war and state building in medieval japan - ferejohn_ john

War and State Building in Medieval Japan

War and State Building in Medieval Japan

Edited by John A. Ferejohn

and Frances McCall Rosenbluth

Stanford University PressStanford, California

Stanford University PressStanford, California

©2010 by the Board of Trustees of the Leland Stanford Ju nior University. All rights reserved.

No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or in any information storage or retrieval system without the prior written permission of Stanford University Press.

Printed in the United States of America on acid- free, archival- quality paper

Library of Congress Cataloging- in- Publication Data

War and state building in medieval Japan / edited by John A. Ferejohn and Frances McCall Rosenbluth. p. cm. Includes bibliographical references and index. ISBN 978- 0- 8047- 6370- 7 (cloth : alk. paper)—ISBN 978- 0- 8047- 6371- 4 (pbk. : alk. paper) 1. Japan—Politics and government—1185– 1600. 2. Japan—History, Military— To 1868. 3. Peasants— Japan—History. I. Ferejohn, John A. II. Rosenbluth, Frances McCall. DS857.W267 2010 952'.025—dc22 2009034670

Typeset by Westchester Book Group in 10/14 Minion

v

WE HAVE INCURRED large debts of gratitude to many people and institutions in the years it has taken us to bring this volume to completion. We owe the great-est thanks to the six historians who were willing to bridge the divide between history and social science to engage with us about the pro cesses of state build-ing in medieval Japan: Th omas Conlan, Karl Friday, Susumu Ike, Tsuguharu Inaba, Pierre Souyri, and Carol Tsang. We stand in awe of their erudition and deep historical knowledge. We are also grateful to Shigekazu Kondo and Nor-iko Kurushima of the Historical Materials Research Institute at the University of Tokyo for participating in the workshop we hosted in Kyoto in March 2004. Terry MacDougall of the Kyoto Center for Japa nese Studies graciously made facilities available to us for the workshop, and the translators of Simul Inter-national did an exceptional job of translating the workshop in real time in both directions. Th e Council on East Asian Studies at Yale University pro-vided funding and the administrative support for the workshop. Teruo Ut-sumi, a postdoctoral fellow in Po liti cal Science at Yale in 2004– 2005, trans-lated the workshop memos from the Japa nese participants into En glish, and Jun and Naomi Saito translated additional materials when it was time to write the book. We are also grateful to Mikael Adolphson, Robert Bates, Wayne Farris, Carol Gluck, Phillip Hoff mann, Hamada Koichi, Margaret Levi, Mark Ramseyer, Conrad Totman, and Mimi Yiengpruksawan for helpful comments and to two anonymous reviewers for close and insightful readings of the en-tire manuscript. Stacy Wagner and Jessica Walsh of Stanford University Press and Barbara Goodhouse of Westchester Book Group provided wonderful editorial and production support. We are grateful to Keizo Goto for helping

Ac know ledg ments

vi Acknowledgments

us procure the rights to the spectacular depiction of the Battle of Nagashino that graces the dust jacket, and to the Nagoya City Museum for granting permission.

Contributors ix

1 War and State Building in Medieval Japan John A. Ferejohn and Frances McCall Rosenbluth

2 Th ey Were Soldiers Once: Th e Early Samurai and the Imperial Court Karl Friday

3 Competence over Loyalty: Lords and Retainers in Medieval Japan Susumu Ike

4 Community Vitality in Medieval Japan Tsuguharu Inaba

5 “Advance and Be Reborn in Paradise . . . ”: Religious Opposition to Po liti cal Consolidation in Sixteenth- Century Japan Carol Richmond Tsang

6 Autonomy and War in the Sixteenth- Century Iga Region and the Birth of the Ninja Phenomenon Pierre Souyri

7 Instruments of Change: Or gan i za tion al Technology and the Consolidation of Regional Power in Japan, 1333– 1600 Thomas Conlan

Contents

viii Contents

Postscript John A. Ferejohn and Frances McCall Rosenbluth

Glossary

Index

ix

John A. Ferejohn is a po liti cal economist and demo cratic theorist who has writ-ten widely in the areas of po liti cal institutions and behavior, judicial politics, and the philosophy of social science. He was the Carolyn S. G. Munro Profes-sor of Po liti cal Science at Stanford University before becoming Professor of Law and Po liti cal Science at New York University in 2009. Ferejohn is a mem-ber of the National Academy of Sciences and the American Academy of Arts and Sciences, and has held fellowships with the Brookings Institution, the Guggenheim Foundation, the Center for Advanced Study at the University of Illinois, and the Center for the Advanced Study of the Behavioral Sciences at Stanford. He received an Honorary Degree from Yale University in 2007 for his contribution to the development of positive po liti cal theory. He sits on the editorial boards of Social Choice and Welfare, Demo cratization, Supreme Court Economic Review, and the Cambridge Press series Philosophy and Law and Economics and Philosophy.

Frances McCall Rosenbluth is a po liti cal economist with a special interest in Ja-pan. She is the Damon Wells Professor of International Politics at Yale Uni-versity. She is a member of the American Academy of Arts and Sciences and has received fellowships from the Fulbright Commission, the Social Science Research Council, the Abe Fellowship Program of the Japan Foundation, and the Council on Foreign Relations. Her books include Financial Politics in Con-temporary Japan (Cornell 1989), Japan’s Po liti cal Marketplace (Harvard 1994, with Mark Ramseyer), Th e Politics of Oligarchy: Institutional Choice in Imperial Japan (Cambridge 1996, with Mark Ramseyer), an edited book on Th e Po liti cal

Contributors

x Contributors

Economy of Japan’s Low Fertility (Stanford 2007), Women, Work, and Power (Yale University Press 2010, with Torben Iversen), and Japan Transformed: Po-liti cal Change and Economic Reform (Prince ton University Press 2010, with Michael Th ies).

Thomas Conlan is a historian of Japan at Bowdoin College. Professor Conlan studied Japa nese history at the University of Michigan (B.A., 1986), Kyoto Uni-versity, and Stanford University (Ph.D., 1998). Conlan’s scholarship focuses on medieval Japa nese history and in par tic u lar on the nature of warfare and the po liti cal role of esoteric (Shingon) Buddhism. His books include In Little Need of Divine Intervention (Cornell 2001); and State of War: Th e Violent Order of Fourteenth Century Japan (Michigan 2003).

Karl Friday teaches at the University of Georgia, where he focuses on Japa nese military institutions and traditions. His Ph.D. is from Stanford in history in 1989. His books include Hired Swords: Th e Rise of Private Warrior Power in Early Japan (Stanford 1992), Legacies of the Sword: the Kashima Shinryu and Samurai Martial Culture (Hawaii 1997), and Samurai, Warfare and the State in Early Medieval Japan (Routledge 2004). Professor Friday’s most recent book is called Th e First Samurai: Taira Masakado and His World (John Wiley & Sons).

Susumu Ike is a professor of medieval and early modern history at Hitotsubashi University in Japan. He is widely known in Japan for his economic history of Japan and for his research on Hideyoshi Toyotomi’s invasion of Korea.

Tsuguharu Inaba is well known in Japan as a proponent of “new cultural his-tory,” which examines po liti cal economy from “the bottom up.” Eschewing older models of history that focused on elite behavior or economic determin-ism, Inaba’s research entails painstaking archival work to understand the lives of common villagers.

Pierre Souyri is a historian of medieval Japan at the University of Geneva. Pro-fessor Souyri received his Ph.D. in history at Paris- Nanterre University in 1984 and taught at the Paris Institute of Oriental languages and civilizations (Inalco) for fi ft een years before taking his current position in Geneva in 2003. Although most of his scholarly work is in French, English- speaking readers are familiar with his highly accessible book on Japa nese history, Th e World Upside Down: Medieval Japa nese Society (Columbia 2001).

Contributors xi

Carol Richmond Tsang is a specialist in the religious movements in medieval Japan that successfully resisted Japa nese territorial consolidation for many de cades. She received her Ph.D. in history from Harvard in 2005 and is the author of War and Faith: Ikko Ikki in Late Muromachi Japan (Harvard East Asia Center 2007).

1

Introduction

Th e Ninja— the lightly armed warrior who operates by stealth and amazing physical prowess to attack powerfully equipped enemies— is a familiar comic book image and heroic action fi gure. It is generally known that the ninja ex-isted sometime in the mists of Japa nese history. Less well understood is that the ninja was but one manifestation of fi erce and extensive re sis tance to en-croaching armies in the dying years of medieval Japan. Local farming com-munities, particularly those in mountain valleys, armed themselves with sim-ple weapons and guerrilla techniques to forestall the trend toward territorial consolidation and centralized taxation. Th e transformation of “ninja” (the “for-bearing ones” or shinobi mono) into warriors with virtually supernatural pow-ers is a recent invention that glorifi es the struggle of humble mountain villages for local autonomy in the late sixteenth century.

Th e world is more familiar with similar events in Eu rope. Th e legend of Wil-liam Tell is of a simple mountain man who inspired Swiss alpine farmers in 1307 to resist domination by the Habsburg Empire. Tell, it is said, was forced to shoot an apple on his son’s head in exchange for freedom aft er he failed to bow to the Austrian governor’s hat placed in the village square. In the Battle of Morgarten in 1315, Swiss farmers armed with rocks, logs, and pikes are said to have crushed the magnifi cent cavalry of Duke Leopold I of Austria in an ambush at Morgarten Pass, pushing countless horses and their riders off a steep mountainside, spear-ing other unfortunates through with pikes, and causing the rest to fl ee in terror.

Swiss pikers from mountain villages managed to protect their land from foreign invaders, thereby assuring Swiss autonomy. Feared and admired the

War and State Building in Medieval JapanJohn A. Ferejohn and Frances McCall Rosenbluth1

2 Ferejohn and Rosenbluth

world over for their ferocity in battle, Swiss fi ghters were recruited into mer-cenary armies throughout Eu rope. Th e Roman Catholic pope chose them for his own guards, a role they continue to serve, at least symbolically, to this day.

Unlike the Re nais sance Italians or the seventeenth- century En glish, the Swiss did not elaborate an indigenous theory of limited government, though their practices of cantonal government with local referenda have endured. Th e Swiss mountain warriors were uneducated farmers and woodsmen scrab-bling out a living in alpine valleys and were unfamiliar with the classical Greek and Roman texts that inspired Italian and En glish antimonarchical theoriz-ing. What distinguishes the Swiss in the forest cantons from farmers else-where— as well as from Swiss farmers in the rolling hills in the north— was not so much a belief in the right to their land, but the formidable terrain that made it possible for them to think they had a chance to preserve their in de-pen dence. Th ere is little wonder that the great plains of Eu rope, which some-times doubled as highways for marauding armies, were populated with seem-ingly weak- kneed farmers who chose instead to exchange their labor for military protection. Japa nese mountain dwellers and Swiss alpine farmers took natu-rally to fi ghting for their freedom, not because they were braver than their lowland counterparts, but because their craggy fortresses gave them the pos-sibility of resisting domination.

Th ree other groups in Japan successfully resisted po liti cal incorporation for centuries. For seafaring pirates (wako), water provided the functional equiva-lent of mountain defense. Th eir ships navigated deft ly through the coastal wa-ters, which they knew better than those who commanded the commercial ships on which they preyed or the government’s ships that pursued them. As Japan’s earliest historical rec ords testify, pirates plagued coastal commerce around the Japa nese archipelago from time immemorial.

Th e po liti cal arm of Buddhism constituted a second group in medieval Japa nese society that managed for centuries to repel the government’s encroach-ing territorial and jurisdictional authority. Buddhist temples, monasteries, and farming communities, oft en heavily armed but also oft en allied with mem-bers of the imperial family, avoided government taxation and regulation un-til Oda Nobunag, one of the unifi ers of Japan, fi nally brought them to heel in the 1580s. Priests protected the tax- free status of temple lands by promising blessings to their patrons, but they would resort to armed defense when nec-essary. In the case of the spectacularly expansive Ishiyama Honganji branch of Jodo shinshu (Pure Land) Buddhism (discussed in detail in Carol Tsang’s

War and State Building in Medieval Japan 3

chapter in this book), thousands of believers were members of a vast Buddhist movement in the province of Kaga. Th ey enjoyed de facto autonomy from Kyoto or local warlords until they were vanquished in 1584.

Less romantic but more successful was the opposition to centralized rule by the territorial domains in the far- fl ung islands of Kyushu and Shikoku or the outer reaches of eastern Honshu, which had consolidated locally around a powerful warlord (daimyō). It was not until the Battle of Sekigahara in 1600 that these great outlying domains were vanquished. Th is battle occurred some 15 years aft er the defeat of mountain villages and religious communities, and only when one of the lords switched sides in the end game to gain spoils from the others. Th e secret to the local domains’ longevity was their attainment of considerable economies of territorial scale through the exchange of security for taxation with which to fund large armies. Th is early set of successful Hobbesian bargains at the local level would infl uence Japan’s constitutional structure for centuries to come, in the form of Tokugawa’s de facto federal system, which was built on semiautonomous domains.

All of Japan, some parts of which were more aff ected than others, suc-cumbed to Tokugawa rule for three centuries before a new government would take tentative steps toward constitutional monarchy in 1868. Although the Meiji oligarchs only cracked open the door to electoral competition, the energetic expressions of free speech and support for democracy by incipient po liti cal par-ties were testament to a latent yearning for self- governance. Th is is not to say that Japan’s freedom- fi ghting past was a continuing legacy that kept alive the potential for re sis tance. Re sis tance or acquiescence in Japan’s early history fol-lowed a pattern of opportunity or necessity. Th e Japa nese accommodation to military rule in the 1930s and 1940s, which was followed by an enthusiastic embrace of democracy from 1945 onward, is better explained by changes in con-straints than by long- standing mental frames.

Th is book relates the tumultuous events of Japan’s medieval and early modern history— roughly 1185 to 1600— to theorizing about war and politics elsewhere. Japa nese resisters and Swiss alpine warriors are exceptions to the general rule that people tend to populate fertile plains where livelihood is the easiest to secure. Th e plains areas were also the favored pathways of invading armies and were used to destroy the food supplies of enemy troops as well as to amass large armies on a battlefi eld. While Japa nese and Swiss holdouts provide a fascinating sideshow, the main story of the emergence of the modern territorial state is a Hobbesian one of distraught peasants exchanging fi nancial


and labor resources for military protection. We do not intend to paint a pic-ture of happy peasants bargaining and contracting for a better life. Rather, we seek to underscore the severe circumstances in which the Japa nese, along with many of the earth’s population, found themselves. As the weak have always known, when life is “solitary, poor, nasty, brutish, and short,” subservience to a protective power may be a lesser evil, even if it is a deeply resented one. Japa-nese and Swiss fi ghters off er a romantic picture of rustic self- governance of the sort that Rousseau contemplated in his discourse on equality. But Rousseau’s world was out of reach for most people. In the embattled lowlands of France, for example, the protection that comes with strong centralized government gave Bodin’s work widespread credibility. Opposition to absolutism was stiff er, and theorizing about limited government was more prolifi c in Italy and En gland, whose indeterminate topography left open a range of po liti cal outcomes.

War, though miserable for those who fi ght and for those whose homes and fi elds are destroyed in the path of battle, can sometimes function as a po liti cal leveler. History provides some dramatic examples of po liti cal rights that have issued from war mobilization, starting with classical Athens and republican Rome. Mobilizing for war can shift the balance of bargaining power away from rulers in favor of those whose resources are required for battle. But much depends on several factors that aff ect how much and to whom rulers need to make concessions in exchange for resources, including whether the people sup-plying resources for war value the protection of a powerful ruler. If communi-ties are confi dent of their ability to protect themselves, they will be willing to fi ght for others only in exchange for something of value such as po liti cal free-doms or, if they are already free, for pay.

Japa nese mountain villagers needed relatively little protection from over-lords because their topography made it possible for them to protect themselves. By contrast, agricultural communities that were located in the crossroads of competing territorial claims were more likely to supply increasingly large rev-enues in exchange for protection. Th eir fear of violence and instability was greater even than their desire for freedom from domination. Th eir willingness to supply resources for large armies lies at the root of Japan’s po liti cal unifi ca-tion in the sixteenth century. Th e same logic accounts for the rise of Eu rope’s large territorial monarchies.


The Rise and Fall of Decentralized Military Rule in Medieval Japan

Th e debates among social and economic historians over the repressive nature of Japa nese feudalism have largely played themselves out as accumulating evidence suggests that farmers retained some leverage in dealing with overlords. We will therefore avoid using the term feudalism altogether. Moreover, farmers’ lever-age varied considerably over time and place. Still underdeveloped, however, is theoretical analysis explaining this variation in leverage both within Japan and between Japan and elsewhere. Th e contributors to this volume establish that, holding all else constant, farmers’ bargaining leverage was inversely related to their vulnerability to military attack and hence to their willingness to pay for protection.

All else is not constant, of course, because there were also more purely eco-nomic sources of farmers’ bargaining power, such as labor scarcity during the early period of land abundance. Japan was settled in the Paleolithic period, tens of thousands of years ago, by hunter- gatherers from the Asian mainland (to which Japan was physically attached by land bridges during the ice age) and fi sher folk from Polynesia who enjoyed land abundance and relatively egalitarian social structures. Th en, in about 300 b.c., waves of immigrants from Korea invaded Ja-pan and pushed the earlier inhabitants into the mountains and outlying islands. Th e new ruling elite or ga nized into po liti cal units (uji) that jostled among them-selves for preeminence. By the eighth century, the uji had imported ideas along with material culture from China and took to calling their leader an “emperor” on the Chinese model. Imperial succession, though sometimes spectacularly con-tested, was usually managed peacefully through negotiations among a co ali tion of leading clans. Unlike many powerful monarchical dynasties in China or Eu-rope, imperial succession rules were loose, allowing for a large number of poten-tial heirs. A signifi cant part of the ruling class derived benefi t from the imperial institution, giving it the structure of an oligarchy rather than an autocracy.

Th e scions of some court families emigrated to the provinces beginning in the late ninth century. Th ey did well for themselves by exploiting their connec-tions to powerful court fi gures and institutions, and by obtaining sinecures as government offi cials or managers of private estates. Th e court, in turn, culti-vated its connection to these emerging military families to help extend the reach of the court into the hinterlands and to protect the court from both inter-nal and external threats.


Access to abundant frontier land, followed by the scramble to clear new arable land out of forests and swamps, aff orded a modicum of bargaining le-verage to farmers who were willing to do this work. Noble families, whose land was exempt from some kinds of taxation, bid up the price of agricultural labor in their eff orts to claim new land for themselves. Farmers oft en chose to work as tenants on this tax- privileged land rather than to till taxable lands allotted to them by the central government. Meanwhile, provincial nobles and other elites increasingly commended their lands to military leaders, who could defend the land from predation by bandits and opportunistic neigh-bors. In the centuries that followed until the sixteenth century, the imperial court became overshadowed by military order provided by one group of war-riors or another. Periods of stability were punctuated dramatically by violent rivalries, until all of Japan— save a few mountain redoubts— became engaged in civil war from 1467.

Th e romantic image of the valiant and honorable medieval samurai keeping the peace is a myth. Among warriors, loyalty to their lord was least common when it was the most valued. Warriors fought alongside their lords when they thought they could win, but they oft en switched sides to join the victors rather than have their land confi scated and reallocated among the winners. Among the farmers whose land was ravaged and lives were destroyed, war was hell.

Territorial Consolidation

Farmers were inevitably drawn into wars among provincial warriors in one way or another. But by the mid- fi ft eenth century, when territorial control of Japan was divided into scores of local domains, two of the most innovative warlords, Oda Nobunaga (1534– 1582) and his general Toyotomi Hideyoshi (1536– 1598) who succeeded him, sharpened the division of labor between farm-ers and warriors (heino bunri) that had already begun to emerge under do-mainal rule. Farmers were to remain on their land to produce food and pay taxes, while only warriors (many who had previously been farmers, jizamurai) would fi ght in battles. Although taxes increased, so did agricultural productiv-ity and economic growth.

Making good on the promise to protect farmers gave these leaders an enor-mous advantage over their opponents. Leaving farmers on the farm, Oda No-bunaga and Toyotomi Hideyoshi created disciplined and skilled armies. Histo-ries of modern warfare herald Maurice of Nassau (1567– 1625) and Gustav Adolf of Sweden (1594– 1632) for building regimented and skilled armies, but Oda


Nobunaga and Toyotomi Hideyoshi were achieving similar success on the other side of the globe. Hideyoshi also carried out extensive land surveys to clarify available assets for taxation, and he dealt gently with former enemies to win their compliance. In the space of less than two de cades, Hideyoshi and Nobunaga reversed the centrifugal movement toward smaller po liti cal units and created signifi cant economies of scale. By the 1580s, they had managed to consolidate about half of Japan’s land mass under unitary rule. Although it remained for Tokugawa to build a co ali tion big enough to fi nish the job, Nobu-naga and Hideyoshi had broken the back of re sis tance to central military control.

