

PALACE

LEA D E R SH I P

1 17

Ministry , another a Chamberlain, and the third a director of Mit­ subishi Heavy I ndustries . Tokudaij i was thus related directly to top court and government leaders , scions of the old feudal domai ns, and leaders in three of the four maj or industrial-fi nancial complexes in prewar Japan . Japan's last Privy Seal ( 1 940-4 5 ) , Marquis Kido Koichi (b. 1 889) , was almost equally well connected . His wife was the fifth daughter of General Count Kodama Gentaro ( 1 852- 1 9 1 6) , Army M i nister from 1 900 to 1 902 . Kido's sister, Yaeko, married Kodama 's fourth son, Kodama Tsuneo, an Army officer and one-time director of Manchuria A i rl i nes . A mong Kido's brot hers-i n-law were Kodama Hideo ( 1 876-1 947) , Mi nister of Overseas Affairs from 1 934 to 1 93 6 and subseq uently three ti mes M i nister of State ; Kodama Kyiiichi , a Home M inistry bureaucrat ; Kodama Kunia and Kodama Sadao, both busi ness executives , and Kodama Tomoo, one time commander of the Japa nese Army i n Taiwan, who was married to the third daughter of B aron General Nakamura Yiij iro, Household Mi nister from 1 920 to 1 92 1 and a protege of Yamagata Aritomo. Kido's younger brother, Wada Koroku , was a famous aeronautical engi neer. Kido's eldest daughter, Takiko, married the eldest son of General Abe Nobuyuki , Prime Mi nister i n 1 93 9-40 . The family relationships of Pri nce Tokudaij i and Marq uis Kido illustrate the iron law of family that li nked Japa n's sociopolitical elites throughout the prewar period . Family connections served to unite merit and ascribed status as well as to link the components of Japanese society laterally across the top. But much too much behavioral cau­ sality has been attributed to family connections in Japan. Family only opened up opportunities ; it did not ensure success . On the contrary , the great use of marriage and adoption may have even assured the in­ fusion of talent and ability into lead ing social and political roles . The four leadi ng officials at court , in summary , had generally achieved high bureaucratic positions outside the palace prior to be­ comi ng palace leaders , were representatives of the military and government bureaucracy as well as the interlocki ng merit and ascribed aristocracy , and termi nated successful public careers " at the side of the emperor. " Why the palace req uired such capable public leaders from the outside is answered in part by the theory of legitimacy , which

118

OFFICES

AND

OFFICERS

OF

P A LA C E

L E A D E R SH I P

made the imperial institution the center of political authority. I f in fact that theory were operative , all acts of government would req uire ra­ tification by the I mperial Will . And if the Imperial Will were to have authoritative meaning, ratification would req uire a high degree of political skill at court-to ensure that the emperor ratified a generally agreed upon policy (a consensus) or that he ratified a partisa n policy promoted by palace leaders in alliance with policy advocates in the government proper. That bureaucrats from the top strata of Japanese political society held or controlled the four leading palace offices from 1 885 to 1 945 is prima facie evidence that such a theory was indeed operative and that government leaders held to the theory of imperial prerogative with tenacity. THE I M PERIAL HOUSEHOLD M I N I STER : C H I E F M ANAG ER OF T H E EM PEROR-IN-COURT If you were to ask me I 'd say that it 's by havi ng the M i nister and Vice M i nister brought i n from the outside that the I mperial Household M i nistry gets its raison d 'etre as a n " I mperial Household M i nistry i n l i ne with the times" in the fi rst place . To cover its tracks , as it were, with persons from within the mi nistry-si nce neither a Kido nor a Hirohata are in fact outside people-a nd not to bri ng i n people from the outside would be tantamount to letti ng the I mperial H ousehold M i nistry dig its ow n grave . If that know-nothing bunch [ i n the mi nistry ] alone had their ow n way, wouldn't they ultimately deprive themsel ves of their ow n raison d 'etre ? Pri nce Saionj i Ki mmochi 1 1

The pol icy o f selecting the Imperial Household Mi nister from out­ side the " know-nothing" career officials at court was consistently pursued from 1 885 to 1 94 5 . All twelve of the Household M i nisters between 1 88 5 and 1 945 , a nd indeed many of the Vice Mi nisters , achieved some measure of success in the government bureaucracy prior to enteri ng the palace . Although only one Prime M i nister was among their number, six (50 percent) had been Mi nisters of State , seven ( 5 8 . 3 percent) had been or became Privy C ouncillors (i nclud i ng 1 1 . As quoted by Baron Harada Kumao, Pri nce Saionji's political secretary, in 1 932 (Harada Kumao, Saior!]i-ko to Seikyoku, 9 vols. [Tokyo: lwanami Shoten, 1 950-56] , 2 : 397).

OFFICES

AN D

OFFICERS

OF

P A LA C E

119

L E A D E R SH I P

two Presidents) , one ( 8 . 3 percent) had served i n one of the leading military positions , and six (50 percent) received imperial appoint­ ments to the House of Peers at one time or another. Only two ( 1 6 . 7 percent)-the last two Household Ministers-were not titled ; only one ( 8 . 3 percent), Iwakura Tomosada, was anything approachi ng a court figure , and he had served for nine years as Privy Councillor before his appoi ntment as Household Mi nister in 1 909. If approximately twenty years i n a given institution is arbitrarily selected as sufficient for defining " career, " the number of I mperial Household Mi nisters whose careers were in the Home M i nistry num­ bered three (25 percent), in the Foreign Mi nistry two ( 1 6 . 7 percent) , and in the Army , the Finance Mi nistry , the Justice Mi nistry , and the court bureaucracy one each ( 8 . 3 percent) . I have classified the remaining three Household M inisters (25 percent) as Restoration bu­ reaucrats, si nce these Household M i nisters had served in a variety of positions in a bureaucracy and political system t hat was of their ow n creation prior to becomi ng Household Mi nisters ; they do not fit any fixed career category . More revealing, however, is the correlation between career back­ ground and length of tenure in office as Household Mi nister: Restoration bureaucrats Home M i nistry bureaucrats Foreign M i nistry bureaucrats J ustice M i nistry bureaucrats Army officers Fina nce M i nistry bureaucrats Court bureaucrats Total

23 years I 5 years 1 3 years 6 years 1 year 1 year 1 year 60 years

Throughout the sixty-year period between 1 885 and 1 945 career Army officers held the mi nistership but one year, i ndicating that the normal practice was for civilian bureaucrats to occupy that post, espe­ cially those from the Home and Foreign M inistries after 1 92 1 . I n general , an I mperial Household Minister termi nated a suc­ cessful public career in that office . There were notable exceptions , however. Ito Hirobumi became Prime Mi nister three ti mes and President of the Privy Council twice after he resigned as Household Mi nister in 1 887 . lchiki Kitokuro became President of the Privy

1 20

OFFICES

AND

OFFICERS

OF

P A LA C E

L EA D E R SH I P

Council in 1 934 after having served a s Household M i nister for eight years . Makino Nobuaki became Privy Seal i n 1 92 5 after four years as Household M i nister; Yuasa Kurahei also became Privy Seal after serving as Household M i nister. Averagi ng five years in office , however, t he Household M i nisters held office independent of Cabi net changes , except in 1 93 6 when assassinations caused changes in bot h court and Cabinet , and were thus "above politics" throughout the prewar period . The average age on assuming office, moreover, was 5 8 . 3 years; ranging in ages from 44 to 67 , the Household M inisters were quite young. As a group they were not venerables given their of­ fice as reward for services rendered but competent bureaucrats ex­ pected to end their public careers "at the side of the emperor. " Perhaps an exami nation of the prescribed duties of the I mperial Household M i nister's office will reveal, at least in part , why Pri nce Saionj i would i nsist on extra-court bureaucrats to admi nister the court bureaucracy . The 1 886 reorganization of the I mperial Household M i nistry merely stated that the H ousehold M i nister was to decide upon all matters regarding the affairs of the imperial house, to supervise the palace staff and staffs to the imperial family and family members, and to manage the aristocracy . The 1 886 act was signed by the Household M inister alone . 12 As early as 1 88 9 , however, the or­ ganic law of the I mperial Household Mi nistry was revised to specify the j urisdiction of the Household M i nister in greater detail . I n ad­ dition to the duties set forth in 1 88 6 , the Household M i nister was em­ powered to issue directives to subjects of the realm regardi ng imperial house matters such as palace ceremonies , festivals, and imperial tours . He was also empowered i n this respect to issue directives to the S uperi ntendent Ge neral of Met ropol i t a n Police (Tokyo) , t he Governors of the Prefectures , and other government officials. 1 3 As chief manager of the public role of the Emperor-i n-Court , therefore , the Household M i nister was directly involved with the government , especially the H ome M inistry , which had j urisdiction over prefectural Governors as well as over the ordi nary police establ ishment . Palace autonomy did not mean the isolation of the court from the government but rather a symbiotic relation between the two i n w hich i nitiative in 1 2 . "Kunaisho Tasshi , " no. I , February 4 , 1 886. 1 3 . Shokuin Roku, 1 894, vol . I , pt . 2 , p. 1 3 .

OFFICES

AND

OFFICERS

OF

PA LA C E

LEA D E R SH I P

121

arranging the public d isplay of the Emperor-in-Court rested with the H ousehold M i nistry , w hile t he act ual arra ngeme nts req u i red government cooperation . The H ousehold M inistry's reorganization of 1 907 further ac­ knowledged the entanglement of court and government that actually occurred i n the exercise of t he mi nistry 's duties . 14 Significa ntly, t he 1 907 act was countersigned by the Prime M i nister and Home M inister as well as by the Household M i nister. 1 5 A paraphrase of t he fifteen articles concerning t he Household M i nister reveals the prob­ lems of relati ng court a nd government that had emerged between 1 886 and 1 907 : 1 . The I mperial H ousehold Minister shall be a direct imperial appointee [shinnin I and shall be responsible for "advising and assisting" ( hohitsu I the throne on all matters regarding the imperial house. 2 . He supervises the personnel of the ministry and concurrently has ju­ risdiction over the aristocracy . 3 . When it is necessary to abolish, revise or enact I mperial House Ordi­ nances [Koshitsu Rei] , the H ousehold Minister prepares appropriate drafts for presentation to t he throne. If the draft concerns the work of other Ministers of State, the H ousehold Minister shall report the draft j ointly with t he Prime Minister or with the Prime Minister and concerned Ministers vf State . 4 . The Household Minister may prescribe regulations necessary to enforce Imperial H ouse Ordina nces and carry out his other duties. If such regulations concern the fu nctions or responsibilities of ot her Ministers of State they shall be passed upon by the Prime Minister and concerned Ministers of State. 5. The Household Minister may promulgate ordinances on matters under his jurisdiction. 6 . He may issue orders and directives to the Superintendent General of Metropolitan Police and to l ocal officials regarding matters under his j urisdiction. 7. I n accordance wit h imperial comma nd , he may dispense relief, awards, and grants. 8 . He reports to t he t hrone promotions and retirements of Household M inistry executive appointees (so nin l , as well as officials treated as imperial appointees (chokuninl and executive appointees . He decides 1 4 . Although t here was one more ministry reorganization prior to the end of the Pacific War, the 1 907 reorga nization was in fact the last major revision regarding the duties of the Household Mi nister. The last prewar reorganization took place in 1 92 1 . 1 5 . Naikaku Kiroku Ka, 1 920, vol . I , pt. 3 , pp. 6-8.

1 22

OFFICES

9. 1 0. 11.

12. 13. 14. 15.

AND

OFFICERS

OF

P A LA C E

L EA D E R S H I P

o n promotions and retirements of official appointees [ hannin] , those treated as official appointees, and other lower ra nking officials. He reports to the throne investitures in rank of Household Ministry personnel and members of the aristocracy . He reports conferrals of decorations to the throne through the Prime Minister. With imperial sanction the Household Minister may create advisers and special officers to the ministry regarding matters under his j uris­ diction . When in il l health or otherwise indisposed the Household Minister may appoint t he Vice Minister to perform the duties of Minister tem­ porarily . B ut in this capacity t he Vice Minister shall not go beyond public promulgations in accordance wit h the I mperial House Law and countersigning in accordance with the Public Forms Ordinance [Koshiki Rei] . The Household Minister may delegate part of his duties to the Vice M i nister and the bureau and section chiefs . He may not obstruct the proceedings of the audit . He may establish, regulate, and abolish divisions or sections within the secretariat and other bureaus and divisions under his j urisdiction. He may establish posts of imperial appointee , executive appointee , and official appointee and stipulate their d uties ; posts of executive ap­ pointee and above require imperial sanction. 16

No later than 1 907 , therefore , the fact of interaction between court and government was officially acknowledged regarding ( 1 ) drafts of Imperial H ouse Ordi nances , (2) regulations established to carry out those ordi nances , (3) police and local officials, predominantly those under the Home Mi nistry , and ( 4) conferrals of decorations , w hich were handled through the Bureau of Decorations in the Prime M inister's Office . The I mperial H ousehold M i nister in 1 907 was Count Tanaka Mit­ suaki ( 1 843- 1 939) , the second of two loyalists from the domain of Tosa to head the household bureaucracy between 1 887 and 1 909 . Tosa had been one of the four maj or domai ns i n the politics of Res­ toration, but its role had been one primarily of negotiation and media­ tion between the two most active restorationist domai ns of C hoshii and Satsuma. B oth Ta naka and his predecessor as Household M inister, Count H ijikata Hisamoto ( 1 833-1 9 1 8 ) , had played nego­ tiator roles in forging the C hoshii-Satsuma alliance on which the 1 868 1 6 . Shokuin Roku, 1 9 1 0, p t . 2 ,

sec .

I , p. 6 1 .

OFFICES

AND

OFFICERS

OF

P A LA C E

L E A D E R SH I P

1 23

Restoration was engi nee red . 17 After the Restorat i o n , more Im­ port a ntl y , Tanaka conti nued his role as negotiator within t he oligarchy , the oligarchy bei ng pri marily from C hoshu and Satsuma . He has been credited , for example, with acting as midwife for Ito Hirobumi 's third Cabi net in January 1 898. 18 After Tanaka became Household M i nister in February of that same year, he remained a middleman to the oligarchy ; that Tanaka " acted as a safety valve i n the political world while in office a s I mperial Household Mi nister i s a n absolutely inescapable fact . " 19 Tanaka was the last true Restoration loyalist , or participant i n the politics of Restoration, to head the household bureaucracy . But he was a loyalist of a particular stamp .20 He did not seek to challenge the C hoshu-Satsuma oligarchs from within by building a court consti­ tuency , as his fellow clansman, Sasaki Takayuki , apparently did as a jiho i n 1 877-7 9 . Nor did he attempt to challenge them from without by building an antioligarch political party, as did another Tosa clansman, ltagaki Taisuke . Tanaka appeared content to be a member of the charmed Restoration ingroup surrounding the throne and to ne­ gotiate among and on behalf of the oligarchs . Si nce the government was i n fact domi nated by leaders from C hoshu and Satsuma well into the twentieth century , it was also logical to have the Household M i nister appointed from a domai n other than those two: such a selection would give the I mperial Household Mi nistry at least the semblance of detachment from alleged factional strife in the government while allowing the oligarchs influence at court as wel l , dependi ng on the person selected as Household M i nister. As late as 1 92 1 Count Makino Nobuaki told 1 7 . Tomita Koj i ro, p. 292; Sawamoto Kenzo, ed . , Hakushaku Tanaka Seisan (Tokyo: Ta­ naka-haku Denki Kanko Kai, 1 929), pp. 1 6 1 -8 3 ; Hijikata-haku, 2d .ed. (Tokyo: Toyo I nsatsu Kyoku, 1 9 1 4) , pp. 240-49. On Tosa 's role i n t he Restoration see M arius B. Jansen, Sakamoto Ryo ma and the Meiji Restoration (Princeton : Pri nceton University Press, 1 96 1 ) . 1 8 . Tomita Koj iro, p p . 298-30 1 ; Sawamoto Kenzo, p p . 426-2 8 . 1 9. Sawamoto Kenzo, p. 430. 20. The variety of Army, police, and court offices held by Tanaka between 1 868 and 1 909 reveals the difficulty i n categorizing the "career background" of all the Restoration leaders. Tanaka achieved the rank of major general in t he Army a nd was at one time t he Superintendent General of Metropolitan Police. He also served as Chief Cabinet Secretary, a post t hat t his predecessor, H ijikata Hisamoto, had also held . Tanaka entered t he palace as a court adviser in 1 89 1 a nd served as Vice M inister of the Household M inistry from 1 895 to 1 897 .

1 24

OFFICES

AND

OFFICERS

OF

P A LA C E

L E A D E R SH I P

Privy Seal Pri nce Matsukata M asayoshi t hat in selecting a successor to Household M inister General Nakamura Yuj iro: We should select as successor to the I mperial Household Minister someone suitable from among persons other than those of Satsuma or Choshu origins. The times have changed . To give the people the im­ pression t hat Satsuma and C hoshu have always monopolized power even in the palace is simply wrong. 2 1

As second-ranki ng members of t he Restoration ingroup, bot h Tanaka and his predecessor were ideal candidates for leadi ng t he court bu­ reaucracy : they possessed Restoration status and were committed as negotiator-mediators to the new government evolved by the oligarchs . Tanaka 's actions as Household M i nister ( 1 898- 1 909) i ndicate t hat the Household M i nister's i nteraction with the government , as sug­ gested by the 1 907 description of t he Household M i nister's duties , oc­ curred largely i n his capacity as chief manager of t he Emperor-in­ Court . I n that capacity Ta naka defended the formal autonomy of t he court with ferocious tenacity, dow n t o the most trifling matters of cere­ mony . He had resigned as Vice M i nister of t he Household Mi nistry i n 1 897 over a dispute with the government concerning who was to an­ nounce t he death of t he empress dowager, t he I mperial Household M inister or the Prime Mi nister. The emperor's view was solicited a nd the Privy Council 's advice asked . Tanaka 's i nsistence that the an­ nouncement be made i n the name of the I mperial Household Mi nister was upheld but at the expense of Prime M i nister M atsukata's friend ­ ship, and Ta naka resigned over t he difficulties caused .22 After becomi ng H ousehold Mi nister in 1 898 Tanaka conti nued t o defend t he formal parity o f court and government w i t h equal stub­ bornness. O n one occasion Prime M i nister Katsura Taro asked Ta­ naka to come t o the Prime M i nister's residence for a discussion. Ta­ naka refused : I ' m not under the Pri me M i nister; the Imperial Household M i nister is i n an independent position outside the Cabinet . Other than H i s Majesty I a m t o b e under n o one's orders . Of course , I have the greatest respect for 2 1 . Shimozono Sakichi , Makino Nobuaki-haku (Tokyo: Jimbunkaku, 1 940) , I ronically, Nakamura's successor w a s Makino, w h o was himself from Satsuma . 2 2 . Sawamoto Kenzo, pp. 422-2 4 .

p.

1 86 .

OFFICES

AND

OFFICERS

OF

PALACE

LEA D E R SH I P

1 25

Prince Katsura as an individual . . I refused because if Prince Katsura had some matter to take up with me he should come here . For me to go on summons would be unacceptable in view of my office . 23 .

.

Katsura then called back and said he would come see Tanaka if Ta­ naka were inconvenienced . Once he received this ritualistic ac­ knowledgment of his " office" from the Prime Mi nister, Tanaka re­ plied that he would go see the Prime M i nister and immediately ordered a carriage .24 How Tanaka served as political mediator among government leaders and participated in policy at this time is not the issue here; his behavior in both i nstances cited above showed extraor­ dinary sensitivity to the symbolic separation of court and government and to the formalistic role of office in palace politics . Preservation of parity also i nvolved the preservation of the court 's fi nancial autonomy from government encroachment . I n 1 902 , for example, Tanaka refused to give imperial lumber to the Home M i nistry free of charge for the reconstruction of the Grand Shrine at Ise. Tanaka argued that si nce the court and the government had separate budgets, and si nce the government already had a budget allocation for rebuilding the shri ne, the Home M i nistry should pay the I mperial Household M i nistry a suitable price . To keep the dis­ t i nction between court a nd government free of confus i o n , he continued , there must be no government inroads on imperial house property . The Cabinet reconsidered the matter and the Home M i nistry subseq uently bought the lumber.25 Household M i nister Ta­ naka was in no sense the servant of the government , and he perceived his office as granti ng hi m great freedom of action agai nst the government and its leaders on court matters , as he defi ned them. If Tanaka kept the government at bay , he also acted firmly with both the court a nd the emperor . J ust after Tanaka had become H ousehold M i nister in February 1 898, Prince-of-the-Blood Komatsu req uested a B udd hist fu neral for t he late Pri nce-of-t he- B l ood Yamashina, as Pri nce Yamashi na had req uested in his wil l . With the support of the Privy Council Tanaka refused the request on the grounds that a member of the imperial house held a position above 2 3 . Tomita Koj i ro, p. 477 . 2 4 . I bid . , pp. 477-7 8 . 2 5 . Ibid . , pp. 3 1 0- 1 1 .

1 26

OFFICES

AND

OFFICERS

OF

PALACE

L E A D E R SH I P

all religions; the funeral service had t o b e one appropriate t o the status and position of the imperial family.26 Tanaka also had a considerable reputation for goi ng agai nst the emperor's wishes and for doing precisely what he believed his office demanded . Tanaka's biographer, Tomita Koj iro, states that hardly anyone dared counter the emperor to his face , not even the Elder Statesmen, Grand Chamberlain , or Privy Seal . Tanaka , however, would argue with the emperor " as a comrade. " 27 On one occasion the emperor is reported to have flared out at Tanaka's stubborn resista nce : "Aren't you a Maj or General ? You 're breaking mil itary discipline if you don't listen to the order of your supreme com­ mander. " Tanaka retorted : " I ' m sorry , I wasn't speaking in my ca­ pacity as Maj or General . I must beg your indulgence on this matter from my position as Imperial Household Mi nister . " 28 I n 1 904 General Yamagata Aritomo was dispatched to the front to review the Russo-Japanese War situation. For this purpose he requested t hat Tanaka obtai n two of the emperor's horses . The em­ peror, claims Tomita, had very little he could truly call his ow n other than his beloved horses . When Tanaka approached the em­ peror with Yamagata's request , the emperor flatly refused . After unsuccessful attempts to obtain the emperor's permission, Tanaka fi ­ nally handed over two of the emperor's horses to Yamagata without a word to the emperor. Overj oyed , Yamagata immediately audienced with the emperor and t hanked him for his generosity .29 Tanaka viewed his relationship with the emperor as one of sole responsibility for advisi ng and assisting the throne on court and household affairs , but " t o obey imperial directives to the letter, without offeri ng the slightest admonition, cannot be called fulfilling the responsibility of advice and assistance , to my way of thinking. " 30 In this i nsta nce " admonition" took the form of disobedience. C ommenti ng o n Tanaka ' s position as I mperial H ousehold M i nister, Tomita argues that a courageous person of Tanaka's in­ tegrity was needed to maintai n t he sepa rat i o n of court and 26. 27. 28. 29. 30.

Ibid . , Ibid . , Ibid . , Ibid . , Ibid . ,

308-09 . 484-8 5 . p . 483. pp. 484-8 7 . p . 47 9 . pp. pp.

OFFICES

AND

OFFICERS

OF

PALACE

L EA D E R SH I P

1 27

government in t he face of clique politicians anxious to achieve their ambitions Y Tanaka evidently demonstrated such courage not only toward the government but also toward the court and t he emperor as wel l . It would appear that for Tanaka the emperor was a fellow Restoration " stalwart , " a comrade i n arms who could be rebuked both from the standpoi nt of Tanaka 's perception of " office" and from the spirit of competitive comradeship that prevailed among the oligarchy . As recognized in the 1 907 orga nic law of the Imperial H ouse­ hold Ministry , palace autonomy obviously did not mean isolation for the emperor as Emperor-in-Court . I mperial house funds were used to emphasize the role of the emperor as social paragon, as symbol of socially desirable styles , activties , a nd goals. To encourage education , for example, an imperial gift of 50,000 yen was granted to Fukuzawa Yukichi in 1 900. In 1 908 Waseda University received 30 ,000 yen from the I mperial Household M i nistry .32 Since the emperor was "the father and mother of his people , " the " standard of public morals , " and the " fundamental basis of institutions and culture , " such uses of privy monies were amply rationalized .33 B ut the emperor was also the " fount of politics . " 34 The vast sums available to the H ousehold M i nister had political uses as well . I n 1 8 9 2 , for e x a m p l e , t h e I m p e r i a l D i et refused t o p a s s t h e government 's naval shipbuilding budget . A t the request o f the Cabinet , which sought to overcome Diet opposition by imperial rescript , the emperor called a meeti ng of the Prime M inister , Privy Councillors , and the Speakers of both houses of the Diet . An im­ perial rescript was subsequently issued to the effect that for the next six years 300 ,000 yen per year would be allocated from the court fund for naval shipbuilding.35 This sum was one tenth the annual allotment to the court from the national treasury . Quite clearly, the oligarchs in government and the leaders at court cooperated in a political maneuver to maintain the supremacy of oligarchic decision making . 31 . 32. 33. 34 . 35.

Ibid . , pp. 306-07 . Ibid . , pp. 329-30. Sawamoto Kenzii, p. 441 . Ibid . Kuroda Hisata , pp. 75-7 7 .

1 28

OFFICES

AND

OFFICERS

OF

PALACE

L E A D E R SH I P

This action also demonstrated the " sacrifices" that the emperor was willing to make for " patriotic" goals. But privy monies were also used for political manipulation that had no such patriotic ra­ tionalization. Hara Kei , for example, recorded a conversation with oligarch I noue Kaoru on December 4, 1 90 1 , i n which I noue stated that the I mperial H ousehold M i nistry had on at least one occasion given oligarch Yamagata Aritomo court funds . W hen Hara spoke with Pri nce Saionj i two days later, Saionj i told him that Yamagata had at one time received the enormous sum of 980 ,000 yen from the Household M i nistry . Hara confessed amazement : it was difficult to believe that such a sum was needed merely to establish a Cabinet , as Yamagata's request to the throne had stipulated . Despite Yama­ gata's reputation for i ntegrity , Hara was firmly convi nced that Yamagata had used this and other sums from the court to bribe members of the H ouse of Representatives . In 1 90 1 Prime Minister Katsura Taro asked H ousehold M i nister Tanaka for a similar "contribution . " Although he had given funds from the privy purse to a number of oligarchs up to that poi nt , Tanaka refused Katsura's request , evidently because Katsura was not an oligarch. Hara specu­ lated that Katsura , like Yamagata, intended to use such court funds , had they been made available, to entice members of the House of Representatives to go along with Cabi net policies .36 Apparently Yamagata conti nued to tap court funds later as well by having his proteges appoi nted Household M i nister from 1 909 to 1 92 1 .37 There exists , t herefore , some evidence to support I noue Kiyoshi's claim that "this extensive imperial house property did play the political and military role that Iwakura [Tomomi] had hoped . " 38 H ousehold M i nister Tanaka was i n fact an agent of the oligarchs brought into the palace from the " outside" to enhance cooperation between court and government i n managi ng the Emperor-i n-State . As i ndicated by Tanaka's refusal to give Katsura court monies and other uncooperative acts already noted , Tanaka cooperated only with the oligarchs (Elder Statesmen) , not with their proteges or other 36. Hara Kei Nikki, 9 vols. (Tokyo: Tokyo Kengen Sha, 1 950), 2 : 454-56. 3 7 . Fukumoto Kunio, Kanryo (Tokyo: Kobundo, 1 959), p. 66. 3 8 . I noue Kiyoshi , p. 94 . Unfortunately , I have been unable to fi nd any further evidence of such political uses and abuses of court funds . It is difficult to believe that t hey were not greater than I have found and t hat t hey did not continue.