Warrior- farmers in mountain hideaways and armed monks in monaster-ies in Kyoto, Kanagawa, and elsewhere held out with remarkable tenacity. For these fortress communities, some by virtue of geography and others by virtue of religiously motivated militarism, the protection that Oda Nobunaga and Toyotomi Hideyoshi aff orded came at a dear price: their autonomy. Th e result was a series of ferocious, village- razing battles in which the trained armies of samurai won. It was not a technologically foregone conclusion, however, for guns were available to both sides. Th e or gan i za tion al and numerical superi-ority of the conquering army was made possible by the taxes of millions of war- weary farmers.

War and State Building in Japan and Eu rope

Following the fall of the imperial court in the late twelft h century, Japan came to be dotted with castles of noble warriors in much the same way as was then happening in Eu rope. Farmers in the surrounding countryside provided la-bor, crops, or both in exchange for protection from invaders provided by the nobility’s cavalry. Common to both Japan and Eu rope was the rather small territorial size of feudal domains aft er the breakup of empires, given the dif-fi culty of protecting large tracts of open terrain with small bands of warriors on horse back. It was not until war once again became endemic, and farmers were willing to pay larger sums for their protection, that military leaders raised armies large enough to command expansive territorial control.

Changes in economies of territorial scale are, of course, also aff ected by fac-tors such as modes of economic production and military technology. But eco-nomics and technology, whether alone or together, leave unexplained substantial parts of the variation in scale economies. Economic theory might suggest, for example, that Eastern Eu rope was dominated by larger fi efs and more per sis tent


serfdom because economies of scale in grain production there are greater than for the crops cultivated in the hillier terrain of Western Eu rope. Th e fi xed costs of maintaining teams of oxen and other farm equipment would be too great for small landholders, making large manors more eco nom ical ly produc-tive. Th ere is no obvious reason, however, why farmers might not have worked out some cooperative arrangement to share expensive livestock and tools.

Another infl uential economic model of serfdom (and slavery) turns stan-dard bargaining logic on its head. Precisely because abundant land- to- labor ra-tios favored peasants in Eastern Eu rope compared to Western Eu rope, po liti cal regimes had to be more oppressive in order to extract economic surplus from peasants. Regimes that mustered coercive power displaced their more timid counterparts because they were able to extract greater eff ort and productivity from peasants. But this model does not explain the origins of coercive power.

Military technology is another possible explanation for changes in econo-mies to territorial size. Th e introduction of stirrups to Eu rope from somewhere in Asia in the late seventh century gave the cavalry an edge over amassed foot soldiers, ushering in an era of feudalism in which only nobles could aff ord the required horses and armor. Castles were easy to defend and hard to destroy, creating diseconomies of scale until the invention of heavy artillery in the mid- fi ft eenth century. With the development of cannon in 1449– 1450, Charles VII of France regained much of Normandy by knocking down 60 fortifi cations— each of which took the En glish a year to build— at a clip of more than one a week. Th e Turks destroyed Constantinople in 1453 with comparable dispatch. Well- regimented armies, equipped with heavy artillery, were now a match for the castles and cavalry of the nobility.

Changes in technology may indeed have shift ed the relative productivity between capital- and labor- intensive modes of warfare, with potential conse-quences for the bargaining leverage of those with capital or labor. While rich nobles thrived in the age of the cavalry, farmers stood to gain from the mili-tary value of foot soldiers before the stirrup and aft er the cannon. But the exposure of farmland to plundering armies meant that farmers were always exploited, even when they were mobilizationally useful. Th ey were too vul-nerable to make use of their relative scarcity as bargaining leverage.

Early modern Poland is instructive here because the inability of Polish peas-ants to assert their natural bargaining advantage conferred by the abundance of labor relative to land has always been something of a mystery to economists. Th e dominance of the Polish landed nobility makes more sense when one re-


members the serial invasions by Magyar and other horse back invaders, against whom the Polish cavalry was quite successful. Th e Polish cavalry’s battle record was so glorious that Poland neglected military innovation and was destroyed by Rus sian and Prus sian armies of foot soldiers in 1794 and 1797.

For farmers the emergence of larger territorial units was a double- edged sword. Although the larger government unit was able to raise more powerful armies and provide greater security, it also meant that farmers lost the benefi t of exit options among multiple po liti cal units. Th is was particularly true in Eastern Eu rope where there were fewer cities to provide absconding peasants with anonymity and alternative employment.

Consolidating territorial size is a function of raising enough revenue to pay for the inputs of war, a problem the economic and technology accounts fail to address. Th e remaining puzzle is identifying the source of this money. Th e medieval Japa nese experience, like the Eu ro pe an one, revolved around an implicit peasant demand for protection. Lords competed for the loyalty of lesser lords, but ultimately the whole edifi ce rested on resources, which were wrested from the land in some fashion. Lords had an incentive to extract from their own base and compete for the loyalty of neighboring farmers. Only where warfare was infrequent or where locals were confi dent of their ability to defend themselves did peasants resist taxes and territorial consolidation.

Territorial consolidation began in those fl at areas that were most vulnera-ble to military invasion and spread as the armies of those areas gained pre-eminence. In France, monarchical control began in the Ile de France and Normandy but was resisted longer in Langue d’Oc and Brittany, and only gained widespread ac cep tance in the wake of harrowing religious violence in the sixteenth century. Elsewhere in Eu rope, taxes were higher and armies were larger in the great fl atlands of Prus sia and Rus sia, in the pathway of steppeland hordes. Big armies can oft en conduct an eff ective defense, but there is the question of paying to feed, train, and equip them. Widespread fear of violence and the demand for protection gave birth to the modern nation- state with territorial control, fi rst in the form of the absolutist state.

Th is recognition of the importance of peasant demand for protection dif-fers from the Marxist suggestion that the landed nobility needed an absolutist state to fi x their status against the onslaught of the proto- industrial bourgeoi-sie. Nor were undefended “church lands” available for confi scation in Japan because monasteries tended to be armed to the hilt. In Japan, the unifi ers robbed their competitors of their lands militarily, whereupon they divided the


lands among their men. In any case, the practice of rewarding loyal warriors with land taken from the vanquished had occurred so many times during Ja-pan’s medieval period that by the mid- fi ft eenth century most of the elite were of shallow vintage. Not that this situation was unique to Japan; the Romans did the same thing following Sulla’s reforms and aft er Octavius’s victories. Th e “proscriptions” are remembered mostly for legal murders, but note that they also served to take lands away from rich opponents and confer them on allies.

In overwhelmingly rural Eastern Eu rope, Rus sian and Prus sian “absolut-ism” seems hardly to have been a response to a threatening rise of the middle class. Peasant fear of violence from marauders is a more consistent theme that runs through all of these cases. Th is is not to deny that peasants were oft en mis-erable, on the verge of starvation, and hardly able to pay heavier taxes for larger armies. But in their desperation, they chose among the available poisons. Th eir choices had signifi cant consequences for the kinds of states that would subse-quently emerge.

Conclusions

Widespread territorial vulnerability and fear of violence created the territo-rial state. Th is book tells that story as it unfolded in medieval and early mod-ern Japan, but it follows a general logic that holds in Eu rope and elsewhere.

Th ere is some irony to the way vulnerability paved the way for strong, hier-archical governments with extensive coercive powers over the subject popula-tion. At least in the short run, mobilization for war could have increased the bargaining leverage of the populace whose resources were needed for war. When foot soldiers are militarily valuable, peasants may profi tably refuse to fi ght unless the leader is willing to off er better terms of exchange. History gives us a number of examples of po liti cal concessions to peasant fi ghters, including fi ft h century b.c. Athens when Cleisthenes granted the masses full participa-tory rights in exchange for their help in ousting the Spartan- installed oligar-chy. In Republican Rome, fi ghting wars for Rome was the ticket to citizen-ship, fi rst for local residents and then for men of conquered lands as well. During the protracted Dutch Revolt against the Habsburg Empire (1568– 1648), ordinary citizens gained the right to participate in politics, even if the rights were substantially retracted aft er the war was won. In modern times, World War I ushered in female suff rage in most rich democracies, World War II launched the civil rights movement, and 18- year- olds gained the right to vote during the Vietnam War. Why does war bring po liti cal rights in one setting and an abdication to absolutist government in another?


Japa nese history provides a number of clues. As the example of the ninja shows, mountain villagers had little use for an absolutist ruler when they could live out their lives without a strong protector. In the lowlands, the bid-ding among nobles for peasant support did in fact raise many farmers through the ranks of warrior status to become lords in their own right. But armies consisted of small bands of cavalry, and competing nobles typically could not aff ord to put entire villages in arms. Compared to classical Greece, which was invaded by the ferocious Persian army, violence in Japan escalated only slowly. We can only speculate whether, had the Mongols landed in full force, the Japa nese people might also have won po liti cal concessions in exchange for emergency mobilization.

Th e piecemeal intensifi cation of violence in Japan worked against broader mobilizational concessions. Once widespread destruction reached an intolera-ble threshold, ordinary people were willing to pay for large armies and a leader strong enough to lead them. Territorial consolidation ended competition among aspiring generals, without which peasant bargaining was attenuated.

In the chapters that follow, historical experts sift through archival materi-als to provide a close look at the choices made by lords and peasants through-out the medieval period and across diff erent parts of Japan. In Chapter 2, Karl Friday reminds us that the samurai came into being from within the imperial system and served the system obediently for over 300 years. Although war-riors possessed a monopoly over the instruments of armed force, and indeed functioned as the court’s protectors and administrators in the provinces, the warrior families were too divided among themselves, even under the Kamak-ura shogunate, to be a viable alternative to court- based rule and legitimacy. When the court collapsed into civil war between the Northern and Southern Courts in 1333– 1335, the Ashikaga shogunate stepped into the breech. But the power of the Ashikaga shogunate rested on the acquiescence of local military lords who claimed growing powers for themselves at the expense of central-ized rule, setting the stage for eventual civil war.

Chapter 3 by Susumu Ike describes the range of relationships between lord and retainer during the decentralized years of medieval Japan, aft er the decline of the imperial authority and before unifi cation under Oda Nobunaga. Com-petition among military leaders may have increased the bargaining leverage of peasants who were needed to feed the soldiers or to join armies as soldiers. But navigating relationships among competing nobles was treacherous business because backing a loser could mean the loss of land and death. Th e civil war that broke out in 1437 among lords competing for territory forced local villagers to


make painful choices, sometimes in favor of retainers against their masters (gekokujo), when the retainer was thought to be better able to lead armies to victory.

In Chapter 4, Tsuguharu Inaba describes the misery of peasant life in war-time and explains the relief with which peasants greeted Oda Nobunaga’s pol-icy of dividing labor between farmers and warriors. Rather than being dis-tressed to have their swords confi scated, many farmers greeted with relief the new leader’s ambition to curb the fi ghting clans. Th roughout these years, Inaba emphasizes, farmers retained their local village councils in which they made their collective decisions to support the emerging centralized regime.

Chapter 5 by Carol Tsang recounts the tenacity with which some religious organizations, such as the Honganji Temple in Osaka, resisted Oda Nobunaga’s pell- mell push toward territorial consolidation. Th e Honganji Temple’s thou-sands of followers in villages across a sizable area in today’s Ishikawa prefec-ture created a structure of local self- rule and po liti cal order to match that provided by local military leaders. Th e villagers who joined in this religious league had little need for the protection off ered by secular lords because they were well armed and well disciplined. Th eir belief that fi delity to their cause would be awarded in the aft erlife empowered them to fi ght fearlessly. But in the end, this league and others like it throughout central Japan were steamrolled by Nobunaga’s and Hideyoshi’s even more powerful military machines. Lack-ing the economies of scale that the centralizers had developed, the most pas-sionately in de pen dent local re sis tance movements were swept aside.

In Chapter 6, Pierre Souyri writes of the farmer- warriors from the mountain valleys of Iga and Koga near Kyoto who tenaciously resisted Oda Nobunaga’s ambition for territorial consolidation. Had these valleys been clustered more closely together rather than spread out along the spine of Honshu, one wonders whether a ninja league might not have prevailed as their counterparts did in Switzerland, with vastly diff erent consequences for Japa nese po liti cal history. An alternative route to domestic tranquility, apart from the militarily imposed peace of an absolutist ruler, might have entailed the sorts of treaties that the Swiss Alpine regions concluded with the lowland cantons.

In Chapter 7, Th omas Conlan addresses the question of how the more suc-cessful military lords managed to consolidate territory. Conlan documents the or gan i za tion al prowess of military leaders of regional domains. Among the many military governors or shugo, the most successful became domain lords or daimyō. For their spectacular success in achieving territorial consoli-


dation, Conlon credits these generals’ methodical approach to gaining public ac cep tance, raising revenues, and building armies, rather than the guns with which they equipped their men.

In Japan, as in modern Eu rope, state building emerged out of the mayhem of warfare. But societal need for order is, by itself, no explanation for how a state capable of providing security materialized in either place. Th e variation within Japan over time and place suggests a mechanism for its emergence that was at work in Eu rope as well. Farmers in the pathway of armies became desperate for protection, even at the cost of their money and freedom. Although less vulner-able populations in hills or islands resisted territorial incorporation that would burden them with taxes to pay for the security of others, farmers on the fertile plains generated enough money and military might to break the re sis tance of these natural fortresses. In Japan, the entire archipelago became as one.

Notes1. “Ninja” is a term of modern origin used to describe rustic fi ghters in ages past

shrouded in myth and mystery. Following forcible po liti cal unifi cation in 1600, these re sis tance fi ghters were among the crack forces the government employed to do their dirty work, including vanquishing the Shimabara Rebellion of 1637 with a ruthless show of force. See, for example, Adachi 1967.

2. According to legend, which fi rst appears in texts dating to the sixteenth cen-tury, Tell shot the apple cleanly, sparing his son’s life and limb. When the Austrian magistrate asked what the second arrow in his quiver was for, Tell replied that he would have shot the Austrian magistrate in the event that his aim had faltered and he shot his son. In the heroic sequence that followed, Tell was thereupon arrested but es-caped and then did shoot the magistrate aft er all. All of this is said to have inspired the Swiss mountain villages to rise in full revolt.

3. Machiavelli’s Discourses built on a tradition of Italian Re nais sance theorizing on liberty during a time when liberty— from both foreign and domestic domination— was under threat (Wootton 2007). Th e En glish carried on the tradition in their re sis tance to monarchs who overstepped traditional boundaries. An authoritative source is Pocock 1975. Th e famous Swiss champion of self- governance, Jean Jacques Rousseau (1712– 1778), wrote much later, having the full advantage of earlier Italian and En glish theorizing.

4. Petrarch (1304– 1374) is said to have revived the study of Roman thought, mak-ing him one of the fathers of the Italian Re nais sance.

5. North and Th omas 1971.6. Murai 1993. We are grateful to an anonymous reviewer for this parallel.7. Adolphson 2001; Adolphson and Ramseyer 2009.


8. Adolphson 2001; Adolphson and Ramseyer 2009.9. We are indebted to an anonymous reviewer for this point.10. In Th e Leviathan (1660), written aft er a brutal civil war in En gland, Th omas

Hobbes urged citizens to invest their governments with strong powers. In Hobbes’s view, exchanging freedom for security was a good choice because life in the state of nature was in any case “solitary, poor, nasty, brutish, and short.”

11. It may be reasonable to consider that Japan’s communitarian culture had the ef-fect of hastening a switch to a new equilibrium in response to new circumstances. But which equilibrium was never a foregone conclusion.

12. Hobbes, Leviathan. Historians will object that farmers are rarely if ever ob-served to consciously make these sorts of deals. We are employing a social science methodology that attributes motivation based on observable patterns of evidence, and assumes only that people were limited in their choices, not that they chose freely or were even aware of alternative courses of action.

13. Jean Bodin, like Hobbes writing against the backdrop of civil strife, favored strong government capable of defending its citizens. Six Books of the Republic (Les Six livres de la République, 1576).

14. In the fourteenth and fi ft eenth centuries the French experienced the ravages of En glish armies, which were ordered to plunder the countryside and show no mercy to enemy and noncombatant alike. Th is brutal tactic, known as chevauche, had the dual advantage to the En glish of relieving the costs of feeding their own troops and depriving the French of supplying theirs (Sumption 2001: 272). Th e sixteenth century witnessed the vicious religious wars, in which Huguenots and Catholics fought with-out mercy for doctrinal control. In the seventeenth century, the Th irty Years’ War brought plundering armies into villages over large swaths of Germany and France.

15. Montesquieu, in eighteenth- century France, mused that En gland’s relative iso-lation may have given it some advantage in developing a theory and practice of limited government (Th e Spirit of the Laws).

16. Sometimes the term absolutism is used to describe the large territorial state, but the implication of absolute control, as many historians have pointed out, is mis-leading (see Henshall 1992; Hoff man and Rosenthal 1997).

17. Japa nese economic history in the 1920s and 1930s centered on a debate be-tween two Marxist variants: the koza- ha (associated with the Communist Party), who thought Japan had failed to achieve a bourgeois revolution by the twentieth century and required forced industrialization by the state; and the rono- ha (associated with the socialists), who thought that industrialization was proceeding on its own and that a socialist society could emerge without a communist revolution. Aft er World War II, left ist intellectuals dominated Japa nese universities because they escaped Occupation purges of the conservative right. Medieval Japa nese history was more theoretical than empirical and focused, as a matter of belief, on the landed elites’ ability to extract eco-


nomic surplus from unfree labor. Tom Scott (1998) documents peasant choices and resourcefulness in the Eu ro pe an context. See also Brown 1974.

18. Ge ne tic tests show that modern Japa nese inhabitants of the northernmost is-land of Hokkaido and the southernmost island of Okinawa share more DNA with each other than with Japa nese living on the central islands of Kyushu, Shikoku, and Honshu (see Imamura 1996).

19. Kiley 1999.20. Piggott 1997.21. Th e Japa nese “estate” was “a group of plots, oft en scattered, that were bound to-

gether under a common proprietor. Th e proprietor, who might be the head of a powerful local family, a member of the aristocracy or the imperial family, or a religious institu-tion, inherited the immunities created by the establishment of the estate and held most of the key powers over the land” (from Duus 1993: 31).

22. Asakawa called these “exempt manors” or “exempt shoen.” See Kambayashi and Hamada 2007; Duus 1993; Yamamura 1974.

23. Hurst 1990; Friday 1994. Berry (2005: 842) points out that Yamamoto Tsune-tomo, the author of the samurai classic, Hagakure, invented a code of abject loyalty dur-ing the Tokugawa peace when this loyalty would never be tested. “To speak of loyalty in these circumstances is deceptive silliness.”

24. Berry 1994.25. Totman 1993: 59; Farris 2006: 223. Farris points out that the warring states’

daimyō led the way in clearing land, fi xing riverbanks, and irrigating new fi elds.26. Roberts 1956; Parker 1988; Lynn 2003; Rogers 1995.27. Duus 1993: 76– 77.28. Conlan, this volume, establishes the insuffi ciency of the technology argument.29. Hideyoshi did fi nish the job of unifying all daimyō under himself, but he

failed to establish a mechanism that would allow him to pass his hegemony on to his heirs. Ieyasu was the fi rst to accomplish that. Berry (1986: 242) points out that the uni-fi ers did not establish a strong, interventionist state beyond what was necessary to carry out their mandate of imposing peace and security.

30. Th e movie Kagemusha depicts the Battle of Nagashino of 1575 as a tragic charge by Takeda Shingen’s cavalry into Nobunaga’s army of musketeers. Historians have recently shown that the Takeda side had as many guns as the Oda forces did. Th e real lesson of Nagashino is that a large force ensconced behind fi eld fortifi cations can defeat a small force attacking it. It was an advantage of defense over attack, given those technologies and strategic ideas. Th at is not to say that that a blitzkrieg could not have worked with the same technology.

31. Domar (1970: 13) credits Kliuchevsky (1937) for the argument that, when Rus-sian nobles competed for scarce labor in the sixteenth century, the government re-stricted the peasants’ freedom. When labor is scarce, it has higher marginal value than


land, and landlords have an incentive to control labor in order to expropriate the full value of labor’s marginal product. Th is has become the standard explanation for estab-lishing oppressive colonial regimes in Latin America. See Martins 1982; Acemoglu and Robinson 2006.

32. White 1962; Rogowski and MacRae 2004.33. Bean 1973: 207.34. Th e stirrup and cannon are meant here as shorthand for a host of reasons why

mass armies may or may not be eff ective. In island territories such as Greece, heavily armored hoplites were no match for light foot soldiers in skillfully maneuvered boats. See, for example, Strauss 2004. In mountainous terrain, horses are never a match for local fi ghters who know how to use terrain in their favor.

35. Frost 2000.36. Eastern Eu ro pe an nobles, on whom the Holy Roman Emperor depended for

protection from the Magyars and others, secured tax concessions for manufacturing operations such as breweries undertaken on their lands. In Poland, once thriving towns along the trade routes and rivers feeding the Baltic Sea fell into decline with the collapse of the Polish- Lithuanian monarchy in the sixteenth century.

37. Richard Bean (1973) notes the importance of what he calls the “administrative technique” of raising taxes, but this only asks the question in another form.

38. Sam Cohn (1999) found that mountain villages in areas surrounding Florence successfully negotiated lower taxes than lowland villages. Th is was thanks to the natu-ral fortifi cation provided by terrain rather than on account of being in a borderland area where the villages could pit competing overlords in a bidding war with each other. Note that lowland borderlands, such as those between France and Holland or France and Germany, paid higher taxes than areas at some distance from military thorough-fares. See Henneman 1976.

39. Anderson 1974.40. As in Eu rope, “there was a per sis tent moral threat insisting that the lord

should grant fi efs and that his men have the right to look elsewhere if he disappoints them.” See Bartlett 1993: 46.

41. To eliminate his enemies and to restore funds to the depleted Roman trea sury, Sulla in 82 b.c. posted the names of men he declared to be “enemies of the state.” Bounty hunters received a reward for murder, and the state got the confi scated prop-erty. Th e Triumvirate repeated these state- sanctioned murders in 43 b.c.

42. Cicero was killed in a.d. 43 in a proscription run by the triumvirate of Octavius, Mark Antony, and Lepidus.

43. Samons 1998.44. Lintott 1999; Cornell 1995.45. Israel 1995.46. Keyssar 2001.


ReferencesAcemoglu, Daron, and James Robinson. 2006. Th e Economic Origins of Dictatorship

and Democracy. New York: Cambridge University Press.Adachi, Ken’ichi. 1967. Taishu geijutsu no fukuryu. Tokyo: Rironsha.Adolphson, Mikael. 2001. Gates of Power: Monks, Courtiers, and Warriors in Premod-

ern Japan. Honolulu: Hawaii University Press.Adolphson, Mikael, and J. Mark Ramseyer. 2009. “Th e Competitive Enforcement of

Property Rights in Medieval Japan: Th e Role of Temples and Monasteries.” Jour-nal of Economic Behavior and Or ga ni za tion, 71, 3: 660– 668.

Anderson, Perry. 1974. Lineages of the Absolutist State. Atlantic Highlands, NJ: Humani-ties Press.

Bartlett, Robert. 1993. Th e Making of Eu rope: Conquest, Colonization, and Cultural Change, 950– 1350. Prince ton, NJ: Prince ton University Press.

Bean, Richard. 1973. “War and the Birth of the Nation State.” Journal of Economic His-tory, 33, 1: 203– 221.

Berry, Mary Elizabeth. 1982. Hideyoshi. Cambridge, MA: Harvard University Press.Berry, Mary Elizabeth. 1986. “Public Peace and Private Attachment: Th e Goals and Con-

duct of Power in Early Modern Japan.” Journal of Japa nese Studies, 12, 2: 237– 271.Berry, Mary Elizabeth. 1994. Th e Culture of Civil War in Kyoto. Berkeley: University of

California Press.Berry, Mary Elizabeth. 2005. “Presidential Address— Samurai Trouble: Th oughts on

War and Loyalty.” Journal of Asian Studies, 64, 4: 831– 842.Birt, Michael P. 1985. “Samurai in Passage: Transformation of the 16th Century Kanto.”

Journal of Japa nese Studies, 11, 2: 369– 400.Blum, Jerome. 1957. “Th e Rise of Serfdom in Eastern Eu rope.” American Historical

Review, 62, 4: 807– 836.Bodin, Jean. 1992/1583. On Sovereignty: Six Books on the Commonwealth (Les Six livres

de la République). Edited by J. H. Franklin. New York: Cambridge University Press.Braudel, Fernand. 1986. Th e Perspective of the World: Civilization and Capitalism. New

York: Harper & Row.Brown, Delmer M. 1993. “Th e Yamato Kingdom.” In Ancient Japan, volume 1 of Th e

Cambridge History of Japan, edited by Delmer M. Brown. New York: Cambridge University Press, pp. 108– 162.

Brown, Elizabeth R. 1974. “Th e Tyranny of a Construct: Feudalism and Historians of Medieval Eu rope.” American Historical Review, 79: 1068– 1088.

Cohn, Sam. 1999. Creating the Florentine State: Peasants and Rebellion, 1348– 1434. Cambridge: Cambridge University Press.

Cornell, Timothy. 1995. Th e Beginnings of Rome: Italy and Rome from the Bronze Age to the Punic Wars. London: Routledge.


Dewald, Jonathan, and Liana Vardi. 1998. “Th e Peasantries of France, 1400– 1789.” In Tom Scott, ed., Th e Peasantries of Eu rope. New York: Longman Press.

Domar, Evsey. 1970. “Th e Causes of Slavery or Serfdom: A Hypothesis.” Journal of Economic History, 30, 1: 18– 32.

Duus, Peter. 1993. Feudalism in Japan, 3rd ed. New York: McGraw- Hill.Fariss, Wayne. 2006. Japan’s Medieval Population: Famine, Fertility, and Warfare in a

Transformative Age. Honolulu: University of Hawaii Press.Friday, Karl. 1992. Hired Swords: Th e Rise of Private Warrior Power in Japan. Stanford,

CA: Stanford University Press.Friday, Karl. 1994. “Fushido or Bull? A Medieval Historian’s Perspective on the Imperial

Army and the Japa nese Warrior Tradition.” History Teacher, 27, 3: 339– 349.Friday, Karl. 2003. Samurai, Warfare, and the State in Early Medieval Japan. London:

Routledge.Frost, Robert. 2000. Th e Northern Wars: War, State and Society in Northeastern Eu-

rope, 1558– 1721. London: Longmans Harlow.Henneman, John Bell. 1976. Royal Taxation in Fourteenth Century France: Th e Cap-

tivity and Ransom of John II, 1356– 1370. Philadelphia: American Philosophical Society.

Henshall, Nicholas. 1992. Th e Myth of Absolutism: Change and Continuity in Early Modern Eu ro pe an Monarchy. New York: Longman.

Hildinger, Erik. 1997. Warriors of the Steppe: A Military History of Central Asia, 500 b.c. to 1700 a.d. Cambridge, MA: Da Capo Press.

Hobbes, Th omas. Leviathan. 1651/1968. London: Penguin Classics.Hoff man, Philip, and Jean- Laurent Rosenthal. 1997. “Th e Po liti cal Economy of War-

fare and Taxation in Early Modern Eu rope.” In John Drobak and John Nye, eds., Th e Frontiers of New Institutional Economics. New York: Academic Press.

Hurst, G. Cameron, III. 1990. “Death, Honor, and Loyalty: Th e Bushido Ideal.” Phi-losophy East and West, 40, 4: 511– 527.

Imamura, Keiji. 1996. Prehistoric Japan: New Perspectives on Insular East Asia. Hono-lulu: University of Hawaii Press.

Israel, Jonathan. 1995. Th e Dutch Republic: Its Rise, Greatness, and Fall, 1477– 1806. New York: Oxford University Press.

Kambayashi, Ryo, and Koichi Hamada. 2007. “Kan’ichi Asakawa as an Economic His-torian, or a Historian of Medieval Japan.” Paper presented at “Japan and the World: A Conference in Honor of the Memory and Legacy of Asakawa Kan’ichi,” Yale University, March 9– 10.

Keyssar, Alexander. 2001. Th e Right to Vote: Th e Contested History of Democracy in the United States. New York: Basic Books.

Kiley, Cornelius J. 1999. “Provincial Administration and Land Tenure in Early Heian.” In Heian Japan, volume 2 of Th e Cambridge History of Japan, edited by Donald H.


Shively and William H. McCullough. New York: Cambridge University Press, 1999: 236– 340.

Kliuchevsky, Vladimir. 1937. Kurs russkoi istorii. Moscow: Gosudarstvennoe sotsail’no- ekonomicheskoe izdatel’stvo. Translated by C. J. Hogarth as A History of Rus sia (New York: Russell and Russell, 1960).

Levi, Margaret. 1997. Consent, Dissent, and Patriotism. New York: Cambridge Univer-sity Press.

Lintott, Andrew. 1999. Th e Constitution of the Roman Republic. New York: Oxford University Press.

Locke, John. 1690/2003. Two Treatises of Government and A Letter Concerning Tolera-tion, edited by Ian Shapiro. New Haven, CT: Yale University Press.

Lynn, John. 2003. Battle: A History of Combat and Culture. Boulder, CO: Westview Press.

Mass, Jeff rey. 1993. Yoritomo and the Founding of the First Bakufu. Stanford, CA: Stan-ford University Press.

Machiavelli, Niccolo. 1970/1517. Th e Discourses. London: Penguin Classics.Martins, Roberto Borges. 1982. “Growing in Silence: Th e Slave Economy of Nineteenth

Century Minas Gerais, Brazil.” Journal of Economic History, 42, 1: 222– 223.Millward, Robert. 1982. “An Economic Analysis of the Or ga ni za tion of Serfdom in

Eastern Eu rope.” Journal of Economic History, 42, 3: 513– 548.Montesquieu. 1949. Spirit of the Laws. New York: Hafner.Murai, Shosuke. 1993. Chusei wajin den [Medieval Pirates]. Tokyo: Iwanami shoten.North, Douglass, and Robert Th omas. 1971. “Th e Rise and Fall of the Manorial Sys-

tem: A Th eoretical Model.” Journal of Economic History, 31, 4: 777– 803.Parker, Geoff rey. 1988. Th e Military Revolution: Military Innovation and the Rise of the

West, 1500– 1800. New York: Cambridge University Press.Piggott, Joan. 1997. Th e Emergence of Japa nese Kingship. Stanford, CA: Stanford Uni-

versity Press.Pocock, J. G. A. 1975. Th e Machiavellian Moment: Florentine Po liti cal Th ought and the

Atlantic Republican Tradition. Prince ton, NJ: Prince ton University Press.Roberts, Michael. 1956. Th e Military Revolution. Belfast: Queens University Press.Rogers, Cliff ord, ed. 1995. Th e Military Revolution Debate: Readings on the Military

Transformation of Early Modern Eu rope. Boulder, CO: Westview Press.Rogowski, Ronald, and Duncan MacRae. 2004. “In e qual ity and Institutions: What

Th eory, History, and (Some) Data Tell Us.” Manuscript, UCLA Po liti cal Science Department.

Samons, Loren J., ed. 1998. Athenian Democracy and Imperialism. New York: Hough-ton Miffl in.

Scott, James. 2009. Th e Art of Not Being Governed: An Anarchist History of Upland Southeast Asia. New Haven, CT: Yale University Press.


Scott, Tom, ed. 1998. Th e Peasantries of Eu rope from the Fourteenth to the Eigh teenth Centuries. New York: Longman Press.

Strauss, Barry. 2004. Th e Battle of Salamis: Th e Naval Encounter Th at Saved Greece— And Western Civilization. New York: Simon and Schuster.

Sumption, Jonathan. 2001. Th e Hundred Years’ War: Trial by Fire. Philadelphia: Uni-versity of Pennsylvania Press.

Tigar, Michael, and Madeleine Levy. 1977. Law and the Rise of Capitalism. New York: Monthly Review Press.

Totman, Conrad. 1993. Early Modern Japan. Berkeley: University of California Press.White, Lynn. 1962. Medieval Technology and Social Change. Oxford: Clarendon Press.Wootton, David. 2007. “Th e True Origins of Republicanism or de vera respublic.” In

Manuala Albertone, ed., Il repubblicanesimo moderno: L’idea di repubblica nella refl essione storica di Franco Venturi. Naples: Bibliopolis.

Yamamura, Kozo. 1974. “Th e Decline of the Ritsuryo System: Hypotheses on Eco-nomic and Institutional Change.” Journal of Japa nese Studies, 1, 1: 3– 37.

21

MEDIEVAL JAPAN WAS TRULY, as Souyri recently styled it, a “world turned upside down.” Th e prevailing forces of the preceding epoch, the Nara (710– 794) and Heian (794– 1185) periods, had been centripetal, and the realm was governed po liti cally, eco nom ical ly, and socially by civil authority and the court nobility. But the medieval age was a centrifuge that spun power, wealth, and eminence away from the capital and into the hands of warriors in the countryside.

Over the past four de cades historians have radically and dramatically re-vised their thinking on how and when the one epoch gave way to the other. It was once common to cite the creation of the fi rst shogunate in the 1180s as the end point of the classical era— equating the birth of this warrior- led institution with the advent of a “feudal” state ruled by warriors. Now, however, scholars have come to view the Kamakura period— the late twelft h, thirteenth, and early four-teenth centuries— as a time of transition between the classical and medieval ages. Th e establishment of the Kamakura regime is now held to signal not an end to the ascendancy of the civil court nobility, but merely the beginning of its decline.

“Th e rise of the samurai” was less a matter of dramatic revolution than one of incremental evolution, occurring in fi ts and starts. Japan’s famous warrior order came into being to serve the imperial court and the noble houses that comprised it as hired swords and contract bows— just one of many by- products of the broad trend toward the privatization of government functions and dele-gation of administrative responsibility that distinguished the Heian polity from its Nara pre de ces sor. Its roots sprang from shift s in imperial court mili-tary policy that began in the middle de cades of the eighth century and picked up momentum in the ninth.

They Were Soldiers OnceThe Early Samurai and the Imperial Court

Karl Friday

2

22 Karl Friday

The Ritsuryo State and the Emperor’s Army

At the dawn of the seventh century, most of Japan had been bound loosely to-gether by a confederation of regional chieft ains, among which one— the royal, or Yamato, house— stood as fi rst among equals. Some of the other houses were entirely dependent on the Yamato for their positions, but the majority had their own geographic bases of power, within which they were largely autono-mous. In theory, these provincial hegemons drew their titles from the court, but in practice, their positions were permanent, hereditary, and only nomi-nally related to the king’s authority. In fact, the principal role of the royal court— and of the countrywide polity— was little more than to serve as a ve-hicle for cooperation among the great houses in matters of “national” con-cern. All of this changed rapidly and dramatically during the seventh century, as this polity gave way to a centralized imperial regime.

Th e changeover accelerated aft er the sixth month of 645, when radicals led by the future Emperor Tenji seized power by hacking their po liti cal opponents to pieces with swords and spears, in the midst of a court ceremony. In the wake of this spectacular coup d’état, Tenji and his supporters introduced a series of centralizing mea sures collectively known as the Taika Reforms, aft er the cal-endar era in which the fi rst were launched. Over the next several de cades, the great regional powers were stripped of their in de pen dent bases and converted to true offi cials of the state, while the Yamato sovereigns were restyled in the image of Chinese emperors, as transcendent repositories of all po liti cal au-thority. Th e result was what historians have come to refer to as the imperial, or ritsuryō (aft er the administrative and penal codes that formed its framework), state.

Th e reformers prevailed through cajolery, cooptation, and coercion, aided in no small mea sure by widespread apprehension over the growing might of Tang China, which had been engaged since the early 600s in one of the greatest military expansions in Chinese history. Specters of Tang invasion fl eets loom-ing over the horizon muted opposition to losses of local or hereditary privilege and promoted support for state- strengthening reforms, as central and provin-cial noble houses set aside their diff erences in the face of a perceived common enemy. For it was obvious to all concerned that the Yamato military or ga ni za-tion was far from equal to the task of fending off the Tang.

It is scarcely surprising, therefore, that the centralization and restructuring of the military constituted a major element of the state reformation pro cess.

They Were Soldiers Once 23

“National armies” of the Yamato state had been cobbled together from forces raised in de pen dently by provincial chieft ains, who then led them into battle under the banner of the Yamato sovereign. Recruitment, training, or ga ni za-tion, and mobilization varied from place to place and from confl ict to confl ict. So did command, which was eclectic, and oft en divided among multiple “Supreme Commanders.”

By the close of the seventh century, the whole of the state’s martial resources— weapons, auxiliary equipment, horses, troops, and offi cers— had been sub-sumed under the direct control of the newly emergent emperor and his court. To be sure, the ritsuryō military system is distinguished from its pre de ces sor more in terms of principle and formal structure than function or personnel. But there was a crucial diff erence, for henceforth centrally appointed offi cers and offi cials oversaw all military units and activities, and direct conscription by the imperial court replaced enlistment of troops through regional chieft ains.

Under the new system, all free male subjects between the ages of 20 and 59, other than rank- holding nobles and individuals who “suff ered from long- term illness or were otherwise unfi t for military duty,” were liable for induction as soldiers, or heishi. Conscripts were enrolled in provincial regiments (gundan), which were structured as militia units, akin to modern national guards. Once assigned and registered as soldiers, most men returned to their homes and fi elds. Provincial governors maintained copies of regimental rosters, which they used as master lists from which to select troops for training; for peacetime po-lice, guard, and frontier garrison duties; and for ser vice in war time armies.

Draftees and Horse men

Th e model for the ritsuryō military system had been Tang China. Contrary to the images that still dominate many pop u lar histories, however, the new institutions— like the rest of the imperial state structure— were not simply a dopted wholesale. Th e architects of the imperial state carefully adapted Chinese practices to meet Japa nese needs and circumstances. At the same time, the system they designed was all too oft en the product of confl icting priorities, and accordingly, incorporated more than a few rather unhappy compromises. Moreover, the original foibles of the system were exacerbated by changing con-ditions: by the mid- eighth century the needs and priorities of the Japa nese state diff ered considerably from those of the late seventh.

One of the diffi culties the government faced was enforcing its conscription laws. Under the ritsuryō polity, military conscription was simply one component

24 Karl Friday

of the state’s tax requirements; induction rosters were compiled from the same population registers that were used to levy all other forms of tax. For this reason, peasant eff orts to evade any of these taxes also placed them beyond the reach of the conscription authorities. Far more important than the reluctance of peasants to serve in the military, however, were the fundamental tactical limitations of the ritsuryō armies.

Like their Tang archetypes, the regiments that formed the backbone of Japa nese imperial armies were mixed weapons- system forces: predominantly infantry, but augmented by heavily armored archers on horse back. Th is infantry- heavy balance was the product of both design and necessity.

Th e ritsuryō military had been contrived in the face of two perceived threats: a Chinese invasion and regional insurrections led by the old provincial chief-tains. Th e architects of the system seized on large- scale, direct mobilization of the peasantry as a key part of the answer to both dangers. Th e apparatus they created enabled the court to corner the market on military manpower and to create loyalist armies of daunting volume, thereby eff ectively closing the door on military challenges to imperial power or authority. An army of imposing numbers was also precisely what would have been needed to fend off a foreign invasion, while the militia structure made it possible for a tiny country like Ja-pan to muster large- scale fi ghting forces when necessary, without bankrupting its economic and agricultural base, as a large standing army would have.

Nevertheless, the court had opted for size at the expense of the elite tech-nology of the age, constructing a force composed primarily of infantry, while the premier military technology of the day was mounted archery. Th e state did try to maintain as large a cavalry force as it could, but eff orts to that end ran afoul of major logistical diffi culties. Foremost among these was the simple truth that fi ghting from horse back, particularly with bows and arrows, de-mands complex skills that require years of training and practice to master. It was just not practical to attempt to develop fi rst- rate cavalrymen from short- term peasant conscripts. Th e court addressed this problem through the straight-forward expedient of staffi ng its cavalry units only with men who had acquired basic competence at mounted archery on their own, prior to induction.

Th is policy had far- reaching consequences for the shape of military things to come in Japan. For if the prerequisite to becoming a cavalryman was skill with bow and horse, cavalrymen could come only from families that kept horses, a practice that did not spread beyond the nobility and the very top tiers of the peasantry until the tenth century or later. Th us only a small portion of


the imperial armies could be cavalry, and that cavalry was composed solely of the scions of elite elements of society.

None of this mattered a great deal initially: the ritsuryō military structure was more than adequate to the tasks for which it was designed. By the middle de cades of the eighth century, however, the po liti cal climate— domestic and foreign— had changed enough to render the provincial regiments anachronis-tic and superfl uous in most of the country. Th e Chinese invasion the Japa nese had so feared simply never materialized. What ever real threat there might have been was gone by the late 670s, when the kingdom of Silla forced the Tang off the Korean peninsula and checked its eastward expansion. Th e likeli-hood of violent challenges to the central polity from the regional nobility had also dwindled rapidly, as former provincial chieft ains came to accept the im-perial state structure as the arena in which they would compete for power and infl uence.

Th e passing of these crises all but ended the need to fi eld large armies and prompted the court to begin restructuring its armed forces. In the frontiers— particularly the north, where the state was pursuing an aggressive war of occupation— large infantry units still served a useful function. But the mar-tial needs of the interior provinces— the vast majority of the country— quickly pared down to the capture of criminals and similar policing functions. Un-wieldy infantry units based on provincial regiments were neither necessary nor well- suited to this type of work; small, highly mobile squads that could be assembled with a minimum of delay and sent out to pursue raiding bandits were far more appropriate to the task at hand. In the meantime, diminishing military need for the regiments encouraged offi cers and provincial offi cials to misuse the conscripts who manned them; they were borrowed, for example, for free labor on their personal homes and properties.

Th e court responded to these challenges with a series of adjustments and general reforms. Th e pattern of the reforms, which began as early as the 730s, indicates that the government had concluded that it was more effi cient to rely on privately trained and equipped elites than to continue to attempt to draft and drill the general population. Accordingly, troops mustered from the ordi-nary peasantry played smaller and smaller roles in state military planning, while the role of elites expanded steadily. Th e provincial regiments were fi rst supplemented by new types of forces and then, in 792, eliminated entirely in all but a handful of provinces. In their place the court created a series of new military posts and titles that legitimized the use of personal martial resources

26 Karl Friday

on behalf of the state. In essence, the court outsourced military and police ser vices, shift ing from a conscripted, publicly trained military to a contract force composed of professional warriors.