OFFICES

AND

OFFICERS

OF

P A LA C E

L E A D E R SH I P

129

lesser political leaders .39 The available evidence suggests that Ta­ naka willingly coordinated the I mperial Will in politics with the policies of the oligarchs , while mai ntai ning the formal separation of court and government . Tanaka's actions as H ousehold M i nister re­ veal the significance of Saionj i ' s caveat that H ousehold Mi nisters be " brought i n from the outside" : a H ousehold M inistry " i n l i ne wit h the times" was one that cooperated with the domi nant political forces at any given moment i n managi ng the Emperor-i n-State , w hile maintai ni ng the autonomy of the Emperor-i n-Court . The "dominant political forces" -in Tanaka's time a handful of Cabi net oligarchs or oligarchs acting as Elder Statesmen-had chief responsi­ bility for political decision making, for defining the official I mperial Will in politics . The H ousehold M i nistry cooperated with the government i n this process . With regard to the Emperor-i n-Court , howeve r , t he palace bureaucracy held t he i nitiative, a nd the government cooperated with t he Household M i nister i n maintaining the autonomous image of the public Emperor-i n-Court . Between 1 909, w hen Tanaka resigned , and 1 92 1 , when Count Makino Nobuaki became H ousehold M i nister, there were four H ousehold M inisters . One was predominantly a palace official (lwakura Tomosada) , one a Home M i nistry bureaucrat (Watanabe C hiaki), one a J ustice M i nistry official (Hatano Taka nao) , and one a career Army officer (Nakamura Yiij iro) . The average term of office between 1 885 a nd 1 945 was five years . The average between 1 885 and 1 909, however, was eight years , while that between 1 909 and 1 92 1 was but three years . B oth the freq uency of changes and the va­ riety of career backgrounds brought to the office of Household M i nister between 1 909 and 1 92 1 suggest that this was a period of transition for the court bureaucracy as it adj usted to t he loss of Res­ toration leaders . I n contrast to the tranq uility that had prevailed at t he side of the emperor during the period 1 885 to 1 909, the period 1 909 to 1 92 1 was one of confusion as the leadership at court strug­ gled to get in line with the new times . From 1 92 1 to 1 945 , however, the I mperial Household Mi nisters , with the possible exception of Ishiwatari Sotaro, were drawn from a 39. Katsura was Yamagata's leading protege (Tetsuo Najita, Hara Kei in the Politics of Compromise, 7905- 7 9 75 !Cambridge, Mass . : Harvard University Press, 1 96 7 ] , p . 8 1 ) .

1 30

OFFICES

AND

OFFICERS

OF

P A LA C E

LEA D E R SH I P

group of civilian bureaucrats whom I have called the constitutional monarchists . M ost important , perhaps , was Ichiki Kitokuro , Household M i nister from 1 92 5 to 1 93 3 . Although lchiki had been a career bureaucrat i n the Home M inistry and served as Education Mi nister i n 1 9 1 4- 1 5 , H ome M inister i n 1 9 1 5 - 1 6 , and Privy Councillor from 1 9 1 7 , his importance did not derive from such na­ tional visibility . In 1 894 lchiki had become professor of law at Tokyo I mperial University while remaining an official in the H ome M inistry . In that same year Ichiki 's most celebrated pupil , M i nobe Tatsukichi, entered the university. It was from Ichiki that M inobe was to acquire the basic ideas of his " organ theory " (kikan setsu) of constitutional monarchy , a theory that gained primacy in Japanese constitutional thought duri ng the 1 920s and which was also accepted by a number of Japan's leading bureaucrats.40 lchiki 's very presence in the palace as chief manager of the Emperor-in-Court suggests the acceptance of his theory , or of M i nobe's more liberal variant , by the court leadership during the 1 920s and early 1 930s. lchiki argued that the "supreme right of rule" over w hich the em­ peror exercised general supervision was lodged i n the state as a legal person. The state exercised its supreme right of rule through " organs" (kika n ) , the highest of which exercised general control . When a monarch was the highest organ of state but governed with the cooperation of other organs, such as a legislature, he was a constitutional monarch. For these reasons , Ichiki concluded , Japan under the Meij i Constitution was a constitutional monarchy . The emperor as highest organ of state was, however, under very little restriction : he could at any time divest another orga n of state of its power. The Imperial Diet , according to lchiki, could not assert its powers over M inisters of State, si nce ministers were responsible in­ dividually to the emperor. W hile placi ng the emperor under law , as embodied i n the constitution, lchiki allowed no government organ to check the emperor.41 I n M i nobe's hands, however, lchiki's organ theory was greatly liberalized . W hile mai ntaining t hat the emperor was the highest organ of state, M inobe i nsisted that the emperor was bound by the 40. Nihon fimbutsu. Shi Taikei, 7 vols. , 3d ed . (Tokyo: Asakura Shoten, 1 963 ) , 7 : 270. 4 1 . Ibid . , pp. 2 7 1 -73.

OFFICES

AND

OFFICERS

OF

P A LA C E

L E A D E R SH I P

131

constitution; were he t o act arbitrarily, the emperor would be vio­ lating Japan's long history as well as the constitution. Moreover, M inobe made the I mperial Diet i n effect the highest organ of state under the emperor . Minobe's theory was ultimately an attack on the independence of the bureaucratic institutions of prerogative and on "transcendental Cabinets . " If pushed to its logical conclusion, his theory j ustified the creation of Cabinets responsible to the elected House of Representatives , w hile still preserving the sanctity of the emperor as sovereign u nder law . Mi nobe never asserted the demo­ cratic pri nciple of popular sovereignty , but his theory might have led to democratic government . The orthodoxy of M inobe's theory , as far as the palace leaders and t heir allies were concerned , is i ndicated by the fact that he received an imperial appointment to the H ouse of Peers in June 1 932 . B y this time, however, the worldwide depression and right­ wing movements in Japan had brought on a reaction that ultimate! y forced M inobe to resign that post . His theory was condemned i n 1 93 5 a s contrary to the ideals o f Japan's national polity-ironically, with t he support of rightists and ultraconservatives i n the very insti­ tution he had so strongly supported : the I mperial Diet . More signifi­ cantly , lchiki was dislodged less than a year later from the presidency of the Privy Counci l , an office he had taken in May 1 934 . The attack on M i nobe became the vehicle for bringing about lchiki 's downfal l , si nce lchiki was allegedly responsible for " the importation of the hateful doctri nes . " 42 B y extension, it also became an attack on court leaders and the emperor's close advisers , ultimately en­ couragi ng the outbreak of assassinations and attempted assassina­ tions that peaked on February 2 6 , 1 93 6 . M iller's a nalysis of the M inobe i ncident of 1 93 5 shows most dramatically that the condem­ nation of M i nobe's theory was the catalyst for a massive attack on the constitutional monarchists in government and court by renova­ tionist a nd militarist bureaucrats in t he government .43 Despite the eclipse of t he constititional monarchists following M inobe's ouster and the assassi nation and attempted assassi nation of 42 . M iller, p. 2 1 8 . 4 3 . Ibid . , pp. 1 96-2 5 3 .

1 32

OFFICES

AND

OFFICERS

OF

P A LA C E

L EA D E R SH I P

the emperor's close advisers i n February 1 93 6 , the constitutional monarchists never lost control of the H ousehold M i nister's office . The Household Minister from 1 93 6 to 1 945 was Matsudaira Tsuneo, a career diplomat w ho had been ambassador to the United States ( 1 925-28) and England ( 1 929-3 5 ) . He was, like most palace leaders of the 1 920s and early 1 930s, pro-Anglo-American and a cir­ cumspect moderate in domestic politics . The forces of extremism never quite managed to capture the Emperor-i n-Court " from the in­ side , " however much they manipulated the imperial symbol outside the palace gates . The court remai ned , in I rie Sukemasa 's exag­ gerated terms , a place "where freedom of speech has been guarded at every turn " : I rie could still claim that there was " no more demo­ cratic a place than this" .

TH E GRAND C HAM B E RLA I N : C H I E F M ESSENG E R IN ATTENDANCE Wel l , I became G rand C hamberlain. How best go about the job? Even on that score I had no idea , but I did hear various thi ngs from the B oard of C hamberlains , learned about His Majesty's daily routine, was helped out by Kawai Yahachi , the Deputy G rand Chamberlain, . . . Suzuki Kantaro44

Even more than the duties of the Chamberlains, those of the Grand C hamberlain were ill-defi ned . After the I mperial Household M i nistry was reorganized in 1 88 6 , the 1 87 1 office of Grand C hamberlai n45 was given two vague and flexible functi ons : ( 1 ) to render assistance at the emperor's side, and (2) to supervise the Chamberlains and Stewards.46 Between 1 886 a nd 1 945 there was no further specification of the Grand C hamberlain's duties , in marked contrast to the considerably detailed regulations to which the I mperial Household Mi nister was subjected after 1 886. When an outsider took up that post , as Admiral Suzuki Kantaro did in 1 92 9 , he could either unearth the precedents 44. Suzuki Kantaro, Suzuki Kantaro Jiden, ed . Suzuki Haj i me (Tokyo: O gikukai , 1 949),

p. 270.

45. " Daj okan Tasshi , " no. 400, 10 August 1 87 1 , in Ho rei Zensho, 1 87 1 . 46. "Kunai Sho Tasshi , " no. 1 , February 5 , 1 886, in Ho rei Zensho, 1 886.

O F FI C E S

AND

OFFICERS

OF

P A LA C E

L EA D E R SH I P

1 33

applyi ng to his new office-or make them himself. Such vagueness was disconcerting to a career mil itary officer like Admi ral Suz uki , but it allowed an incumbent considerable latitude to i nterpret the meaning of rendering " attenda nce at the emperor's side . " For example, Cham­ berlain Irie Sukemasa writes of Admiral Suzuki 's tenure as Grand C hamberlain from 1 92 9 to 1 93 6 , " Since Makino and Saito as Privy Seals did not come to their offices but once a week or once a month, it was natural that the Grand Cham berlain should come to ha ndle part of the Privy Seal 's work. " 47 I n terms of j urisdiction and behavior the Grand Chamberlain, at least by 1 92 9 , played a highly flexible role in the politics of relating emperor and government , despite the fact that he was in charge of the " i nner side" of t he court and presumably a purely court figure . The career backgrounds of the personnel appoi nted Grand Cham­ berlain between 1 885 and 1 945 were q uite different from those of t he H ousehold M i nisters . There were ten G rand Chamberlai ns during the sixty-year period : only two (20 percent) had been Mi nisters of State and only one ( 1 0 percent) had been Prime M i nister. The two M i nisters of State (Hatano Takanao and Katsura Taro) and the Prime M i nister (Katsura Taro) , moreover, held the post of Grand C hamberlain a total of less t han six months, indicati ng the irregu­ larity of such appoi ntments. Two (Katsura Taro and Suzuki Kan­ taro) did become Prime M i nister after serving as Grand C ham­ berlai n, but once agai n this was not the norm . I n contrast to the H ousehold Mi nisters , the Grand C hamberlains were frequently drawn from the court and the military . Whereas t he Household M inisters numbered only one mil itary ( 8 . 3 percent) and one court figure ( 8 . 3 percent) , the Grand C hamberlains numbered four (40 percent) w ho had served in high mil itary posts, five (50 percent) whose careers had been largely military , and three (30 percent) career court figures . O nly three Grand Chamberlains (30 percent) had served at one time or another on the Privy Council ; seve n H ousehold M i nisters ( 5 8 . 3 percent) had done so. A striki ng pattern emerges over time i n the career backgrounds of the Grand Chamberlains between 1 885 and 1 945 , a pattern that reflects the bureaucratization of palace leadership from the outside. 47 . l rie Sukemasa, jiju to Paipu,

p.

9 9 . M y emphasis.

1 34

OFFICES

AND

OFFICERS

OF

PALACE

L E A D E R SH I P

I n the years 1 885 t o 1 929 the Grand Chamberlains were draw n al­ most exclusively from the court ; from 1 92 9 to 1 945 they were exclu­ sively retired admirals of the Navy . Calculating career background (twenty years or more) agai nst length of service as Grand C ham­ berlai n, the following institutional emphases emerge : Court Navy Army•• Foreign M inistry Total

36 years 1 7 years 5 years 2 years 60 years

Additionally, it was a rare coi ncidence for the Grand Chamberlai n and the Household Minister to be drawn from the same institutional constituency between 1 885 and 1 94 5 . Only in 1 909 did this occur, when both were drawn from the court bureaucracy . This fact sug­ gests that there was a carefully contrived balance of i nstitutional representatives at the side of the emperor, not merely a mosaic of status, especially after 1 92 9 . Like the Household M i nisters , however, the Grand Chamberlains after 1 927 generally termi nated successful public careers on the " outside" by serving at court . The average age of the Grand Cham­ berlains on assumi ng office was 6 1 . 1 years , rangi ng from 46 to 7 1 . The average term of office was six years . It is important to note , however, that the Grand Chamberlain throughout the Meij i period was one person, Pri nce Tokudaij i Sanenori .49 The very length of his service , from 1 87 1 to 1 9 1 2 , suggests the high degree of tranq uility at the emperor's side if not in the outside political arena . Tokudaij i was the only palace official , moreover, to have held three of the four leading palace offices : he was concurrently I mperial Household 4 8 . The Army official was Takatsukasa H i romichi ( 1 855- 1 9 1 8) , Grand Chamberlai n from 1 9 1 2 to 1 9 1 8 . Alt hough Takatsukasa achieved the rank of major general he was basically a court person of Court Noble li neage . In 1 896 he was Aide-de-Camp to t he crow n pri nce (Shokuin Roku, 1 896, p. 1 38). In such cases it is obviously difficult to separate court from extra-court careers. 4 9 . Ij iri Tsunekichi lists four persons as G rand Chamberlain during the period 1 87 1 -84 , with Tokudaij i becoming Grand C hamberlain in 1 884 (lj i ri Tsunekichi, Rekidai Kenkan Roku [Tokyo : Choyo Kai , 1 925 1 ) . The Imperial Household Agency , however, has Tokudaij i listed a s Grand Chamberlain from 1 87 1 to 1 9 1 2 and lists h i m as serving concurrently as Im­ perial H ousehold Secretary from 1 87 1 to 1 884. Without being able to account for the dise­ repa ncy between t he lj iri and Agency records, I have accepted the Agency roster as official.

OFFICES

AND

OFFICERS

OF

PALACE

L EA D E R SH I P

1 35

Secretary from 1 87 1 to 1 884 and concurrently Privy Seal from 1 89 1 to 1 9 1 2 . Despite his length of service at court , Prince Tokudaij i has enj oyed singular obscurity . There is no biography of t his venerable figure; 50 most biographical dictionaries refer to him si mply as " Em­ peror Meij i ' s Grand Chamberlai n . " He has also been referred to as "a classic court official " who, in constantly attending the Emperor Meij i , " was so humble and diligent that no one ever saw him stand in a fully upright position. " 5 1 Apparently Tokudaij i was a perfectly resistance-free conduit between t he emperor and the oligarchs . He was the emperor's messenger on all important matters of state, such as Cabi net formations and questions of foreign policy , but was never credited with offeri ng his own views, advisi ng the throne or the ol igarchs , or blocking access to the throne .52 In 1 87 8 , he was accused by jiho Sasaki , as we saw in chapter 2 , of catering to the emperor's every whim and of being weak. The meager references to him that ap­ pear in the biographies of the Meij i oligarchs credit him with no other function than a punctilious execution of liaison between the emperor and the oligarchs . Even as hostile a critic of Meij i "absolutism" as I noue Kiyoshi in­ dicates that Tokudaij i was merely a faithful servant to the emperor . I n 1 88 1 Prince Saionj i Kimmochi , one of Tokudaij i ' s younger brothers , became president of the Toyo jiyu Shim bun, a liberal newspaper which numbered Nakae C homin among its outspoken critics of the oligarchs . Sanj o Sanetomi , Iwakura Tomomi , and Tokudaij i -all t h ree C ourt N oble restorati o nists-attempted to dissuade Saionj i from the venture, but to no avail . Finally the oligarchs took the problem to the emperor. "The emperor, too, was greatly alarmed a nd ordered Tokudaij i to have his younger brother quit the newspaper . " B ut Saionj i held fi rm until a "clear" imperial command was issued agai nst his participation.53 In the absence of 50. There is no list i ng, for example, i n Takanishi Koshi's bibliography of Meij i biogra­ phies, Shiseki Kaidai (Denki Hen) (Tokyo: Meij i Shoin , 1 935) . 5 1 . Watanabe lkuj i ro, Meiji Tenno t o Hohitsu no Hitobito (Tokyo: Chigura Shobo, 1 938), p. 1 2 5 . 5 2 . l nada Masatsugu, "Taiheiyo Sensa Boppatsu to Tenno Genro oyobi Jiishin n o Chii," Nihon Gaiko Gakka i , ed . , Taiheiyo Sensa Genin Ron (Tokyo: Shimbun Gekkan Sha, 1 953), p. 3 1 . I noue Kiyoshi, p. 1 1 6 . 5 3 . I noue Kiyoshi , p p. 79-80, 1 1 6 .

1 36

OFFICES

AN D

OFFICERS

OF

PALACE

L E A D E R SH I P

evidence to the contrary , Tokudaij i appears t o have been merely a liaison between emperor and oligarchs and a faithful executor of oligarchic wills, and not extremely effective at the latter. He was evi­ dently an opi nionless member of the Restoration ingroup, which fought over policies on a person-to-person, face-to-face basis but stood united on the privileged position of the Emperor-i n-C hambers . O nce Tokudaij i left his posts as Grand Chamberlain and Privy Seal in 1 9 1 2 , however, there was i nstability at t he side of the em­ peror. Between 1 9 1 2 and 1 92 1 there were no less than four Grand C hamberlains, or one every 2 . 3 years . The average term of office being six years between 1 88 5 and 1 945 , the period 1 9 1 2 to 1 92 1 , as for the Household Ministers , was the most unstable in the history of the Grand C hamberlain's office. It was in these years that court leadership was in effect transferred from men of the court to bu­ reaucrats from the outside in all four leading palace offices . The confusion i nvolved i n the transition is well illustrated by General Prince Katsura Taro's brief tenure as Grand C hamberlai n and Privy Seal from August 1 3 to December 2 1 , 1 9 1 2 . The precedent for holding both posts concurrently had , of course , been set by Katsura's predecessor, Tokudaij i Sanenori . B ut there was a vast difference between a life-long civil-military official a nd protege of oligarch Yamagata Aritomo, w ho was twice Prime M i nister before entering the court , and a life-long court personage, w ho knew no other than palace life. Public response to Katsura's appointment as Privy Seal and Grand C hamberlain was immediate and sharply critical : the previous Privy Seals had been the " flower of the court aristocracy , " while Katsura was of " rustic" origi ns a nd thereby an affront to court office . A "tidal wave" of rumor and gossip engulfed Katsura and he became t he focal point of public censure, controversy and slander : 54 The public believed that Katsura's entrance i nto the Imperial Household , as chamberlain and keeper of the imperial sea l , was a plot of the Yamagata faction to dominate the Imperial Household and establish control over the new Emperor. 55 54. Tokutomi I ichiro, Koshaku Katsura Taro Den, 2 vols. (Tokyo : Ko-Katsura Koshaku Kinen J igyo Kai, 1 9 1 7) , 2 : 5 96-97 . 5 5 . Tetsuo Najita, p. 93.

OFFICES

AND

OFFICERS

OF

P A LA C E

L E A D E R SH I P

1 37

Even as astute a political leader as Hara Kei , Home M i nister at the time of Katsura's appointment , concurred i n this widely held view . H ara wrote i n his diary on the day of Katsura's appoi ntment : Katsura Taro was appointed Grand Chamberlain and concurrently Lord Keeper of the Privy Sea l . I had been unclear about the meaning of what he had said yesterday regarding court and government , but his conversation became understandable as the result of this appoi ntment : it was evident that he had taken part in a scheme , a plot of the Yamagata faction to garner the Privy Council along with the court wholly i nto its hands. 56

Despite these perceptions , quite the opposite reason appears to have motivated Yamagata to engi neer Katsura 's entra nce into an of­ fice of court leadership. Yamagata and other members of his group were fearful that Katsura , though Yamagata's leadi ng protege , would take an i ndependent course from the Yamagata main line and found a political party to counter the predominant influence of the Seiyukai i n the Diet . The Seiyukai president was Saionj i Kimmochi and Hara Kei was Saionj i ' s chief lieutena nt . Between 1 90 1 and 1 9 1 2 Saionj i had alternated with Katsura as Prime Mi nister. B oth Saionj i and Yamagata agreed , for different reasons , t o " retire" Katsura into the palace, thus bri ngi ng Katsura's active political career to an end . Y a m ag a t a hoped by t h i s m e a s u re t o o bt a i n h i s g o a l of "transcendental Cabinets " composed of officials above party and not responsible to the Diet ; Saionj i , who was Prime M inister at the time, hoped to rid himself of his and Hara Kei 's political rival . Katsura , the victim of collaboration between patron and rival , was evidently quite unhappy about his " i mprisonment " in the palace .57 B ut Katsura did not stay imprisoned for long. After some four months as Privy Seal-G rand Chamberlain, he emerged to become Prime Mi nister for the third time on the very day , December 2 1 , 1 9 1 2 , that he resigned his palace offices . His exit from the palace was accompa nied by a public outcry of equal vehemence to that when he was appoi nted :

( I t ] was being said generally of Katsura : He had entered the I mperial Household i n August 1 9 1 2 to establish the control of Yamagata's faction 5 6 . Hara Kei Nikki, ed . Hara Keiichiro, (Tokyo: Fukumura Shuppan Kabushiki Kaisha , 1 96 5 ) , 3 : 24 5 . 5 7 . Tetsuo Najita, p p . 92-95 . Tokutomi l ichiro, Koshaku Katsura Taro Den, p p . 593-96. Hara Kei Nikki ( 1 965), 3 : 24 5 .

1 38

OFFICES

AND

O FFICERS

OF

PALACE

LEA D E R SH I P

there; o n the basis o f his influence in the I mperial Household, h e had destroyed the Second Saionj i Cabi net and t hen had issued an imperial rescript to himself to become prime mi nister for the third time. 58

Katsura's reemergence into active politics , from palace leader to Prime Minister, was "deeply resented by the public. " 59 If the dis­ tinction between court and government was violated by Katsura 's en­ trance into court office it was more than equally violated by his exit. Not since Ito Hirobumi 's first Cabi net in 1 885 had there been such a public clamor over manipulati ng the throne for political purposes . Although a court personage was eventually selected to replace Katsura as Grand Chamberlain and an imperial pri nce appoi nted as Privy Seal , there was some q uestion as to when the successors would be selected . H a ra Kei was deeply suspici ous of Katsura : on December 1 8 , 1 9 1 2 Saionj i told Hara that Katsura had said he i ntended to appoint no successor as Privy Seal for the moment . In his diary Hara comments , " . . . maybe there was no person for this [ post] , and maybe he [ Katsura] would keep it open as a place for him to take refuge in again later. " 60 Here, at least in Hara 's mind , was direct manipulation of palace office by political partisans for political purposes . More important , however, is the institutionalized nature of the conflict over Katsura's appoi ntment and subseq uent actions . On the theoretical level , it was a conflict between the concepts of " party" versus " transcendental" cabi nets . On the i nstitutional level it was a conflict over the composition of the Japa nese executive between the I mperial Diet , specifically the H ouse of Representatives , and the bu­ reaucracy, specifically the ministries , operati ng on the pri nciple of executive imperial prerogatives . On the personal level it was a con­ flict between oligarch Yamagata and his proteges on the one side and Saionj i and his proteges on t he other, specifically protege Katsura versus protege Hara . B oth the theoretical and the personal aspects masked the profoundly i nstitutional nature of t he conflict . It is i n this context that Dietman Ozaki Yukio's famous denunciation of 5 8 . Tetsuo Najita, p. 1 1 6 . 5 9 . Ibid . , p. 1 1 7 . 60. Hara Kei Nikki ( 1 965 ) , 3 : p. 274.

OFFICES

AND

OFFICERS

OF

P A LA C E

LEA D E R S H I P

1 39

Katsura's use of the emperor in politics on February 5 , 1 9 1 3 ought to be evaluated : They [ Katsura and the oligarchs] always mouth "loyalty" and "patrio­ tism" but w hat t hey are actually doing is to hide themselves behind the Throne, and shoot at their political enemies from their secure ambush. (Applause . ] The Throne is their rampart . Rescripts are their missiles.61

Ozaki pitted Diet against bureaucracy and attacked the assumption of t he bureaucracy that it rightfully monopolized the I mperial Will . The uncertai nties i ntroduced i nto Japanese politics as a result of the emergi ng i nstitutional competition underlying Katsura's ap­ pointment are obtusely suggested by Hara on August 1 4 , 1 9 1 2 : " Katsura's appoi ntment had been proposed by Yamagata; they might have satisfied t heir ambitions by this , but the future of bureau­ cratic politics when it comes to this cannot possibly be long. " 62 If a leading practitioner of " bureaucratic politics" had to assume court office in order to protect his and his faction's power from the institu­ tional challenge of Diet politicians , then he and his faction, ac­ cording to Hara, revealed their actual weakness . They had to take " refuge" in the palace. Conversely, palace leadership ought not to be i n the hands of active political partisans but i n the hands of neu­ tral negotiators who were able to keep faction out of the palace : I heard from Saionj i the truth about Grand Chamberlain Tokudaiji' s resignation . Although t h e domain cliques had coveted that office, trying any number of schemes up to that time, Tokudaij i had dedicated himself and been sincere in his service to the throne . 63

According to Hara , therefore , Tokudaij i as Privy Seal and Grand C hamberlain was the ideal type for the post : neutral , above faction, and "sincere . " As we have seen, Tokudaij i was a master of liaison to the poi nt of obscurity . Hara's actions while Katsura was Privy Seal and Grand Cham­ berlain, however, reveal that Hara expected Katsura to act as a non­ partisan negotiator, despite t he longstanding competition between the two over political power. Throughout Katsura's brief tenure in 6 1 . Speech b y Ozaki Yukio, H ouse o f Representatives, a s published i n The Japan Mail, February 1 5 , 1 9 1 3 , pp. 1 97 -98; cited in Scalapino, p. 1 94 . 62. Hara Kei Nikki ( 1 965 ) , 3 : 24 5 . M y emphasis. 6 3 . Ibid.

1 40

OFFICES

AND

OFFICERS

OF

PALACE

L E A D E R SH I P

high palace office Hara attempted t o enlist his efforts a s mediator between the Saionj i Cabi net and the Army to prevent a direct confrontation between the two which might bri ng about Saionj i 's resignation. On August 1 8 , 1 9 1 2 " Katsura made it clear that from now on he would not participate in politics " 64 but it was Hara himself w ho attempted to reinvolve him . On November 1 6 , for example, Hara met with Katsura to discuss the Army 's demand for two new divisions and to enlist Katsura's efforts as a negotiator. At first Katsura declined , argui ng that in his present office he could not speak on such issues . B ut Hara pressed him, and he fi nally agreed to arbitrate on the Cabinet's behalf. Although Hara doubted that Katsura would be si ncere in his efforts,65 he clearly bel ieved that the office of Privy Seal-Grand C hamberlain was no bar to participation as a negotiator in the political process . On the contrary , Hara ex­ pected the Privy Seal-Grand C hamberlain to act as an arbitrator , especially when it came to the survival of a Cabi net , regardless of the political preferences of the individual holdi ng the offices . Within the court Katsura also " tutored " the new Taisho Emperor and coordi nated court and government functions . His biographer im­ plies that one of Katsura's main functions, like that of the jiho i n the 1 870s, was to cultivate the imperial virtue : Although the office of Privy Seal , a completely noble and pure office admi nistering the i mperial seal in service to the t hrone, was no more than a court office servi ng t he throne i n attendance , there was, i n regard to the present E mperor [Taisho] , a certain necessity for someone with qualifica­ tions as a master teacher in politics to be Privy Seal .66

Like the young Meij i Emperor i n the 1 870s, the new Taisho Em­ peror i n 1 9 1 2 had to be brought i n line with the times-his " virtue" made responsive to the dominant political forces of the day . Katsura was particularly concerned with devising a regular series of lectures to the emperor on basic political information and current events : essentials of the constitution , "the reason why Japan's national polity was unsurpassed in the world , " other political systems of the world , ideal emperors of antiq uity , .Japanese military preparedness 64 . Ibid . , p. 247 . 6 5 . Ibid . , p. 2 62 . 6 6 . Tokutomi Iichiro, Koshaku Katsura Taro Den, 2 : 5 9 5 .

OFFICES

AND

OFFICERS

OF

P A LA C E

L EA D E R S H I P

141

a nd that of the Western nations . Military lectures were t o be given by the chiefs and deputy chiefs of the armed services ' General Staffs .67 In his coordinati ng duties Katsura had considerable contact with the H ome Mi nister, Hara Kei , in regard to imperial outi ngs and im­ perial amnesties for political prisoners .68 The Privy Seal-Grand C hamberlain was therefore a key figure in controlling the political i nformation reachi ng the emperor as well as a liaison-coordi nator of court functions that i nvolved government cooperation. After Katsura's brief and stormy period in high court offices , no one was ever again to be Grand C hamberlain and Privy Seal concur­ rently. Nor was a ny ex-Prime M i nister ever to be appoi nted Grand C hamberlai n . Between 1 9 1 2 and 1 927 there were but three Grand C hamberlains : two were of Court Noble origin and one was related to the shogunal house of Tokugawa . All three were predominantly court figures, alt hough one achieved the ra nk of maj or general (Takatsukasa H i romichi) a nd a not her served as G overnor of Saitama Prefecture ( O gimachi Sanemasa) . The period between 1 9 1 2 a nd 1 927 thus represents a return to the tradition that the Grand C hamberlain was to be a neutral figure from the flower of the as­ cribed aristocracy . Si nce the Privy Seals from 1 9 1 4 a nd 1 92 2 were aged but powerful Elder Statesmen and the H ousehold Mi nisters from 1 9 1 0 on were competent bureaucrats from the outside , it would appear that there was no need for powerful men i n the office of Grand Chamberlai n . I n 1 92 7 , however, Chinda Sutemi , an aged career diplomat who had been educated in the United States, was made Grand Cham­ berlain in a fi nal break with the precedent of neutral and aristocratic Grand Chamberlains . His successor as Grand Chamberlain in 1 92 9 , Admiral Suzuki Kantaro, was appoi nted directly from the post of C hief of the Navy General Staff. From 1 92 9 to 1 945 the post of Grand Chamberlain was occupied exclusively by Navy admirals . Not only does t he appoi ntment of Suzuki reveal the bureaucratiza­ tion of the Grand Chamberlain's office from the outside; it also reveals the efforts of partisan palace advisers , the constitutional monarchists, to balance the instituti onal i nfluence of the Army at the 67 . Ibid . , pp. 602-03 . 68. Hara Kei Nikki ( 1 965) , 3 : 2 5 9 .

1 42

OFFICES

AND

OFFICERS

OF

PALACE

L E A D E R SH I P

emperor's side . The Chief Aide and a maj ority o f Aides had always been ra nki ng Army officers whom the Army had nomi nated . By 1 92 9 , moreover, the Army had mou nted its claim to be the leading institution of prerogative and thereby entitled to declare the Imperial Will in both domestic and i nternational affairs . With a Navy ad­ miral at the emperor's side, the palace leaders could pit the i nstitu­ tional interests of the Navy agai nst those of the Army without frontally violati ng the prerogative of supreme command . And si nce the palace leaders had sole responsibility for appoi nting the Grand Chamberlai n , as they did not i n the case of the C hief Aide and Aides , they could choose a Navy officer " i n line with the times"­ and their partisan purposes-who would be useful i n controlling the military, i ncluding the Navy itself. Finally, because the office of Grand Chamberlain was q uite possibly the most ambiguous and flexible of the four palace offices , its i ncumbent had considerable dis­ cretion i n the performance of his vague duties-on his ow n initiative or at the behest of partisan colleagues . The appoi ntment of Admiral Suz uki and his subsequent actions as Grand Chamberlain from 1 92 9 to 1 93 6 ill ustrate these assertions. Al­ though Suz uki's successors were also admirals, there is no i ndication that they became involved i n politics as palace partisans , as did Suzuki . Si nce the palace partisanship of the constitutional monar­ chists was drastically curtailed after the assassi nations of February 2 6 , 1 93 6 , the Grand Chamberlains who succeeded Suzuki were far more neutral figures . Unity in government was also restored in some measure after 1 936 by the Army and its all ies in the bureau­ cracy and I mperial Diet . It might be wel l , then, to concl ude our dis­ cussion of the Grand C hamberlain with Admiral Suzuki and his relation to palace involvement in politics . When Grand C hamberlain Chi nda Sutemi died in office , the I m­ peri al H ousehold M i nister, lchiki K i t okuro , persona l l y pressed Suzuki into the post of Grand C hamberlai n , despite the Navy's reluctance to part with Suzuki and Suzuki 's ow n hesitations. lchiki had consulted with his palace colleague and fellow constitutional monarchist , Privy Seal Makino Nobuaki , and the emperor himself prior to pressuri ng Suzuki to accept . The reasons why the palace selected Suzuki , as given in Suzuki's biography , were : ( 1 ) the Chief

OFFICES

AND

OFFICERS

OF

P A LA C E

L E A D E R SH I P

1 43

Aides had always been Army generals, (2) there were no "suitable" candidates in the civil bureaucracy , and (3) the only other Navy pros­ pect was from Satsuma and would not be acceptable because Privy Seal Makino was also from Satsuma .69 M ore important , however, was the congeniality of Suzuki 's views with those of the two constitu­ ti onal monarchi sts , H ousehold M i nister lchiki a nd Privy Seal Makino, who managed the palace under Pri nce Saionj i 's direction.70 It is claimed , for example, that Suzuki "detested " politics and that he believed military men should stand completely clear of political activity . He is also alleged to have believed that by the late 1 920s military officers were clearly violati ng the Emperor Meij i 's " I nstruc­ tions to Soldiers and Sailors , " which forbade military i ntervention in politics .71 Despite bei ng a career Navy officer, Suzuki apparently sup­ ported Mi nobe's theory of constitutional monarchy .72 In terms of his institutional constituency and political views , therefore , Suzuki was ideally suited to the partisan purposes of the constitutional monar­ chists in their struggle against the militarists and renovationists . That Suzuki served the interests of the constitutional monarchists is well illustrated by his alleged i nvolvement in the fall of the Ta­ naka Cabinet in 1 92 9 and his clear i nvolvement in the London Naval Treaty controversy of 1 930. 6 9 . Suzuki Kantaro Den (Tokyo: Suzuki Kantaro Denki Hensan l i nkai, 1 960), pp. 1 01 -02 . It is i nteresti ng to note that Suzuki's wife had served as a companion to the emperor when he was age three to fourteen ( 1 904- 1 5 ) . Suzuki was therefore l inked to the palace long before he became a candidate for G rand C hamberlain (Kanroj i Osanaga, p. 28). 70. A career bureaucrat i n the H ome M inistry, lchiki was regarded as a protege of Yama­ gata Aritomo (Roger F. Hackett, Yamagata A ritomo in the Rise of Modern japan, 1838-1922 !Cambridge, Mass . : Harvard University Press, 1 97 1 ] , p. 268) . As we have seen, however, lchiki was the originator of the M inobe theory of constitutional monarchy . He was also ac­ cused by conservative household officials of "disrespectful acts" toward the imperial house and the late Taisho Emperor a nd was i ntensely disliked by court conservatives, t he Army, and right-wing groups who considered his views on const itutional monarchy to be the source of his disrespect . Public rumor had it t hat lchiki did not even k now the Japanese national anthem (Nezu Masashi , p . 1 36). Makino was a cosmopolitan bureaucrat whose career had been largely diplomatic. Foreign M inister in Admiral Yamamoto Gombei's first Cabinet ( 1 9 1 3-1 4), Makino had served in both Saionj i Cabi nets, first as M inister of Education ( I 906-08) and then as M inister of Agriculture and Commerce (I 9 1 1 - 1 2 ) . As Privy Seal, M.� kino was also the subject of wild rumors, such as having fl i rted with the empress in t he palace forests, which were evidently believed by right­ wing societies (ibid . , p. 1 37 ) . Makino was sixty-three when he became Privy Seal in 1 92 5 . 7 1 . Suzuki Takeshi, Shusenji Saisho Suzuki Kantaro-o: fusan Kaiki ni Omou (Tokyo: Jimbutsu J idai Sha, 1 960) , pp. 20-2 2 . 7 2 . Nakase juichi , Kindai ni okeru Tenno Kan (Tokyo: Sanichi Shobo, 1 963), p . 2 7 .

1 44

OFFICES

AND

OFFICERS

OF

PALACE

LEA D E R SH I P

Tanaka G iichi 's Cabi net resigned in J uly 1 92 9 . The cause o f his fall was his disposition of the C ha ng Tso-lin affair. C hang Tso-li n , Warlord o f Manchuria, had been killed b y Japanese Army officers and their conspirators in J u ne 1 92 8 . The Army , in which Prime M i nister Tanaka had spent his entire career before t u r n i ng politician, and t he political party of which he was president , the Seiyiikai , wished to hide the facts by passing off the murder as the work of Nationalist Chinese soldiers . Pri nce Saionj i , his colleagues in the palace , and the leaders of t he other mai n political party, the M i nseito, were aware that the facts were widely know n abroad and could not , nor should not , be so clumsily buried . They therefore argued for full disclosure of the truth and for strict action agai nst the Japanese officers i nvolved-courts-martial . While the Army and its allies i nsisted that admission of the facts and strict punishment would tarnish the Army a nd the emperor, Saionj i and his com­ patriots argued that international trust in Japan would be seriously undermi ned if Japan were unwilling or unable to mai ntain her ow n house in order, and that t his would be an even more serious affront to the emperor. Shortly after Chang's murder, Prime Minister Tanaka reported to the emperor that the plotters were evidently Japanese officers sta­ tioned in M anchuria. I f that were the case , Tanaka continued , strict measures would be taken. Tanaka also told the Privy Seal and Pri nce Saionj i the same thing i n the strictest confidence .73 Sub­ sequent i nvestigation by the Army itself verified beyond a doubt that Japanese Army officers were responsible.74 In the face of Army and Seiyiikai pressure to hide the affair as much as possible and to deal lightly with the officers i nvolved , Tanaka fi nally reported to the throne i n M ay 1 92 9-almost a year after Chang's murder and five months after Suzuki's appoi ntment as Grand Chamberlai n-that the plotters were not Japanese but that because the murder took place i n an area u nder Japanese guard the responsible officers would b e dis­ cipli ned " admi nistratively" -transferred to different posts. The em7 3 . Suzuki Haj i me, ed . , Suzuki Kantaro jiden (Tokyo : Jij i Tsiishin Sha, 1 968), p. 254. Okada Keisuke, Okada Keisuke Kaikoroku (Tokyo: Mainichi Shimbun Sha, 1 950) , p . 37 . Harada Kumao, I : I . 74. Okada Keisuke, p. 3 7 .

OFFICES

AND

OFFICERS

OF

PALACE

L E A D E R SH I P

1 45

peror poi ntedly remarked that this report was i n flat contradiction to w hat Tanaka had originally reported . According to the recollections of Admiral Okada Keisuke , Navy M i nister in Tanaka 's Cabi net , the emperor told Tanaka when he at­ tempted to explain the contradiction: " There is no need for an explanation . " Tanaka then returned to meet his Cabi net Mi nisters . They urged that Tanaka attempt to arra nge another audience and explain w hy the government had made its fi nal disposition of the Chang Tso-lin affair in the way it did . B ut when Tanaka went to see Grand Chamberlain Suzuki to arrange the audience , claims Okada , Suzuki said : " I shall pass your req uest on, but I fear that it might be to no avail . " Having lost the emperor's confidence , Tanaka decided to resign.75 Pri nce Saionj i 's secretary , Harada Kumao, gives a slightly different account . After the emperor had scolded Tanaka for contradicting hi mself, Harada relates , the emperor retired i nto the inner palace and told Grand Chamberlain Suz uki , "I simply don't u nderstand what Prime M i nister Tanaka says . I don't want to hear him out agai n . " Harada then claims that " since the Grand Cham­ berlain was new to his post and not yet used to it , " Suzuki told Ta­ naka exactly what the emperor had said .76 Neither Okada nor Harada were present when the alleged con­ versations took place and their records are obviously hearsay . Suzuki's biographers claim that he played no part i n Tanaka 's resignation: [Tanaka's resignation) . . . had no relation whatever to the work of the G ra nd C hamberl a i n . The matter was between t he Army and the government . Matters i nvolving the mil itary fell under the j u risdiction of the Chief Aide-de-Camp; if it were a matter i nvolving politics , the Lord Keeper of t he Privy Seal would have been responsible for renderi ng advice a nd assista nce in attendance. However, it was publicly bruited about that Grand C hamberlain Suz uki had gotten i nvolved and that for this reason the Cabi net may have fallen.77

I n his autobiography Suzuki also claims that he was i n no way i n­ volved : Then, si nce the Army M i nister made a report to the throne that was 75. I bid . , pp. 40, 4 1 . 7 6 . Harada Kumao, I : I I . 7 7 . Suzuki Ka ntaro Den, p. 1 07 .

1 46

OFFICES

AND

OFFICERS

OF

PALACE

L E A D E R SH I P

completely different from w hat t he Prime M i nister had previously reported to the throne , the Emperor poi ntedly i nquired about the disc­ repancy between t he two reports w hen Prime M i nister Tanaka audienced . That is w hat M r . Tanaka told me U ust ] after he withdrew in humble respect from the presence of t he E mperor . Although [ M r . Tanaka ] also said he was goi ng to resign, I, as Grand Chamberlain, could make no reply w hatever to this. As far as the Prime M i nister's report to the Throne is concerned , t he Grand Chamberlain [ Suzuki himself, that is] was not , of course , in attenda nce . I haven't t he slightest notion about the nature of the conversation between His M ajesty a nd the Prime M i nister. The Prime M i nister merely told me i n confidence of his resolve [to resign] . This is t he cause for t he resignation of the Tanaka Cabi net . Later I heard the fol low ing about the state of affairs at t he time: Various contentions seethed in the Cabi net . There were even some who held that it would not do for the Pri me M i nister to decide at his own dis­ cretion that the Cabi net should resign : they wanted to report on the situation to the Throne once again and dispose of the matter without re­ signing. Mr. Tanaka refused , sayi ng that his will had been broken and t hat he simply couldn't do it . That was probably w hat did it. Two or t hree Cabi net ministers came to see the Grand C ha mberlain [ Suzuki] and I think said somethi ng to the ef­ fect that it would be a good thing if I were to be a go-between-t hat is, use my good offices between His Maj esty and the Prime M i nister. I remember havi ng refused , tel l i ng them : For me t hat would be improper . The [ office of] Grand C hamberlain is not t hat kind of position. The Grand Chamberlain [ Suzuki ] merely heard w hat the Prime M i nister vol unteered to let out and left it at that ; [the Grand Chamberlain] is unable to do anything more than that .78

B ut if Suzuki was not actively involved in bri ngi ng the Tanaka Cabi net dow n , he did very little to help the Prime Mi nister stay in of­ fice . Although it was quite proper for him to refuse to mediate on be­ half of the Cabi net , si nce that was the role of the Privy Seal , the net result was to assist Pri nce Saionj i , his palace partisans , and opposition leaders in the House of Representatives i n ousting Tanaka . One of Saionj i ' s pri ncipal allies was Privy Seal Makino, who was allegedly maneuvering " behind the scenes " against Tanaka because he, like Saionj i , was opposed to " military diplomacy" in China .79 It would 7 8 . Suzuki Hajime, pp. 254�5 5 . Note that Suzuki's account states t hat t he Army M inister, not Tanaka, made the contradictory report . 7 9 . Nezu Masashi, p. 99.

OFFICES

AN D

OFFICERS

OF

PALACE

L E A D E R SH I P

1 47

also appear that the emperor himself, who had been sharply critical of Tanaka 's "contradictory" actions , was one of Saionj i ' s maj or allies . As a result , Grand Chamberlain Suzuki was accused of partisan activities at the emperor's side. Suz uki 's biographers claim that they have no idea w here t he " fact s " " got twisted , " givi ng rise t o "groundless misunderstandings . " 80 B ut there are a t least three basic factors that combi ned to create such " misunderstandi ngs" : ( 1 ) the ambiguity of the Grand Chamberlain's role in law and practice, (2) the " privatized " nature of Japanese decision making, and (3) the i ntensity of competition among t he institutions of imperial pre­ rogative and their coalitions of i nstitutionalized elites to declare the I mperial Will in politics . I noted at the outset of this discussion of the Grand C hamberlai n that his office, as defined by I mperial H ousehold M i nistry regula­ tions, was perhaps the most ambiguous of the four leading palace of­ fices . I n practice, moreover, that office between 1 89 1 a nd 1 9 1 2 was held concurrently by the Privy Seal . Si nce the Privy Seal was the emperor ' s chief political adviser in t he palace , t hi s role was i nevitably merged with that of the Grand C hamberlai n duri ng those years . We have also seen that C hamberlain I rie considered it " natural" that the Grand Chamberlain in the 1 930s s hould have acted as Privy Seal when the Privy Seal was not at the palace . Moreover, if the Grand C hamberlai n was a powerful representative from one of the governmental i nstitutions of prerogative, rather tha n a neutral and discreet court bureaucrat from the high aristocracy , he w ould very probably be viewed as a " polit ical part i s a n " by government leaders and the public at large . This was certainly true of Katsura Taro in 1 91 2 , and it was equally true of Suzuki Kantaro from 1 929 to 1 93 6 . It is highly revealing , for example, t hat Grand C hamberlain Suzuki was asked by some of Tanaka's Cabinet Ministers to act as "go-between" for t he emperor a nd the Prime M i nister in order to save the Cabi net ; in the eyes of political actors outside the palace, the Grand C hamberlain was a potential political negotiator, despite Suzuki 's proper disclaimer that the office of Grand Chamberlain was " not that kind of position . " Finally, the 80. Suzuki Kantaro Den, p. 1 08 .

1 48

OFFICES

AND

O FFICERS

OF

P A LA C E

L E A D E R SH I P

Grand Chamberlain scheduled audiences for the Prime M i nister and M i nisters of State . If, for whatever reason, a Prime M i nister or M i nister of State was unable to report to the throne, or his report was delayed , the Grand C hamberlain could be, and occasionally was, accused of " blocking access to the throne . " Si nce there was n o record of w hat the Grand Chamberlain actually said or what he thought and did i n the course of arranging audiences , such accusations could not be refuted or substantiated . Even if Suzuki disclaimed , as he certai nly did , any partisanship i n scheduling audiences , he could be accused of disi ngenuousness by those who considered themselves i nj ured . Given the verbal and closed nature of the prewar decision-maki ng process, as the disposition of the Chang Tso-lin affair ill ustrates , it is no wonder that the " facts" " got twisted " and that leadi ng court and government officials were subject to " misunderstandings . " M ore important than the ambiguity of the Grand Chamberlain's office , therefore , was the nature of the decision­ maki ng process itself. Privatized decision making on issues of the gravest national import-such as C a b i net format ions, Cabi net resignations , and basic foreign and domestic policy directions­ characterized the Japanese pol itical process throughout the prewar period . Since there were no public transcripts of these exchanges and since the entire process was kept from public scrutiny, there was no way to verify the actions of the men i nvolved . Only the decisions were know n ; how , why, and even by whom decisions were made were therefore matters of endless speculation. As a result , charges and cou ntercharge s , rumor mongeri ng, u ncert a i nty , a nd " m i s u nder­ standings" i nfused the entire political process . The confusion was exacerbated i n the 1 920s and 1 930s by the dedicated efforts of Japan's extensive and energetic mass media industry in unearthing and pub­ licizing any and all political rumors . Had there been records , had those records been public, had voting for public officials been a critical part of the political process, had open discussions and voting in the legislature been a decisive factor i n shaping political decisions , rumor might have played only a Drew Pearson role in politics . But such a public or " socialized " process of decision making was neither prac­ ticed nor considered desirable in prewar Japan. Ultimately all deci­ sions, both ideally and in practice , were to be made by the "emperor's

OFFICES

AND

OFFICERS

OF

PALACE

LEA D E R SH I P

1 49

advisers" communicating verbally among themselves and issui ng thei r decisions in the name of the emperor from the presumably secure sanctuary of the imperial palace . Privatized decision making and the flexibility of lead ing palace of­ fices were consistent features of Japanese politics throughout the entire prewar period . Conseq uently, Japanese politics had always been marked by rumor, uncertainty, and " misunderstandi ngs . " Along with the politicization of society, 8 1 however, the i nstitu­ tionalization of imperial prerogatives and political elites added new dimensions to these traditi onal by-products of Japanese politics during the late 1 920s and early 1 930s , produci ng an i ntensity and extensiveness of rumor mongeri ng and a degree of u ncertainty i n basic policy directions that challenged the effectiveness of Japan's traditional system of privatized decision maki ng. The Chang Tso-lin affair, for example, profoundly affected the disposition of institutions competi ng to declare the Imperial Wil l . If the Japanese officers who had plotted and executed C hang's murder were court-martialed and severely punished , both the Army 's China ambitions and its i nstitutional right to declare the Imperial Will i n an i ncreasingly wide arena would have been seriously j eopardized . The institutional strength of the Army in politics was ultimatel y demonstrated by the fact that Pri nce Saionj i ' s views did not prevail , but it took the Army a year of pressure and behi nd-t he-scenes ma­ neuveri ng to achieve its purposes . And Tanaka's Cabi net collapsed as a result . Grand Chamberlain Suzuki was i nevitably i nvolved i n the rumor mongering and back-stage maneuverings because he scheduled audiences for the Prime M i nister. Regardless of the " facts , " Suzuki was held at least partially responsible for Tanaka 's fall by Tanaka 's political party, the Seiyuka i . One of t he i nstitutions of prerogative, or its components , had been adversely affected by a decision reached verbally i n the closed corridors of the i mperial palace . During and after the 1 920s the i nstituti ons of prerogative and thei r K 1 . The expansion of mass media after the 1 870s, the emergence of mass demonstrations . o nd popular movements after 1 905 , the proliferation of right-wing societies after 1 9 1 8 , and the enfranchisement of all adult males in 1 92 5 are some of the promi nent i ndications of such politicization .

1 50

OFFICES

AND

O FFICERS

OF

PALACE

L E A D E R SH I P

el ites i ncreasingly " twisted " the " facts " t o serve their partisan and parochial interests . As the i nstitutional competition for political power widened and i ntensified , so did the rumor mongering. Allega­ tions of violati ng one or another prerogative, of interferi ng in a given i nstitution's alleged decision-maki ng domai n, of schemi ng behi nd the scenes , of engaging i n political activities improper to one's office, were used to vault one i nstitution and its allies i nto political power agai nst a multipl icity of competi ng institutions and their allies . Goverment by rumor occurred precisely because of the intensely privatized nature of decision making; such rumor mongeri ng and its attenda nt un­ certai nties i n policy directions were vastly i ntensified when the paroc­ hial elites of the plural i nstitutions of prerogat ive replaced the autono­ mous i ndividuals of Restoration leadership at the top of Japanese political society. I nstitutional rat her than " personal" resources were brought to bear in the struggle for political power, yet the system for resolving conflicts remai ned based on verbal decisions and face-to-face relations centered on the imperial palace. I f G rand C hamberl a i n S uz u k i ' s a l l eged i nvolvement in t he collapse of Tanaka Giichi 's Cabi net illustrates palace partisanship i n politics o n the one hand , therefore, i t also illustrates the nature and effect of privatized decision maki ng in an environment of institu­ t i onalized prerogati ves a nd parochial e l i tes . B ut t he partisa n activities of Prince Saionj i and his constitutional monarchist allies i n the palace-the " palace group" as they have been called b y one Japanese political scientist82-and the problems of privatized de­ cision making in this period are much more clearly illustrated by the London Naval Treaty controversy of 1 930. Whereas the C hang Tso-lin affair basically i nvolved only the Army , the Prime M i nister and his Cabi net , and Pri nce Saionj i and his allies , the London Naval Treaty brought to bear a complicated set of i nstitutional forces : the Navy M i nistry, which admi nistered the naval establishment ; the Navy General Staff, w hich commanded the Navy 's operati ng forces ; the Foreign Mi nistry , w hich managed the prerogative of treaty maki ng; the Prime M i nister and his Cabi net , who were responsible for setti ng overall national policies ; the Sup.

82 . M atsushita Keiichi , Sengo Minshushugi no Tenbo (Tokyo: N i hon Hyoron Sha, 1 965 } , 2 1 9.