Capital and Countryside

Th e institutions and procedures of the early imperial state had their roots in an intriguing blend of idealism, wishful thinking, pragmatism, and compro-mise. Few survived wholly intact for more than a few de cades, although the spirit of the system— particularly the principle of centralized authority cul-minating in the emperor— remained strong into the fourteenth century and beyond.

Th e ritsuryō provisions for governing the countryside had been calculated, above all, to establish centralized control and to produce revenue for the court. Th ey cast the whole country as a giant manor on which the farming population served as peasant tenants and over which the court— the imperial house, the major religious institutions, and the great noble houses— functioned collec-tively as lord.

Th is basic conception of the proper order of things changed little, but the court’s priorities did shift , particularly from the late eighth century onward, evolving from a strident assertion of direct central oversight to an emphasis on maintaining centralized authority while delegating responsibility for many of the workaday functions of government.

As a result, landholding, tax collection, and social structure in the prov-inces came to bear an ever- paler resemblance to the letter of the ritsuryō law. A new gentry emerged in the countryside, absorbing a burgeoning share of both the revenues produced and the responsibilities of governance there. Nev-ertheless, the medieval world of nearly- autonomous barons contesting among themselves with minimal regard for the wishes or prerogatives of the court was still very far off . Indeed, the ties binding capital to countryside grew stron-ger, not thinner, during the Heian period.

Th is odd mixture of centrifugal and centripetal development transpired because the provincial gentry and the court nobility embraced fundamentally diff erent notions of what constituted wealth and power. Th e Kyoto aristocracy viewed wealth in terms of rice and other products collected, and power as a function of administrative oversight. Th e provincials, on the other hand, saw both in terms of hands- on control over productive resources— lands and peo-ple. As long as they were allowed a share of the profi ts from the land, the gentry


were content to acquiesce to a central authority and to focus not on ultimate own ership but on actual control— management of the agricultural pro cess, col-lection of taxes and rents, and other domainal sorts of authority.

One shortcoming of the ritsuryō landholding and tax structure was that it eff ectively presumed a countryside populated by a largely undiff erentiated mass of peasant house holds. Th e peasants were to be assigned— or leased— fi elds to cultivate, from which they would draw their livelihoods and in ex-change for which they would pay taxes and rents to sustain the emperor, the court, and local offi cials. But the theory behind this system failed to account for diff erences in individual ability: Some peasants proved to be better— or simply luckier— farmers than others. Some were able to make a profi t year aft er year; others found themselves incapable of making ends meet, let alone meet-ing their tax obligations.

Th us some peasants grew wealthy, while others were forced to become de-pendents of neighbors, who increasingly or ga nized the populations around their homes under their own control. A gap was developing between the struc-ture of provincial society as envisioned by the architects of the imperial state and reality in the countryside. As the gap widened, tax collection through the procedures set down in the ritsuryō codes became increasingly diffi cult.

In a remarkably pragmatic eff ort to reconcile old premises to new realities, the court responded with an updated paradigm: Henceforth, tax collection would be a problem between the central and provincial governments, rather than one between the court and individual subjects. Revenue quotas were set province by province, and governors were made accountable for seeing that they were met— making up shortfalls out of their own pockets, if necessary. Th e means by which the taxes were actually collected were left largely to the discretion of the governors, who in turn delegated most of the burden to local elites charged with assembling what ever revenues were deemed appropriate from the specifi c locales in which they had infl uence.

Governors and local managers alike welcomed such policy mea sures as opportunities for increasing their personal wealth and power. In a matter of de cades, the new tax structure turned everyone involved— except peasant cultivators— into tax farmers collecting revenues beyond their assigned quo-tas and pocketing the surplus.

Even as the state was redefi ning the nature of this fundamental aspect of the relationship between its central and provincial organs of government, the aristocracy and religious institutions in the capital were also working to create

28 Karl Friday

and expand private sources of revenue and infl uence outside the offi cial gov-ernment structure. Such eff orts became a per sis tent source of trouble for pro-vincial residents and offi cials from the middle de cades of the ninth century onward.

Council of State edicts from this period railed against “temples, shrines, princes and offi cials behaving like peasants, contesting over peasant paddy fi elds, stealing moveable property, ignoring governors and district magis-trates, and using their prestige and infl uence” to intimidate provincial resi-dents. “Agents of temples, shrines, government offi cials, and great houses,” we are told, were “confi scating boats, carts, horses and men by force,” “robbing tax shipments,” and “causing much suff ering for the people.”

Provincial governors, for their part, demonstrated a less- than- noble incli-nation to regard their provinces chiefl y as revenue- producing resources, and became increasingly indiff erent to the needs and welfare of residents. Problems with gubernatorial abuses of power began to appear by the mid- 800s, and by the late tenth century, the court was receiving a steady stream of petitions de-manding the impeachment of rapacious governors. Th e most famous of these, fi led in 988 by the “district offi cials and taxpayers” of Owari province, accuses the governor of tax fraud, extortion, nepotism, murder, and a host of other greater and lesser crimes. He was said to have collected nearly three times the prescribed amount of taxes, oft en beginning his collection eff orts as much as four months ahead of the proper schedule and using agents who stole even the furnishings from the lodgings provided for them as they made their rounds. His relatives and followers, it was charged, were allowed to seize land through-out the province and operate it as private, tax- exempt holdings. Th ey also “commandeered” horses and cattle from provincial residents, and then sold them back several days later for three to fi ve times their fair price.

Provincial elites, too, were fl exing their po liti cal and economic muscles. One of the defi ning characteristics of the imperial state was the monopoliza-tion of power by the nobility of the capital. Regional nobles elsewhere, who had once been the peers of this group, were— in principal— relegated to the po liti-cal and social back benches. In practice, however, the old regional chieft ains never fully surrendered their wealth or their familial authority in the agricul-tural villages. By the early Heian period, their descendants and other rural elites were challenging gubernatorial power in myriad subtle— and many not so subtle— ways. Most of these challenges took place within the bounds of the law. More overt re sis tance, when it did occur, was generally small scale, but it


did sometimes involve violence. From the turn of the ninth century onward, the court issued a steady stream of complaints about armed marauders “burn-ing people’s homes and using weapons to rob,” “striding about the villages opposing government offi cials and intimidating the poor,” “harming public morals,” and interfering with tax shipments— stealing not only the tax goods themselves but the horses and boats used to transport them.

Life in the provinces during the mid- Heian period, then, was increasingly dominated by a competition for wealth and infl uence among multiple groups, including provincial- resident elites, provincial offi cials, and the “tem-ples shrines, princes and offi cials” of the court. At the axis of this competi-tion were the middle- ranked court nobles whose careers centered on appoint-ments to provincial government offi ces. Such career provincial offi cials (zuryō) forged alliances with the loft y aristocrats (kugyō) above them. At the same time, many found that they could use the power and perquisites of their of-fi ces, and the strength of their court connections, to establish landed bases in their provinces of appointment and to continue to exploit the resources of these provinces even aft er their terms offi ce expired.

Heian society was rigidly stratifi ed, characterized by functionally un-bridgeable gulfs of station separating the top tier of court aristocrats, the lower and middle- level nobles who served as provincial governors, and the residents of the countryside. A complex system of court ranks, created by the ritsuryō codes and similar in purpose to modern civil ser vice rank systems, classifi ed both the central and the provincial nobility. A man’s rank determined his sta-tus, his eligibility for government posts, and his place in the complex protocols for offi cial and social events. Each stratum in this hierarchy had access to spe-cifi c types of government posts, rights over land, and forms of income.

More intriguingly, the rights and privileges of each stratum were sealed from below as well as from above, so that high- ranked courtiers were as eff ectively barred from provincial or local posts as provincial elites were from becoming top court offi cials. Th is arrangement generated intense competition and rivalry among peers, including siblings, who contested in circumscribed and ever- more- crowded arenas for the same baskets of fruit. But it also formed a basis for coop-eration between members of diff erent strata because neither party could chal-lenge the prerogatives of the other, and each member of the alliance could aid the others in obtaining rewards for which he was himself ineligible.

Over time, connections between hereditary status and offi ce- holding be-came progressively deeper and more fi rmly entrenched, and eligibility for

30 Karl Friday

posts became limited to smaller and smaller numbers of houses. As the pros-pect that descendants of par tic u lar families would hold the same posts gen-eration aft er generation became increasingly predictable, many offi ces— and the tasks assigned them— became closely associated with certain houses, and key government functions came to be performed through personal, rather than formal public, channels, rendering “public” and “private” rights and re-sponsibilities more diffi cult to distinguish.

Court society and the operations of government came to be dominated by a handful of high- ranked courtiers and their families, the most powerful of which was the Fujiwara sekkanke or Regents’ House, whose heads served as hereditary regents or chancellors to emperors. Men of the provincial governor class and other low- and middle- ranked nobles would attach themselves to Fujiwara re-gents or other se nior courtiers, who would become their patrons and sponsor their advancement at court. In exchange, the clients served on the patron’s house hold staff and vouchsafed his interests in the course of their offi cial duties in the posts he obtained for them. Th is patron- client relationship was rooted in mutual need and worked to the advantage of both parties: It permitted top courtiers to exploit the administrative talent and the private wealth of provin-cial governors in the administration of their private aff airs, and it assured lower- ranking nobles of continued appointment to lucrative posts.

Alliances were also developing between se nior court fi gures and provincial elites, as the elites learned to use their connections— real or pretended— with the court fi gures to gain greater autonomy from the provincial government. By the late 800s, court edicts complained that the great houses and religious insti-tutions of the capital were “by- passing the governor and issuing private house edicts directly to district magistrates and functionaries within the provincial government” as well as “sending agents, who led followers through the prov-inces” disrupting peasant house holds and government business. Th e edicts also indicated that “natives and drift ers claiming to be house men of princes and government offi cials did not fear the authority of the governor and did not obey the injunctions of district magistrates.” Th e “natives and drift ers” cited in these documents included a new group of provincial residents.

For members of the career provincial governor class, the heightened com-petition, factionalism, and atmosphere of hereditary prerogative that charac-terized court politics during the ninth and tenth centuries meant diminishing prospects in the capital for themselves and their children. At the same time, many were fi nding that they could use the power and perquisites of their


offi ces, and the strength of their court connections, to establish landed bases in their provinces of appointment and to continue to exploit the resources of these provinces even aft er their terms of offi ce expired.

As former offi cials settled in the countryside, they quickly became powers to contend with, competing with subsequent governors, district magistrates, and other, older gentry houses for resources and infl uence. Among other things, the newcomers seem to have been continuing to collect taxes— especially back taxes— just as if they were still offi cials. Ironically then, the settlers, driven out of the capital by waning opportunities there, were creating similar personal networks in the provinces and through them establishing themselves at the top of provincial society, thereby establishing a new, provincial- level system of hierarchical alliances.

Th e central government initially saw the emigration of ranked nobles to the countryside as a threat to provincial order and attempted to prohibit it. Court documents from the late eighth to the late ninth century railed against “offi cials who have fi nished their terms of offi ce, and sons and younger broth-ers of princes and court offi cials,” who were, “settling down in their [former] areas of jurisdiction, where they hindered agriculture, gathered up the peas-antry like fi sh, and constructed plans for their own evil gains.”

By the mid- tenth century, however, the court had given up this eff ort and was instead coming to terms with the immigrants. Former provincial offi cials and their children began to gain appointments to assistant governorships and other provincial offi ces.

Nevertheless, few of the nobles who “settled down” in the countryside ac-tually abandoned life in the capital for a provincial existence. Some individu-als and branches of families became more thoroughly committed to rural life than others, but most were still careful to maintain their ties to the capital. Th ey could not aff ord to do otherwise, for to cut themselves off completely from the court would have meant severing themselves from the source of offi -cial appointments and from critical personal connections. Th is would thereby have ended all hope of maintaining the family’s social and po liti cal position— even in provincial society. Typical exurban provincial offi cials built homes and held packages of lands scattered about the countryside, which provided them with income. At the same time, they maintained extensive contact with po liti cal aff airs in the capital and oft en maintained homes there, to which they shipped most of the profi ts from their rural enterprises. To provincial governors and their families, Kyoto was the source of the human and physical

32 Karl Friday

resources that made their provincial business activities possible, as well as the marketplace for the goods they obtained in the country. By the same token, they used their provincial activities to reinforce their foothold within the po-liti cal and offi cial world of the capital.

Th e carpetbaggers competed with longer- established provincial elites, but they also formed alliances with them. Most importantly, they intermarried with provincial families. Edicts forbidding this practice make it clear that provincial offi cials were taking wives and sons- in- law from provincial elite houses with considerable frequency.

The New Warrior Order

Th e expansive social and po liti cal changes taking shape during the Heian pe-riod spawned intensifying competition for wealth and infl uence among the premier noble houses of the court, which in turn led to a private market for military resources, arising in parallel to the one generated by evolving gov-ernment military policies. State and private needs thus intersected to create widening avenues to personal success for ambitious young men with military talents. Provincial elites and lower- ranking court nobles, therefore, increas-ingly turned to military ser vice as a path to the entourages of powerful aristo-crats and lucrative government posts.

Compelled by a need to defend themselves and their prerogatives against outlawry and armed re sis tance, as well as by the desire to maximize the prof-its that could be squeezed from taxpayers, many provincial governors began to include “warriors of ability” among the personal staff s that accompanied them to their provinces of appointment. Some also took up arms for them-selves and established reputations as military troubleshooters.

By the tenth century, military ser vice at court and ser vice as a provincial offi cial had become parallel and mutually supportive careers for many zuryō, resulting in the emergence of the group Japa nese scholars have dubbed the miyako no musha, or “warriors of the capital.” Th ese were men of the fourth or fi ft h court rank, who curried the patronage of the higher nobility and recogni-tion by the state by serving as bodyguards, police, and soldiers.

Two of the most illustrious warrior names of the Heian age, Taira and Minamoto, both had their beginnings in eff orts to prune the imperial family tree and dispose of extraneous princes and princesses. Th ere were seventeen major lines of Minamoto and four of Taira, all descended from diff erent em-perors. Branches of two of these— the Seiwa Genji deriving from Emperor


Seiwa (r. 858– 876), and the Kammu Heishi, descended from Emperor Kammu (r. 781– 806)—became famous as warrior houses.

Seiwa had nineteen sons, the descendants of nine of whom bore the Mina-moto surname. Th e lines that produced samurai claim descent from Seiwa’s sixth son, Prince Sadazumi, through his son Tsunemoto, who received the Minamoto name just a few months before his death, in 961. Tsunemoto fa-thered nine sons, at least three of whom became warriors of some repute. Th e most important of these was Mitsunaka. It is to Mitsunaka and the progeny of his three eldest sons, Yorimitsu, Yorichika, and Yorinobu, that historians gen-erally refer when they speak of the Seiwa Genji.

Th e Kammu Heishi comprised four main lineages, each descended from a diff erent prince. Th e line that produced warriors began with Kammu’s eldest son, Katsurahara, through his son Takami and grandson Takamochi. Ac-cording to tradition, Takamochi, who was presented with the Taira surname and appointed assistant governor of Kazusa in 889, acquired both the ap-pointment and the name as rewards for his military heroism in suppressing an attempted coup d’état at court. While this story is probably apocryphal; the fact that Takamochi’s descendants believed in such a venerable military tradition for their house is noteworthy.

From the late tenth to the early twelft h century, warriors of Minamoto de-scent dominated the military world of the capital owing to a combination of martial prowess and a hereditary client relationship with the Fujiwara regents. Genji preeminence peaked during the eleventh century with the careers of Yoriyoshi (995– 1082) and his son Yoshiie (1041– 1108), but then waned consider-ably over the next few generations. During the late 1090s, Taira Masamori, the head of a theretofore relatively minor branch of the Kammu Heishi, managed to establish for himself and his family a patron-client relationship with succes-sive retired emperors (in) similar to the one that existed between the Mina-moto and the Fujiwara Regents’ House. By the second de cade of the twelft h century, the strength of this new alliance brought the Taira to full parity of prestige with the Minamoto at court. Th is situation lasted until the 1150s.

Provincial warrior leaders were, broadly speaking, of two main types of pedigree: descendants of cadet branches of miyako no musha houses that had established bases in the provinces; and the scions of families deriving from the old provincial nobility or other longtime provincial elites. In practice, however, both groups intermarried and interacted so thoroughly as to become functionally indistinguishable. A large percentage of families that should

34 Karl Friday

probably be included in the second group appear instead in the genealogies of the fi rst, owing to marriage or other ties.

Marriages among the court aristocracy were polygamous or serially mo-nogamous, and usually involved not just separate bedrooms but separate residences. Women most commonly continued to live in their birth homes or in other residences provided by their fathers, while husbands lived in houses of their own and visited as frequently as circumstances and inclinations al-lowed. Children reckoned descent primarily from their father and took his surname. But they were usually raised in their mother’s home and inherited much of their material property from her. When, moreover, the bride’s family was of signifi cantly higher station than the groom’s, the children— and some-times the new husband— oft en adopted the surname of the bride’s father. Zuryō sent to work in the provinces took their marriage customs with them. As a result, court- derived surnames like Taira, Minamoto, and Fujiwara gradually supplanted those of the older provincial noble families.

Warbands, Marshals, and Sheriffs

Th e history of Japa nese military or ga ni za tion is a tale dominated by dialectics between personal and institutional authority, and between localized and cen-tralized sanctions, jurisdictions, and structures of command. By mid- Heian times, warfare and law enforcement had become the preserve of a warrior or-der that armed, trained, and or ga nized itself. But armies were raised and re-tained with the backing of state authority— sometimes allocated, sometimes borrowed, and sometimes fabricated— through most of the 1300s.

From the late eighth century onward, the court slowly groped and experi-mented its way toward a system that centered on commissioning professional warriors with new military titles legitimating their use of private martial re-sources on behalf of the state— a principle that would characterize military aff airs in Japan down to the modern era. Th e evolution of military posts dur-ing the Heian period and beyond refl ects the emergence of the samurai across the same span of time. Th e two pro cesses were, in fact, reciprocal: Deputizing provincial elites and members of the middle and lower central aristocracy in-evitably had a catalytic eff ect on the development of private martial resources under these leaders, which in turn led to the introduction of new assignments and the modifi cation of existing ones.

Th e cornerstones of the Heian military system were all in place by the mid- tenth century. Although the system continued to evolve and adapt to the


ever- changing sociopo liti cal reality it served, the cardinal features persisted well into the medieval era.

Standing out among these features was the bifurcation of or gan i za tion al principles, and therefore the degree of cohesion, that characterized warrior associations and confederations. At the tactical level, military units formed around personal relationships and personal connections. At the strategic level, however, the or ga ni za tion of warfare remained closely integrated with the framework of the ritsuryō state and surprisingly obedient to visions of centralized, public authority formulated in the early eighth century.

Commissions like Ōryōshi (“Envoy to Subdue the Territory”), tsuibushi (“Envoy to Pursue and Capture”), kebiishi (“Investigators of Oddities”), and tsuitōshi (“Envoy to Pursue and Strike Down”), though extra- codal, held true to the spirit of the eighth- century military system in most respects. Appoint-ments and compensation alike came from the center, following principles and procedures that closely paralleled those specifi ed for comparable posts under the ritsuryō codes. Th ese essential similarities enabled the court to retain exclu-sive authority over— and at least general control of— military aff airs through-out the Heian period. But the new offi ces were also fundamentally diff erent from their ritsuryō antecedents on one critical point: Th ey were premised not on the existence of a publicly conscripted pool of manpower over which the offi cer’s commission gave charge, but on the appointee’s ability to recruit troops for himself.

Curiously, the court established no statutory guidelines for draft ing or oth-erwise raising troops immediately aft er its dismantling of the provincial regi-ments in 792. For most of the ninth century, troop mobilizations remained grounded in public authority but were conducted through ad hoc means. Re-sponsibility for mustering fi ghting men as the need arose rested with provincial governors. Th e specifi c manner in which this mobilization was to be accom-plished varied from case to case, but it generally involved draft ing the necessary manpower on the basis of the general corvée obligations required of all imperial subjects.

Th is notion of public military ser vice and induction based on public duties remained alive throughout the early medieval era, but by the middle of the tenth century, recruitment had become largely privatized, with “government” troops enlisted and mobilized through private chains of command. Th e phe-nomenon that made this possible was the predilection of warriors to arrange themselves into bands and networks.

36 Karl Friday

Warriors began forming gangs by the middle of the ninth century and perhaps even earlier. By the third de cade of the tenth century, private military networks of substantial scale had begun to appear, centered on major provin-cial warriors like Taira Masakado, who, we are told, could charge into battle “leading many thousands of warriors,” each themselves leading “followers as numerous as the clouds.”

Although the court initially opposed these developments, it embraced them as soon as it realized that private military organizations could serve as useful mechanisms for conscripting troops when it needed them. By the mid- tenth century, the government had begun to co- opt private arrangements be-tween warriors, transferring much of the responsibility for mustering and or ga niz ing the forces necessary for carry ing out military assignments to war-rior leaders. Th ese leaders could in turn delegate much of that responsibility to their own subordinates.