O FF I C E S

AND

OFFICERS

OF

PALACE

L E A D E R SH I P

151

preme War Council , which advised the emperor on basic military policies; and the Privy Council , which advised the emperor on treaty ratification. Behi nd each stage in policy formation, negotiation, and ratification in w hich these institutions were i nvolved was the firm hand of Pri nce Saionj i who, at eighty years of age, was determi ned to see the London Naval Treaty through to a successful concl usion . As in the C ha ng Tso-lin affair, the necessity for Japan to act responsibly in international affairs was uppermost in Prince Saionj i ' s mind . A t t h e outset o f t h e negotiations in London, which began for­ mally in January 1 930, the Navy hard-liners had publicly taken a "70 percent or bust " stand : Japan had to have a minimum tonnage ratio of 70 percent with the American fleet in all categories of auxiliary combatant ships , from heavy cruisers to submarines . Saionj i was severely critical of this position. If Japan were completely unwilling to compromise, the London Conference would fail and the responsibility for its failure would rest securely with Japan. Saionj i therefore argued that Japan should lead the conference to a successful conclusion, increasing Japan's international stature and impressing the Great Powers that Japan intended peace . In alliance with Great B ritain and the United States , continued Saionj i , Japan could share in wielding "the baton of command . " By insisting on a 70 percent ratio Japan would " throw away its grip on the handle" and "j oin the ranks of France and Italy . " Saionj i concluded that there was nothing to .Japan's interest in parting company with Great Britain and the United States : in fact , closer ties between Japan and these two Great Powers were desirable . 83 Given Saionj i ' s pro-Anglo-American stance , what resources did he have to implement his views? First , Prime M i nister Hamaguchi and his Cabinet were overwhelmingly of Saionj i 's opi nion. As president of the Minseito in the House of Representatives , the Prime M inister had a strong base of support . Second , Japan's chief delegate to the conference was ex-Prime Minister Wakatsuki Reij iro, a pro-Saionj i bureaucrat with w hom the Prime M i nister had been very closely a ssociated . Wakatsuki was responsible for signing t he America n ! D . Harada Kumao, 1 : 1 7- 1 8 . It should be noted t hat Saionj i , like all Japanese leaders of l i lt" modern peri od , was deeply concerned with Japan's international prestige and national

'"·•·ngt h. He differed drastically from the militarists and other right-wing leaders on how that pn·sr igc and strengt h were to be established , maintained , and used .

1 52

OFFICES

AND

OFFICERS

OF

PALACE

L E A D E R SH I P

compromise proposal of March 1 930, w hich ultimately led t o the successful conclusion of the London Naval Treaty .84 Third , Saionj i had powerful support withi n the Navy itself, most notably from Ad­ mirals Okada Keisuke and Saito Makoto. As a Supreme War Councillor, Okada was i n a position to check those "Navy Elders" w ho opposed any compromise on t he 70 percent ratio. Pri nce Saionj i had urged Okada t o act a s t he treaty's " matchmaker. " Okada pursued that role by helping to control his hard-li ne colleagues i n the Navy General Staff a nd t he Supreme War Council from March to October, w hen Japan finally ratified the compromise treaty .85 Ad­ miral Saito, at that time Governor General of Korea, was strongly in favor of the treaty and in complete accord with Pri nce Saionj i . Al­ though he was unable to stay i n Tokyo to keep pressure on his hot­ headed colleagues in the Navy , Saito lent public support to the treaty by, for example , issui ng a statement to the newspapers praising the results of the conference and criticizi ng the actions of the Navy General Staff as " most disagreeable. " 86 Fourth, the Foreign Min­ istry was led by Baron Shidehara Kij uro, a long-ti me opponent of " military diplomacy " in China and a strong pro-treaty partisan . Shidehara met w i t h Admiral Okada a t least once and urged Okada to help bring the treaty to a successful conclusion. Fifth, Pri nce Saionj i had Prince Konoe Fumimaro in t he House of Peers to do his biddi ng in that component of imperial prerogative. 57 Finally, Saionj i had his supporters at the side of the emperor : Privy Seal Makino, Household M i nister lchiki , a nd Grand Cham­ berlain Suzuki . Throughout the treaty controversy Saionj i was i n constant touch with all three, usually through his tireless secretary , Harada Kumao. I n March 1 930, for example, Saionj i told Harad a : " It is extremely important a t this j uncture t hat the I mperial H ousehold M i nister and Grand C hamberlain, along with the Lord Keeper of the Privy Seal, give us their complete understanding on this matter. " 88 After the American proposal had been approved by the Japanese government in Apri l , the problem of Privy Council ra84. 85. 86. 87. 88.

Ibid . , p. 1 9 . Suzuki Kantaro Den, p . 1 1 0 . Okada Keisuke, pp. 44-5 5 . Harada Kumao, 1 : 67 . Harada Kumao, 1 : 49-50. I bid . , pp. 2 6 , 2 1 -2 2 . Ibid . , p . 2 8 .

OFFICES

AND

OFFICERS

OF

PALACE

L E A D E R SH I P

1 53

tification arose . At that time Saionj i told his secretary : "The Privy C ou ncil problem will ultimately come out all right if those at the side of the Emperor, like the Lord Keeper of the Privy Seal and the Grand Chamberlain, hold fast . " 89 A nd hold fast they did . Grand C hamberlai n Suzuki was very much i nvolved in keeping Admiral Kato Kanj i , Suzuki 's successor as C hief of the Navy General Staff, in l i ne, although the exact nature of Suzuki's i nvolvement is a matter of controversy and speculation. Suzuki heard from the C hief Aide-de-Camp, for example, that Kato had requested an audience for April 1 . Prime M i nister Hamaguchi had also req uested an audience , via Grand Chamberlain Suzuki , for that same day to obtain imperial sanction for the government 's ap­ proval of the American compromise . Suzuki suspected trouble: Kato probably wished an audience in order to denounce the government 's position on the treaty. Suzuki therefore called Kato to his official residence. Did Kato intend a contrary report to the throne ? When Kato replied that he did , Suzuki told him that this would cause His Majesty great difficulty : Kato should give careful consideration to the consequences of putti ng the emperor in a strait between the Prime M inister, w ho was responsible for politics , and the Chief of Staff, who was responsible for supreme command . As a result of his talk with Suzuki, Kato agreed to withdraw his April 1 report to the throne : The matter was most simple. As a friend , and as a senior in the Navy , Suzuki employed honest counsel so that no fault would lie with Kato and so that no trouble would be caused His M aj esty . Kato, too, understood the good will of his senior and withdrew his report to the throne of April 1 . Therefore, there should have been nothing that would become an issue , but . . go .

.

According to his ow n recollection, Suzuki called Kato dow n sharply , telling h i m in effect that the size o f the naval establishment w a s no busi ness of the Navy General Staff, which was in charge of opera­ tions, not admi nistration. As Chief of the Navy General Staff, Kato must go along with the Prime M i nister and the Cabi net on w hich the Navy Mi nister sat .91 89. Ibid . , p. 5 2 . 9 0 . Suzuki Kantaro Den, p p . 1 1 2 1 3 9 1 . Suzuki Haj i me, pp. 256-57. -

.

1 54

OFFICES

AND

O FFICERS

OF

PALACE

L EA D E R SH I P

This account b y Suzuki and his biographers , however, i s but one of at least three differi ng accounts of the same event . Admiral Okada states in his memoirs that Kato came to see him on April 1 . Saying that his request for an audience on that day had apparently been blocked by someone close to the throne, Kato told Okada to find out from Grand Chamberlain Suzuki what had happened . Okada saw Suzuki immediately and was greatly relieved to hear that Suzuki had not blocked Kato's report . Since the emperor's schedule had already been filled for that day , Suzuki i nformed Okada, it would have been very difficult to arrange an audience for Kato.92 Suzuki also told Harada Kumao that he had not blocked Kato's report . Poi nting out that reports by the Chiefs of Staff had no relation to the Grand Chamberlain, Suzuki argued that such reports fell under the j uris­ diction of the C hief Aide-de-Camp. Since the emperor simply had no time that day , said Suzuki , Kato's report had to follow that by the Prime Mi nister.93 Suzuki's action on April 2 , the day after he had allegedly coun­ seled Kato to withdraw his report to the throne, is also the subj ect of conflicting accounts. On April 2 Kato made a grudgi ngly con­ ciliatory report to the t hrone , telling the emperor that the Navy would make do with the treaty, though this would be difficult.94 A newspaper later alleged that Grand Chamberlain Suzuki was i n at­ tendance duri ng Kato's report , in place of the C hief Aide whose duty it was to attend such reports by officers of the Supreme Com­ mand . Harada categorically denies this , but Suzuki 's biographers are equally categorical in claiming, on the contrary , that Suzuki was in attendance , taking the "completely unprecedented " action of re­ placi ng the Chief Aide.95 Whatever the facts , Suz uki 's behavior was subject to the same ki nd of " misunderstandi ngs" that arose regardi ng his alleged in­ volvement i n Tanaka Giichi's resignation less than a year earlier. He became the target of censure by antitreaty forces i n the govern­ ment and the public at large as they mobilized to defeat the treaty's 92 . 93 . 94 . 95 .

Okada Keisuke, p p . 55-56. Harada Kumao, I : 48 . Suzuki Kantaro Den, p. 1 1 4 . Harada Kumao, I : 1 05 ; Suzuki Kantaro Den, p. 1 1 4 .

OFFICES

AND

OFFICERS

OF

P A LA C E

L EA D E R SH I P

1 55

ratification i n the Privy Council .96 I n J une , for example, a reporter came to see Harada with the following story . The C hief Secretary to the Privy Cou ncil had said that the C hief Aide had told him that the Grand C hamberlain and Privy Seal had prevented the C hief of t he Navy General Staff from reporti ng his criticisms of the treaty to t he throne. In so doing, they had bypassed the C hief Aide, and therefore the C hief Aide had submitted his resignation. The re­ porter then told Harada that the C hief Secretary to the Privy Coun­ cil had concluded that the Privy Seal , Grand C hamberlai n , and H ousehold M i nister should " accept responsibility" for this state of affairs by resigning their offices .97 Thus a n alleged event on April 1 was relayed to Harada in J u ne by a newspaper reporter who heard about it from a treaty opponent who in turn heard about it from the C hief Aide w ho, as the responsible official at t he emperor's side regardi ng military reports , was allegedly abused by his fellow palace officers . A n inevitable product of prewar Japan's privatized decision­ making process , hearsay issui ng from the palace inundated the public with uncertai nties as i nstitutionalized actors and their allies manipulated unverifiable rumors to serve their preferences on issues of basic importance to t he direction of Japan's foreign a nd domestic policies . And given the oligarchic disunity i n government , the palace officers were bound to be partisan, w hether they were active or in­ active , si nce they managed the i mperial sanction of policies on which there was no effective consensus in the government . Though distorted and contradictory , the evidence does sustain the conclusion t hat Grand C hamberlain Suzuki was actively in­ volved as a palace partisan in seeing the treaty through to a success­ ful conclusion: I n this situation i t w a s inevitable that Grand Chamberlain Suzuki, too, having been the previous Chief of the Navy General Staff and being one of the leading officials at the emperor's side in the palace, would not be allowed to stand aloof, l ooking on indifferently . 98

The inherent flexibility of the Grand C hamberlain's role allowed '!6 . Suzuki Kantaro Den, p. 1 1 3 . Harada Kumao, I : 4 8 . 97 . Harada Kumao, 1 : 1 04-05 . 98. Suzuki Kantaro Den, p. 1 1 1 .

1 56

OFFICES

AND

OFFICERS

OF

PALACE

L EA D E R SH I P

Suzuki considerable latitude t o use his office a s h e saw fit . And be­ cause he was Kato's predecessor as Chief of the Navy General Staff he could talk with Kato "as a friend , and as a senior in the Navy " ­ possibly without violati ng his official role a s Gra nd Chamberlain i n his o w n mind but certainly not serving a s a neutral liaison betwee n government and court i n the minds of those adversely affected . More remarkable is the argument that , as one of the " lead ing officials at the emperor's side in the palace , " Suzuki could not remain aloof. As we have seen, the traditional role of the Grand C hamberlain was precisely such al oofness . The conseq uence of i nsti­ tutional fragmentation i n the government and of bringing a powerful bureaucrat from that i nstitutional world i nto the palace as Grand Chamberlain, however, was that the office of Grand Chamberlain became something quite different from the courtly office it had been in Tokudaij i Sanenori 's day . The Grand Chamberlain was no longer a frictionless conduit between a transcendental court and a united oligarchy i n government . He was one of three leadi ng palace officers serving the partisan purposes of the constitutional monarchists as they attempted to achieve their policy preferences by controll i ng the palace and manipulati ng it agai nst t he forces of extremism in a n en­ vironment of unstable government coalitions of institutionalized elites . The fourth palace leader was not a constitutional monarchist . B ut the actions of the C hief Aide-de-Camp i n the 1 930s i nvolved par­ tisanship nonetheless-a partisanship that more often than not contradicted the efforts of the constitutional monarchists who were his palace colleagues .

T H E C H I E F A I DE-DE-CA M P : C H I EF M I L ITARY ADVISER IN ATTENDANCE One other personal adviser, if he can be called so, on the Emperor's staff was the chief aide-de-camp ( jiju bukanch o ) . Certainly he was the least helpful member, for he seems to have functioned as the eyes and ears of the General Staff rather than as the Emperor's trusted aide . . . . The chief aide-de-camp was at this time [ 1 93 1 ] functioning as a General Staff infor-

O FFICES

AND

O FFICERS

OF

PALACE

L E A D E R SH I P

1 57

mant in the Palace rather than as a discreet and devoted assistant to the Emperor i n his difficult task of coordination . Yale Candee Maxon99

The unique position of the Office of Aides-de-Camp among the of­ fices of court has been described i n chapter 3 . The post of Chief Aide-de-Camp was explicitly recognized in a revision of Army regu­ lations on October 1 0 , 1 87 9 , although a set of regulations applying exclusively to the Office of Aides did not materialize until April 1 , 1 89 6 . Aides had acted as messengers between the emperor and the organs of military command as early as 1 87 5 , 100 however, and prior to the formal creation of the Office of Aides in 1 896, Aides had been acknowledged members of the General Staff. I n 1 893 , for example , the wartime Imperial Headquarters incl uded six Imperial Aides as well as both service ministers and the Chief of the High Command . 101 The Aides were the military equivalent of the C hamberlains, and the C hief Aide functioned as the liaison between the emperor and the military leadership, much as the Grand C hamberlain did between the emperor and the civilian leadership . More importantly, the C hief Aide was the military equivalent of the Privy Seal . He advised and assisted the emperor on mil itary affairs in regular attendance , much as the Privy Seal did on civil affairs . B oth emissary and adviser func­ tions were implicit in the 1 896 regulati ons discussed in chapter 3 : the Chief Aide-and even the Aides-reported and replied to the throne . Between April 1 896 and November 1 945 , when the Office of Aides was abolished , there were but eight Chief Aides , or one every 6 . 3 years . Measured by personnel changes , therefore , the office of C hief Aide was the most stable of the four leadi ng court offices : there was a new Household M i nister every 5 years , a new Grand Cham­ berlain every 6 years a nd a new Privy Seal every 4 . 6 years . For three of the four court offices , the Meij i period wit nessed the highest degree of personnel stability, but for the Chief Aide, the greatest stability occurred between 1 9 1 3 a nd 1 93 3 . There were two C hief 99. Yale Candee Maxon, Control of Japanese Foreign Policy: A Study of Civil-Military Uwalry, 1930- 1 945 (Berkeley and Los A ngeles : University of California Press, 1 95 7 ) , p. 54. 1 00 . Matsushita Yoshio, pp. 205-06. 1 0 1 . l naba Masao, "N ihon no Sensii Shidii: Sono Kikii to J issai ( 1 ) , " Kokubo 10 no. 7 ( M a rrh 1 962 ) : 7 9 .

1 58

OFFICES

AND

OFFICERS

OF

P A LA C E

LEA D E R SH I P

Aides from April 1 896 t o November 1 9 1 3 (9 years average) , two from November 1 9 1 3 to April 1 933 ( 9 . 5 years average), and four between April 1 93 3 and November 1 94 5 ( 3 . 3 years average) . The greatest instability at the emperor's side i n military affairs thus came i n t he period when the mil itary , particularly the Army , made its bid for i nstitutional hegemony over government decision making. As the Army's demand s on the government i ncreased in scope and intensity , it became i ncreasi ngly difficult to find a Chief Aide satisfactory to both the Army a nd the court , and this evidently produced frequent changes . Only w hen Japan settled down to mobilizing for total war under military leadership after 1 93 9 did the office of C hief Aide re­ stabilize. Until the 1 930s the Chief Aide, i nvariably an Army general , was notable mostly for his lack of visibility in the politics and operations of the military bureaucracy . The first Chief Aide, General Viscount Okazawa Kiyoshi ( 1 834-1 908) , was from the restorationist domain of Choshu . Prior to his appoi ntment he had served as Vice Minister of the Army and concurrently C hief of the Mil itary Affairs B ureau ( 1 8 9 1 -92 ) . He had also been a member of the staff at I mperial H e a d q u a r t e r s d u r i ng t he S i n o -J a p a nese W a r of 1 8 9 4 - 9 5 . Okazawa's successor, however, was General B aron Nakamura Sa­ toru ( 1 854- 1 92 5 ) , a famed swordsman of very little importance in the military bureaucracy . Havi ng served as Aide to the crow n pri nce in 1 895 and as Aide to the emperor in 1 896, Nakamura was ap­ pointed C hief Aide in 1 908 without any experience in maj or Army positions . When he resigned as C hief Aide in 1 9 1 3 , Nakamura was appoi nted to the Supreme War Council . Composed of mil itary venerables , the Supreme War Cou ncil was the military's eq uivalent of the largely civilian Privy Council . Nakamura's successor, General Uchiyama Koj iro ( 1 859-?), also held no Army posts of significance ; like Nakamura he had served previously as an Aide-de-Camp . With the appoi ntment of Nara Takej i as Chief Aide in 1 922 , the first of the Army 's " C hi na hands" came to the palace . In add ition to havi ng served as Director of the M ilitary Affairs B ureau , the first to have held one of the maj or Army offices si nce Okazawa, Nara had commanded the Japanese garrison in China. I ncluding Nara, four of the five C hief Aides from 1 922 to 1 945 had seen service in China :

OFFICES

AND

OFFICERS

OF

P A LA C E

L E A D E R SH I P

1 59

H onj o Shigeru , C hief Aide from 1 933 to 1 93 6 , had been commander of t he Kwantung Army w hen the Manchurian I ncident broke out i n 1 93 1 ; Hata Shunroku, Chief Aide briefly in 1 93 9 , had been com­ mander of the Shanghai forces in 1 93 8 , and Hasunuma Shigeru , C hief Aide from 1 93 9 to 1 9 t-.::i , had at least some experience i n C hi na . O nly Usami Okiie, C hief Aide from 1 936 to 1 93 9 , ap­ parently had no experience in the China theater. Since the Army's dema nds for policy control were i nextricably i nterwove n w i t h military actions on the C hinese mai nland , it is notewort hy that Army officers with Chi na experience served as C hief Aides for twenty of the twenty-three years from 1 922 to 1 94 5 . Like other palace leaders , the Chief Aides, with one exception, were all ending their active bureaucratic careers . The average age upon appointment was 5 6 , ranging from 5 1 to 62 , and the average age on leaving office was 62 . 1 , rangi ng from 54 to 7 4 . O nly one achieved high military office after serving as C hief Aide : Hata Shunroku became Army M inister in 1 93 9 directly after a brief tour of duty as C hief Aide. Until the 1 930s, therefore, the C hief Aides were largely drawn from the second level of the Army bureaucracy ; almost all terminated their active military careers as Chief Aides to the emperor . I n contrast to the position of Aide-de-Camp , the office of Chief Aide was not a steppi ng-stone to higher military office, with the one exception of Hata Shunroku. Two Chief Aides in the 1 930s did , however, become Privy Councillors . That Nara Takej i in 1 937 and Honj o Shigeru in 1 940 should have received such appoi ntments is evi­ dence of the weight that the Army as an institution carried in the councils of state i n the late 1 930s. Although the Army held the i nitiative in appoi nti ng the Chief Aide , it would be misleadi ng to label the C hief Aide as the " spy" of the General Staff in the palace throughout the prewar period . In the first place, the Chief Aides from 1 896 to 1 922 were neither maj or figures nor focal poi nts of conflict resolution. When Hara Kei , as Saionj i ' s H ome M inister in 1 9 1 2 , attempted to negotiate the Army's demand for two new divisions , for example, he went to Privy Seal Katsura Taro, not to Chief Aide Nakamura Satoru . Hara's diary records no contact with the Chief Aide nor does it mention any ac­ t i ons taken by the Chief Aide in connection with the dispute . It

1 60

OFFICES

AND

OFFICERS

OF

P A LA C E

L E A D E R SH I P

would appear that the C hief Aide was a loyal servant of the Meij i oligarchy u ntil 1 9 1 3 and a neutral figure until approximately 1 92 8 . Second , leadership over the military was not yet in the hands of generals and admirals w ho were entirely the products of the m i l i t a r y ' s r i g i d l y i ns t i t u t i o n a l i z e d c ha n ne l s of recrui t m e n t , education, and career advancement-channels that were so all-en­ compassing and exclusive that a general or admiral tended to equate his i nstitution's particular interests with the general national i nterest (Imperial Will ) . 102 The first C hief Aide to have graduated from the M ilitary Academy , for example , was Uchiyama Koj iro, Chief Aide from 1 9 1 3 to 1 92 2 . Thereafter every C hief Aide was a Military Academy graduate, and most were graduates of the Military Staff Col lege as wel l . Finally, the Army's involvement in C hi na did not reach the poi nt of no return until 1 928-3 1 , begi nning with the murder of Manchurian warlord Chang Tso-lin in June 1 92 8 and made irrevocable by the Manchurian I ncident i n September 1 93 1 . 103 O nly when the i nstitutionalization of military leadership combined with an irrevocable commitment in C hi na did the office of Chief Aide become a vehicle for Army activism in the palace . The tenure of General Honj o S higeru as C hief Aide from 1 93 3 to 1 93 6 is t he most extreme example of i nstitutional ized Army representation at the side of the emperor. 104 Honj o was one of five generals produced by the ni nth class of the Military Academy , which was graduated in 1 897 . His fellow generals were Araki Sadao, a leading spirit in the Army's Imperial Way Faction ; Mazaki Jinzaburo , also a leader of the Imperial Way Faction and deeply im­ plicated in the coup of February 26, 1 93 6 ; Matsui lwane, com­ mander i n C hina during the rape of Nanking in 1 937 and ha nged as a war criminal after the war; a nd Abe Nobuyuki . 105 Two of t he five 1 02 . On t he exclusiveness of Army education and its effects on Army leadership see Alvin D . Coox, " Year o f t h e Tiger," Orient/ West 9, n o . 4 Ouly-August 1 964) : 34-7 1 ; and Maruyama Masao, Thought a nd Behaviour, pp. 1 3- 1 5 . 1 03 . On t he Army's role in Manchuria see Sadako N . Ogata , Defiance i n Manchuria: The Making of japanese Foreign Policy, 1931-1932 (Berkeley and Los Angeles: University of California Press, 1 964) . 1 04 . Honjo's diary, cited earlier, is t o m y knowledge the only primary source t o have been published on the activities of any C hief Aide-de-Camp. N or have I d iscovered biographies or autobiographies of any of the C hief Aides. My discussion of t hat office and its personnel is t herefore at best fragmentary . 1 05 . Honj o Shigeru , p. i .

OFFICES

AND

OFFICERS

OF

PALACE

L E A D E R SH I P

1 61

were prominent Army politicians . Araki and Abe both served as Cabinet M i nisters and Abe became Prime M i nister in 1 93 9 for a brief period . Three of the five-Honj o , Mazaki, and Matsui-had seen extensive service in China. As noted above , General Honj o had been commander of the Kwantung Army w hen the M a nchurian In­ cident broke i n September 1 93 1 . Honj o's appointment as Chief Aide in 1 93 3 , t herefore , brought into t hat pal ace office mil itary prominence , i nstitutionalized career, and China experience . To this was added his association with the allegedly extremist Army group , the Imperial Way Faction . As Chief Aide, Honj o played four i nterrelated roles : ( 1 ) adviser to the emperor; (2) defender of the i nstitutional i ntegrity of the military, primarily t hat of the Army , at the emperor's side; (3) liaison between the emperor and the military ; and (4 ) consultant and confidant to the other t hree palace leaders . 106 The vagueness of " reporting and replyi ng to the throne on military matters" allowed the C hief Aide considerable discretion in the performance of these functions . The role of adviser to the emperor in Honj o's case in­ volved replying to rhetorical q uestions by the emperor as much as it did the renderi ng of advice . On April 1 8 , 1 93 3 , twelve days after he had been appointed C hief Aide, Honj o was summoned by the em­ peror and asked w hether a direct imperial command should not be issued to the Kwantung Army to halt its advance in the Jehol area of Manchuria. Honj o guessed t hat the emperor was pointing out by such a question that it would be a breach of faith for Japan to continue advancing while declaring to t he foreign powers that Japan would not advance into China's interior. Fearing that this kind of im­ perial command would lead to u ndesirable consequences , he advised the emperor to wait . Honj o then retired from the emperor's presence 1 06 . Unfort unately, the particulars of Honj o's appointment are not known . Nezu M asashi argues that his appointment was a personal reward by t he emperor for service i n Manchuria (Nezu Masashi, p. 1 1 6) . There is as little evidence to support this view as there is for the ("()ntrary proposition that by being i n the palace Honj o would be " neutralized" as a politically artive military leader. The biographers of Suzuki Kantarii argue that Honjo became Chief Aide at the recommendation of General Araki Sadao, who was Army M inister when Ho�ii was appoi nted ; subsequently, Honj o allegedly reported everything t hat went on in t he palace lo Araki (Suzuki Ka ntaro Den, p. 1 20) . Nezu supports t he " reward" t heory of appoi ntment , Suzuki 's biographers the "spy" i nterpretation. Both require modification, although the latter i s apparently more accurate. Honjo was forced to resign i n 1 936 because of complications •·a used by his son-in-law's participation in the February 26 coup.

1 62

OFFICES

AND

OFFICERS

OF

P A LA C E

L E A D E R SH I P

and went to see the Deputy Chief o f the Army General Staff, General Mazaki J i nzaburo. Honj o told Mazaki that what Mazaki and others had reported to t he throne conflicted with the actual situation in China. As a result , the General Staff issued an order to t he army in the field to pull back. 107 In t his i nstance the Chief Aide advised the emperor to wait , which the emperor did, and then saw to it t hat the conditions provoking t he emperor's rhetorical q uestion were relieved . O n January 2 6 , 1 934 the emperor told Honj o that accordi ng to the newspapers t he Army and Navy Mi nisters had stated to t he Diet t hat t here was nothi ng wrong in military men discussi ng and studying politics . Commenting, the emperor stated that even re­ search might lead to evil i nfluences if it went too far. That afternoon, i n response to i mperial summons , Honj o apprised the emperor of the meaning of the ministers' remarks . First , he argued , officers must know about politics at t he national and l ocal levels i n order for t hem to command reservists and veterans as well as their own troops effec­ tively . Without such knowledge, officers would be unable to possess t he deep sympathy for their men that is required for leadership : communication problems would develop between officers and men . Second , under present conditions, when the strength o f the nation must be i ncreased and concentrated on prepared ness for war, domestic politics must be shaped toward t his end . Those loyal to na­ tional defense s hould petition t hrough proper channels, such as the Army Mi nistry , for the remedy of political defects i n this regard . Senior officers must know about politics for morale purposes and for unity i n t heir views on national defense. Finally, although officers ought not to be diffident about politics , t hey must not participate i n politics directly; direct political action was prohibited b y military regulations . In response to Honj o's rationalization of the military's political attitudes , the emperor remarked that such an i nterpretation lay on the side of moderation and was acceptable. 108 Honjo's " ad­ vice" to the throne i n t his instance reveals an attempt by the C hief Aide to mollify t he emperor w hile defending the i ntegrity of the 1 07 . Honj o Shigeru, p . 1 59 . 1 08 . Ibid . , p p . 1 82-83 .