Historians attempting to generalize about Heian military alliances must, however, not lose sight of the fact that they are dealing with a variety of inter-active but diverse entities. During the Heian period, private martial entou-rages were assembled by retired emperors and top courtiers, by monastic in-stitutions, by career provincial offi cials, and by provincial residents of many levels of status. Although all such organizations shared an obvious similarity of purpose, and warrior groupings at various levels were further knit together into networks of alliances, both the networks and their component parts var-ied enormously in scale, complexity, and cohesiveness from place to place and from time to time.

Private military organizations during the Heian period tended to be patch-work assemblages of several types of forces. Leading warriors in both the prov-inces and the capital maintained relatively small, core bands of fi ghting men who were direct economic dependents of the warriors, lived in homes in or very near the warriors’ compounds, and were at their more or less constant disposal. Manpower for these martial entourages could be drawn from a vari-ety of sources. Some troops were simply hired mercenaries; others were sons or close relatives of the or ga ni za tion’s leader; still others were conscripted from among the residents and cultivators of lands over which the leader exercised some degree of control.

Just how small these components were is diffi cult to ascertain, for few reli-able sources record the numbers of followers under a given warrior’s direct command. Th ose that do, moreover, indicate substantial variation from one


samurai leader to another. At the same time, they also suggest that the core units from which early medieval military forces were compiled averaged around a half- dozen or so mounted warriors, augmented by varying numbers of foot soldiers. Some were much smaller. But even the largest numbered only in the high teens or low twenties.

For major campaigns, samurai also mobilized the cultivators, woodsmen, fi shermen, and other residents of the lands in and around the estates and dis-tricts they administered. Such men were, strictly speaking, not under the war-rior’s control, but they oft en leased land from him, borrowed tools and seed from him, and conducted trade at his compound, making his residence an im-portant economic center for them. By exploiting what ever po liti cal and eco-nomic leverage they could bring to bear on these semidependents, warriors could assemble armies numbering in the hundreds. Larger armies had to be knit together through networks based on alliances of various sorts between samurai leaders of diff erent sociopo liti cal status. Th is technique made it possible for war-riors to assemble forces many times the size of their core organizations.

Th e incentive to build or belong to military organizations of larger and larger scale was a natural consequence of the same factors that induced men to take up arms or create warrior bands in the fi rst place. Th e most obvious way for a warrior to augment his personal corps of followers was to go into partnership with his peers. Genuinely lateral alliances were problematic, how-ever, for the corollary to the factors and principles that encouraged vertical cooperation between members of diff erent sociopo liti cal strata was that the interests of men of similar station were generally at odds with one another. Th is tended to exercise a divisive infl uence, rendering lateral alliances unsta-ble and therefore diffi cult to maintain for long periods, with the result that, for most of Japa nese history, horizontal cooperation between coequals has tended to be an ephemeral phenomenon.

Warrior leagues that were in essence alliances between coequals did come into existence during the early medieval period, most notably in the southern part of Musashi. But the predominant or gan i za tion al pattern for warrior cooperation was hierarchical, centering on fi gures whose prestige enabled them to serve as rallying points for alliances between warriors of lesser pedi-gree. Peer rivalries could be readily subsumed and transcended in warrior networks that were focused on men of overarching status.

Even so, the standout feature of Heian era warrior alliances was their fragility, a condition that refl ected the amorphous nature of the lord- vassal

38 Karl Friday

bond during the period. For, unlike other forms of consociation, such as the land- commendation arrangements set forth between provincial elites and the leading courtiers and religious institutions of the capital— the pro cess by which estates (shōen) were formed— alliances between warriors were not supported by legal contracts. Th e exchange of obligations that accompanied warrior partnerships during Heian times was far less palpable, and the na-ture, extent, and duration of these obligations much less precise, than in me-dieval times.

Formal arrangements under which specifi ed benefi ces were off ered in re-turn for defi ned military ser vices were slow to develop in Japan because the ability of warrior leaders to manipulate any carrot or stick approach in order to recruit, maintain, and control followers was closely circumscribed by their relatively weak po liti cal circumstances. Even the most powerful warriors of the age occupied only intermediate positions in the sociopo liti cal hierarchy and were dependent on connections with the higher echelons of the court to maintain their po liti cal and economic positions. Th eir autonomy in matters of governance and landholding was limited, which meant that they lacked the right— and therefore the means— to reward or punish their own troops directly.

In Heian times, warriors remained essentially mercenaries, off ering their skills and ser vices in exchange for long- term patronage of their careers by court powers- that- be, or for more immediate rewards. While such activities oft en brought perquisites over lands and peoples, and sometimes involved the transfer of properties hitherto administered by warriors on the losing side of a confl ict, Heian samurai were rarely, if ever, able to specify the size or the particulars of rewards for themselves. Any transfers of lands were accom-plished indirectly, through the agency of the court and in accord with the niceties prescribed by the court- centered legal system.

Consequently, Heian military alliances tended to be nebulous and short- lived. Th e larger the or ga ni za tion, the more ephemeral it tended to be. On oc-casion, illustrious warriors like Minamoto Yoshiie or Yoshitomo were able to construct martial networks that extended across multiple provinces, but until the 1180s, no such or ga ni za tion survived the death of its found er.

But then, in 1180, Minamoto Yoritomo, a dispossessed heir to a leading samurai house, adeptly parlayed his own pedigree, the localized ambitions of provincial warriors, and a series of upheavals within the imperial court into the creation of a new institution— called the shogunate, or bakufu, by historians— in


the eastern village of Kamakura. In so doing, he unwittingly set in motion developments that would in due course overwhelm the old polity.

The First Shogunate

Th e events that led to the birth of Japan’s fi rst warrior government began in 1159, when Yoritomo’s father, Yoshitomo, joined a clumsy attempt to seize control of the court. In the resulting Heiji Incident (named for the calendar era in which it occurred), Yoshitomo was defeated by his longtime rival Taira Kiyomori, who then executed Yoshitomo’s allies and relatives, and exiled his sons— including Yoritomo, who was 13 at the time.

For the next two de cades, Kiyomori’s prestige and infl uence at court grew steadily. In 1171 he arranged to marry his daughter, Tokuko, to the emperor. In 1179 he staged a coup d’état, seizing virtual control of the court. Kiyomori reached the height of his power in 1180, when his grandson (by Tokuko) as-cended the throne as Emperor Antoku. Th at same year, however, a frustrated claimant to the throne, Prince Mochihito, provided Yoritomo with an oppor-tunity for revenge.

Yoritomo was by any reckoning an unlikely champion: in 1180 he had even less going for him than typical warrior leaders of his age. His father’s misad-venture two de cades before had cost him the career as a government offi cial and warrior noble he would otherwise have enjoyed, and doomed him instead to an obscure life as a minor provincial warrior. He held no government posts, led no warband of his own, and controlled no lands; his one and only asset was a shaky claim to leadership among his surviving relatives. And so, being unable to work within the system, Yoritomo instead hit on an ingenious end run around it.

He used Mochihito’s call to arms to rescue the court from Kiyomori as a pretext to issue one of his own, declaring a martial law under himself across the eastern provinces and declaring that, in return for an oath of allegiance to himself, henceforth he (Yoritomo) would assume the role of the court in guar-anteeing what ever lands and administrative rights an enlisting vassal consid-ered to be rightfully his own.

In essence, Yoritomo was proclaiming the existence of an in de pen dent state in the east, a polity run by warriors for warriors. But he took pains to portray himself as a righ teous outlaw, a champion of true justice breaking the law in order to rescue the institutions it was meant to serve. His initiative touched off a groundswell of support, as well as a countrywide series of feuds

40 Karl Friday

and civil wars loosely justifi ed by Yoritomo’s crusade against Kiyomori and his heirs.

In the course of this so- called Gempei War (the name of which derives from the Sino- Japanese readings for the characters used to write “Minamoto” and “Taira”), however, Yoritomo revealed himself to be a surprisingly conser-vative revolutionary. Rather than maintain his in de pen dent warrior state in the east, Yoritomo negotiated a series of accords that gave permanent status to the Kamakura regime, trading reincorporation of the east into the court- centered national polity for formal court recognition of many of the powers he had seized.

Th e regime that this agreement established— the Kamakura shogunate— was a kind of government within a government that was both part of and dis-tinct from the court in Kyoto. It acted as the main military and police agency of the court, exercised broad governing powers in eastern Japan, and held special authority over the warriors, scattered countrywide, that it recognized as its formal vassals. Aft er the Jōkyū War of 1221, an ill- fated attempt by a re-tired emperor, Go- Toba, to eliminate the shogunate, the balance of real power shift ed steadily toward Kamakura, and away from Kyoto. By the end of that century, the shogunate had assumed control of most of the state’s judicial, military, and foreign aff airs.

Following Yoritomo’s death in 1199, control of the regime fell into the hands of his in- laws, the Hōjō family. It was, in fact, the Hōjō, not Yoritomo himself, who made the new regime a shogunate— that is, a government under a seii taishōgun (“Great General for Subduing the Barbarians”). Th is already venerable title had hitherto been a temporary, war time commission given to offi cers who led imperial armies in the east. Yoritomo held it briefl y from 1192 to 1195; his son and successor, Yoriie, held it for just under a year, beginning in late 1202. In 1203, however, Yoritomo’s widow, Masako, her father, Hōjō Toki-masa, and her brother, Yoshitoki, deposed Yoriie and replaced him with his brother, Sanetomo. Th ey legitimized this new power arrangement by having Sanetomo appointed seii taishōgun. Th us Sanetomo, who is remembered as the third Kamakura shōgun, was actually the fi rst to hold that title from the beginning to the end of his reign (1203– 1219). But he and the six shōgun who followed him in the offi ce were fi gurehead rulers (many of them were children at the time they took offi ce), while the Hōojō ran things behind the scenes as hereditary directors of the shōgun’s private chancellery.

Th e samurai- centered military and police system of the Heian era had rep-resented a curious mixture of public and private parts: It depended on private


resources— private training and acquisition of skills, private recruitment and mobilization, and private equipment— for its operations; but it was or ga nized and directed through centralized, public principles. In most respects, the or-ga niz ing principles of Kamakura period armies and military campaigns re-mained true to this pattern. Th e key diff erence was, of course, the introduction of the shogunate, which, aft er 1183, stood as an intermediary between the court and warriors throughout the country. But in its military capacity, if not in its judicial and managerial roles, the new institution represented less a usurpation or an intrusion into the system than an adaptation or outgrowth of it. With re-spect to the court and state military functions, the shōgun simply assumed roles and duties that had hitherto been spread among the various Minamoto and Taira warrior leaders. Th at is, the shōgun—or rather the institution he symboli-cally headed— became in eff ect the lone surviving miyako no musha, carry ing out the court’s law enforcement and national defense jobs by mobilizing per-sonal retainers, in much the same way that Heian samurai leaders had. Militar-ily, the shogunate was in essence simply a corporate warband leader writ large.

What was new, however, was the permanence of the commission Yoritomo and his successors held, as well as the sheer size of the vassal corps they led and the degree to which they were able to rationalize and institutionalize both. For unlike twelft h- century military offi cers, whose command authority lasted only for the duration of a specifi c mission, the thirteenth- century shogunate exercised an ongoing and more or less exclusive jurisdiction over warfare. It also introduced new mechanisms for or ga niz ing and directing its house men, as well as an unpre ce dented clarity to the reciprocal obligations that bound them.

In its relationships to the imperial court and to samurai in the country-side, the fi rst shogunate is perhaps best understood as a kind of warriors’ union. Before its creation, warriors in the provinces were merely local govern-ment administrators or caretakers for estates that belonged to court nobles or temples. Th e court had kept them po liti cally weak by playing them against one another. By insulating an elite subgroup of the country’s provincial war-riors from direct court control or employ, however, the shogunate signifi cantly changed the rules of the game. Initially, this merely served to vault Yoritomo (and aft er him, the shogunate) into prominence, but in the long run, it created a mechanism for unraveling the fabric of centralized authority.

Th e eastern warriors who answered Yoritomo’s call to arms in 1180 came to him because they were frustrated by the limitations of their traditional place in the landholding and governing systems. Yoritomo exploited those frustrations

42 Karl Friday

to build his original vassal band, and then, through his military and diplo-matic eff orts over the next fi ve years, extended this or ga ni za tion across the rest of Japan. As his regime developed, Yoritomo kept himself indispensable to both his men and the court by making himself the exclusive intermediary be-tween them, insisting that all calls to ser vice, all rewards, and all disciplinary matters involving Kamakura vassals pass through him. Th is arrangement be-came permanent— hereditary—under the Hōjō aft er Yoritomo’s death.

Th e existence of the shogunate, therefore, rested on two competing obliga-tions: On the one hand, it had a mandate from the court to maintain order in the provinces— to keep its own men under control and to use them to defend the court. Th is is what made the regime legal and formed the basis of its na-tional authority. On the other, the shogunate’s ability to carry out this mandate depended on the continuing support of its followers, which in turn hinged on its support of their ambitions for greater freedom from court control.

Th ese competing obligations led the shogunate to adopt a policy of mini-mal action whenever it was called upon to resolve a dispute between one of its vassals and some court noble or temple. It never acted unless asked (by one or both of the parties involved), it kept no rec ords of its own on the proceedings or the results of such lawsuits, and it rarely imposed penalties more severe than ordering the defendant to cease and desist from what ever it was that he had been doing to cause the complaint.

Kamakura vassals across the country quickly learned to take advantage of this situation, manipulating their special status to lay stronger and more per-sonal claims to their lands— and the people on them. Th ey did this in much the same way that a per sis tent dog will sometimes crawl under the blankets and take over most of its own er’s bed.

Th e fi rst step in the pro cess usually involved a warrior exceeding his au-thority in some small way, such as keeping more than his agreed upon share of the rents or taxes collected on the lands he administered. Because the war-rior was a Kamakura vassal, the estate own er could not discipline the man himself; he could only complain to the shogunate. Th e shogunate, however, was less concerned with abstract matters of right and wrong than with keep-ing order and minimizing bad feelings on either side. Aft er hearing argu-ments and examining documents submitted, its rulings more oft en than not involved some kind of compromise between the claims of the two parties. Th is meant that the warrior got less than he had tried to seize, but also that the estate own er would, from then on, be forced to accept less than he had before


the dispute began. Th is sort of warrior behavior became increasingly frequent and blatant as the Kamakura period wore on. Settlements were followed by new violations and new compromise settlements, as the same warriors pushed the boundaries of their legal obligations over and over, generation aft er gen-eration. Th e pro cess worked like a ratchet: Each new settlement represented a net gain for warriors and a net loss for the court.

Th us, through a gradual advance by fait accompli, real power over the coun-tryside bled steadily from the center to the hands of local fi gures, and a new warrior- dominated system of authority absorbed the older, courtier- dominated one. By the second quarter of the fourteenth century, this evolution had pro-gressed to the point where the most successful of the shogunate’s provincial vassals had begun to question the value of continued submission to Kamakura at all. Th e regime fell in 1333, as the result of events spawned by an imperial suc-cession dispute.

. . .

While historians once equated the appearance of the samurai with the onset of “feudalism,” more recent scholarship emphasizes the continued— indeed, the enhanced— integration of the center and peripheries during the late Heian period. It is now clear that military power in Japan became privatized and decentralized long before governing power did. Court enfranchisement of private warriors from very early on worked, paradoxically, to minimize con-nections between military and po liti cal power for many centuries thereaft er. Unlike medieval Eu rope, where knights and military lordship arose more or less in tandem, Heian Japan remained fi rmly under civil authority. Th ere was no power vacuum into which incipient warlords could rush, and little warrior class- consciousness emerged to incite a warrior revolution.

Indeed, Heian samurai at all levels in the sociopo liti cal hierarchy thought of themselves as warriors in much the same way that modern corporate executives view themselves as shoemakers, automobile manufacturers, or magazine dis-tributors: Th ey identifi ed more strongly with their nonmilitary social peers than with warriors above or below them in the hierarchy, just as executives tend to identify more closely with their counterparts in other fi rms and other indus-tries than with the workers, engineers, or secretaries in their factories, design workshops, and offi ces. While the descendants— both genealogical and institu-tional— of the professional warriors of Heian times did indeed become the masters of Japan’s medieval epoch, until Minamoto Yoritomo’s epoch- making

44 Karl Friday

usurpation of power in the late twelft h century, the samurai remained the ser-vants, not the adversaries, of the court and the state.

Notes1. Pierre François Souyri, Th e World Turned Upside Down: Medieval Japa nese So-

ciety (New York: Columbia University Press, 2001).2. See Mikael Adolphson, Edward Kamens, and Stacie Matsumoto, eds., Heian

Japan: Centers and Peripheries (Honolulu: University of Hawaii Press, 2007); Joan R. Piggott, ed., Capital and Countryside in Japan, 300– 1180: Japa nese Historians In-terpreted in En glish (Ithaca, NY: Cornell University East Asia Series, 2006); and Jeff rey P. Mass, ed., Th e Origins of Japan’s Medieval World: Courtiers, Clerics, War-riors, and Peasants in the Fourteenth Century (Stanford, CA: Stanford University Press, 1997).

3. Th e following account of the origins of the samurai is condensed primarily from Karl Friday, Hired Swords: Th e Rise of Private Warrior Power in Early Japan (Stanford, CA: Stanford University Press, 1992). For an alternative view, see William Wayne Farris, Heavenly Warriors: Th e Evolution of Japan’s Military, 500– 1300 (Cam-bridge, MA: Harvard University Press, 1992).

4. For more on the po liti cal structure and the events of this period, see Bruce Bat-ten, “Foreign Th reat and Domestic Reform: Th e Emergence of the Ritsuryō State.” Monumenta Nipponica 41, no. 2 (1986): 199– 219; Cornelius J. Kiley, “State and Dynasty in Archaic Yamato,” Journal of Japa nese Studies 3, no. 1 (1973): 25– 49; Inoue Mitsusada with Delmer M. Brown, “Th e Century of Reform,” in Ancient Japan, vol. 1 of Th e Cam-bridge History of Japan, edited by Delmer M. Brown (New York: Cambridge Univer-sity Press, 1993), 163– 220; Joan R. Piggott, Th e Emergence of Japa nese Kingship (Stan-ford, CA: Stanford University Press, 1997).

5. Batten, “Foreign Th reat and Domestic Reform,” 10– 14.6. See, for example, the forces described in Nihon shoki 591 11/4 and 602 2/1 (Shin-

tei zōho kokushi taikei [Tokyo: Yoshikawa kōbunkan, 1985]).7. Ryō no gige, in Shintei zōho kokushi taikei (Tokyo: Yoshikawa kōbunkan, 1985), 192.8. Detailed accounts of the ritsuryō tax system appear in Dana Robert Morris,

“Land and Society,” in Heian Japan, vol. 2 of Th e Cambridge History of Japan, edited by Donald H. Shively and William H. McCullough (New York: Cambridge University Press, 1999), 183– 235; Morris, “Peasant Economy in Early Japan, 650– 950” (Ph.D. disser-tation, University of California, Berkeley, 1980); and Torao Toshiya, “Nara Economic and Social Institutions,” in Ancient Japan, vol. 1 of Th e Cambridge History of Japan, ed-ited by Delmer M. Brown (Cambridge: Cambridge University Press, 1993), 415– 452.

9. Ryō no gige, p. 183.


10. Ryō no gige, pp. 185, 193, 195– 196; Shoku Nihongi 709 10/26, 714 11/11 (Shintei zōho kokushi taikei [Tokyo: Yoshikawa kōbunkan, 1986]); Suzuki Takeo, “Heian jidai ni okeru nōmin no uma,” Nihon rekishi no. 239 (1968), 42– 55.

11. For a detailed treatment of the “pacifi cation campaigns” in the northeast, see Karl Friday, “Pushing Beyond the Pale: Th e Yamato Conquest of the Emishi and Northern Japan,” Journal of Japa nese Studies 23, no. 1 (1997): 1– 24.

12. See, for example, Shoku Nihongi 704 6/3; or 753 10/21 daijōkampu in Ruijū sandaikyaku, in Shintei zōho kokushi taikei (Tokyo: Yoshikawa kōbunkan, 1983), 2: 553.

13. A detailed account of this pro cess appears in Friday, Hired Swords, 45– 69, 122– 166.

14. Morris, “Peasant Economy,” 35, 44– 45.15. For an extensive and detailed elaboration on this point, see Adolphson, Ka-

mens, and Matsumoto, Heian Japan.16. Cornelius J. Kiley explores this central idea in detail in “Provincial Adminis-

tration and Land Tenure in Early Heian,” in Heian Japan, vol. 2 of Th e Cambridge History of Japan, edited by. Donald H. Shively and William H. McCullough (New York: Cambridge University Press, 1999), 236– 340.

17. Yoneda Yūsuke, Kodai kokka to chihō gōzoku (Tokyo: Kyōikusha, 1979), 115– 118; Yoshie Akio, “Shōki chūsei sonraku no keisei,” in Kōza Nihonshi, vol. 2, edited by Rekishigaku kenkyūkai and Nihonshi kenkyūkai (Tokyo: Tokyo daigaku shuppankai, 1970), 112– 114.