OFFICES

AND

O FFICERS

OF

PALACE

LEA D E R S H I P

1 63

military's political views. His remarks also show the degree of politicization that had taken place withi n the Army and his sym­ pathy with it. Honj o again defended the Army's political i nterests to t he em­ peror on February 8, but with less success. The emperor told Honj o t hat the military's i nterest in agrarian problems and sympat hy with rural distress should go only so far; t he farmers also had their pleasures and the aristocracy their pains. The emperor pointedly re­ marked that the Taisho Emperor's illness was provoked by t he cares a nd restraints of being emperor. At this veiled rebuke, Honj o was apparently contrite, " u nable to restrai n his emotion and awe . " 109 On occasion the C hief Aide and t he emperor collided directly. At one point the emperor told Honj o poi nt blank of the Army's obstruc­ tion between emperor and Cabinet duri ng t he Manchurian I ncident . H onj o felt compelled to remonstrate with the emperor and replied i n effect that n o such thing could ever happen . 1 10 M ore importa nt , however, was t he emperor's constant conflict with his C hief Aide over M inobe's theory of constitutional monarchy . As often as the emperor explai ned the theory and castigated the Army's opposition to it, H onj o j ustified t he Army's stance : the Army believed the em­ peror to be god manifest ; to make t he emperor a person would be troublesome for "troop education and supreme command . " 1 1 1 When t he emperor told the ranki ng naval Aide-de-C a m p, ldemitsu Mambei , that t he military was contradicti ng imperial wishes re­ garding M i nobe's theory , ldemitsu replied that the emperor should " tra nscend " such debates . 1 12 Throughout t he controversy over M inobe's t heory , which raged in t he Diet and elsewhere from March i nto October 1 93 5 , "advice" to the emperor from the C hief Aide as well as t he ranki ng naval Aide amounted to staunch defense of t he military's position and outright contradiction of the emperor's views. 1 13 Honj o's actions during the M i nobe affair illustrate how much the office of Chief Aide-de-Camp had been transformed i nto a 1 09 . I 1 0. 111. 1 12. 1 1 3.

Ibid . , Ibid . , I bid . , Ibid . , Ibid . ,

p. 1 85 . p . 207 . p. 204 . p. 2 1 1 . pp. 203-3 1 .

1 64

OFFICES

AND

OFFICERS

OF

P A LA C E

LEA D E R SH I P

channel for institutional representation a t the emperor's side and a vehicle for countering opinions at the palace contrary to those of the military , including those of the emperor hi mself. Honj o also used his office as military adviser to the throne to protect as best he could the interests and i ntegrity of the Army in the aftermath of the February 26 I ncident of 1 93 6 , a coup attempt by young officers w hich brought i nt o q uest i o n t he capacity or willi ngness of the senior Army officers to control their younger colleagues . On March 6, 1 93 6 , shortly after t he coup had been brought under control , the emperor asked Honj o whether he should not dissolve the regiments that had spaw ned the " i nsurgent officers and men . " Honj o replied that the matter was now under investi­ gation by those i n positions of responsibility and poi nted out to the emperor that to dissolve the "dishonorable" regiments would be a great rep r i m a nd t o t he e nt i re m i l i t a r y e s t a b l i s h m e n t . N ot surprisingly, "those in responsible positions" subsequently decided not to abolish the concerned regiments . Because the regimental colors had been bestowed by the emperor, however, the Office of Aides-de-Camp thought it improper merely to present such a de­ cision to him without first aski ng his view . Consequently, the Army M inistry , the I nspectorate of Mil itary Education, and the Army General Staff reconsidered and decided to ascertain the emperor's views on the issue of dissolution. The burden of negotiating a solution satisfactory to both Army and emperor now lay sq uarely on C hief Aide Honj o . On March 1 4 the Section Chief for Mil itary Affairs went back and forth between the Aides' Office and the Army M i nistry ; he emphasized that if by any chance the emperor should press for dissol ution of the tai nted regiments the Army authorities would be in serious straits. Havi ng reported this to the emperor and heard his views, Honj o asked the Director of the Mil itary Affairs B ureau, General l mai Kiyoshi , to come to Honj o's office : it was the emperor's wish to permit the retention of the tai nted regiments . On March 1 7 the Army M i nister and Army C hief of Staff met with the emperor at the palace . They reported , over their j oi nt signatures , that : ( 1 ) t he coup was an un­ precedented disgrace and a blot on the bright record of the I mperial Army , in particular of the regi ments concerned ; (2) although

OFFICES

AND

OFFICERS

OF

PALACE

LEA D E R SH I P

1 65

dissolving the four regiments i nvolved was deemed unavoidable i n view of regimental and officers' rules o f conduct , those who planned a nd participated in the i nsurgency were but a small fraction of these regi ments ; a nd ( 3 ) t he tai nted regiments w i s hed to redeem themselves . The emperor replied : "Then all is wel l . Guard agi nst t he future . " After severely reprimanding their regiments, two of the four commanders came to visit the C hief Aide on March 2 0 , expressi ng their gratitude and asking that their appreciation be reported to the throne . 1 14 W hile the dissolution issue was pending, Honj o was also set to work on what the emperor was to say formally to the Army M i nister w hen he proffered his apologies for the i nsurrection. The i nteraction among palace officials i n arrivi ng at the public I mperial Will in politics as well as the C hief Aide's efforts on behalf of his i nstitu­ tional constituency are both illustrated by the reprimand process. I n late February the Office of Aides began t o i nvestigate the substance of a suitable reprimand , the procedures i nvol ved , and the locus of responsibility for the i ncident . Honj o asked the Privy Council President and former I mperial Household M i nister, Ichiki Kitokuro, about procedures for the reprimand . Ichiki replied that on relatively routine matters it had been customary si nce the time of Meij i to relay the emperor's words precisely as he had stated them ; on mat­ ters of grave import , however, the emperor's statement was to be " advised and assisted " on by the Privy Seal . Honj o agreed that consultation with the Privy Seal was req uired and the Office of Aides proceeded accordi ngly . 1 15 The Army M i nister apologized i nformally on February 2 9 and formally on March 3. B ut because a new Privy Seal had not yet been appoi nted to replace Privy Seal Saito Makoto, one of the key senior leaders assassi nated in the February 26 coup , the emperor did not reply to the Army Mi nister. Honj o urged the necessity of a prompt imperial response . W hen he was summoned by the emperor on March 4, Honj o was told that if the reprimand were too strong the Privy Seal would once again be menaced by radical rightists . Yet some discipli nary action had to be taken, the emperor implied , be1 1 4 . Ibid . , 1 1 5 . Ibid . ,

pp. pp.

290-9 1 . 2 9 1 -92 .

1 66

OFFICES

AND

OFFICERS

OF

P A LA C E

L EA D E R SH I P

cause for h i m personally the murder o f his most trusted senior statesmen and military officers caused the greatest anguish . Moreover, such behavior went against the constitution, violated the Emperor Meij i ' s I nstructions (to soldiers and sailors) , and stai ned the national polity . On March 5 Honj o was again summoned by the emperor, who was concerned whether his words, given through the ordi nary channels, would have any effect at all on officers who seemed so willing to defy the constitution . The Office of Aides fi nally prepared a suitable reprimand and on March 8, after the New Privy Seal, Yuasa Kurahei, had agreed to and signed it, Honj o obtai ned the emperor's sanction. The repri­ mand was severe : the Army had frequently and repeatedly staged ominous i ncidents, finally giving rise to the recent disaster which ut­ terly contradicted the I nstructions to Soldiers and Sailors , tarnished the history of Japan, and caused i nsufferable anxiety. The military was ordered to clean house thoroughly and to guard against such a disgrace occurring again . 1 1 6 B ut Honj o's labors were not at an end. One hour after the em­ peror had apparently sanctioned the text he summoned Honj o once agai n : the text did not i nclude t he phrase " stained the national polity . " W hat was the Army's i nterpretation of t he national polity , given its i nsistence on "clarification of the nati onal polity " ? Honj o replied that Japan's national polity " lay i n an emperor of one l i ne unbroken for ages eternal who ruled the state i n obedience to the be­ queathed i nstructions of the imperial ancestors and the imperial founder . " B ut , Honj o conti nued , t he fearsome i ncident of late " can­ not . . . in any way be said to have stained the national polity . " 1 1 7 Despite the emperor's poi nted questions, the phrase "stai ned the na­ tional polity " was left out of the official reprimand . Honj o had once again succeeded in compromisi ng t he severity of action contemplated agai nst his i nstitutional constituency . O n March 1 0 t he new Army M i nister, G e neral Terauchi Hisaichi , was summoned to the palace specifically to receive the formal imperial reprimand . Terauchi was ordered by the emperor to make the i ntent of the reprimand thoroughly know n to subordi nate 1 1 6 . Ibid . , p . 292 . 1 1 7 . lbid . , pp. 2 92-93 .

OFFICES

AND

OFFICERS

OF

PALACE

L E A D E R SH I P

1 67

personnel i n the Army . To avoid future misunderstanding, Honj o gave the Army M inister a copy of the emperor's words. At four o'clock that afternoon the emperor summoned Honj o and asked him how the Army M i nister planned to transmit t he reprimand to the Army . Honj o relayed the emperor's q uestion to the Army M inister, w ho took it u nder consideration immediately _ u s The responsible au­ thorities of the Army , with the participation of Aide-de-Camp Sakai , met on March 1 9 and decided that ( 1 ) for the emperor to issue such an imperial rescript of reprimand to the entire Army was without precedent ; (2) a rescript with such words as " tarnished the history of our country , " w hen the conduct in question was that of a small segment of the Army , would leave a stain on the Army that would be difficult to remove ; (3) a rescript of reprimand should only be given after all other efforts had failed , and the Army's leaders had not yet taken all alternative measures ; therefore, (4) there should be no im­ perial rescript at this time. Havi ng received this decision from his responsible subordi nates , the Army M i nister asked Honj o the following day about distributing the emperor's reprimand to Army personnel . The Office of Aides consulted the I mperial H ousehold M i nistry secretaries and the Privy Seal 's Office on t he nature of imperial rescripts and declarations; the Army Minister was then advised that the intent of t he imperial reprimand should be written into the Army M i nister's instructions to t he Army but that it would be advisable to avoid distributing copies of the reprimand to the Army . Copies could , however, be given to high-ranki ng officers , includi ng division commanders . 1 19 The emperor's reprimand to the Army M inister i n chambers was therefore not transformed into a public imperial declaration to the entire Army . Throughout his tenure as Chief Aide Honj o reconciled the con­ flicti ng demands of Army and emperor, includ i ng those of the em­ peror's "advisers , " as much in favor of t he Army as possible. W hat was "possible" depended on ( 1 ) the emperor's stance a nd to whom that stance was known, (2) the i ntransigence of the Army , (3) the precedents and procedures available to rationalize given courses of Ibid . , 1 1 9 . I bid . ,

1 1 8.

p.

293. 293-94 .

pp.

1 68

OFFICES

AND

OFFICERS

OF

P A LA C E

LEA D E R SH I P

action, and (4) the views of other palace leaders w ho shared in the process of conflict reconciliation. The emperor's stance in politics was conditioned by his civilian advisers , especial ly palace leaders and others , such as Prince Saionj i , who held to M i nobe's theory of constitutional monarchy or sympathized with i t . The sharpness of the conflict between Chief Aide Honj o and the emperor over the M inobe theory i ndicates how partisan t he emperor's stance was . Honj o was therefore restrai ned by t he emperor to the extent that the emperor's actions a nd views had to be taken i nto account by the C hief Aide i n dischargi ng his function of advice a nd assistance . The i ntransigence of Honj o's i nstitutional constituency was ap­ parently t he most important factor i n Honj o's actions; he felt constrai ned to " remonstrate" with the emperor w hen basic Army i nterests were threatened . W hile " remonstration" for Household M inister Tanaka M itsuaki in t he early twentieth century was largely based on his personal role conception and his person-to­ person relations with the oligarchs and their proteges , Chief Aide Honj o's remonstration in the 1 930s was based primarily on institu­ tional considerations . It is important to repeat t hat the C hief Aide was an Army officer on active duty ; both Ho�o and his predecessor, Nara Takej i , for example, went i nto the Reserves after they resigned as C hief Aides . The fact t hat the C hief Aides were active-duty of­ ficers appoi nted on Army initiative rei nforced t heir loyalty to the Army . Also quite obvious in Honj o's behavior was a great concern for procedure and precedent . Honj o was meticulous i n consulting the Privy Seal , the President of the Privy Council , a nd household of­ ficials w henever formal action was to be taken by the emperor, as i n­ dicated by his actions i n the aftermath of the February 2 6 I ncident . Precedent was useful i n mitigating the emperor's actions against the Army i n t hat critical period ; a formal reprimand to the Army, for example, would have been "without precedent . " B ut procedures and precedents worked two ways. Precedent had to be sought for actions on the part of the C hief Aide, as well as the emperor, and the observance of procedures quite obviously allowed palace leaders some i nfluence over the behavior of the C hief Aide . Once he became one of the four leading palace officers , Honj o was surrounded by

OFFICES

AND

OFFICERS

OF

PALACE

LEA D E R SH I P

1 69

a different set of procedures and precedents from those in the Army , and a different set of colleagues . Among his palace colleagues , the one with the greatest influence on Honj o appears to have been Grand Chamberlai n Suzuki Kan­ taro. I t was Suzuki who initially i nformed Honj o of the norms of conduct regardi ng the throne for both civilian and military leaders . This he did by discussi ng improprieties i n the behavior of Prime M inister Tanaka Giichi in 1 92 9 and , on instructions from the em­ peror, the contradictory actions of the Chief of the Navy General Staff duri ng the London Naval Treaty controversy of 1 930. 120 Al­ though Chief Aide Honj o and those i n the Office of Aides were the sole formal liaisons between the palace a nd the military commands, H onj o was careful to keep himself i nformed of the Privy Seal 's views . On one occasion, for example, he ascertai ned the Privy Seal 's opinion, via Grand Chamberlain Suzuki , on the revision of reporting procedures for the Navy General Staff. 121 The Household M i nister was also i nvolved on occasion. When i n J anuary 1 934 a request came from Pri nce-of-the-Blood Asaka for an imperial review of the Palace Guards (Konoe Shidan) , Honj o busied himself with precedents for such a review and obtai ned the " under­ standi ng" of the H ousehold M inistry as well as t he B oard of C ham­ berlains . 122 The H ousehold M i nister and Chief Aide also worked together when t he Army obj ected to the emperor's reviewi ng J apanese troops with the emperor of M a nchuria during the latter's visit to Japan. The Household M i nister i nsisted that courtesies to a visiting head of state fel l under his j urisdiction, not t he Army's. Through Honj o , however, a compromise was reached w hereby spe­ cial courtesies were to be rendered by the Army to the Japanese em­ peror. The issue took two weeks to settle. 123 Honj o also kept palace leaders i nformed , at least sporadically, of Army criticisms of those " at the emperor's side" or "close to the t hrone . " Around September 1 93 5 , for example, H o nj o told 1 20. Ibid . , pp. 1 60-62 . Suzuki t hus conti nued his partisan activities o n behalf o f the consti­ tutional monarchists. I n the instances cited above he was used i n an attempt to bri ng the Chief Aide i nto l i ne with palace policy duri ng 1 93 3 . 1 2 1 . Ibid . , p p . 1 63-67 . 1 2 2 . Ibid . , pp. 1 79-80. 1 23 . Ibid . , pp. 201 -03 .

1 70

OFFICES

AND

OFFICERS

OF

P A LA C E

L EA D E R SH I P

Household M i nister Yuasa Kurahei that a segment of the Army was angry over the " weak" advice given the emperor by the Senior Retainers, i ncluding the emperor's palace advisers . Yuasa told Honj o that he, as Household M i nister, was never asked political questions by the emperor, although some felt that the Household Minister should speak his mind on political issues . Grand C ham­ berlain Suzuki added that the emperor never asks the Privy Seal questions about military affairs . Honj o believed , however, t hat the Privy Seal might properly be consulted on the d isti nctions between civil and military matters . 124 Honj o did, it appears , act as a colleague among fellow palace leaders in coordi nating the emperor's role as E mperor-i n-State . It is much too simple to classify even as extreme an example as C hief Aide Honj o as a military " spy" at the emperor's side . Although he tenaciously defended the policies a nd interests of t he military, espe­ cially the Army , Honj o was meticulous in observing palace customs , procedures , and precedents. He kept the emperor and palace leaders i nformed of opi nion within the Army . If he behaved as the representative of his i nstitutional constituency at t he emperor's side , he also demonstrated an awareness of his role as a palace person, as one of those i nvolved in the coordination of t he I mperial Will i n politics . H i s actions reveal the complexity of pressures o n him as military adviser a nd liaison to the throne, pressures that derived from his dual capacity as institutional representative and palace ne­ gotiator. As such, he was both the voice of military demands at the side of the emperor and t he instrument of palace communication and negotiation with the military . Havi ng an active-duty status . within t he Army, having served a lifetime in that constituency , and having i n fact been given his palace office by that constituency , the C hief Aide was, however, a v ery dif­ ferent palace officer from his palace colleagues . He was espec· lly set on apart from his civilian counterpart who advised the emper " politics" : the Lord Keeper of the Privy Seal .

5

1 24 . Ibid . , pp. 224-2 5 .

OFFICES

AND

OFFICERS

OF

PALACE

L E A D E R SH I P

171

THE LORD KEEPER OF THE PRIVY SEA L : C H I EF POLITICAL ADVISER IN ATTENDANCE Because Count Makino also felt deep concern over his i ncapacity i n of­ fice, [ his i nfirmities] bei ng such that it was completely impossible for him to take part even i n court events at the year's end and the year's begi n­ ning, Pri nce Saionj i , Prime M i nister Okada, H ousehold M inister Yuasa , Grand C hamberlain Suzuki , and others came to an agreement among themselves and fi nally brought about [ M akino's resignation as Privy Seal ] . Former Prime M i nister Viscount Saito was made his successor by direct imperial appointment because his character and career were considered most fit for dischargi ng the heavy responsibilities of rendering advice and assistance in regular attendance on the Emperor, given that the viscount was already one of t he Senior Retainers . Although some gave heed to his connections with t he present Cabi net , si nce the viscount had been involved as midwife of the Okada Cabinet , in fact there was absolutely no political significa nce in this. To kyo Asahi Shim bun125

However accurate as a picture of political reality, these comments on the appointment of the new Privy Seal in 1 935 reflect idealized perceptions about the right man to be Privy Seal and the proper nature of his role. A Senior Retai ner (ex-Prime Minister) , Viscount Saito was most " fit" to render advice and assistance to the throne , which he would have done i ndirectly in any case as one of those consulted by Pri nce Saionj i w hen a Cabinet collapsed during the 1 930s. There was " no political significance" -undue political par­ tisanship or unhealthy political connections-in the fact that he had been " midwife" to the Okada Cabi net in J uly 1 934, despite Vis­ count Saito's pro-Saionj i stance in politics and his long association with Okada in the Navy . As we shall see , the Privy Seal had always acted as liaison a nd negotiator between the throne and the political world outside the palace gates . That Saito had midwifed the Okada Cabi net prior to his appointment as Privy Seal was evidence that he 1 2 5 . December 27, 1 93 5 , as qu oted in Shishaku Saito Makoto Den, 3 vols. (Tokyo: Zaidan l l oj i n Saito S hishaku Kinen Kai , 1 94 1 ) 3 : 646. ,

1 72

OFFICES

AND

OFFICERS

OF

P A LA C E

L E A D E R SH I P

was i n fact able t o perform one of the Privy Seal 's most important functions : political broker to the throne on Cabi net formations . As indicated by the appoi ntment of Senior Retai ner Saito i n 1 93 5 , t he Privy Seals genera l l y were expected t o be accomplished statesmen , and i n fact they were the most prestigious of the palace leaders in career and status. Of the thirteen Privy Seals between 1 885 and 1 945 , four (30 . 8 percent) had been Prime M i nisters ; nine (69.2 percent) had served as M i nisters of State; seven ( 5 3 . 8 percent ) , i ncludi ng t w o presidents , had been Privy Councillors ; a nd a l l but one (92 . 3 percent) were at one time members of the House of Peers . Two ( 1 5 .4 percent) were Elder Statesmen . The only imperial pri nce among all four lead ing palace officers from 1 885 to 1 94 5 , Fushimi Sadanaru , served as Privy Seal . All the Privy Seals were titled ; among them they i ncluded one pri nce-of-t he-blood (7 .7 percent ) , five pri nces (38 . 5 percent) , one marq uis (7 .7 percent) , two counts ( 1 5 . 4 perce nt ) , two viscounts ( 1 5 . 4 percent ) , and two barons ( 1 5 . 4 percent) . Averaging sixty-five years of age upon assuming office , ranging from forty-eight to eighty-two, the Privy Seals were also on the average the most senior members of the palace leadership. The average term of office was 4.6 years , although two Privy Seals served but one day in office , thus distorting the average dow nward . The Privy Seal , like all palace leaders , was expected to " serve a lifetime at the side of the emperor" once appoi nted to leading court office . And most of them did ; eight of the thirteen died within two years of retirement . Althouth many Privy Seals did not serve twenty years in any one given i nstitutional constituency prior to appoi ntment , their maj or career backgrounds were : restorati onist , three (23 . 1 percent); H ome M inistry , three (23 . 1 percent) ; Army , two ( 1 5 .4 percent) ;{and court , Education M i nistry , Foreign M i nistry , Navy , and Agric�ure and Commerce Mi nistry , one each ( 7 . 7 percent) . Calculati ng career backgro\lnd agai nst length of service, however, it is apparent that court , restoration, and Foreign Mi nistry officials held the post of Privy Seal 7 5 percent of the time between 1 885 and 1 945 : 126 1 26 . Four of the thirteen Privy Seals served less than six mont hs : Hamao Arata ( Education M i nistry), Katsura Taro (Army), Saito Makoto (Navy), and lchiki Kitokuro (Home M i nistry).

OFFICES

AND

OFFICERS

OF

P A LA C E

Court Restorationist Foreign M i nistry H ome Ministry Agriculture and Commerce M i nistry Army Total

1 73

LEADERSH I P

20 14 II 7

years years years years

6 years 2 years 60 years

That only three military officers should have received appointments , excludi ng Elder Statesman General O yama lwao, and that they served a total of but two years i ndicates that the office of Privy Seal was very much a civilian post . The military officers who did hold the post, moreover, were senior officers who had either served as Prime M inister (two) or were members of the imperial house (one) . The only time a court person served as Privy Seal was from 1 89 1 t o 1 9 1 2 , when Pri nce Tokudaij i Sanenori served concurrently as Grand Chamberlain and Privy Seal . Like the other offices of court leadership, therefore , the office of Privy Seal also underwent bureau­ cratization from the outside once the oligarchic solidarity of the Meij i period ended . The official duties of the Privy Sea l , the initial conception of his role i n 1 885-as revealed by the appoi ntment and actions of the first Privy Seal, Pri nce Sanj o Sanetomi-and the ac­ tions of subsequent Privy Seals reveal the flexibility of that most im­ portant of court offices as it responded to changing times throughout the period 1 885- 1 94 5 . A s was the case with the Grand C hamberlai n, the duties o f the Lord Keeper of the Privy Seal were briefly defi ned and the j urisdic­ tion of his office diffuse . The office of Privy Seal was created withi n the palace in December 1 88 5 . The Privy Seal was to ( 1 ) take custody of the I mperial Seal and the Great Seal of State, (2) render advice and assistance in attendance on the emperor, and (3) preside over the proceedings of the Court Advisers (Kyuchu Kom onkan) . 1 21 The ef.

1 2 7 . " Daj okan Tasshi , " no. 68, December 2 2 , 1 88 5 , in Hii rei Zensho, 1 885, 2 : 1 ,043 . Palace advisers, origi nally fifteen i n number, were appoi nted for meritorius service and were to advise the throne on ceremonies stipulated in household regulations (ibid . ) . B y 1 907 the fifteen Court Advisers·, as well as the I mperial H ousehold Mi nistry Consultants (Kunai Shii Goyogakari) , fell under t he Imperial Household M i nister's office . In 1 94 3 , however, there were three special "consultants" (goyiigakari) to the Privy Seal : one a Privy Councillor, one a n I mperial Household M i nistry Councillor (Sanjikan) and the third a Bureau of Legislation ( :ouncillor ( Prime Minister's Office) (Shokuin Roku, July 1 943, pt . 2, pp. 1 4- 1 5 ) .

1 74

OFFICES

AND

OFFICERS

OF

P A LA C E

LE A D E R SH I P

fective regulation governi ng the Privy Seal 's office from 1 907 until 1 942 added that the Privy Seal was to take charge of imperial edicts , imperial rescripts , and matters regarding palace documents, while deleting the provision on presidi ng over the meetings of Court Advisers . In addition to two secretaries of executive appointee rank , the Privy Seal was assigned a C hief Secretary of imperial appoi ntee rank to assist him in his duties . 128 Of the four leading palace officers only the Household M i nister and the Privy Seal were empowered to "advise and assist " (hohitsu) the emperor, as were the Mi nisters of State ; the C hief Aide , however, was permitted to " report and reply to" the throne, which allowed him similar i nfl uence, and the gene­ rality of " renderi ng attendance at the emperor's side" allowed the Grand Chamberlain influence as wel l . B ut except for precedents and the "trends of the time , " the Privy Seal , unlike the Household Minister, was under no restriction regarding advice and assistance . If the Privy Seal 's was a disti nctive office for assisting the emperor by renderi ng advice on both national affairs and imperial household matters , the manner in which the function of advice and assistance was fulfilled varied considerably throughout the period from 1 88 5 to 1 94 5 . We have seen, for example , that from 1 89 1 to 1 9 1 2 Privy Seal-Grand C hamberl a i n Tokudaij i was a frictionless conduit between emperor and ol igarchs . The first Privy Seal , Pri nce Sanj o Sanetomi ( 1 8 3 7 -9 1 ) , was , however, a n act i ve member o f the oligarchy . His role as Japan's first Privy Seal has been the subj ect of considerable speculation and controversy. One opinion is that the of­ fice of Privy Seal was created especially for Sanj o as a place for him to retire in dignity . According to this i nterpretation, Ito Hirobumi wished to fi nd a place for Sanj o commensurate with the post of Chancellor, a post from which Sanj o would be forced to resign when the cabi net system was i naugurated in 1 88 5 . 129 Sanj o has also been characterized as a weak member of the oligarchy . 130 Even w hile Sanj o was Privy Seal , moreover, he was attacked for faw ning ser1 28 . "Kiishitsu Rei ," no. 4 , November I , 1 907, as amended in 1 9 1 0, 1 9 1 4 , and 1 9 1 7 , in Naikaku Kiroku Ka , 1 94 2 , vol . I, pt. 3, chap. I , sec. 3 , p . 20. 1 29. l nada Masatsugu, p . 35; I noue Kiyoshi, p. 1 03 . 1 30. Tsuji Kiyoaki, Nihon Kanryo Sei no Kenkyu (Tokyo: Kiibundii, 1 952), p . 70.

OFFICES

AND

OFFICERS

OF

PALACE

LEADERSHI P

1 75

vilely on the Ito Cabinet . 131 Thus the office of Privy Seal may have been created as a device for kicking the ineffectual Sanj o upstairs . I noue Kiyoshi disagrees with such a n ad hoc interpretation of oligarchic i ntent behind the creation of the Privy Seal 's office . He argues that the office of Privy Seal , which he believes should have been a government rather than a court office, was deliberately placed under the emperor and outside the government to ensure that the emperor could be used to rescue the government oligarchs in time of need , as in confrontations with the I mperial Diet . As such, the office of Privy Seal was intended to be the ultimate stronghold of the "em­ peror system's bureaucratic-military despotism . " 132 It is not necessary to subscribe to I noue's theory of despotism to agree with him that the office of Privy Seal was intended to be pi­ votal in t he politics of oligarchic decision maki ng. Sanj o ' s experiences a s a Restoration leader made him a valuable member of the oligarchy and ideally suited to serve oligarchic unity as Privy Seal . Prior to the creation of the cabi net system and the office of Privy Seal in 1 885 , Sanj o had been instrumental in coordinating personnel a nd policy . Throughout the process leadi ng to t he i nauguration of t he cabi net system, for example, he had worked closely with Ito in reachi ng and re-reaching compromises . At no point , according to Ito's biographer, was Sanj o anythi ng but fair and impartial i n his attitudes toward this maj or revision of government structure, a revision that would deprive him of his office as Chancellor. If oligarch Ito was the leader and initiator of policy , oligarch Sanj o was the key policy negotiator. As harmonizer of ol igarchic wills, moreover, Sanj o did not hesitate to offer suggestions to Ito that did not always conform with I to's views . 133 A high Court Noble outside the Satsuma-Choshu group of leaders , Sanj o was also the chief link between Ito and the emperor. As an oligarch, Sanj o was a member of the government elite; a s a Court Noble he had ac­ cess to the ernperor and was familiar with the ways and customs of court . Sanj o was t herefore ideally suited to render " advice and 1 3 1 . Shumpo Kotsui Shoka i , 2 : 5 1 9 . 1 32 . I noue Kiyoshi, pp. 1 02 -03 . 1 33 . Shumpo Kiitsui Shokai, 2 : 44 1 -84 .

1 76

OFFICES

AND

OFFICERS

OF

P A LA C E

L E A D E R SH I P

assistance" a t the emperor's side a s negotiator, not initiator, withi n the oligarchy . After he became Privy Sea l , Sanj o conti nued to be an invaluable member of the oligarchy as it weathered the drafting of the Meij i Constitution between 1 88 5 and 1 889 and the opening of the first I m­ perial Diet in 1 890. When in 1 88 9 , for example, no agreement could be reached among the oligarchs on a successor to Kuroda Kiyotaka as Prime M i nister, Sanj o became Pri me M inister pro tempore . For two months , from October 2 5 to December 24, 1 88 9 , Sanj o served as both Prime M inister and Privy Seal while engaged in negotiations that ultimately led to t he formation of a Cabi net under Yamagata Aritomo. 134 Rather than "weakness " Sanj o demonstrated great skills as a negotiator, making his office of Privy Seal that of " Prime M inister in reserve. " Sanj o also stood staunchly behi nd the Prime M i nister, believing it to be his duty as Privy Seal to do so until the Pri me Mi nister himself had decided to resign. In the spri ng of 1 887 Prime Mi nister Ito Hirobumi was attacked for blurri ng the disti nction between court and government as well as for failing to revise the unequal treaties between Japan and the Western powers . Despite the possible tarnish on his career that his support of Ito might have caused , Sanj o refused to give in to " public opinion . " Sanj o stood aloof, insisting that any q uestion of resignation had to be delayed until Kuroda Kiyotaka had returned from abroad and all the oligarchs could assemble to reach agreement on a course of action. 135 Given Sanj o's career both before and during his tenure of office as Privy Seal , it might even be argued that he was one of the crucial figures in negotiating the Restoration settlement to a successful conclusion. It is tempting to speculate that the i nnocuous Tokudaij i was able to serve as both Privy Seal and Grand Chamberlai n from 1 89 1 to 1 9 1 2 because Sanj o, as Privy Seal , had been so successful i n consolidati ng oligarchic unity around the Meij i Emperor : a skilled negotiator was no longer required as Privy Seal after Sanj o's death in 1 89 1 . There were but two Privy Seals from December 1 885 to July 1 34 . Ibid . , p p . 698-705 , 1 ,007 . 1 35 . Ibid . , pp. 5 1 9-20.

OFFICES

AND

OFFICERS

OF

PALACE

LEA D E R SH I P

1 77

1 9 1 2 , the terms of Sanj o and Tokudaij i averaging 1 3 . 3 years . As in the case of the Household M i nister and the Grand Chamberlain, however, the end of the Meij i era was also the end of personnel stability in the Privy Seal's office . From August 1 9 1 2 , when ex­ Prime Mi nister Katsura became both Privy Seal and Grand Cham­ berlai n , to March 1 92 5 , when Makino Nobuaki was appoi nted Privy Seal , there were six Privy Seals, or one every 2 . 1 years . An ex­ Prime Mi nister, a prince-of-the-blood , two Elder Statesmen, and two ex-Ministers of State , in that order, held the post of Privy Seal , i l l ustrat i ng the transit i o n from restorat ionist to bureaucrat ic leadership at the palace . The transition was one of trial a nd error, posing considerable dif­ ficulties both i n government and i n court . The difficulties involved in Katsura Taro's entrance i nto and exit from the office of Privy Seal in 1 9 1 2 have already been described . Katsura's successor, Pri nce-of­ the-Blood Fushimi Sadanaru , caused problems of the reverse order . Rather than an active faction politician like Katsura , Fushimi was an imperial prince . His appoi ntment was i nterpreted as renderi ng the office of Privy Seal impotent precisely because of his status . W hen t he issue of Household M i nister Watanabe C hiaki ' s resignation arose in early 1 9 1 4 , for example, the Privy Council had to affirm Privy Seal Fushimi 's powers , as an imperial pri nce, to countersign the resignation, givi ng rise to fears at the time that the office of Privy Seal might "stagnate . " Consequently, the Elder Statesmen, affirming the need for a " full-time" Privy Seal , decided that one of their number, General O yama Iwao, should replace Fus­ himi . 136 I mplicit in their action was the proposition that an imperial prince had to be " above politics , " yet the Privy Seal's office i nvolved political action-includi ng even such details as the countersigni ng of ministerial appointments and resignations. And these acts, however formal , were potentially partisan. If an imperial pri nce were Privy Seal , his acts would ipso facto affect the impartiality of the imperial family-and the emperor. A Privy Seal had to act discreetly. By virtue of his station, Prince Fushimi could not act discreetly, a nd it is 1 36 . Gensui Koshaku Oyama lwao (Tokyo: O yama Gensui Den Kanko Kai , 1 93 5 ) , pp. H23-24 .

1 78

OFFICES

AND

OFFICERS

OF

P A LA C E

L E A D E R SH I P

significant that h e was the only imperial pri nce t o serve m any leading court office from 1 885 to 1 945 . 137 From 1 9 1 5 to 1 922 the office of Privy Seal was occupied by two aged Elder Statesmen, O yama lwao a nd Matsukata Masayoshi . Their appoi ntments suggest that the office of Privy Seal , i n the tradition of Sanj o Sanetomi , belonged to experienced and infl uential statesmen who had no further ambitions in government and were thus ideally suited to act as impartial but effective negotiators i n the political process . Si nce the Elder Statesmen advised the throne on maj or national policies as well as on the appoi ntment of the Prime M i nister, moreover, having one of their number as Privy Seal made this process of advice simpler and solidified the concept that such basic advice to t he t hrone was t he prerogative of t he Elder Statesmen . Count Hirata Tosuke was appoi nted to succeed Pri nce Matsukata Masayoshi in 1 92 2 . Hirata was a protege of Yamagata Aritomo. Next to Yamagat a , appare nt l y , H i rata was best know n t o Matsukata, and it was a t Matsukata's recommendation that Hirata , at the age of seventy-three , was made Privy Seal . 138 Hirata was not a complete stranger to the ways of court . I n 1 9 1 9 he had been appointed I mperial Household M i nistry Consultant to assist the I mperial House Economic Council in the adj ustment of im­ perial house properties . That council had traditionally i ncluded many of Japan's leading statesmen ; Ito Hirobumi, Yamagata Ari­ tomo, and Matsukata M asayoshi , for example , had at one time or another advised on court fi nances as members of the counci l . W hen the Elder Statesmen became few i n number and aged , younger scholars and officials were appoi nted Consultants to assist in the council's work. As Consultant , Hirata i nvestigated the fi nances of foreign royal houses , heard the views of men in the world of finance , and consulted with court officials . 139 1 37 . We have seen t hat one of t he objections to Katsura was his " rustic" origi ns, and this may have been important i n t he selection of Prince Fushimi to replace Katsura . After Fushimi, however, no person of even Court N oble origi n was to be Privy Sea l . If "status" was one of the criteria i n selecti ng t he Privy Sea l , it q uickly gave way to "ability, " as proven by t he successful bureaucratic careers of all those who served as Privy Seals from 1 9 1 5 on . 1 38 . Hakushaku Hirata Tosuke Den (Tokyo: Hirata-haku Denki Hensan Jimusho, 1 927 ) , p. 1 67 . 1 39. Ibid . , pp. 1 64-6 5 .

OFFICES

AND

OFFICERS

OF

P A LA C E

L E A D E R SH I P

1 79

The various posts for advising the court on one matter or another were devices for recruiti ng government leaders into the court , on the one hand , a nd for keeping leaders available to reenter t he government after they had resigned or been forced to resign from their government posts, on the other. Under pressure over the treaty revision issue, for example, I noue Kaoru had resigned as Foreign Minister i n 1 887 only to be appoi nted Court Adviser; he emerged from the palace less than a year later as M inister of Agriculture and Commerce . 140 Leading court officials also retired i nto palace advisory posts. Tanaka M itsuaki , for example, became an I mperial House Economic Adviser after he retired as Household Minister in 1 909, as did ex-Household M i nister lchiki Kitokuro when he was forced to resign as President of the Privy Council in 1 93 6 . The palace, with its multiplicity of advisory posts, thus served as a sanctuary for Japan's leading statesmen and bureaucrats . For Hirata, however, advisi ng at court was evidently a device for acquainting the protege of an Elder Statesman with the ways of court in preparation for high court office . The fundamental role of the Privy Seal as political negotiator was continued by Hirata . Shortly after he had become Privy Seal in Sep­ tember 1 922 , dissension between the government and the Privy Council arose over the Si no-Japanese postal treaty. The issue was basically a procedural one, involving t he timing of the Privy Council's advice on treaty ratification. The task of mediation fell to Hirata as Privy Seal . The I mperial Diet took up the issue i n December 1 92 2 , causi ng further complications b y subjecti ng the treaty to public criticism, and the issue required five months to settle. Throughout those five months Hirata was guided i n his task of me­ diation by two negative pri nciples : ( 1 ) not to leave future complica­ t i ons after the treaty had been ratified , and (2) not to "wound the face" (i� ure the pride) of either the government or the Privy Council . 141 I f in Sanj o's case the role of Privy Seal was to negotiate among the powerful men of the oligarchy , Hirata's role was to mediate among institutions , in this i nstance the Cabinet and the 1 40. S humpo Kiitsui Shiikai, 2 : 5 5 3 . 1 4 1 . Hakushaku Hirata Tosuke Den, p p . 1 68-69.

1 80

OFFICES

AND

OFFICERS

OF

P A LA C E

L E A D E R SH I P

Privy Council . Hirata was less concerned with personal reputations than with the prestige and status of institutions i n conflict . Duri ng Hirata 's tenure as Privy Seal there were no less than three Cabi net changes . When Prime M i nister Kato Tomosaburo died i n office o n August 2 5 , 1 92 3 , the Regent " put the q uestion" concerning a successor to the Elder Statesmen . 142 Saionj i went to the imperial villa at Hayama to respond to the throne . There he met Privy Seal H i rata a nd told him t hat he had consulted w i t h Matsukata Masayoshi , the only other remai ning Elder Statesman, and that they wished to recommend Yamamoto Gombei as Kato's successor . Asked his opinion , Hirata replied that w homever the Elder Statesmen recommended , he deemed appropriate. The imperial command to form a Cabi net fell to Yamamoto on August 28, 1 92 3 . 143 I n contrast to his role as institutional mediator in the postal treaty controversy , Privy Seal Hirata's role in Cabi net formations was evidently one of liaison between the throne and the elder Statesmen, although they solicited his view . 144 From 1 885 to 1 945 the selection of the Prime Mi nister was a care­ fully preserved imperial prerogative . Because it was an imperial pre­ rogative , however, the appointment of the Prime Mi nister was to be advised upon by various political leaders via the Privy Seal . During their active political leadership in the Cabinet from 1 885 to 1 90 1 , the oligarchs usually agreed to have one of their number serve as Prime M i nister. In 1 892 , for example, four oligarchs-Kuroda Kiyotaka , Yamagata Aritomo, I noue Kaoru , a nd Ito Hirobumi­ gat hered t o select a s uccessor to Prime M i nister M atsukata Masayoshi . 145 As a result , Ito succeeded Matsukata and I noue be­ came Ito's Home Mi nister. After 1 90 1 , however, the ol igarchs retired from positions of direct leadership in the Cabi net , preferri ng to work behind the imperial screen. As the Council of Elder Statesmen ( Genro Kaigi) , they " occupied a supreme position in Japanese politics up to the 1 920's . " 146 Under these circumsta nces the 1 42 . The present emperor became Regent for his father in 1 92 1 . 1 43 . Hakushaku Hirata Tosuke Den, pp. 1 69-70. 1 44 . U nfortunately , H i rata's biographer does not t reat H i rata's role in the succeedi ng cabi net changes . 1 45 . Chitoshi Ya naga , Japan since Perry (New York : McGraw-Hill Book Compa ny, 1 949) , p. 2 1 9. 1 46 . Scalapino, p. 1 50.

OFFICES

AND

O F FI C E R S

OF

P A LA C E

L EA D E R S H I P

1 81

imperial prerogative of appointment was in fact exercised by the Elder Statesmen; the Privy Seal was primarily a liaison betwee n them and the throne i n this function. In 1 924 Pri nce Matsukata Masayoshi died , leavi ng Pri nce Saionj i a s the last Elder Statesma n. Si nce palace leadership was al most ex­ clusively in the hands of Pri nce Saionj i 's partisans , a measure of stability was restored in the Privy Seal 's office . Between 1 92 5 and 1 93 6 there were but two Privy Seals, or one every 5 . 5 years , and both were constitutional monarchists . I n 1 93 2 , however, Pri nce Saionj i , who was over eighty, bega n to make provisions for an advi­ sory system after his deat h. He therefore asked Kido Koichi , the Chief Secretary to the Privy Seal, to examine and revise proced ures for replying to the throne on appoi nting Prime M i nisters . Kido sub­ sequently drafted a plan w hereby the Senior Retai ners would meet to advise the t hrone w hen a Prime M i nister res igned . The conference was to be composed of ex-Prime M i nisters (Senior Retai ners) , the President of the Privy Counci l , the Privy Seal , and the Speakers of bot h houses of the I mperial Diet . Household Minister lchiki Kitokuro, however, deleted the two Speakers from the proposed conference. 147 This conference , as modified by lchiki , met with Saionj i in the palace in J uly 1 934 to assist him in selecting a successor to Prime M i nister Saito Makoto. 148 As a consultant in this process , the Privy Seal gai ned an increased voice in exercising the imperial prerogative of appoi ntment . From 1 93 6 , when Privy Seal Saito Makoto was assassi nated , to November 1 94 5 , w hen the office of Privy Seal was abolished , there were two Privy Seals, averagi ng five years in office . Neither were of Senior Retainer status , yet it was during this time that the Privy Seal acq uired primary responsibility for recommending new Prime M inisters to the throne . 1 49 Yuasa Kurahei, Privy Seal from 1 93 6 to 1 940, had been a career Home Mi nistry official whose highest posts before becomi ng Privy Seal were Civil Governor of Korea ( 1 925-27) a nd I mperial Household Mi nister ( 1 933-36) . His successor, Mar-

�

ezu 1 47 . f?kada Keisuke (Tokyo: Okada Taisho Kiroku Hensan Kai , 1 956), pp. 233-M asash1 , p. 13 I . 1 4 8 . Harada Kumao, 3 : 347-48 . 1 4 9 . l nada Masatsugu, p. 93 . I nada places the date of the Privy Seal's primacy in recom­ n w nd i ng Prime M i nisters precisely at May 1 93 7 .

1 82

OFFICES

AND

OFFICERS

OF

PALACE

L E A D E R SH I P

quis Kido Koichi , had served in two Cabi nets between 1 937 and 1 93 9 prior to his appoi ntment as Privy Seal in 1 940. With Pri nce Saionj i 's death in 1 940, Privy Seal Kido became solely responsible for replying to the throne on basic policy directions and Cabi net changes . The importance of the Privy Seal's office thus varied according to the composition of the forces competing to declare the I mperial Will and the nature of the emperor's senior advisers . The flexibility of the office , al ong with that of the other leading palace offices , allowed the Privy Seal and his palace colleagues considerable ma neuverability to cope with any political situation. The Pri vy Seal, however, was the most i mportant palace officer in rel a t i ng emperor to politics throughout the period 1 885 to 1 945 . As i nitially conceived by the oligarchs and as developed by the first Privy Seal from 1 885 to 1 89 1 , the office of Privy Seal was to be the chief link between court and government . It is not surprising, therefore , that throughout the entire prewar period the Privy Seals as a group were the most highly statused in terms of titles , the most accomplished in terms of prior government careers , and on average the most senior of the four leading palace officers .

THE NATUR E AND ROLE OF PALAC E LEADERS H I P IN PREWAR J A PANESE POLITICS

Table 9 summarizes data on each of the four leadi ng palace officers from 1 885 to 1 945 and demonstrates once again the primacy of the Privy Seals in palace politics . In contrast to the diversity and stature of the Privy Seals, and indeed the Household M i nisters and Grand Chamberlains, the Chief Aides were representatives of only one in­ stitutional constituency , the Army, achieved no titles higher than vis­ count , and on the average were the most j unior of the four palace leaders . As table 9 also reveals , the Chief Aide's was the only office that remai ned strictly autonomous ; no Chief Aide ever held any other position of palace leadership. There was considerable interchange among all the other offices : some officials held two posts concurrently; others transferred from one palace office to another. And Chief Aides

1 83

OFFICES AND OFFICERS OF PALACE LEADERSH IP

TABLE 9. A COMPARATIVE PORTRAIT OF THE FOUR LEADING PALACE OFFICIALS BY OFFICES, 1 885-1 945 Privy Chief Grand Household Seals A ides Chamberlains Ministers 1 00 % ) 1 00 %) ( 1 3 1 00 %) (8 1 00 % ) ( 1 0 (12 =

=

=

=

Average age on taking office

58 . 3

61 . 1

56

65

Range of ages

44--6 7

46-7 1

5 1-62

48-82

6.3

4.6

Average years in office High offices held: Prime Minister Minister of State Privy Council President Privy Councillor Top military office Supreme War Councillor Ambassador House of Peers One o r more of the other top four palace offices Prior career against years in court office: Restorationist Army Navy Foreign Ministry Home Ministry Justice Ministry Finance Ministry Agriculture and Commerce Ministry Court Titles eventually achieved: Prince Marquis Count Viscount Baron None

6

5 I (8 . 3 % ) 6 (50% )

2 ( 20 % ) 2 (20 % )

0 ( 12 . 5%)

4 (30 . 8 % ) 9 ( 69 . 2 % )

2 ( 1 6 . 7%) 7 (58 . 3 % ) (8 . 3 % )

( 1 0% ) 3 (30 % ) 2 (20% )

0 2 (25 % ) 3 (37 . 5 % )

2 ( 15 . 4% ) 7 (53 . 8 % ) 3 (23 . I % )

0 2 ( 1 6 . 7%) 1 0 ( 83 . 3 % )

4 ( 40 % ) ( 1 0% ) 6 ( 60 % )

4 (50 % ) 0 0

4 ( 33 . 3 % )

3 (30 % )

0

23 0 13 15 6

0 I

2 0 4 I 3 2

( 16. 7%) ( 33 . 3 % ) (8 . 3 % ) (25 % ) ( 16. 7%)

2 ( 1 5 . 4% ) I (7 . 7 % ) 1 2 (92 . 3 % )

5 (38 . 5 % )

0 5 17 2 0 0 0

0 49 0 0 0 0 0

14 2 0 II 7 0 0

0 36

0 0

6 20

0 0 0

6 I 2 2 2 0

3 ( 30 % ) 0 3 ( 30 % ) ( 1 0% ) I ( 10% ) 2 (20% )

( 12 . 5%) 4 ( 50% ) 3 ( 37 . 5 % )

( 46 . 2 % ) (7 . 7%) ( 15 . 4% ) ( 1 5 . 4% ) ( 1 5 . 4% )

1 84

OFFICES

AND

OFFICERS

OF

PALACE

LEA D E R SH I P

simply did not move in the same circles a s did other palace leaders , even if those others had also been career military officers . General Nakamura Yuj i ro received two imperial appointments to the House of Peers prior to becoming I mperial Household Minister in 1 920. No Chief Aides were appointed to the House of Peers . General Katsura Taro had been Prime M i nister tw ice before he became Privy Seal­ Grand Chamberlain in 1 9 1 2 . Admiral Saito Makoto had also been Prime M inister before his appointment as Privy Seal ( 1 93 5 ) . Admiral Suzuki Kantaro became President of the Privy Council ( 1 944) and Prime M inister ( 1 945 ) after having served as Grand Chamberlai n . N o Chief Aide, however, served a t any time a s Prime M inister o r as President of the Privy Counci l . Unlike the Chief Aides , therefore , those military officers who held other leading palace offices were " ge n e ra l i s t s " a n d " n e g o t i a t o r s , " n o t d e d i c a t e d i n s t i t u t i o n a l spokesmen. Like their civilian colleagues , they were part o f the elabo­ rate network that linked Japan's political leaders laterally across the top of the plural institutions of government after imperial prerogatives had been i nstitutionalized . The Chief Aides , however, conceived of themselves as career Army officers who had no relation to politics ex­ cept when the Army's interests were involved or when their role as court officials compelled collaboration with the other palace leaders . The bureaucratization of the four leading palace offices over time and t he nature of the evolved i nstitutional balances at the emperor's side are illustrated i n table 1 0 . The periodization is by reigns , w hich, as I poi nted out earlier, corresponds roughly with the maj or stages i n the evolution of palace leadership. During the Meij i period court persons were the maj or component of palace leadership for the only time in the history of palace pol itics from 1 885 to 1 945 . That Restoration leaders were the second component is evidence that court leadership was under the control of the oligarchs, bringing tran­ quility at the side of the emperor throughout the Meij i period . From 1 9 1 2 to 1 92 6 , however, tranquil ity at the side of the emperor was l ost . The diversity of career backgrounds brought to high palace of­ fice during that period i ndicates no firm pattern of oligarchic control or of balanced i nstitutional representation at court . After 1 92 6 , however, a fairly definite pattern of bureaucratic representation at the emperor's side emerged . Four bureaucratic in-

OFFICES AND OFFICERS OF PALACE LEADERSH I P

1 85

TABLE 1 0 . BUREAUCRATIZATION AT THE SIDE OF THE EMPEROR : CAREERS PRIOR TO COURT OFFICE AGAINST YEARS IN COURT OFFICE Household Minister Meiji Period (1885-1912} : Restorationist Army Navy Foreign Ministry Home Ministry Justice Ministry Finance Ministry Agriculture and Commerce Ministry Court Taisho Period (1912-1926} : Restorationist Army Navy Foreign Ministry Home Ministry Justice Ministry Finance Ministry Agriculture and Commerce Ministry Court Showa Period (1926-1945} : Restorationist Army Navy Foreign Ministry Home Ministry Justice Ministry Finance Ministry Agriculture and Commerce Ministry Court N OTE :

Grand Chamberlain

Chief A ide

Privy Seal

Total

23 0 0 0 3 0 0

0 0 0 0 0 0 0

0 16 0 0 0 0 0

6 0 0 0 0 0 0

29 16 0 0 3 0 0

0

0 27

0 0

0 20

0 48

0 I 0 4 3 6 0

0 5 0 0 0 0 0

0 14 0 0 0 0 0

8 2 0 2 3 0 0

8 22 0 6 6 6 0

0 0

0 9

0 0

0 0

0 9

0 0 0 9 9 0

0 0 17 2 0 0 0

0 19 0 0 0 0 0

0 0 0 9 4 0 0

0 19 17 20 13 0

0 0

0 0

0 0

6 0

6 0

Because only those who served in office six months or more are listed, 5 of the

: 1 7 persons who held one or more of the four top palace offices are excluded from the

l abulation : Privy Seal-Grand Chamberlain Katsura Taro (Army}, Privy Seal Hamao Arata (Education Ministry}, Privy Seal Saito Makoto (Navy}, Privy Seal lchiki K itokuro (Home Ministry}, and Chief Aide Hata Shunroku (Army ) .