18. Morris, “Peasant Economy,” 163– 205; Yoneda Yūsuke, Kodai kokka, 145– 146.19. 896 4/2 daijōkampu, in Ruijū sandai kyaku 2:606; 835 10/18 daijōkampu, quoted

in 867 12/20 daijōkampu; 894 7/16 daijōkampu in Ruijū sandai kyaku 2:623– 624.20. Heian ibun, edited by Takeuchi Rizō, 15 vols. (Tokyo: Tōkyōdō, 1965), doc. 339.

Mo rita Tei, Zuryō (Tokyo: Kyōikusha, 1978), 50– 52, 107– 117, 136– 138.21. Ishimoda Shō, Kodai makki seijishi jōsetsu (Tokyo: Miraisha, 1964), 43– 44.

Ruijū kokushi 798 2/1 (5 vols., Shintei zoho kokushi taikei. Tokyo: Yoshikawa ko-bunkan, 1986); Shoku Nihon kōki 838 2/9, 2/10, 2/12, 839 4/2, 850 2/3, 857 3/18 (Shintei zōho kokushi taikei. Tokyo: Yoshikawa kōbunkan, 1987); Ruijū sandai kyaku 2:614 (840 2/25 daijōkampu), 2:620– 621 (891 9/11 daijōkampu), 2:623– 624 (867 12/20 daijōkampu), 2:640– 641 (867 3/24 daijōkampu), 2:565 (Kōzuke no kuni ge, quoted it 899 9/19 daijōkampu); Nihon Montoku tennō jitsuroku 857 3/16, 858 2/22 (Shintei zōho kokushi taikei. Tokyo: Yoshikawa kōbunkan, 1984); Sandai jitsuroku 861 11/6, 862 5/20, 866 4/11, 870 12/2 (2 vols., Shintei zōho kokushi taikei. Tokyo: Yoshikawa kōbunkan, 1986). Ishi-moda also cites the cases of a district magistrate in Tsushima who led 300 men in an attack on the governor in 857, and two district magistrates in Iwami who led 270 peas-ants in revolt against the “misrule” of the governor.

46 Karl Friday

22. Th is and similar phrases appear in numerous documents from the ninth cen-tury onward, lamenting the depredations of central court fi gures in the provinces. See, for example, Ruijū sandai kyaku, vol. 2, pp. 623– 624 (835 10/18 daijō kampu, quoted in 867 11/20 daijō kampu); pp. 606 (896 4/2 daijō kampu); pp. 617– 618 (905 8/25 daijō kampu).

23. Cornelius J. Kiley, “Estate and Property in the Late Heian Period,” in Medieval Japan: Essays in Institutional History, edited by John W. Hall and Jeff rey P. Mass (New Haven, CT: Yale University Press, 1974), 110– 111.

24. For more on this phenomenon, see Takahashi Masaaki, Bushi no seiritsu, Bushizō no sōshutsu (Tokyo: Tōkyō daigaku shuppankai, 1999), 13– 20; or John W. Hall, Government and Local Power in Japan, 500– 1700: A Study Based on Bizen Prov-ince (Prince ton, NJ: Prince ton University Press, 1966), 116– 128.

25. G. Cameron Hurst III, Insei: Abdicated Sovereigns in the Politics of Late Heian Japan 1086– 1185 (New York: Columbia University Press, 1976), 19– 35, discusses this system in detail.

26. Ruijū sandaikyaku 2:617– 618 (905 8/25 daijōkampu); 2:638 (Settsu no kuni ge, quoted in 860 9/20 daijōkampu).

27. Kitayama Shigeo, Taira Masakado (Tokyo: Asahi shinbunsha, 1993), 23– 28.28. Ruijū sandaikyaku 2:619– 621 (797 4/29 daijōkampu, quoted in 891 9/11 daijōkampu;

842 8/15 daijōkampu, quoted in 895 11/7 daijōkampu). Th e court issued repeated prohi-bitions against zuryō establishing residential bases in their provinces, in areas as far apart as Kyushu and Kazusa. See, for example, Ruijū sandaikyaku 2:619– 621 (891 9/11 daijōkampu; 895 11/17 daijōkampu); Sandai jitsuroku 884 8/4.

29. Kiley, “Provincial Administration,” 335.30. Hodate Michihisa, “Kodai makki no tōgoku to ryūjū kizoku,” in Chūsei

Tōgokushi no kenkyū, edited by Chūsei tōgokushi kenkyūkai (Tokyo: Tōkyō daigaku shuppankai, 1988), 7– 12; Takahashi Masaaki, Kiyomori izen (Tokyo: Heibonsha, 1984), 14.

31. Ruijū sandai kyaku 1:302 (744 10/14 kyaku); 1:303 (868 6/28 kyaku).32. Chōya gunsai, in Kokushi taikei (Tokyo: Yoshikawa kōbunkan, 1964), 525;

Heian ibun doc. 339; Konjaku monogatarishū 19.4, 28.2 (Vols. 21– 24, Nihon koten bun-gaku zenshū, [Tokyo: Shōgakkan, 1971]). Friday, Hired Swords, 82– 85.

33. Karl Friday, Hired Swords, 81– 89; Farris, Heavenly Warriors, 169– 177.34. Oboroya Hisashi, Seiwa Genji (Tokyo: Kyōikusha, 1984), 21– 27, discusses this

phenomenon in detail. For a discussion in En glish, see William H. McCullough, “Th e Heian Court, 794– 1070,” in Heian Japan, vol. 2 of Th e Cambridge History of Japan, edited by Donald H. Shively and William H. McCullough (New York: Cambridge University Press, 1999), 39– 40; or Friday, Th e First Samurai: Th e Life and Legend of the Warrior Rebel, Taira Masakado (New York: John Wiley & Sons, 2008), 35.


35. Sompi bummyaku, in Shintei zōho kokushi taikei (Tokyo: Yoshikawa kōbunkan, 1983) vols. 3 and 4. Th e term Genji derives from the Sino- Japanese reading of the surname Minamoto. Similarly, Heishi comes from the Sino- Japanese reading of the surname Taira.

36. Sompi bummyaku, 3:57– 62. Th e Seiwa Genji have been the subject of numer-ous studies by Japa nese historians. Oboroya Hisashi, Seiwa Genji; Yasuda Motohisa, Bushi sekai no jōmaku (Tokyo: Yoshikawa kōbunkan, 1973); Noguchi Minoru, Genji to bandō bushi (Tokyo: Yoshikawa kōbunkan, 2007); and Motogi Yasuo, Minamoto No Mitsunaka Yorimitsu: sassei hōitsu Tomoie no shugo (Tokyo: Meburuva shobō, 2004) are especially recommended.

37. Sompi bummyaku 4:1– 4. Some scholars have questioned whether the warrior Taira are truly an off shoot of the Kammu Heishi. Th ey note that the principal source of the belief that they were is the Heishi genealogy in the Sompi bummyaku, which contains numerous errors and is generally considered to be the least reliable section of the work. Yet, as Yasuda Motohisa has observed, we have no means to correct most of these errors, and therefore we have little choice but to accept the general outline of the genealogy. See Yasuda Motohisa, Bushi sekai no jōmaku, 68– 69; Takahashi Masaaki, “Masakado no ran no hyōka o megutte,” Bunka shigaku no. 26 (1971): 26; or Mo rita Tei, Zuryō, 139– 140. Th e defi nitive study of the early Heishi is Takahashi Masaaki, Kiyomori izen.

38. Friday, Hired Swords, 98– 99; Farris, Heavenly Warriors, 188– 189; Fukuda Toyo-hiko, Tōgoku heiran to mononofu- tachi (Tokyo: Yoshikawa kōbunkan, 1995), 6– 7.

39. Fukuda Toyohiko, Tōgoku heiran to mononofu- tachi, 6– 7; Farris, Heavenly Warriors, 188– 189; Morris, “Land and Society,” 198– 199; Kiley, “Provincial Admin-istration,” 281. For more on Heian marriages and family structure, see William Mc-Cullough, “Japa nese Marriage,” Harvard Journal of Asian Studies, no. 27 (1967): 103– 167; Peter Nickerson, “Th e Meaning of Matrilocality: Kinship, Property, and Poli-tics in Mid- Heian,” Monumenta Nipponica 48, no. 4 (1993): 429– 468; Ivan Morris, “Marriage in the World of Genji,” Asia, no. 11 (1968): 54– 77; Wakita Haruko, “Mar-riage and Property from the Perspective of Women’s History,” Journal of Japa nese Studies 1, no. 2 (1984): 321– 345; Hattō Sanae, “Kazoku to kyōdōtai,” Rekishi hyōron no. 424 (1985): 14– 23; Hattō, “Sekkanki ni okeru zuryō no ie to kazoku keitai,” Nihon rekishi no. 442 (1985): 1– 18; Sekiguchi Hiroko, “Kodai kazoku to kon’in keitai,” in Koza Nihonshi 2, edited by Rekishigaku kenkyūkai and Nihonshi kenkyūkai (To-kyo: Tokyo daigaku shuppankai, 1984), 287– 326; Tabata Yasuko, “Kodai, chūsei no ’ie’ to kazoku: yōshi o chūshin to shite,” Tachibana jōshi daigaku kenkyū kiyō no. 12 (1985): 41– 67.

40. Karl Friday, Samurai, Warfare and the State in Early Medieval Japan (London: Routledge, 2004), 34– 62, discusses this issue in detail.

48 Karl Friday

41. See, for example, Sandai jitsuroku 862 5/20, or 883 2/9.42. Shōmonki, vol. 2, Shinsen Nihon koten bunko, edited by Hayashi Rokurō (Tokyo:

Gendai shichō sha, 1975), 101. For more on Masakado, see Friday, Th e First Samurai.43. Friday, Hired Swords, 70– 97. Th e warrior alliances of the Heian and Kama-

kura periods are commonly referred to by Japa nese historians as bushidan (literally, “warrior groups”), but, for several reasons, I fi nd it advisable to avoid use of the term. For details, see Hired Swords, 93– 95.

44. Friday, Hired Swords, 93– 98. Mikael S. Adolphson, Teeth and Claws of the Buddha: Monastic Warriors and Sōhei in Japa nese History (Honolulu: University of Hawaii Press, 2007) off ers a lively analysis of monastic military institutions.

45. Shōmonki, 83; Konjaku monogatarishū 25.5; Heian ibun docs. 372 and 4652; Kamakura ibun, edited by Takeuchi Rizō (Tokyo: Tōkyōdō, 1971), docs. 11115, 12275 (see Th omas Conlan, In Little Need of Divine Intervention: Scrolls of the Mongol Inva-sions of Japan [Ithaca, NY: East Asia Program, Cornell University, 2001], 216, for a translation of this document) and 12276. Both Farris, Heavenly Warriors, 335– 343, and Conlan, Divine Intervention, 261– 264, off er illuminating discussions of the sizes of early medieval fi ghting forces.

46. See, for example, Honchō seiki 941 11/5 (Kokushi taikei [Tokyo: Yoshikawa kōbunkan, 1964]) or Fusō ryakki 940 2/8 (Kokushi taikei [Tokyo: Yoshikawa kōbunkan, 1965]).

47. Th is basic issue has been discussed in other contexts by Endō Yukio, “Shuba no tō no kōdō to seikaku,” in Nihon kodai shi ron’en, ed. Endō Yukio sensei shōju kinen kai (Tokyo: Kokusho kangyōkai, 1983), 13– 15; and by Miyagawa Mitsuru with Cornelius J. Kiley, “From Shōen to Chigyō: Proprietary Lordship and the Structure of Local Power,” in Japan in the Muromachi Age, ed. John W. Hall and Toyoda Takeshi (Berkeley: University of California Press, 1977), 89– 107.

48. Th e so- called seven leagues of Musashi are discussed in detail by Yasuda Mo-tohisa, Bushi sekai no jōmaku (Tokyo: Yoshikawa kōbunkan, 1973), 28– 38; and by Ishii Susumu in Kamakura bushi no jitsuzo (Tokyo: Heibonsha, 1987), 91– 94.

49. Friday, Hired Swords, 112– 113. Some scholars have described warrior alli-ances as analogous to land- commendation agreements (see, for example, Yasuda Motohisa, “Bushidan no keisei,” in Iwanami kōza Nihon rekishi kodai 4 [Tokyo: Iwanami shoten, 1962], 127– 128), but the absence of legal paperwork in the warrior alliances represents a crucial— and fundamental— diff erence between the two phe-nomena. Commendation instruments exist in abundance, but one searches in vain for a single document formalizing a military alliance prior to the agreements issued by Minamoto Yoritomo in the 1180s. As in the case of patron/client relationships be-tween court nobles, a warrior entering the ser vice of another presented his new mas-ter with his name placard (myōbu). Th ere is, however, no evidence that the ju nior


party to the arrangement ever received any written confi rmation in exchange. For examples of warriors off ering myōbu as gestures of submission, see Heian ibun doc. 2467 or Konjaku monogatarishū 25.9.

50. On this point see Jeff rey P. Mass, Warrior Government in Medieval Japan: A Study of the Kamakura Bakufu, Shugo and Jito (New Haven, CT: Yale University Press, 1974), 33– 35, 45– 54.

51. Jeff rey P. Mass, Yoritomo and the Founding of the First Bakufu (Stanford, CA: Stanford University Press, 1999), 13– 62 discusses the vicissitudes of Taira and Mina-moto eff orts to create lasting warrior networks. In- depth treatments of the factors that defi ned the cohesiveness of Heian military networks appear in Friday, Hired Swords, 98– 121; and Friday, Samurai, Warfare and the State, 53– 62.

52. See Jeff rey P. Mass, Th e Development of Kamakura Rule 1180– 1250: A History with Documents (Stanford, CA: Stanford University Press, 1979) and Lordship and In-heritance in Early Medieval Japan: A Study of the Kamakura Sōryō System (Stanford, CA: Stanford University Press, 1989).

ReferencesAdolphson, Mikael S. Teeth and Claws of the Buddha: Monastic Warriors and Sōhei in

Japa nese History. Honolulu: University of Hawaii Press, 2007.Adolphson, Mikael, Edward Kamens, and Stacie Matsumoto. Heian Japan: Centers

and Peripheries. Honolulu: University of Hawaii Press, 2007.Batten, Bruce. “Foreign Th reat and Domestic Reform: Th e Emergence of the Ritsuryō

State.” Monumenta Nipponica 41, no. 2 (1986): 199– 219.Brown, Delmer M., and Inoue Mitsusada. “Th e Century of Reform.” In Ancient Japan,

vol. 1 of Th e Cambridge History of Japan, edited by Delmer M. Brown, 163– 220. New York: Cambridge University Press, 1993.

Endō Yukio. “Shuba no tō no kōdō to seikaku.” In Nihon kodai shi ron’en, edited by Endō Yukio sensei shōju kinen kai. Tokyo: Kokusho kangyōkai, 1983, 3– 17.

Farris, Wayne. Heavenly Warriors: Th e Evolution of Japan’s Military, 500– 1300. Cam-bridge, MA: Harvard University Press, 1992.

Friday, Karl. Th e First Samurai: Th e Life and Legend of the Warrior Rebel, Taira Ma-sakado. New York: John Wiley & Sons, 2008.

Friday, Karl. Hired Swords: Th e Rise of Private Warrior Power in Early Japan. Stanford, CA: Stanford University Press, 1992.

Friday, Karl. “Pushing Beyond the Pale: Th e Yamato Conquest of the Emishi and Northern Japan.” Journal of Japa nese Studies 23, no. 1 (1997): 1– 24.

Friday, Karl. Samurai, Warfare and the State in Early Medieval Japan. London: Rout-ledge, 2004.

50 Karl Friday

Hall, John W. Government and Local Power in Japan, 500– 1700: A Study Based on Bi-zen Province. Prince ton, NJ: Prince ton University Press, 1966.

Hattō Sanae. “Kazoku to kyōdōtai.” Rekishi hyōron, no. 424 (1985): 14– 23.Hattō Sanae. “Sekkanki ni okeru zuryō no ie to kazoku keitai.” Nihon rekishi, no. 442

(1985): 1– 18.Hodate Michihisa. “Kodai makki no tōgoku to ryūjū kizoku.” In Chūsei Tōgokushi no

kenkyū, edited by Chūsei tōgokushi kenkyūkai. Tokyo: Tōkyō daigaku shuppankai, 1988.

Hurst, G. Cameron III. Insei: Abdicated Sovereigns in the Politics of Late Heian Japan, 1086– 1185. New York: Columbia University Press, 1976.

Ishii Susumu. Kamakura bushi no jitsuzo. Tokyo: Heibonsha, 1987, 91– 94.Ishimoda Shō. Kodai makki seijishi jōsetsu. Tokyo: Miraisha, 1964, 43– 44. Ruijū kokushi

798 2/1 (5 vols.), Shintei zoho kokushi taikei. Tokyo: Yoshikawa kobunkan, 1986.Kiley, Cornelius J. “Estate and Property in the Late Heian Period.” In Medieval Japan:

Essays in Institutional History, edited by John W. Hall and Jeff rey P. Mass. New Haven, CT: Yale University Press, 1974.

Kiley, Cornelius J. “Provincial Administration and Land Tenure in Early Heian.” In Heian Japan vol. 2 of Th e Cambridge History of Japan, edited by Donald H. Shively and William H. McCullough. New York: Cambridge University Press, 1999, 236– 340.

Kiley, Cornelius J. “State and Dynasty in Archaic Yamato.” Journal of Japa nese Studies 3, no. 1 (1973): 25– 49.

Kitayama Shigeo. Taira Masakado. Tokyo: Asahi shinbunsha, 1993, 23– 28.Masaaki, Takahashi. Bushi no seiritsu, Bushizō no sōshutsu. Tokyo: Tōkyō daigaku

shuppankai, 1999.Masaaki, Takahashi. Kiyomori izen. Tokyo: Heibonsha, 1984.Mass, Jeff rey P. Th e Development of Kamakura Rule, 1180– 1250: A History with Docu-

ments. Stanford, CA: Stanford University Press, 1979.Mass, Jeff rey P. Lordship and Inheritance in Early Medieval Japan: A Study of the Ka-

makura Sōryō System. Stanford, CA: Stanford University Press, 1989.Mass, Jeff rey P., ed. Th e Origins of Japan’s Medieval World: Courtiers, Clerics, War-

riors, and Peasants in the Fourteenth Century. Stanford, CA: Stanford University Press, 1997.

Mass, Jeff rey P. Warrior Government in Medieval Japan: A Study of the Kamakura Ba-kufu, Shugo and Jito. New Haven, CT: Yale University Press, 1974.

Mass, Jeff rey P. Yoritomo and the Founding of the First Bakufu. Stanford, CA: Stanford University Press, 1999.

McCullough, William H. “Th e Heian Court, 794– 1070.” In Heian Japan, vol. 2 of Th e Cambridge History of Japan, edited by Donald H. Shively and William H. Mc-Cullough. New York: Cambridge University Press, 1999, 20– 96.


McCullough, William H. “Japa nese Marriage.” Harvard Journal of Asian Studies, no. 27 (1967): 103– 167.

Miyagawa Mitsuru, with Cornelius J. Kiley. “From Shōen to Chigyō: Proprietary Lord-ship and the Structure of Local Power.” In Japan in the Muromachi Age, edited by John W. Hall and Toyoda Takeshi. Berkeley: University of California Press, 1977, 89– 107.

Mo rita Tei. Zuryō. Tokyo: Kyōikusha, 1978, 50– 52, 107– 117, 136– 138.Morris, Dana Robert. “Land and Society.” In Heian Japan, vol. 2 of Th e Cambridge

History of Japan, edited by Donald H. Shively and William H. McCullough. New York: Cambridge University Press, 1999.

Morris, Dana Robert. “Peasant Economy in Early Japan, 650– 950.” Ph.D. dissertation, University of California, Berkeley, 1980.

Morris, Ivan. “Marriage in the World of Genji.” Asia, no. 11 (1968): 54– 77.Motohisa, Yasuda. Bushi sekai no jōmaku. Tokyo: Yoshikawa kōbunkan, 1973.Motogi Yasuo. Minamoto No Mitsunaka. Yorimitsu: sassei hōitsu Tomoie no shugo.

Tokyo: Meburuva shobō, 2004.Nickerson, Peter. “Th e Meaning of Matrilocality: Kinship, Property, and Politics in

Mid- Heian.” Monumenta Nipponica 48, no. 4 (1993): 429– 468.Noguchi Minoru. Genji to bandō bushi. Tokyo: Yoshikawa kōbunkan, 2007.Oboroya Hisashi. Seiwa Genji. Tokyo: Kyōikusha, 1984.Piggott, Joan R., ed. Capital and Countryside in Japan, 300– 1180: Japa nese Historians

Interpreted in En glish. Ithaca, NY: Cornell University, East Asia Series, 2006.Piggott, Joan R. Th e Emergence of Japa nese Kingship. Stanford, CA: Stanford Univer-

sity Press, 1997.Sekiguchi Hiroko. “Kodai kazoku to kon’in keitai.” In Koza Nihonshi 2, edited by Rek-

ishigaku kenkyūkai and Nihonshi kenkyūkai. Tokyo: Tokyo daigaku shuppankai, 1984: 287– 326.

Souyri, Pierre François. Th e World Turned Upside Down: Medieval Japa nese Society. New York: Columbia University Press, 2001.

Tabata Yasuko. “Kodai, chūsei no ’ie’ to kazoku: yōshi o chūshin to shite.” Tachibana jōshi daigaku kenkyū kiyō, no. 12 (1985): 41– 67.

Takahashi Masaaki. “Masakado no ran no hyōka o megutte.” Bunka shigaku, no. 26 (1971): 25– 44.

Torao Toshiya. “Nara Economic and Social Institutions.” In Ancient Japan, vol. 1 of Th e Cambridge History of Japan, edited by Delmer M. Brown. Cambridge: Cambridge University Press, 1993, 415– 452.

Toyohiko, Fukuda. Tōgoku heiran to mononofu- tachi. Tokyo: Yoshikawa kōbunkan, 1995.

Wakita Haruko. “Marriage and Property from the Perspective of Women’s History.” Journal of Japa nese Studies 1, no. 2 (1984): 321– 345.

52 Karl Friday

Yasuda Motohisa. “Bushidan no keisei.” In Iwanami kōza Nihon rekishi kodai 4. Tokyo: Iwanami shoten, 1962, 132– 60.

Yoneda Yūsuke. Kodai kokka to chihō gōzoku. Tokyo: Kyōikusha, 1979, 115– 118.Yoshie Akio. “Shōki chūsei sonraku no keisei.” In Kōza Nihonshi, vol. 2, edited by

Rekishigaku kenkyūkai and Nihonshi kenkyūkai. Tokyo: Tokyo daigaku shup-pankai, 1970, 105– 130.

53

Introduction

It is hard for modern Japa nese, accustomed to po liti cal and social stability, to imagine the chaos that characterized life in medieval Japan. Th is is particu-larly true for the Warring States (sengoku) period from 1467, when the Ōnin War broke out, to 1615, when the Toyotomi family was ruined in the Battle of Summer at the Osaka castle and the peace of the Tokugawa was consolidated. But it was true for some time before full- fl edged civil war broke out as well.

Th e increasing economic self- suffi ciency and autonomy of the peasants eroded control by local lords and eventually led to an amalgamation of estates into larger territorial units throughout the Japa nese archipelago. In par tic u-lar, the archetypal models of the daimyō domains were fi rst established pri-marily in the Tokai and Chugoku areas, which are not too far from Kyoto. In more distant areas, the rivalry of local estate own ers (kokunin) persisted for a long time, probably because peasants were more deeply reliant on materials supplied by estate own ers. In contrast, in areas surrounding Kyoto, there were robust network grassroots pacts among local residents (kokunin ikki) that re-sisted intrusion by centralized po liti cal leadership. Interestingly enough, the consolidation into territorial domains took longer to develop in both the hin-terlands and in Kyoto’s vicinity.

Th e territorial extension of daimyō rule intensifi ed rivalry among them-selves. But fi ghting at every opportunity was not always the best course of ac-tion for staying in the game. Daimyō oft en shift ed their alliance partners and redrew their territorial boundaries as a way to keep opponents off balance. On occasion, one side exterminated the other, but in most cases the evolutionary

Competence over LoyaltyLords and Retainers in Medieval Japan

Susumu Ike

3

54 Susumu Ike

pro cess of winnowing out the weak led to per sis tent rivalry among those who survived. In due course, rivalry among powerful domains gradually engulfed the hinterlands as well.

Th is chapter explores the dramatic disruptions in po liti cal power that rocked the world of medieval inhabitants, and how those shift s shaped the relationships between lords and their retainers. Th e next section reviews the historical development of the territorial domains and their relations with the central authority, starting from the transition to the Kamakura shogunate (1185– 1333) until the collapse of the Ashikaga shogunate in 1573. Th e third sec-tion documents the weak bonds of loyalty between the local estate own ers and their retainers, contrary to more romanticized views of medieval Japan. Th is is followed by an analysis of emerging local territorial states in the late medi-eval period, as exemplifi ed by Uesugis in Echigo. Th e fi ft h section examines the pro cess of territorial consolidation during the Warring States (sengoku) period. As local domains were gradually consolidated, the nature of military confl icts shift ed from internal strife to battles among domains. Th e sixth sec-tion explores conditions that eventually contributed to Toyotomi Hideyoshi’s unifi cation of Japan. Th e concluding section discusses changes and continuity between the Warring States period and the Tokugawa shogunate.

Kamakura and the Search for Order

Even during the heyday of the imperial court in Kyoto, the central government had only minimal power to prevent private wars (shisen) among neighboring domains over border lands or within families over succession. A typical exam-ple was the Shohei- Tenkei War in the tenth century, more commonly known as the Taira- no Masakado’s Rebellion. It started as an internal battle within the Taira clan in the frontier region in Kanto far to the east, and the central govern-ment did not intervene in the battle while Masakado fought against his uncles, Kunika and Yoshikane. However, Masakado attacked a kokuga (the govern-ment’s provincial offi ce) of the province of Hitachi (now, the prefecture of Ibaragi) and began to call himself a new emperor. Only then did the govern-ment regard Masakado as a rebel. Unable to dispatch a sizable army of its own to quell the rebellion, the emperor put a bounty on Masakado’s head. Two months later, Masakado’s cousin Sadamori, whose father Masakado had killed, defeated Masakado in battle and took his head to Kyoto to receive his reward.

In the dying years of the imperial court, between the twelft h and four-teenth centuries, warrior clans fought each other for de facto control of Japan.

Competence over Loyalty 55

Th e Minamoto’s defeat of the Taira clan in 1185 established the Minamoto family as the offi cial protectors of the court. Although the emperor hardly had a choice, he granted Minamoto no Yoritomo the right to appoint his retainers, or gokenin (“house men”) as military governors (shugo) in the provinces and military stewards ( jito) in both public and private landed estates. Th us began the Kamakura shogunate (1185– 1333), in which the Minamoto family and their Hojo regents ruled Japan from the eastern city of Kamakura far from Kyoto, near today’s Tokyo.

In 1221, some two de cades aft er Yoritomo’s death, the imperial court raised an army to challenge the Hojo regents who were ruling in the Minamoto name. Th e Hojo successfully defended their control of the Kamakura shogu-nate in a military contest against the imperial troops. Upon achieving mili-tary victory, the Hojo seized lands from the vanquished nobles and imperial forces and redistributed them to their own vassals, consolidating their control of the Kamakura government.

In order to solve confl icts peacefully, the Kamakura shogunate established a judicial system. Because of limited capacity, however, it relied entirely on self- reporting. Th is is the source of the proverb to the eff ect that “Even if a dead body lies down in front of the police offi ce or the court, the police will not take any action unless someone specifi es a criminal and accuses him to the police” ( gokuzen no shinin uttaenakunba kendannashi).

Since criminal cases were treated in this rather offh and way, it is not dif-fi cult to imagine how unsatisfactorily many civil cases were handled. Ac-cording to studies on a history of the judicial system, the lawsuit procedures were as follows. A plaintiff submitted a letter of complaint. If the letter was accepted, the shogunate issued a letter of question to order a defendant to defend himself. Th e defendant then submitted a letter of petition for de-fense. Th is pro cess was repeated three times. Th en, the two confronted each other face to face and, aft er that, a sentence (saikyo) was fi nally handed down. Th e system seems reasonable if these procedures were actually fol-lowed. However, it was the task of the plaintiff , and not the court, to send a defendant a letter of question. In addition, when either a plaintiff or a de-fendant referred to laws enacted by the shogunate in order to justify his po-sition, he had to prove that those laws were really enacted. More problemati-cally still, even if a sentence was pronounced, the shogunate did not have a system of enforcement. Th us, the prevailing party in a lawsuit could only claim the legitimacy of his right, based on a letter of sentence (saikyo- jo) issued by

56 Susumu Ike

the shogunate. In order to recover his right in practice, he had to resort to force.

Th e shogunate took seriously threats to its regime, but played a limited role in maintaining social order in the Kamakura period, doing little to inter-vene in confl icts between private parties. Wars of attrition among disputants would have become common if not for the resort to mediation by powerful individuals (chunin- sei). However, mediation oft en failed. Th e Soga brothers, Goro and Juro, for example, murdered Kudo Suketsune in retaliation for kill-ing their father at a hunting game hosted by Minamoto- no Yoritomo at the foot of Mount Fuji. In praising the avenging brothers, Yoritomo endorsed the widely practiced custom of revenge similar to the fehde in medieval Eu rope.

Th e Kamakura shogun recognized as his retainers (gokenin) the lords of local estates who pledged fealty to him. Th e shogun guaranteed the local lord’s land rights either by assigning him the position of steward of an es-tate ( jito- shiki) or by ratifying his position of estate offi cer already assigned by the estate own ers (sokan- shiki). Th e Kamakura shogunate typically pre-served existing assignments in these positions, provided they were loyal to the regime.

Given the weakness of centrally supplied law and order, local lords and their bands of warriors took matters into their own hands. Warrior retainers were offi cially part of the master’s house hold rather than owning their own estates and formed a kinship- like band for self- protection. Th e master’s house hold di-rectly managed dozens of hectares of lands, using not only slave- like direct la-borers (genin) but also neighboring peasants of very small operating units (menka). Furthermore, house holds had workshops for food, clothes, and agri-cultural tools, and were highly self- suffi cient. In the unit of the house hold, kin members and retainers controlled tenant workers and protected landed estates. Local social order was maintained through the coexistence of these large oper-ating units, which also placed neighboring peasants under their control and protection.

A master controlled his house hold, and even the government could not easily intervene in its internal aff airs. For example, Article 26 of the Goseibai- shikimoku (Jōei Formulary) enacted by the Kamakura shogunate recognized a master’s right to his land (honryo) and his right to pass it along to his chil-dren. A parent’s intention was given priority over that of the shogun.

In 1333, the Hojo family lost its grip on the Kamakura shogunate, in part because expensive preparations against Mongolian invasions had consumed


enormous resources. Central tax revenues were minimal, forcing the shogu-nate to rely on the loyalty and support of its vassals, but the Hojo had no more lands with which to reward the local lords for their expenditures and work. With the shogunate in this weakened state, Emperor Go Daigo declared war on the shogunate, and his general, Nitta Yoshisada, raised an army that de-stroyed the Kamakura forces. In spectacular battles fought over the next few years, the emperor’s and clan armies vied for hegemony. Both sides were so weakened by the eff ort that in the end Ashikaga Takauji, a former Kamakura vassal, emerged from the shadows and established a new shogunate under his own name.

Th e Ashikaga government gave local jurisdiction to military generals (shugo), who became daimyō or domain lords. Th ese local lords were given authority to punish illegal activities such as harvesting someone else’s rice plants (karida- rozeki). Th e government also established a law called kosenbosen- ho to restrict private wars. Th e term kosen means initiating a fi ght, while the term bosen means protecting oneself against it by force. In the beginning, only the aggressors were punished. In time, the daimyō agreed to the kenka- ryōseibai- hō under which both aggressors and defenders were regarded with equal severity and private wars were prohibited altogether.

At the same time, grassroots movements emerged within local society to resolve confl icts peacefully. Local residents might form a pact (kokujin- ikki) to solve territorial confl icts, particularly regarding territorial boundaries, and troubles in markets and towns, through peaceful discussions rather than with force.

Th ese local pacts emerged as a consequence of the breakdown of the house hold system. Early pacts were oft en negotiated at the instigation of lo-cal lords attempting to re unite a clan against the growth and in de pen dence of branch families in the clan. But for several reasons, local social order be-came more urgent, leading to regionwide pacts. Th e in de pen dence of branch families entailed the division of landed estates, disrupting sharecropping ar-rangements. Moreover, with the development of new lands peasants became increasingly self- supporting and dissatisfi ed with labor duties on a lord’s estate. As the estate system loosened, stipends became attached to the lands them-selves rather than to the offi cial positions (shiki) associated with protecting the land. Estate own ers began to accept farmers’ right to farm in exchange for land taxes (nengu). Peasants began to purchase commercially available goods such as agricultural tools rather than rely on materials supplied by the lord. Small

58 Susumu Ike

farmers began to thrive within local society, and the problem of boundaries among operating units became increasingly serious. In addition, the develop-ment of commerce increased the stake in smoothly functioning markets. Th e formation of communitywide social pacts arose in response to these problems.

Ultimately, locally initiated attempts to maintain peace failed. Among domain lords, the temptation to take advantage of a neighbor’s weakness proved too great. Within domains, impending threats from neighbors desta-bilized succession plans that kin and retainers feared would fail to put the strongest heir in place as leader. Th e result was full- scale civil war (the Ōnin War) between 1467 and 1477, and continued jostling among domains for an-other hundred years aft er that.

The Warring States Period

Aft er the Ōnin Civil War, the shogunate’s control became limited to the Kinai region around Kyoto and its neighboring provinces. In the other regions, the vassal relationship between the shogun and local lords disappeared. Although some retainers still responded to the shogun’s request for military mobiliza-tion, their willingness to serve was not based on their loyalty to him or an expectation of acquiring land as a reward, but for other reasons such as secur-ing the po liti cal leadership in local society or securing the right to trade with China. In the Kyoto region and its neighboring provinces, the powerful Hoso-kawa family gained control of the region, while direct retainers of the shogun (hokoshu) remained loyal to him in order to have him protect their land rights. Except for very close retainers, loyalty to the shogun evaporated once the shogun lost the power to guarantee land own ership.

An even more important change was the breakdown of the house hold (so-ryo) system of bequeathing landed estates to kin members and retainers. With the chaos of civil war, relatives and warrior retainers remained loyal to a house hold head only as long as he could protect their lands, rather than risk all through unconditional loyalty. As retainers began to intervene in the se-lection of a new master at the time of house hold succession, either through council meetings or by resorting to force, the lord- retainer relationship became more mutual. Fights for house hold succession became common, but retainers had considerable say in the outcome because they shared a desire for someone with the ability to provide peace and security.

Despairing of protecting their lands alone, a growing number of warrior lords turned to emerging warrior leaders known as kokushu, or Warring


States daimyō. A master’s ability to protect land rights was an important con-dition for attracting retainers, which must explain how rarely one encounters retainers expressing feelings of guilt for abandoning a master who lost his ability to protect land rights. Sometimes, masters (daimyō) and retainers were ordered to leave their ruling regions together for new regions. Less fre-quently, close retainers died together with their masters in battles between Sengoku daimyō. Th ere is the famous story dating to 1701, when a master of the Asano family of Ako in the province of Banshu (now, the prefecture of Hyogo) was ruined, leaving his retainers without land or possessions. Legend has it that forty- seven of the retainers committed collective suicide aft er avenging their dispossessed lord by killing his rival. Th e story has lasting appeal, per-haps because a show of this kind of loyalty was so rare. Far more oft en, re-tainers surrendered to the victors in order to secure their land rights.

The Emergence of Local Territorial States

During the Warring States period, the failure of local estates to keep the peace resulted in the emergence of larger po liti cal units controlled by domain lords. An example of the frustrations with the status quo comes from Echigo (now, the prefecture of Niigata) in the early sixteenth century. Uesugi, the military governor of Echigo, had a strong connection to the Ashikaga shogunate, since the mother of Ashikaga Takauji came from the Uesugi family. While Uesugi resided in Kyoto and participated in the central government, he was titular ruler of his provincial domain. Because of Uesugi’s absentee rulership, local lords were highly in de pen dent, and local social order was maintained by agreements among neighboring lords and their mediation (kinjo- no- gi). Not surprisingly, this local practice sometimes broke down. Occasionally, territo-rial confl ict occurred between bordering lords, and one party would invade the other’s territory. Alternatively, a neighboring lord might intervene in a dispute over fi shery rights, but his mediation could fail if it was favorable to the one party over the other. Confl icts among neighboring lords were some-times unyielding to local mediation altogether.

Other evidence of local disharmony comes from the province of Suruga, now the prefecture of Shizuoka. Th e Imagawa kanamokuroku was a code of laws, enacted by the Imagawa, Sengoku daimyō of the province of Suruga in 1526. Imagawa is commonly depicted as a weak daimyō because the clan rap-idly came to ruin aft er Oda Nobunaga defeated Imagawa Yoshimoto in the Battle of Okehazama. However, the Imagawa code of laws was one of the most

60 Susumu Ike

comprehensive and sophisticated among Sengoku daimyō law codes. It in-cludes many articles regarding the settlement of confl icts through judicial pro cess.

Article 1 concerns the rights of own ership and management associated with cultivated lands, and it is important for studying social structure during that historical period. Article 2 deals with confl icts on boundaries among fi elds and mountains and prescribes severe punishment to the party losing a lawsuit. Th e punishment was confi scation of a third of the party’s land. Later, retainers claimed that the punishment was too harsh, and it was soft -ened. But the prescription of such severe punishment shows how seriously boundary disputes were taken. Article 3 focuses on redevelopment of barren land destroyed by fl ooding and other natural disasters (kawanari, uminari no chi). Th is indicates that the extension of development of barren land was one of the major causes of boundary confl icts. Th e house laws of the Yuki (the Yuki- shi shinhatto) of the province of Shimosa (now, the prefecture of Ibaragi), enacted in 1556, also includes an article about a boundary confl ict and men-tions that development of fi elds and mountains from two sides, by extending to the boundary, could cause a boundary confl ict (Article 58).

Boundary confl icts of the day were closely related to the entrepreneurial cultivation of new lands by ordinary peasants. Where property rights or com-mon usage were not well established, competing claims to lands ensued. In order to maintain and expand their territories, local lords competed with one another to protect peasants living within their borders.

To see how the demand for local social order generated new local public authority, consider the case of Mori, a kokujin lord of the province of Aki (now, the prefecture of Hiroshima). Th e Mori house hold rapidly expanded its territory from the mid- sixteenth century on and grew into the daimyō of the Chugoku region. Before then, however, Mori was only a county- level lord, with his authority based on local social pacts. According to a letter of oath submitted in 1532, thirty- two in de pen dent lords with their own retainers swore allegiance to Mori Motonari. Confl icts among them were to be solved cooperatively, and the authority of the master was limited to punishing off enders.

Mori’s hand was strengthened in 1550 when he expelled the Inoue clan from the pact for invading a neighbor’s territory (the Inoue- shu chubatsu- jiken). Following the incident, 237 retainers submitted a written oath to Mori, in which they agreed that confl icts among them would be subject to Mori’s jurisdiction. Th ey forswore territorial expansion and confl ict resolu-


tion by self- redress. Th e expulsion of the Inoue clan succeeded because many of the retainers hoped that Mori’s fi rm hand could rein in cheaters in a way that their horizontal pacts had failed to do. Local lords elevated Mori to the position of a powerful regional lord, in hopes that a central authority would be more eff ective in providing social order than the prevailing system of self- help and vigilante justice.

Consolidation became a trend as in region aft er region, a dominant family arose. Th e Uesugi (later, the Nagao) managed Echigo, the Imagawa managed Suruga, and this pattern was replicated in many provinces. Th e Imagawa, for example, punished both sides of a territorial squabble with capital punish-ment. Th e Imagawa code also included detailed rules regarding the judicial system and clearly demonstrated a trend toward fi rm rule by the regional pub-lic authority.

War in Medieval Japan

Th e proximate causes of the Ōnin War (1467– 1477) were disputes over house hold succession in the Shiba and Hatakeyama families, both powerful military gov-ernors. By the fi ft eenth century, the Ashikaga shogunate was internally di-vided and could not intervene eff ectively in the succession confl icts of re-tainer families. As luck would have it, the shogun family itself was internally divided over the selection of an heir. Furthermore, Hosokawa and Yamana, who were called shukuro and had important policymaking roles in the shogu-nate, were at loggerheads. Th e po liti cal system of the shogunate was dysfunc-tional for all practical purposes, and confl icts could be solved only by force. Th e Ōnin War ended when Hosokawa Katsumoto and Yamana Sozen died and their respective heirs chose to settle up, at least temporarily. But the underly-ing structural problem of widespread incentives to grab what ever they could from neighbors remained unsolved.

Because of the general anxiety about social peace, retainers of powerful families had become actively involved in their masters’ house hold succession. Furthermore, the positions they took were oft en complicated by their own stakes in various territorial disputes. Th is situation oft en made succession problems intractable, and the recognition of a house hold heir by the shog-unate authority could no longer break logjams.

Aft er the Ōnin War, the Hosokawa and Hatakeyama continued to fi ght each other for lordship over the Kinai region. Amazingly, in 1493, twenty years aft er the end of the Ōnin War, Hosokawa Masamoto, a powerful regent

62 Susumu Ike

of the shogunate, discharged the shogun himself. In the pro cess, Hosokawa began to consolidate his control over the Kinai region. Th e same process— regional consolidation— continued in the rest of Japan, with the Hojo moving in on the southern Kanto region.

Th e formation of a new local public authority brought a change from the existing relatively egalitarian order as seen in the local social pacts (kokujin ikki) to vertical order by a strong authority. In some places, re sis tance to such authority emerged, and such re sis tance was forcefully quelled. Th e example of the expulsion of the Inoue clan by Mori of the province of Aki was typical. Although this incident did not develop into an internal war, there were indeed many retainers in Mori’s band of retainers who supported the Inoue clan. On numerous occasions, internal wars broke out. Nagao Tamekage, a lieutenant governor of the province of Echigo, defeated Governor Uesugi in battle but faced a rebellion by local lords who objected to his power play. Date Uemune of the province of Mutsu (now Fukushima prefecture) provoked a revolt by a band of retainers when he attempted to reinforce his control. Harunobu (later Takeda Shingen) revolted against his father, Takeda Nobutora of the province of Kai (now, the prefecture of Yamanashi), when Nobutora sought to expand his power.