1 86

OFFICES

AND

OFFICERS

OF

PALACE

L EA D E R SH I P

stitutions o f imperial prerogative were represented , and i n fairly close balance : the Army , Navy , Home Mi nistry , and Foreign Ministry . These four, plus the Fi nance Mi nistry , were the major components of the prewar policy-maki ng process . Why the Finance M inistry had only one of its representatives at the emperor's side , and for only one year in the entire period 1 885 to 1 945 , is a matter for speculation. As noted in chapter 3, the court was fi nancially au­ tonomous and had no relation to the Ministry of Finance . More im­ portantly, perhaps, the maj or issues facing Japan in terms of na­ tional defense, domestic controls, and foreign policy were managed primarily by t he Army , N a vy , Home M i nistry , a nd Foreign Ministry . The Army and Navy exercised the emperor's prerogatives of supreme command and military administration. By the 1 930s they received the lion's share of the national budget in order to carry out a rather extensive conception of " national defense . " The H ome M inistry controlled local government , shrines a nd temples , and the regular police force . With the M inistry of Education, the Home M inistry was involved in maintai ni ng orthodox t hought in Japanese society-but the Home M inistry had the instruments of force to mai ntai n that orthodoxy . The Foreign M inistry exercised the em­ peror's prerogatives in diplomacy and was the primary source of in­ formation about the "trends of the times" internationally. These four bureaucratic structures dealt with the mai n substantive issues of domestic and foreign policy and were therefore of key importance to the palace in its negotiation and ratification roles . Duri ng the 1 930s the inner Cabinet was composed of the Prime Minister, Army M i nister, Navy Minister, Foreign M inister, H ome Minister, and Finance M i nister, on whom all had budgetary de­ mands. The inner Cabi net was most important in maki ng and maintai ni ng national policies until the Liaison Conference system was adopted in 1 937 . 150 When Prime M inister Saito Makoto resigned in July 1 934, for example , he attempted to ensure the con­ tinuity of his policies in t he succeeding Cabinet by having his Army , ISO. Liaison Conferences combined part of the inner Cabinet with t he leaders of t he General Staffs in an effort to produce unity between High Command a nd Cabinet policies . For a description of the Liaison Conference, see Ike Nobutaka, Japan 's Decision for War: Records of the 1941 Policy Conferences (Stanford : Stanford University Press, ! 967), pp. xv-xvi .

OFFICES

AND

OFFICERS

OF

P A LA C E

L E A D E R SH I P

1 87

Navy , and Foreign M inisters carried over and his Minister of Agri­ culture and Forestry made Home M i nister. 151 It appears to be no ac­ cident , therefore, that the mai n components of bureaucratic politics in government were represented at court on an almost one to one basis. And this balance of key institutional forces at the emperor's side remained intact from the late 1 920s u ntil the end of 1 945 . Despite t he i nstitutionalizat i o n of i mperia l prerogatives and political elites outside the palace gates and the accompanyi ng institu­ tional balances at the emperor's side that had occurred by 1 92 9 , the palace remai ned the sole referent for all government decisions and the central institution through which the I mperial Will was declared in politics . Selection of the Prime M inister from 1 885 to 1 945 , for example , was up to "the palace . " This meant those who advised the throne at any given time : the oligarchs , Elder Statesmen, Senior Retainers, Privy Seal . These advisers reached their decisions on the basis of their personnel and policy preferences and the "trends of the times , " as they evaluated them in consultation with other political leaders . B y the late 1 920s Pri nce Saionj i was t he emperor's key adviser as to w ho was best able to form a Cabinet t hat could and would "con­ trol t he situation" outside the palace gates . Pri nce Saionj i ' s nomi­ nation was based on his policy preferences , whom he deemed most able to realize those preferences , a nd w hat was possible, both i n terms of policies and personnel , given the state o f i nstitutional com­ petition to declare the I mperial Will at a ny given moment . As we saw i n the resignation of Tanaka Giichi in 1 92 9 and the 1 930 London Naval Treaty controversy, Saionj i had strong preferences i n domestic and international politics . He favored " party" Cabinets when possible-which meant appoi nti ng as Prime M inister the leader of a maj or political party i n t he House of Representatives , not Cabinets composed entirely of party men or Cabinets elected by the House of Representatives . More importantly, he was guided in his nomi nation by his strong views on foreign policy : the necessity of controll i ng the Japanese Army in China, and the desirability of maintaining A nglo-American amity. Tanaka Giichi , as a retired 1 5 1 . Okada Keisuke, p . 2 3 1 .

1 88

OFFICES

AND

OFFICERS

OF

P A LA C E

LEA D E R SH I P

Army general respected b y that i nstitution o f imperial prerogative and as president of the leading political party in the House of Representatives , appeared to be an ideal nomi nation as Prime M inister in 1 92 7 . Saionj i evidently believed that Tanaka , having support i n two maj or i nstitutions of imperial prerogative, would be able to deal effectively with Army adventurism in Manchuria and northern China . The Chang Tso-lin affair proved that Tanaka was unable to do so, and Saionj i and his allies in the government and i n the palace forced Tanaka to resign . Saionj i nomi nated as Tanaka 's successor Hamaguchi Osachi , president of the ot her leading political party i n the House of Representatives . As we saw earlier, Hamaguchi proved willing and able to do Saionj i 's bidding during the London Naval Treaty con­ troversy of 1 930. Although Saionj i was successful i n obtai ning his treaty , however, he lost his Prime Mi nister: withi n a mont h after imperial sanction of the treaty, Hamaguchi was critically wounded by an assassin. After a brief caretaker government in 1 93 1 under Wakatsuki Reij i ro, Saionj i nomi nated l nukai Tsuyoshi to form a Cabi net . An aged party politician who had served i n the House of Representatives conti nuously si nce 1 890, I nukai had succeeded Ta­ naka Giichi as president of the Seiyiikai in 1 92 9 . But withi n six months of taking office , l nukai was assassi nated . Unable to persist in his party Cabinet policy , Saionj i turned to the Navy . Between the assassi nation of I nukai in 1 932 and the coup at­ tempt of February 26, 1 93 6 , Saionj i selected two senior naval officers as Prime M i nisters : Admiral Saito Makoto a nd Admiral Okada Keisuke . As al ready d i scussed , both were open supporters of Saionj i ' s position on the London Naval Treaty and involved in con­ trolling antitreaty forces within the Navy . As moderates, however , both were out of line with the trends of the times as far as mil itarists and right-wing leaders were concerned . Okada narrowly escaped assassination on February 2 6 , 1 93 6 ; Saito, who had become Privy Seal in late 1 93 5 , did not . With this attack on moderate government leaders and the em­ peror's palace officials , Saionj i 's partisan activities became all but fu­ tile. I n early 1 937 , for example , he recommended as Prime M i nister a retired general of the Army , Ugaki Kazunari , si nce he was aware

OFFICES

AND

OFFICERS

OF

P A LA C E

LEA D E R SH I P

1 89

of t he " unlikelihood of the Army being satisfied with anythi ng less than a cabi net headed by a general . " 152 Ugaki had cooperated i n reducing the size of the standing army during the 1 920s . The Army t herefore deemed him i nappropriate a nd p revented him from formi ng a Cabi net by the simple device of refusi ng to nomi nate an Army M i nister. 153 Through his power to advise the throne on the appointment of the Prime M inister and on basic political issues , Saionj i had managed to brake the forces of extremism from 1 92 1 to 1 93 6 , but with decreasing effectiveness after 1 93 0 . He had also managed to keep constitutional monarchists domi nant among the emperor's advisers during that period . The only leading palace office that el uded Saionj i , as we have seen, was the office of C hief Aide-de-Camp . Ironically, it was duri ng the domination of the constitutional monar­ chists at court that Japanese fascism became the domi na nt force in government . Professor Maruyama Masao divides the evolution of Japanese fascism i nto three period s : preparation ( 1 9 1 9-3 1 ), matu­ ration ( 1 93 1 -36), and consummation ( 1 936-4 5 ) . The union of militarist and renovationist political leaders with rightist societies , which occurred during the preparation period , produced the rash of assassi nat i o ns , attem pted coups , terrorist act i vities , and public agitation during the maturation period that ultimately vaulted the military into the center of political power. 154 While the palace be­ came t he main i nstitution of political moderat i o n , t herefore , Japanese society and the government proper moved in the opposite direction-toward militarism, virulent ethnocentrism, and bureau­ cratic fascism. U nd e r s u c h p r e s s u r e s fr om the " o u t s i d e , " an " I m p e r i a l Household M i nistry i n line with the times " , which Pri nce Saionj i s o strongly insisted upon, could n o longer b e one led b y himself and his " pro-A nglo-A merican , " " l iberal , " " pro-party" partisa ns. To be " in l i ne with the times" after 1 930 mea nt foll owing the lead of the reno­ vationists and mil itarists in the instituti ons of imperial prerogative . 1 52 . Robert J . C . Butow, Tojo and the Coming of the War (Princeton: Pri nceton University 1 96 1 ) , p. 8 8 . 1 5 3 . I bid . , pp. 88-90. 1 54 . Maruyama Masao, Thought and Behaviour, pp. 26-34.

Press,

1 90

OFFICES

AND

OFFICERS

OF

P A LA C E

LE A D E R S H I P

The constitutional monarchists in court fi nally gave way t o the tradi­ tionalists , j ust as the moderates in government were forced i nto the background by the renovationists and mil itarists . The traditionalists were concerned above all w i t h t he preservat ion of i mperial transcendence , so seriously threatened by the terrorist activities of 1 930-3 6 . This meant havi ng the Emperor-i n-State ratify as accurate a consensus of national opinion as possible, regardless of where that consensus led the nation. It also meant that the emperor would not be permitted to persist in open support of constitutional monarchist policies , at least to the extent that his palace advisers would once again be threatened by assassins . 155 As managers of the Emperor-in­ Court and coordinators of the public Emperor-i n-State, therefore , the traditionalists in the palace presided over the consummation of Japanese fascism by movi ng away from the constitutional monarchist position toward a " neutral" eval uation of the "trends of the times" as the basis for advisi ng the throne. O ne of the most promi nent handmaidens of bureaucratic fascism i n government was Pri nce Konoe Fumimaro. Viewed as "a renova­ tionist [ kakushin-teki] politician bringing harmony to the opposi­ tions among the military factions , bureaucrats , and political parties after the Manchurian I ncident [ 1 93 1 ] , " Konoe was a hereditary member of the H ouse of Peers , becoming its Vice President in 1 93 1 and its President i n 1 93 3 . 156 He held no other posts i n government prior to becomi ng Prime M i nister in 1 937 . Duri ng his three Cabi nets between 1 937 and 1 94 1 , Konoe presided over events and decisions of such magnitude as the commencement of the war i n China ( 1 937 ) , the National General M obilization B i l l ( 1 93 8 ) , the Axis alliance ( 1 940) , and the inauguration of the I mperial Rule 1 5 5 . Assassination plots continued to be di rected against certain palace leaders and advisers after 1 936, but t hey were quickly discovered and suppressed . That t hey were quickly put down, in contrast to the 1 930-36 period, suggests that assassination of t he emperor's palace advisers was no longer needed in order for the mil itarists and renovationists in government to have their way (see Butow, Tojo, pp. 1 28-29). 1 56 . Dai jimmei jiten, 2 : 587 . It is claimed that from the time w hen he went to the 1 9 1 9 Paris Peace Conference , Konoe was held i n great "affection" by Prince Saionji (ibid) . During the 1 930 London Naval Treaty controversy, Konoe was able to do Saionji's biddi ng by stifling antitreaty movements i n the H ouse of Peers (Harada Kumao, I : 2 1 -22). But Konoe's chief political role was not so much that of policy advocate, either pro-Saionj i or promilitarist, as it was that of political matchmaker or negotiator. Apparently he moved with the " trends of the times" rather than attempting to control those trends.

OFFICES

AND

OFFICERS

OF

P A LA C E

L E A D E R SH I P

191

Assistance Association ( 1 940) , of which he became president . 157 I n October 1 94 1 General Toj o Hideki replaced Konoe as Prime Minister. Toj o was prepared to lead the nation in the event of war with the United States . Despite his role in preparing the road for such a war, Konoe, apparently, was not . 158 The leadi ng traditionalist at court was Marq uis Kido Koichi , one of Konoe's closest associates . With Kido's appoi ntment as Privy Seal in J u ne 1 940 the traditionalists replaced t he constitutional monar­ chists as the emperor's chief palace advisers . Since some constitu­ tional monarchists , most notably Household M i nister Matsudaira Tsuneo, remained at the emperor's side , Kido's appoi ntment was neither a sudden transition nor a decisive break with the past . Nor, as we shall see in the next chapter, was Kido an advocate of renova­ tionist or militarist policies . Rather, as chief manager of the Em­ peror-i n-State from 1 940 to 1 94 5 , Kido restored the emperor's "traditional " role as ratifier of personnel and policies "in line with the times . " Kido was able to perform this role because , as the em­ peror's chief adviser after Prince Saionj i ' s death in 1 940, he took the consensus that led to his recommending candidates for Prime Minister to the emperor; he also took the consensus regarding national and international trends at any given moment , on the basis of which he rendered advice to the throne on political policies. Despite the vast social , economic , and political changes that occur­ red d u r i ng t he course of Japa nese modernization, t he palace remai ned the key mechanism through which the I mperial Will was declared in politics from 1 88 5 through 1 94 5 . It was able to remain so because Saionj i 's caveat that the Imperial Household Mi nister must come from "the outside" to keep the mi nistry " i n l i ne with the times" was applied with equal rigor to all four palace leaders . The high bureaucratic achievement of those serving in the four key palace offices , the reflection in their career backgrounds of the changi ng political forces outside the palace gates , and the flexibility inherent in those four offices attest to the critical role played by palace I S7 . Konoe's three cabinets were J une 1 937-January 1 939, J uly 1 940-July 1 94 1 , and j u ly -October 1 94 1 . For discussions or Konoe's role in t he 1 930s and early 1 940s, see Borton , pp. 347-69, and Butow , Tojo, pp. 77-3 1 5 . I SH . Butow , Tojo, p. 3 1 4 .

1 92

OFFICES

AND

OFFICERS

OF

PALACE

LEA D E R SH I P

leadership in managing the Emperor-i n-State s o that the I mperial Will would be responsive to the "trends of the times" -not only duri ng the "enl ightened " modernization of the Meij i period but also duri ng the periods of institutional pluralism and of bureaucratic fas­ cism that followed . The key palace officer in mai ntai ning the cen­ trality of the palace in politics during the period of bureaucratic fas­ cism was Japan's last Privy Seal , Marquis Kido Koichi .

CHAPTER FIVE

KIDO KOIC HI AND THE MATRIX OF PALACE POLITICS

has dealt with the theory of legitimacy on w hich the prewar Japanese state rested , the court structure for managi ng palace autonomy, and the nature of palace leadership from 1 885 to 1 94 5 . The purpose of this chapter is to exami ne the network of association and communication that li nked court and government duri ng the 1 930s and early 1 940s . It was during this pe­ riod that Cabi nets led by moderate bureaucrats and party leaders gave way to Cabi nets led by renovationists in the civil and military bureaucracies and in the Imperial Diet , especially the House of Peers . And it was during this period that court leadership passed from the hands of the constitutional monarchists i nto the hands of t he traditional ists . The subject of our analysis is Kido Koichi . Kido is especially suitable for study because he has left an extensive diary of his political associations and activities duri ng the two periods he was i n t he palace : 1 930 t o 1 937 and 1 940 t o 1 945 . 1 More a n appoi ntment hook than a political commentary or record of substantive decisions , Kido's diary allows a comprehensive statistical reconstruction of the OuR

D I SCU S SI O N T H U S

FAR

1 . Kido Kiiichi , Kido Koichi Nikki, 2 vols. (Tokyo: Tiikyii Daigaku Shuppan Kai , 1 966) . This record of 1 ,257 pages is supplemented by a volume of related documents and Kido's

postwar comments on the diary : Kido Nikki Kenyu Kai , Kido Ko ichi Kankei Monjo (Tokyo: J ',-,kyi> Daigaku Shuppan Kai, 1 966) .

1 93

1 94

KIDO

AND

MATRIX

OF

P A LA C E

POLITI C S

patterns of association that linked court and government in t he case of one of the court's most important political personages . Kido's associations and behavior will be exami ned i n 1 932 , 1 93 6 , and 1 94 1 . These three years have been chosen for two reasons . First , all three were years of crisis as bureaucratic fascism matured and consolidated i n government . On M ay 1 5 , 1 932 the last prewar party Prime M inister was assassinated . The Shanghai crisis and the formation of the puppet state of Manchukuo contributed to making 1 932 an important year in Japanese foreign policy . In 1 936 the most extensive coup attempt si nce Saigo Takamori 's 1 877 revolt was mounted agai nst "evil advisers" close to the throne . As i n 1 932 , a new Cabi net came into bei ng as the result of violence . Also i n 1 93 6 the Anti-Comi ntern Pact was concluded with Nazi Germany. I n 1 94 1 two Cabi nets were formed and war was declared agai nst the United States . I n all three years the palace was deeply involved i n Cabinet formations and foreign policy decisions . Second , Kido held different palace posts i n each of these three years . His associations and behavior in each office reveal three dif­ ferent roles he played i n the operation of palace politics : the " secretary , " the " bureaucrat , " and the " negotiator . " As Chief Secretary to the Privy Seal in 1 932 , Kido acted as a political liaison and i nformation broker. His chief associate was another "secretary , " Baron Harada Kumao. Harada was Pri nce Saionj i ' s " public" secretary and was also a member of Kido's i ngroup-a small knot of aristocrats who collectively facilitated Kido's i nformation-gathering role. B oth his role as liaison and his efforts to gather information as a "secretary " were primarily for the benefit of his superior , Privy Seal Makino Nobuaki . As Privy Seal , Makino was one of the key " negotiators" in the prewar political process : he was one of those experienced statesmen responsible for maki ng recommendations to the throne on political appointments and basic national policies . As Director of Peerage Affairs i n 1 93 6 , Kido was a " bu­ reaucrat " -a p u b l i c offi c i a l resp o n s i b l e for t he d a y - t o- d a y management of a specific official j urisd iction. Although Kido was concurrently Chief Secretary to the Privy Seal until J une 1 3 , 1 93 6 ,

KIDO

AND

MATRIX

OF

PA LA C E

P O LI TI C S

1 95

his associations and behavior throughout that year illustrate with sufficient clarity the role of the bureaucrat in palace politics . A bu­ reaucrat might use his office to build a political faction and advocate policies , since policy i nitiatives emerged from within the bureau­ cracy . M ost prewar bureaucrats , however, were ci rcumspect administrators content to play a neutral role i n politics . This was ap­ parently the case with Kido. His chief associates in 1 936 were fellow bureaucrats i n the palace w ho had official duties that i nvolved Kido's official duties . His diary gives no evidence that he used his office to build a political constituency at court to press his palace colleagues on policy matters , or that he used his office to consolidate a policy­ advocating coalition of government and court leaders . I n 1 94 1 Kido was a " negotiator. " As Lord Keeper of the Privy Seal he was responsible to the emperor for negotiating Cabi net formations as well as advising on policies emanating from the i nsti­ tutions of imperial prerogative. Like Sanj o Sanetomi , Japan's first Privy Seal , Kido was the key negotiator in presenti ng unified government policy to the emperor for ratification. In that capacity, Kido's primary associates were the emperor, the M inisters of State , and the other three leadi ng palace officers . Kido's behavior changed markedly i n terms of his patterns of association in each of the three years , showing a remarkable sensitivity to the political roles he was called upon to play . A n exami nation o f h i s associates reveals not only that others performed similar roles in both court and government but that there were other roles essential to the successful operation of Japan's privatized political system : the " ratifier, " the " policy advocate , " and the " instrument of pressure . " The behavior of Kido and his associates was understandable , if not predictable or productive of "good " policies , given the i nstitutional role of the palace as the center of the ratification process i n prewar Japanese politics . Our discussion will conclude with a critique of Maruyama M asao's typology of political personality in prewar Japan in order to demonstrate the primacy of the bureaucracy in the political process and the effectiveness in that process of the political roles revealed by Kido and his associates .

1 96

KIDO

AND

MATRIX

OF

P A LA C E

P O L I TI C S

K ID O 'S A P POINTM ENT AS C H I EF SECRETARY TO THE PRIVY SEAL

Kido Koichi was appoi nted Lord Keeper of the Privy Seal on June 1 , 1 940. Not only was he the last Privy Seal (the office was abolished on November 2 4 , 1 94 5 ) ; he was also the only Privy Seal to have served previously as Chief Secretary to the Privy Seal . Kido became C hief Secretary on October 2 8 , 1 930, after some fifteen years in the Ministry of Agriculture and Commerce and the M i nistry of Com­ merce and I ndustry . Like other prominent leaders in the palace bu­ reaucracy , he was brought in from "the outside . " For the C hief Secretary previous governmental experience and wide associations were apparently desirable, j ust as they were for the Privy Seal himself. The manner in which Kido was brought i nto the palace ill ustrates the high degree of informality that prevailed in the top echelons of Japan's prewar polity . On Sunday , August 1 7 , 1 930, Kido found himself at the golf course with Prince Konoe Fumimaro and Nagaike Choj i , a director of the Bank of Japan . While resting during the match, Konoe mentioned casually that he would like Kido to become Chief Secretary . 2 Kido replied that he had no positive desire to be Chief Secretary but would not decl ine if the circumstances were fa­ vorable . He would leave everything to Konoe . 3 Kido had begun to wonder what Konoe had in mind when he did not hear from him for well over a month . Then, on the evening of September 2 9 , Konoe called : Privy Seal Makino Nobuaki wished to see Kido at nine o'clock the following morning. 4 Accordingly, Kido visited Makino at the Privy Seal's official residence on September 30. Since Kido had tenta­ tively agreed with Konoe to become Chief Secretary , he assented when Makino asked him. Following his interview with Privy Seal Makino, Kido went to meet I mperial Household Minister Ichiki Kitokuro . Ichiki expressed his hope that Kido would accept the post of Chief Secretary and also asked him to serve concurrently as Counselor 2 . The i ncumbent Chid Secretary , Viscount Okabe Nagakage, had met Kido on July 23 and talked about accepti ng elect ion to the House of Peers ; he subseq uently resigned as C hief Secretary (Kido Nikki Kenyii Kai, p. 98). 3 . Kido Koichi , I : 3 3 ; Kido Nikki Kenyii Kai, p. 98. 4. Kido Koichi, I : 3 8 .

KIDO

AND

MATRIX

OF

P A LA C E

P O LI TI C S

1 97

(Sanjikan) to the I mperial Household M i nistry . Kido replied that he had no objection . 5 Commenting many years later on his appoi ntment , the "turni ng point in my destiny , " Kido said that at the time he was somewhat worried about working on a person-to-person basis with Makino, since Makino was over seventy years old . 6 When he met Makino for the first time on September 30, however, he was greatly impressed and instantly made up his mind to accept . There was no " old man's stubbornness" in Makino ; on the contrary , he was a man of " l ibera l " views, and Kido thought h e would b e a b l e to learn from h i m . Kido was not , howeve r , as e nthusiastic ab out I mperial H ousehold Minister Ichiki , who seemed more withdrawn than Makino. Sum­ ming up his impressions of the two men who were to be his supe­ riors , Kido observed : "The Imperial Household Mi nister was, from a look at him , a person of the wise bureaucrat type, but I felt that the Privy Seal had more charm as a person. " 7 Neither in this later commentary nor in the diary at the time of his appoi ntment are there any q ueries or statements about the nature of the office Kido was to assume . He was concerned solely with the personalities he would be working with, not the official duties of the Chief Secretary to the Privy Seal . His i nterviews with both Makino and lchiki were not "j ob interviews" as commonly understood in the West . Obviously, no examination was required . On October 27 Kido had lunch at the Tokyo Club with Konoe , as he had done freq uently i n the past , and learned that his transfer to the I mperial Household M i nistry would be approved by the Cabi net meeti ng of October 2 8 . Kido arra nged his affairs at the M i nistry of Commerce and I ndustry accordingl y : was this to be the fi nal day i n a fifteen-year period o f his life ? Around noon o n October 2 8 Ki no­ shita M ichio of the I mperial Household M i nistry telephoned . Kido's presence at the mi nistry was desired at 1 : 30 p . m . When Kido ar­ rived at the mi nistry he received his appoi ntment from the Imperial H ousehold M i nister : "Appoi nted Chief Secretary to the Lord Keeper of the Privy Seal and concurrently Counselor to the I mperial 5 . Ibid. 6. Kido Nikki Kenyii Kai, 7 . Ibid . , p . 99.

p.

98.

1 98

KIDO

AND

MATRIX

OF

P A LA C E

P O L I TI C S

Household M i nistry . " 8 Accompanied by a mi nistry officia l , Kido then made his round of greeti ngs to the Privy Seal's Office, the B oard of C hamberlains, and the admi nistrative bureaus and divi­ sions of the mi nistry . That afternoon and duri ng the next few days he paid his respects to Pri nce C hichibu , one of t he emperor's three brothers , and to other royal houses . He then made his farewells to officials in the M i nistry of Commerce and I ndustry .9 I n addition to the i nformality with which Kido was i nitially proposed as Chief Secretary , several other aspects of his appoi ntment are worthy of note. Kido had been recommended to the post by bot h Prince Konoe and Viscount Okabe. 10 B oth were aristocrats and friends of long standing. Okabe was not only Kido's predecessor i n office; h e was also a graduate o f the Peers School , a s was Kido. Al­ though five years Kido's senior at the Peers School , Okabe had worked with Kido since graduation to form the Cherry Blossom Fraternity (Sakura Yu kai) , a golfing clubY Okabe was also a charter member of the Eleven Club (ju ichikai) , formed the eleventh day of the eleventh month in the eleventh year of the reign of Em­ peror Taisho ( 1 922) . Composed of a number of aristocrats, it was intended to be something approaching a literary society. As the years passed , most of the club members became members of the House of Peers . Some became Vice M i nisters of State; some , like Kido and Okabe , even became M i nisters . Accordingly , the Eleven Club be­ came a forum for the discussion of current political issues , meeting regularly at Kido's home . 12 Konoe Fumimaro was evidently a member of the club but attended only infreq uently. Thus the spring­ board for Kido's appoi ntment was an aristocratic ingroup cl osely i n­ volved i n clubs , golfing, and the House of Peers . Another aspect of Kido's appoi ntment is the official ritual of ap­ proval by persons in official positions whom Kido did not know socially or officially. Kido looked over and was looked over by bot h the Privy Seal and the Imperial Household M i nister . His ap­ pointment was passed upon by the Cabinet . Formal appoi ntment 8. 9. 1 0. II. 12.

Kido Kiiichi , I : 42. Ibid . , pp. 42-43 . Kido Nikki Kenkyii Kai, Ibid . , p. 98. Ibid . , p . 97 .

p.

96.

KIDO

AND

MATRIX

OF

P A LA C E

P O LI T I C S

1 99

was made by the I mperial Household M i nister. Thus , constituted of­ ficials disposed formally of an appointment proposed informally by a casual social grouping. Such recruit ment procedures were ap­ parently standard for the leading palace offices , with the possible ex­ ception of the C hief Aide-de-Camp . Also noteworthy is the fact that the Privy Seal's office was , at least in this instance, under the j oi nt j urisdiction of the Privy Seal and the I mperial Household M inister. The I mperial H ousehold M inister i nterviewed the candidate for C hief Secretary to the Privy Seal and made the formal appoi ntment . Though the Privy Seal's Office was an adj u nct palace office , its af­ fairs were apparently as much under the I mperial Household Minister as under the direct supervision of the Privy Seal . The procedures involved i n Kido's appoi ntment indicate that a person neither sought directly nor campaigned for a leading palace position . In Kido's case , everything was " left to Konoe , " who i nitiated Kido's candidacy and arranged his official interviews. This pattern of non-office-seeki ng was standard for palace personnel . I nitiative, advocation, and approval all lay beyond the candidate's direct control , although he could presumably veto his ow n nomi­ nation. Fi nally, Kido's appointment illustrates an emphasis on personal as opposed to official concerns by the candidate. As already men­ tioned , Kido was far more concerned about his personal relations with his future official superiors than about the duties of his future office . Possibly Kido knew what the j ob entailed as the result of close association with his predecessor in office , Viscount Okabe . The fact remains, however, that Kido agreed informally to accept the office and " kept his promise" to Konoe after j udgi ng the Privy Seal 's per­ sonality, not the duties of the office or his technical qualifications for it . 13 THE SECRETARY

When Kido became Chief Secretary to the Privy Seal i n 1 930, he brought to that position a network of sociopolitical associations , such 13. Kido did, however, express some reservations about his previous status : he had not hcen an important official in the Mi nistry of Commerce and I ndustry (Kido Nikki Kenkyii Kai, pp. 98-99) .

200

KIDO AND MATRIX

OF

PALACE POLITI CS

as t he Eleven Club, that would prove useful i n dischargi ng his secre­ tarial duties . It was these associations , plus new official connections , that made Kido one of the liaisons between court and government and one of the information sources for palace leaders . Table 1 1 classifies Kido's associations throughout 1 932 according TABLE 1 1 . KIDO KOICHI ' S ASSOCIATIONS IN 1 932 : FREQUENCY AND MODE OF COMMUNICATION Total• (I) Mode of Communicationd

No.

Ten or Moreb (2)

Nine or Less• (3)

%

No.

%

No.

7.6 2.6 1.1 1 .4 0.8 1 .6

31 15 5 0 7 II

6.2 3 .0 1 .0 0 1 .4 2.2

47 12 6 14 I 5

%

OFFICIAL

Office Official residence Official conference Imperial lecture Kido's reports Reports to Kido

78 27 II 14 8 16

Official subtotal

! 54

1 5 . 0%

1 44 75 314 55 71 12 2

14.0 7.3 30 . 6 5 .4 6.9 1 .2 0.2

79 40 1 85 9 51 0 0

15.8 8.0 37 . 0 1 .8 10.2 0 0

65 35 129 46 20 12 2

12 . 3 6.6 24 . 5 8.7 3.8 2.3 0.4

4 1 40

0.4 13 .6

46

0.2 9.2

3 94

0.6 17.8

56

5.5

20

4.0

36

6.8

873

85 . 0 %

69

13 .8%

85

8.9 2.3 1.1 2.7 0.2 0.9 16. 1%

INFORMAL

Home Informal meeting Meals Banquets Eleven Club Sixth Day Club Mist Club Tuesday Luncheon Club Golf Social and miscellaneous Informal subtotal Total

I , 027

1 00 %

43 1 500

86 . 2 % 1 00 %

442 527

83 . 9 % 1 00 %

Kido Koichi, I : 1 26-2 1 0. • Based on a total of I ,027 meetings with 297 persons Kido records having met in 1 932. Average : 3.5 meetings per person. b Based on 500 meetings (48. 7 % of the total) with the 18 persons Kido met 10 times or more. Average : 2 7 . 8 meetings per person. souRcE :

KIDO

AND

MATRIX

OF

PALACE

P O L I TI C S

201

Based o n 5 2 7 meetings (5 1 . 3 % o f the total) with the 2 7 9 persons Kido met 9 times or less . Average : 1 . 9 meetings per person. d The categories devised under Mode of Communication are mutually exclusive. For example, meetings tabulated under Official conference are not listed under Office, even though such meetings took place in official places. Official conference designates meetings with two or more people at an office to discuss official matters. If such a meeting took place at X's office, X would be tabulated under Office and all others present would be tabulated under Official conference. The same applies to Informal meeting and Home. If Kido met Okabe and Konoe at Harada's home, for example, Okabe and Konoe would be tabulated under Informal meeting and H arada under Home. Social and mis­ cellaneous refers to weddings, trips, funerals, parties, social calls. Kido' s reports and Reports to Kido refer almost exclusively to telephone conversations. Imperial lecture refers to Kido's contact with persons giving formal presentations to the throne. Although the emperor was obviously present, I have not included him in the statistics as having met with Kido because the emperor and Kido merely listened to the presentation, apparently without exchanging words. Golf has been listed separately because of the importance of this pastime to Kido. An avid golfer, Kido often recorded his matches and scores more meticulously than the subject, much less the content, of his political meetings. Banquets refers to formal social and state dinners attended by Kido. They have been listed under the informal category because most were social affairs and Kido recorded nothing of official importance at any of these ceremonial functions. Obviously, the basic division of "official" and " informal" that I have employed is somewhat arbitrary. The subdivisions within the categories do reveal, however, the specific kinds of sociopolitical activities in which Kido was absorbed. N OTE : There are several opportunities for error and distortion in compiling such statistics. First, there are several persons with the same surnames whom Kido records only by surname and with no other identification, such as title or official position. There are several Kodamas, for example, some of whom were Kido's relatives and some not. It was not always clear, even using the index of persons at the end of vol. 2, which one was being referred to. Second, Kido does not mention all the people he met. For example, he frequently mentions attending a "Tuesday Luncheon Club," but without naming those whom he met there. This applies to official meetings of Imperial Household Counselors as well. Despite technical problems in using Kido's diary in this manner, the diary provides thorough data amenable to statistical reconstruction of Kido's network of association. Since it is more an appointment book than a record of policy making or ex post facto reflections on events, Kido's diary can be used to demonstrate the nature of palace office and the matrix of communication and association supporting palace politics without relying on the author's value judgments about events, policies, and decisions. Harada Kumao's dictated record and Honjo Shigeru's diary, covering all or part of the period recorded by Kido, are selective accounts in which the authors exercised their judgments and prejudices regarding issues and personnel ; both are to some extent revised accounts, not on-the-spot records. Although both are extremely valuable, they do not lend themselves to the same kind of objective analysis possible with the Kido diary. Kido's wife and daughters, doctors and dentists, and various tutors-calligraphers, etc.-who came to Kido's home for various lessons are excluded from the tabulations in table I I . Such associations were neither numerous nor relevant to Kido's role. •

202

KIDO

AND

MATRIX

OF

P A LA C E

P O LI T I C S

to frequency and mode of communication. Kido recorded having met a total of 2 97 persons (column 1 ). Of these 2 97 , he met 1 8 ten times or more (column 2 ) and 279 ni ne times or less (column 3 ) . Meeti ngs in all three frequency groups are divided into two basic categories , "official" and " informal , " each of which is subdivided according to w here or how the meeti ngs took place . Mode of communication is thus defined by place or circumstances of meeting. The purposes of classifying Kido's association i n this fashion are to analyze his political role as a secretary in terms of frequency of association and mode of communication i n 1 932 and to make comparisons later on with his role as a bureaucrat in 1 93 6 and as a negotiator i n 1 94 1 . The most conspicious aspect of Kido's associations i n 1 932 , im­ mediately apparent in table 1 1 , is the great predominance of i nformal over official modes of communication, regardless of frequency of association : 85 percent of Kido's meetings were i nformal , only 1 5 percent official . Variation according to frequency (columns 2 and 3 ) was a mere 2 . 3 percent , although t here was a tendency for frequency of association to be related positively to i nformal modes of communi­ cation. This is especially true for Kido's meetings at meals : 3 7 . 0 percent o f Kido's meetings with those h e met ten times o r more occur­ red at meals, while only 24 . 5 percent with those he met nine times or less took place in that setti ng. When Privy Seal Maki no i nterviewed Kido in 1 930 he told him that as Chief Secretary he would have a great deal of free time. Did this mean that Kido was able to transform this legally undefined and apparently undemanding office i nto a vehicle for socializing with his aristocratic cronies on government pay and "company time" ? Or does the role of the secretary i n Japanese politics reveal why in­ formal should domi nate official modes of communication ? Why does Kido note t hat 1 930 was a turning poi nt in his career, that after 1 930 he was swamped in political activity ? A second aspect of Kido's associations i n 1 932 is the presence of a key group of persons freq uently met . Eighteen persons met Kido ten times or more during that year. Those eighteen met him an average of 2 7 . 8 times each, whereas those who met him nine times or less averaged a mere 1 . 9 meetings (table 1 1 ) . M oreover, there was considerable variation between the two frequency groups in modes of

KIDO

AND

MATRIX

OF

P A LA C E

P O LI T I C S

203

communication . Kido's meetings at meals, homes , and the Eleven Club account for 63 .0 percent (3 1 5 meetings) of all his associations with the eighteen persons in the ten or more group but only 40 . 6 percent (2 1 4 meeti ngs) of his associations with those h e met ni ne times or less . Conversely, those meeti ngs involving almost purely social and recreational activities-golf, banquets, social a nd miscella­ neous-account for 33 .4 percent ( 1 7 6 meetings) of Kido's total meeti ngs with those in the nine or less group but only 1 5 .0 percent (75 meetings) of those in the ten or more group. It might be wel l , therefore , to begin our analysis o f Kido's activities and role i n 1 932 by determini ng w ho the eighteen persons were , w hat their social and political positions were, and what their domi nant modes of com­ munication with Kido were (table 1 2) . These eighteen may be grouped i nt o dist i nct type cl usters : t i tled aristocrats (eleven) , political contacts i n the i nstitutions of imperial prerogative (four) , members of Kido's family (two) , a nd golfi ng cronies (one ) . The two members of Kido's family , his brother Wada Koroku and eldest son Takazumi , held no political positions and were outside the network of Kido's aristocratic associations . Their meeti ngs with Kido in­ volved social and family activities of no significance to Kido's political office or role . Kato Kyohei was a promi nent busi ness leader whose meeti ngs with Kido all i nvolved golfi ng. Like Kido's brother and son, Kato had no demonstrable relation to Kido's political activities . The remai ning fifteen persons , however, formed the core of Kido's secretarial network-one of the associational matrices that li nked court and government , the emperor and his palace officials with the leaders of political power and status outside the palace gates . As revealed i n table 1 2 , eleven titled aristocrats were the center of Kido's associations in 1 932 , accounti ng for 82 percent of Kido's five hundred meeti ngs with the eighteen persons he met ten times or more . Who were these aristocrats a nd what roles did they play in the accomplishment of Kido's secretarial functions ? Ten of the eleven aristocrats were currently in the House of Peers , as was Kido himself. All ten, again like Kido, had inherited their titles; they were all members of the new hereditary aristocracy that had matured under the Meij i constitutional order. None had been

204

KIDO

AND

MATRIX

OF

P A LA C E

POLITICS

TABLE 1 2 . PERSONS IN CONTACT WITH KIDO KOICHI TEN OR MORE TIMES, 1 932 ( 1 8 PERSONS )

Person

Harada Kumao

BARON

Okabe Nagakage

VISCOUNT

Konoe Fumimaro COUNT Makino Nobuaki couNT Sakai Tadamasa COUNT Kuroki Sanji Wada Koroku K ato Kyohei PRINCE

Takagi Yoshihiro MARQUIS Inoue S aburo Suzuki Teiichi BARON

Oda Nobutsune Inukai Takeru

VISCOUNT

Sasaki Yukitada

MARQUIS

Position

Sec'y to Prince Saionji Kimmochi ; Member, House of Peers Member, House of Peers ; Kido's predecessor as Chief Sec'y to Privy Seal Member, House of Peers Lord Keeper of the Privy Seal Member, House of Peers Member, House of Peers Kido's brother Director, Mitsubishi Trading Co., Ltd. , etc. Member, House of Peers Member, House of Peers Lt. Col. , Military Affairs Bureau Member, House of Peers Member, House of Representatives ; son of incumbent Prime Minister Inukai Tsuyoshi and a secretary to him Member, House of Peers

Number of Meetings

Dominant Mode of Communication Number

Mode

1 35

Meals

53

( 39 . 3 % )

67

Meals

28

(4 1 . 8 % )

57

Meals

29

( 50 . 9 % )

41

Official residence Meals

15

( 36 . 6 % )

15

(62 . 5 % )

24

18 16

Eleven Club Each (Each 33 . 3 % ) 8 or Golf 6 ( 33 . 3 % ) Social 1 2 ( 75 . 0% ) Golf

15

Meals

13

( 86 . 7 % )

13

Meals

6

(46 . 2 % )

13

Meals

9

( 69 . 2 % )

13

Eleven Club

8

(6 1 . 5 % )

12

Informal meetings

4

( 33 . 3 % )

II

Eleven Club

8

( 72 . 7 % )

24

KIDO

AND

MATRIX

OF

P A LA C E

205

P O LI T I C S

TABLE 1 2-Continued

PI!Tsons

Tani Masayuki

Fujinuma Shohei VISCOUNT

Uramatsu Tomoaki (Tomomitsu) Kido Takazumi

Total

Position

Chief, Asia Bureau of the Foreign Ministry Chief, Metropolitan Police Member, House of Peers

Kido's eldest son

Numbi!T of Meetings

Dominant Mode of Communication Numbi!T

Mode

II

Meals

9

(81 . 8 % )

10

Meals

6

(60 . 0% )

10

Eleven Club

9

( 90 . 0 % )

10

Golf or Social

Each (Each 5 50 . 0 % )

500

imperially appoi nted to the House and none had earned h i s title a s the result o f i ndividual contributions to state and society. The spri ngboard of Kido's political activity in t 932 , therefore , was a group of hereditary aristocrats in the House of Peers . 14 Eight of the ten Peers were also members of the Eleven Club, Kido's i ngroup within the aristocracy . These eight accounted for 68.2 percent of all Kido's meeti ngs with those he met ten ti mes or more , thus forming the core of Kido's associates in t 932 . I ncluding Kido, there were fourteen members of the club in 1 932 _ 15 Rangi ng i n 1 4 . M arquis Kido served in t he H ouse of Peers from 1 9 1 7 t o 1 94 5 b y right of hereditary Iitle. The political i mporta nce of the House of Peers and its hereditary aristocrats i n the matu­ ration and consummation of Japan's bureaucratic fascism is suggested by Kido's political career and by that of one of his closest political associates , Pri nce Konoe Fumimaro. Having held no positions of leadership outside the House of Peers, Konoe was the first aristocrat of Court Noble origins to serve as Prime M i nister since Pri nce Saionj i in 1 9 1 1 - 1 2 . Konoe served as Prime M inister three times between 1 937 a nd 1 94 1 . M arquis Kido was t he first purely hereditary Peer to serve as Privy Seal si nce Pri nce-of-the- B l ood Fushimi Sadanaru in 1 9 1 3- 1 4 . 1 5 . The fourteen members were: M arquis Kido Koichi, Baron Harada Kumao, V iscount Okabe Nagakage , Prince Konoe Fumimaro, Count Sakai Tadamasa , Count Kuroki Sanj i , Vis­ munt Oda Nobutsune, Marquis Sasaki Yukitada, V iscount Uramatsu Tomoaki (Tomomitsu), V iscount Soma Taketane, Ma rq uis Matsudaira Yasumasa , Marquis Hirohata Tadataka, ( :ount Arima Yoriyasu, and Count Ya nagisawa Yasutsugu.

206

KIDO

A ND

MATRIX

OF

PA LA C E

POLITICS

age from thirty-nine to forty-eight , all fourteen had inherited their titles . Most had graduated from Kyoto I mperial University; the re­ mai nder were graduates of Tokyo I mperial University. These two universities were the elite institutions of higher education in prewar Japan. In short , the Eleven Club was a group of approximately the same age , social station, educational background , and base of political operations. It was the inner ri ng of Kido's concentric rings of aristocratic association that reached ultimately into the palace , to Kido's direct superior, Privy Seal Count Makino Nobuaki . Count Makino was the only aristocrat Kido met ten times or more who was not a member of the House of Peers at that time . And he was the only aristocrat i n the ten-or-more frequency group who achieved his highest title as the result of merit . B ut si nce only eight of the thirteen Eleven Club members met Kido ten times or more in 1 932 and the frequency of Kido's associa­ tion with these eight varied from 1 0 to 1 3 5 meetings , something other than mere ingroup fraterniz ing would appear to be involved . The central person in Kido's 1 932 network of associations was Baron Harada Kumao, a member of the Eleven Club who was also Prince Saionj i ' s political secretary .16 Kido's meeti ngs with Harada were twice as numerous as those with any other person Kido met during 1 932 and comprised 1 3 percent of his total meeti ngs in that year. Harada's association with Kido, as with the other members of the Eleven Club, was of long standing. For example, when Harada was appoi nted Secretary to the Prime Mi nister in J une 1 924 (a post he held for two years) , Kido, Viscount Okabe , and Pri nce Konoe 1 6 . Saionj i's private or personal secretary was Nakagawa Koj lirii. Consequently, Harada had been referred to as Saionji's " public" secretary (Harada Kumao, 1 : i i ) . Born in 1 88 5 , Harada w a s o f Kido's age group a n d , like Kido, a descendant o f t he merit aristocracy created during the Meij i peri od . A graduate of Kyoto I mperial University, as was Kido, Harada served with the Bank of Japan ( 1 9 1 6-22) a nd traveled in Europe and America under the aus­ pices of the I mperial H ousehold M inistry ( 1 922-24). In 1 92 6 he became a member of the Sumitomo Company, one of the four great fi nancial and i ndustrial combines in prewar Japan. That same year he became Pri nce Saionj i 's public secretary . I n 1 93 1 he was elected by his fellow aristocrats to serve i n the H ouse of Peers. Like Kido, t herefore, Harada is an example of the aggl utination of status, political position, and busi ness connection that characterized the prewar aristocracy . Harada died on February 26, 1 946, ten years to the day, ironically, after the coup w hich did so much to destroy Saionj i 's influence in politics (Ibid . , " Bekkan," p. 389; Shiigiin-Sangii n , Gikai Seido Shichiju Nen Shi [Kizokuin-Sangiin Giin Meikan] , " Da nshoku Giin" section, p. 78).

KIDO

AND

MATRIX

OF

P A LA C E

POLITICS

207

were all i nvolved . 17 I n addition to bei ng aristocratic associates , however, Kido and Harada were more importantly connected by virtue of their secretarial roles . Because Saionj i was the highest political adviser in the Japanese state from 1 924 to 1 940, Harada, as his political secretary , had almost carte blanche access to all civil and military leaders . His position was one of an unofficial post to a semiofficial person, the last Elder Statesman. 18 Kido's position as Chief Secretary to the Privy Seal was that of an official post to an of­ ficial person. The two secretaries were the key liaisons i n the official process that commenced when the " imperial q uestion" (gokamon) was put. When a Cabi net resigned , the emperor would ask the Privy Seal about appoi nting a new Prime M i nister. The Privy Seal would then reply that the Elder Statesman-Pri nce Saionji-should receive the " imperial question, " meaning that the Privy Seal would as­ certain Saionj i 's views as to a suitable successor. In this fashion, the process of consultation, negotiation, resolution, and reply to the throne would be set i n motion . Kido and Harada, in their secretarial roles , were brought toget her as liaisons i n this critical process of responding to the throne when a " political change" (seihen) occur­ red-that is, when a Cabi net resigned . O ne simple i ndicator t hat H arada a nd Kido were brought together primarily by virtue of their pol itical roles is the very fre­ quency with which the two met : 1 3 5 ti mes throughout 1 93 2 . That the frequency of association was not merely i ngroup socializing is further substantiated by contrasting the number of meetings between the two before and after Kido became C hief Secretary . During the first six months of 1 930 Kido and Harada met only four times-at t he " usual " Monday lunches at the Tokyo Club with fellow aristocrats such as Konoe and Okabe . 19 After Kido was appointed Chief Secretary on October 2 8 , 1 930, he saw Harada 1 0 times in the 1 7 . Harada Kumao, I : vi ; " Bekka n , " p. 390. 1 8 . Harada was not officially appoi nted by the government or court to be Saionj i ' s «' ichi a s Di-

rector of Peerage Affairs, 220, 2 3 1 -33; im­ portance to Kido Kiiichi as Privy Seal , 2 5 1 , 254; and palace leadership, 1 1 6 Arita Hachirii, 233 Army : balance w i t h the Navy at court , 1 4 1 -4 2 ; and the Chang Tso-lin affair, 1 44 , 1 49, 1 50, 1 60; and the Chief Aide-de­ C a m p , 1 42 -4 3 , 1 5 8 - 6 8 , 1 7 0 , 1 82 - 8 4 ; Control Faction o f (q . v . ) ; a n d t h e Feb­ ruary 26 Incident , 1 64-67 , 282; and the Hamaguchi Cabinet , 1 03 ; I mperial Way Fact ion of (q . v . ) ; and the I mperial Will, 95 , 1 0 1 , 1 42 ; and military representation at the emperor's side, 60, 8 1 - 8 4 , 90, 1 4 1 -42, 1 60-6 1 , 1 82-84 , 1 85-86; and the October I ncident, 92-95 ; as an "Outlaw" organization, 282; participation in pol itics, 1 62-63; rivalry with the Home M inistry, I 00- 1 , 292n, 3 1 8 ; and Saionj i's second Cabinet , 1 04, 1 40; "subjective tendencies" of, 264, 266; and Tanaka Giichi , 1 44 , 1 87-88; a n d t h e Tiij ii Cabinet , 1 03 ; and Ugaki Kazunari , 1 88-89 ; see also Aides­ de-Camp; I mperial prerogative (of su­ preme command) Board of Chamberlains, see Chamberlains (Board of) B r i t i s h pol it ics u n d e r G eorge I l l : a n d prewar Japa nese pol itics, 3 1 4 - 1 6 B u reaucrac y : a n d .J apa nese pol it ical

343

344

I NDEX

B ureaucracy ( Cont. ) modernization , 1 0; as an organization in the national polity, 2 8 1 -82 ; primacy in politics of, 3-4 , 1 0, 9 1 , 1 06, 1 95 ; see also "Bureaucrat " ; Japanese (bureaucratic) fas­ cism; " Policy advocate " ; Renovationists " B ureaucra t " : Kido Kiiichi as, 1 94-9 5 , 220-37, 275-309; a s a political role, 9 , 1 94-95 , 222, 288, 324 Cabinet (informal) , 1 7- 1 8 , I Bn ; as the direct advisory group to the emperor, 22 Cabinet : as a coalition of institutional elites, 1 02-5 ; composition of, 3 1 5- 1 6 ; creation of the Cabinet system, 2 4 , 1 7 5 ; inner Cabinet , 1 86-87 ; a n d the palace, 1 05 ; "responsibility, " 268, 268n ; role of, 1 02 ; and S a nj ii 1 75-76; Sanetom i , " t ra nscenden t a l " C a b i n e t s , 9 8 , 1 3 1 , 1 37-38, 325 ; see also Hamaguchi Osachi ; House o f Representatives; Okada Keisuke; Political parties; Saionj i Kimmochi ; Tiijii Hideki Ceremonies, Board of the : and the Cham­ berl ains, 7 8 ; evolution of, 30-3 1 ; and foreign dignitaries, 64 Ceremonies, Office of the, 30 Chamberl a i n s , 7 4 - 7 9 ; and the A i d es-de­ Camp, 8 1 , 84, 86; and the aristocracy , 74-7 5 ; Board of, 76, 78-79, 1 69 ; and the I mperial Guards, 94 ; and the inner side of the Imperial Household M inistry, 63; see also Grand Chamberlain Chang Tso-li n : murder of, 1 44 , 1 60 Chang Tso-lin affair, 1 44-4 5 , 1 48-5 1 ; and the Army (q . v . ) ; and the emperor, 1 44-4 5 , 1 46-47 ; and Tanaka Giichi, 1 44 , 1 88 Chichibu Yasuhito ( Pri nce-of-t he- B l o od ) : political involvement of, 304-7 Chief Aide-de-Camp, S i n ; abolition of the office of, 2; autonomy of, 1 42 , 1 82 ; career backgro u n d s of, 1 4 2 , 1 5 8 - 5 9 ; C h i n a experience of, 1 58-59, 1 6 1 ; duties and role of, 8 1 -82 , 1 05 - 6 , 1 5 7 , 2 6 5 , 2 6 7 ; as General Staff informant (spy), 1 56-5 7 , 1 59 , 1 6 1 n , 1 70; a n d the Gran!! Cham­ berlain, 1 42-43 , 1 53 , I S S, 1 57 , 1 69, 1 70; and Hara Kei, ! 59 ; Hasunuma Shigeru

•

a s , 1 59, 262 , 265-67 , 269-70; Honjii Shi­ geru as, 1 60-70, 304 , 305 ; and the Im­ perial Household Minister, 1 69-70; and the London Naval Treaty, 1 53-5 5 ; as a " n egot i a t o r , " 1 7 0 , 2 7 4 ; as P r i v y Counc i l l ors, 1 5 9 ; and the P r i v y Sea l , 1 65-70; and Privy Seal Kido Kiiichi , 262, 265-67 , 269-70 ; Saionj i Kimmochi's lack of control over, 1 89 Chinda Sutemi (Grand Chamberlain), 1 4 1 Chiishii (doma i n ) : a s a decision-making model for the prewar Japanese political process, 320-23 , 324, 329; and the Im­ perial Household Minister, 1 23-24 ; and the Meij i ol igarchs, 1 02 n , 1 23 ; and the Meij i Restoration, 1 22-23 Civil Government (Dajo kan ) , 1 7 ; and the Government of Rites (jingikan ) , 28-30, 32 Clique, see Faction Club(s), see Eleven Club; Sixth Day Club; Tuesday Luncheon Club Confucianism : and the role of the emperor, 26, 33; and Shinto rites, 33 Consensus , see Emperor; " Negot iator " ; Palace; Palace leadership; Trad itional ists Const itutional ists : as " policy advocates , " 325-26, 326n Constitutional monarchists: control of the palace, 1 07 , 1 1 1 - 1 2 , 1 8 1 , 1 89 ; and the February 26 I ncident , 330-3 1 ; and the Grand Chamberlain, 1 4 1 -4 3 ; and the Im­ perial Household M inister, 1 29-32 ; and imperial transcendence, 326-27 ; influence on the emperor, 1 68 ; and the London Naval Treaty, 1 50, 1 56; and Minobe's t h e o r y of c o n s t i t u t i o n a l m o n a rc h y , 1 1 1 - 1 2 ; a s " negotiator(s)" (q . v . ) ; political partisanship of, 1 1 1 - 1 3 , 1 4 1 -42, 1 50, ! 56 ; a n d t h e t r a d i t i o n a l i s t s , 1 1 2 - 1 3 , 1 90 , 326-29 , 330; see also Ichiki Kitokuril ; M a k i n o N o b u a k i ; S a i o nj i K i m m o c h i ; Suzuki Kantarii Constitutional monarchy, see Constitutional monarchists; Emperor; Minobe Tatsuk ichi Control Fact ion (Army), 99; and the Fch­ ruary 26 Incident , 286-87 ; and intrainsl i ­ tutional factionalism, 325 C o u rt , see E m p e r o r - i n - C o u rt ; l m p