Although a demand for peace favored territorial consolidation, the new concentration of power generated a power struggle for domainal control. Th e primary locus of this struggle was among a master, his sons, and his brothers. Th e result was frequent rebellion, or gekokujo, in which a retainer took power from his master. Competent and powerful administrators were favored for their ability to keep social order, but the struggle for house hold control pro-vided frequent tests of this ability.

Succession struggles and regional consolidation interplayed in complex ways. In the struggle between Uesugi and his lieutenant Nagao for Echigo, retainers who were experiencing territorial disputes with one another lined up on opposite sides. Ultimately, however, everyone wanted to be on the win-ning side to keep their land, and so alliances could switch quickly as victory came into view.

With territorial consolidation, warrior lords were prohibited from solving confl icts by force and could not expand their lands by invading their neigh-boring lands within their domains. Th e next stage in Japan’s Warring States period began in the second half of the sixteenth century when the newly con-solidated domains began attacking each other.


Th e written oath that Mori’s band of retainers submitted aft er the purge of the Inoue clan contained many articles concerning war. Th ey prescribed war readiness and a pledge to serve bravely in battle as needed. Th ey also included a procedure by which a soldier could appeal to high- ranking retainers in the event he felt his reward for ser vice in battle was insuffi cient. Th e oath was quite explicit about the connection between ser vice and reward. It was a con-tract between Mori and his band of retainers to unite toward external ex-pansion through war, and it clearly showed rewards such as lands to be given for mobilization. In addition, they also imposed martial law, which prohibited soldiers from disobeying commanders’ orders in war and from retreating from a battlefi eld without permission.

In 1555, fi ve years aft er the purge of the Inoue clan, Mori fought in the Is-land of Itsukushima in the province of Aki against the Sue who governed the province of Suo (now, the prefecture of Yamaguchi) and a couple of its neigh-boring provinces. Sue had 20,000 men in arms compared with Mori’s 3,000. Although Mori’s army was vastly inferior, they defeated the Sue with the help a surprise attack at night, whereupon General Sue Harutaka committed suicide. Mori’s army was an or ga nized army with high morale, giving Mori the plat-form to become a powerful daimyō with great infl uence over the Chugoku and northern Kyushu regions. Mori’s band of retainers also increased their lands due to rewards and became so arrogant that their master, Mori Motonari, de-plored their change in character.

As territory became consolidated, the nature of war also changed from internal struggles to interdomainal war. Th e Battle of Kawanakajima, the most famous military struggle in the entire Warring States period, was such a case. Th e Uesugi of the province of Echigo fought the Takeda of the province of Kai over a period from 1553 to 1564 in Kawanakajima of the northern Shinano (now the prefecture of Nagano). Local lords in the northern Shinano asked Uesugi to help defend them against Takeda who, having achieved control of Kai, was thought to have his eyes set on Shinano. Th e Battle of Okehazama in 1560 was a defensive victory by Oda Nobunaga of the province of Owari against Imagawa Yoshimoto of the province of Suruga, who advanced to Owari through the province of Mikawa (both now the prefecture of Aichi).

Because of these battles for territorial expansion, neutral zones among re-gional lords began to disappear and domainal boundaries became well de-marcated. In some cases, one side destroyed the other, while, in other cases, a kind of balance of power was established and a division of domains called

64 Susumu Ike

kuniwake was agreed upon. In the Eastern region, po liti cal and military alli-ances were created among the Imagawa, Takeda, Hojo, and Uesugi. Later, they were called the koso- domei (the Takeda and Hojo alliance), etsuso- domei (the Uesugi- Hojo alliance), and sangoku- domei (the Takeda, Imagawa, and Hojo alliance). Broad co ali tions set the groundwork for even bigger wars.

National Unity

In the Kinai region, the Ashikaga shogunate still existed. In order to restore his authority, the shogun tried to intervene in territorial confl icts between the daimyō and arranged temporary truces between Mori and Amako of the prov-ince of Izumo (now the prefecture of Shimane) and Mori and Ōtomo of the province of Bungo (now the prefecture of Oita). But ultimately, the shogun lacked enforcement power. A peace agreement was easily abandoned as soon as a po liti cal situation changed, and the Amako was destroyed by Mori.

It was Oda Nobunaga’s conquest of the Kinai region and its neighboring provinces that disrupted the delicate equilibrium among domains. Nobunaga advanced to Kyoto on the pretense of helping Ashikaga Yoshiaki to enter Kyoto, and along the way, crushed the Miyoshi, the Rokkaku and Asai of the prov-ince of Omi (now, the prefecture of Shiga), and the Asakura of the province of Echizen. Th e shogun, alarmed at Nobunaga’s power move, exiled him from Kyoto. Undaunted, Nobunaga went on to defeat powerful groups of resisters, including monasteries arrayed against him. He also defeated the Takeda of the province of Kai, which had thrown its forces behind the shogun to stop Nobunaga. Oda Nobunaga thus established a vast domain, which extended northward to the provinces of Etchu (now the prefecture of Toyama) and Noto (now the prefecture of Ishikawa), eastward to the provinces of Suruga, Kai, and Shinano, southward to the province of Kii (now, the prefecture of Wakayama), and westward to the provinces of Bizen (now the prefecture of Okayama) and Inaba (now the prefecture of Tottori).

Several reasons have been proposed explaining why Oda Nobunaga achieved this magnitude of success, including his military talent, victory in the Battle of Okehazama, and accidents such as the death of Takeda Shingen, his most threatening rival. It has also been argued that Nobunaga managed to unite the warrior class through his direct confrontation with pop u lar movements such as ikko- ikki. Oda Nobunaga also trained warrior specialists, separating them from peasants whom he left to till the land. However, Oda was merely follow-ing a practice that daimyō lords before him had initiated. More original to Oda Nobunaga was his logistical move of controlling an eco nom ical ly advanced


region like Kinai. In fact, immediately aft er entering Kyoto, Oda Nobunaga placed Sakai, the center of commerce, under his control and strengthened ties with wealthy merchants. Wealthy merchants supported Oda Nobunaga (and Toyotomi Hideyoshi, later) and played a positive role in purchasing goods as his representative offi cers. Merchants supported the unifi ers because they valued po liti cal stability in a broad region since it increased their profi t from trade with remote areas. Particularly, in those days, the problem of currency was serious and hindered the smooth trade of goods and currency. Merchants desired central authority over a large geographic area for sound currency circulation for their trade.

Oda Nobunaga destroyed hostile daimyō, gave lands to his close retainers such as Shibata Katsuie and Hashiba Hideyoshi, and appointed them rulers of larger regions. However, he did not allow them to establish a lord and vassal rela-tionship with local lords in their ruling regions. Instead, he or ga nized those lords as his direct retainers. He created a daimyō territorial domain over the Kinai region and its neighboring provinces, even though it was very large.

Oda Nobunaga was killed by a disgruntled retainer at Honno- ji in 1588. Th e unifi cation of the nation was taken over by Hashiba (Toyotomi) Hideyo-shi, who became Oda Nobunaga’s successor. Hideyoshi conquered Chugoku, Shikoku, Kyushu, Kanto, and Ou and completed the unifi cation of the nation. He was successful in this eff ort largely because of his ability to mobilize enor-mous armies with the resources acquired from taxes and loans from the Kinai region. Toyotomi raised armies on the order of a hundred thousand men, and secured armor and supplies with the help of wealthy merchants. Th e Hojo, whom Hideyoshi vanquished, were good at the conventional strategy of hold-ing a castle. Hideyoshi’s army transported a large amount of rice from the Kinai and Tokai regions and besieged the Hojo’s castle with two hundred thou-sand soldiers for more than three months. Th e Hojo fi nally surrendered when they were on the point of starvation.

Like daimyō before him, Toyotomi Hideyoshi prohibited lords under his rule from using force to solve territorial confl icts between them, and he or-dered them to obey the judgment of judicial offi cials. Both the Shimazu of the province of Satsuma (now, the prefecture of Kagoshima) and the Hojo, who became the targets of Hideyoshi’s suppression, resorted to arms in territorial confl icts at that time, justifying his continued use of force against them. When he defeated them, Toyotomi rewarded his warriors with the newly ac-quired lands, creating the so- called Toyotomi daimyō including Ishida, Fuku-shima, and Kato. Th e Toyotomi regime recognized each daimyō’s territorial

66 Susumu Ike

rights, including those of the Shimazu who surrendered. But for those who did not surrender, Toyotomi’s power was suffi cient to require compliance. Toyotomi Hideyoshi failed to conquer Korea and died of illness, aft er which the government became unstable. But Tokugawa Iyeyasu’s conquest in 1600 and his regime’s consolidation in 1615 fi nished a job that Oda Nobunaga and Toyotomi Hideyoshi had nearly completed themselves.

Conclusion

Competition among local lords resulted in the territorial consolidation of Ja-pan. But the question that arises is why rivalry among the territorial domains was not sustainable over a long period. Why did the system in which hetero-geneous forms rule and authority coexisted fail to endure? Why did the strug-gle among warlords eventually lead to unifi cation and the establishment of the Tokugawa shogunate?

Some historians have argued that the microcosm of a nation already estab-lished at the domain level provided foundations of legitimacy for national unifi -cation by Oda Nobunaga, Toyotomi Hideyoshi, and Tokugawa Ieyasu. Not only is the ontology of this nation- state logic suspicious, but it is also unclear why the warlords’ legitimacy would extend directly to the national level. Other historians have pointed instead to Toyotomi Hideyoshi’s overweening ambition and success. Not only did Toyotomi vanquish his enemies, but he also launched his armies on a large- scale if ultimately unsuccessful military invasion of Korea, a clue to the scope of his ambition. But ambition alone— or even ambition, ruth-lessness, and singleness of purpose— is unlikely to have divided Toyotomi from other men. Th e picture comes into clearer focus when we notice not only the men of military ambition, but the vast demand for peace among the population of Japan as well. Eventually, even in Japan’s hinterland, the demand for peace trumped the clientelistic relations between estate own ers and peasants, creating fewer but larger territories. Even as Hideyoshi’s sword hunt of 1588 helped to ce-ment his primacy as a warlord, it also managed to ramp down the levels of vio-lence to which the Japa nese farmers had long been subjected.

Notes1. Susumu Ike, Daimyō Ryokokusei no kenkyu [Research on the Territorial Sys-

tem of Daimyō], in Daimyō- ryogoku- sei no Kenkyu [Studies on the Daimyō Domain System] (Tokyo: Azekura shobo, 1995), pp. 66– 68.


2. Aft er Yoritomo died in 1199, eff ective power in the shogunate passed into the hands of the Hojo family, who were the family of Yoritomo’s wife, Masako.

3. Th is episode is known as the Jokyu Disturbance.4. Th is incident was conveyed to people later as the Soga story.5. An early fourteenth- century document called the Sata- miren- sho included in vol.

25 of Zokugunshorui- ju explains the lawsuit procedures of the Kamakura shogunate.6. Th is right was technically possessed by the king of the nation, that is, the

shogun.7. Soryo- sei teki bushidan, in which kin members and retainers (ie- no- ko, roto)

were or ga nized under a master called soryo.8. Th e Meiji government, which ostensibly came to power in 1868 on a platform of

imperial restoration, made national heroes of the imperial generals, Nitta Yoshisada, Kusunoki Masashige, and Kitabatake Akiie, who battled shogunate forces on behalf of the emperor. Th e Meiji government also put its ex post stamp of approval on Em-peror Go Daigo, who refused to acquiesce to Ashikaga’s rule. In defi ance of Ashikaga and his chosen emperor Komyo, Go Daigo established a rebel “Southern Court” in the mountains of Yoshino, but he was by then a paper tiger. In telling symbolism, Ashi-kaga moved the capital back to Kyoto and built for himself an opulent castle that dwarfed the emperor’s.

9. Th e term ikki originally means unifying the ki (michi wo hitotsu nisuru) and thus refers to a group of people with the same objective or their united act to pursue that ob-jective. Th e kokujin means a person who resides in the province. Putting these terms together, the kokujin ikki was an equal- basis- unifi cation of local lords based on the regional relations.

10. See Jeff rey Mass, Lordship and Inheritance in Early Medieval Japan: A Study of the Kamakura Soryo System (Stanford, CA: Stanford University Press, 1989). For the term kerai, the Chinese characters were commonly used from the Muromachi period on.

11. In those days, this ability was called kiryo.12. Due to the separation of the warrior and peasant classes, the relationship be-

tween a warrior lord and local society became weak, and only those warrior lords who were incorporated into the hierarchical lord and vassal system with the shogun as its top could be given feudal land rights.

13. Th e story was memorialized in the kabuki play, Kanedehon Chūshingura, and later in the classic 1941 Mizoguchi movie, Th e Forty- Seven Samurai. Th e war time gov-ernment must also have liked the emphasis on self- sacrifi ce.

14. Susumu Ike, Daimyō- ryogoku Keisei- ki ni Okeru Kokujin- so no Doko [Koku-jin Lords in the Formation of the Daimyō Domain System], Chap. 2 of Part 3 of Daimyōōryogoku- sei no Kenkyu [Studies on the Daimyō Domain System] (Tokyo: Azekura shobo, 1995).

68 Susumu Ike

15. Chusei Seiji- shakai- shiso Ge [Social and Po liti cal Th oughts in the Medieval Period, No. 2], vol. 21 of Nihon Shiso Taikei [A History of Japa nese Th oughts] (Tokyo, Japan: Iwanami shoten, 1972). Article 15 deals with a confl ict caused by a newly created irrigation channel that had to run through someone’s property. Irrigation water is es-sential for rice fi elds, and this is also a confl ict related to development.

16. Dai- nihon Kobunsho Mori- ke Bunsho [Japa nese Old Documents: Th e Docu-ments of the Mori Family], No. 396.

17. Ibid.18. Mori Motonari Inoue- shu Zaijo- sho (Mori Motonari’s letter about the crime

of the Inoue clan). In Dai- nihon Kobunsho Mori- ke Bunsho [Japa nese Old Documents: Th e Documents of the Mori Family], No. 398.


20. Th e original statement is “rihi wo ronzezu,” meaning whether or not there is a reason does not matter.

21. Th is is called the Bunsei Po liti cal Disturbance (Bunsei no seihen) from the era when the incident broke out.

22. At that time, since Shogun Ashikaga Yoshimasa did not have a son, his brother, Yoshimi, was determined to be his successor by becoming his adopted son. However, his wife, Hino Tomiko, had Yoshihisa later and tried to let him succeed to the house hold, which created a confl ict with Yoshimi.

23. Th is episode is called the Meio Po liti cal Disturbance.24. Also at about that time, Ise Nagauji (Hojo Soun) caused a stir when he re-

moved an offi cial, the Horikoshi Kubo. Th e Horikoshi Kubo, a member of the Ashi-kaga clan, was a replacement of the Kamakura Kubo, who was an original administra-tor of the Kanto region of the Ashikaga shogunate but revolted against the shogunate. Th e Horikoshi Kubo was so called because he could not enter Kamakura and stayed in Horikoshi in the province of Izu (now the prefecture of Shizuoka), instead.

25. Mountainous regions resisted territorial incorporation. Even the most ambi-tious daimyō found it diffi cult to extend his po liti cal authority into mountain valleys where the topography worked against invasion and where the inhabitants resented intruders. As a result, taxes in those regions tended to be lower than in the plains.

26. Kyoroku- Tenmon no ran. Nagao was killed at a boat party when one of Uesugi’s retainers threw Nagao off a boat on Lake Nojiri in today’s Nagano prefecture. Possibly as a result of their struggle on board the boat, the retainer drowned with Nagao.

27. Date- shi Tenmon no ran.28. Mori Motonari Jihitsu- shojo (Mori Motonari’s handwritten letter). In Dai-

nihon Kobunsho Mori- ke Bunsho [Japa nese Old Documents: Th e Documents of the Mori Family], No. 418.


29. Specifi cally, gekokujo included the overthrow of a master by a powerful re-tainer and that of a shugo by a lieutenant shugo. Nagao’s failed usurpation of Uesugi’s position was an example of the latter.


31. Sue was originally the lieutenant shugo of Shugo Ouchi. Around the same time as the purge of the Inoue clan, Sue masterminded a coup d’état, replacing his master Ouchi Yoshitaka with Ouchi’s cousin, Ouchi Yoshinaga. Th is coup d’état was also an example of gekokujo.

32. Ashikaga Yoshiaki was Shogun Yoshiteru’s younger brother and was confi ned for a while by Miyoshi when Miyoshi killed Yoshiteru. Yoshiaki escaped, however, and waited for a chance to enter Kyoto. Th e Miyoshi family ruled the Kinai region af-ter the Hosokawa.

33. Th e ikko- ikki mobilized common believers and thus had very large numbers of people in arms. Furthermore, their morale was strong due to their religious motiva-tion. Although the ikko- ikki is considered to have been a threat to Nobunaga, their main purpose was to protect their religious domain rather than to expand territori-ally. Th ey surrendered to Nobunaga once he promised the continuation of their reli-gious or ga ni za tion. Th e Honganji Temple and its sect of Buddhism has remained the biggest religious power in Japan to the present.

34. Naohiro Asao, Shogun Kenryoku no Soshutsu [Th e Creation of Shogun Au-thority] (Tokyo: Iwanami shoten, 1994).

35. Th e term erizeni referred to an act of denying par tic u lar currencies, while several currencies were circulated. In those days in Japan, copper coins imported from China were used. Since their value was not determined by the value of material, as was the case with precious metal coins, this kind of problem occurred (Zenika: Zen- kindai Nihon no Kahei to Kokka [Coins: Th e Currency and Nation in Pre- modern Japan], edited by Susumu Ike (Tokyo: Aoki shoten, 2001).

36. A revolt of 1582 by his retainer, Akechi Mitsuhide.37. Th e weakening of the Ming Empire inspired Toyotomi to imagine a diff erent

order in Asia. For details of his invasion of Korea, refer to Ike Susumu in vol. 13 of A History of Japan: Tenka Toitsu to Chosen Shinryaku [Th e Unifi cation of the Nation and the Invasion of Korea], edited by Susumu Ike (Tokyo: Yoshikawa Kobunkan, 2003).

38. See, for example, Katsumata Shizuo, Sengoku- ho [Th e Law of the Warlords] (Tokyo: Iwanami, 1976), p. 208.

39. See, for example, Fujiki Hisashi’s discussion of the so- buji in Toyotomi Heiwa-rei to Sengoku Shakai [Th e Peace of Hideyoshi and the Japa nese Society During the Warring States Period] (Tokyo: Tokyo Daigaku Shuppankai, 1985).

70 Susumu Ike

ReferencesAsao, Naohiro. 1994. Shogun Kenryoku no Soshutsu [Th e Creation of Shogun Author-

ity]. Tokyo: Iwanami shoten.Hisashi, Fujiki. Toyotomi Heiwarei to Sengoku Shakai [Th e Peace of Hideyoshi and the

Japa nese Society during the Warring States Period]. Tokyo: Tokyo Daigaku Shup-pankai, 1985.

Ike, Susumu. Daimyō- ryogoku Keisei- ki ni Okeru Kokujin- so no Doko [Kokujin Lords in the Formation of the Daimyō Domain System]. Chap. 2 of Part 3 of Daimyō- ryogoku- sei no Kenkyu [Studies on the Daimyō Domain System]. Tokyo: Azekura shobo, 1995.

Ike, Susumu. Daimyō Ryokokusei no kenkyu [Research on the Territorial System of Daimyō]. In Daimyō- ryogoku- sei no Kenkyu [Studies on the Daimyō Domain Sys-tem]. Tokyo: Azekura shobo, 1995, pp. 66– 68.

Ike, Susumu. Nihon ShiTenka Toitsu to Chosen Shinryaku [Th e Unifi cation of the Na-tion and the Invasion of Korea], vol. 13, edited by Susumu Ike. Tokyo: Yoshikawa Kobunkan, 2003.

Ike, Susumu, ed. 2001. Zenika: Zen- kindai Nihon no Kahei to Kokka [Coins: Th e Cur-rency and Nation in Pre- modern Japan]. Tokyo: Aoki shoten.

Mass, Jeff rey P. Lordship and Inheritance in Early Medieval Japan: A Study of the Ka-makura Soryo System. Stanford, CA: Stanford University Press. 1989.

Shizuo, Katsumata. Sengoku- ho [Th e Law of the Warlords]. Tokyo: Iwanami, 1976.

71

Introduction

Autonomous communities are known to have developed in medieval Japan. Th ere is considerably less agreement as to whether or not the capacity for local governance survived Japan’s transition to the modern age. According to the traditional view, towns and communities achieved considerable autonomy in the late Middle Ages but were beaten back by po liti cal repression, particularly upon the establishment of a strong central government in 1600. According to a revisionist view, however, communitarian impulses not only survived but thrived through the transition to the modern territorial state. Local commu-nities managed to maintain or even expand their autonomy, forming the basis of a robust modern society capable of demo cratic self- governance.

Th e characterization of Japan as a typical “Asiatic nation” governed by a centralized authority was not only a common view among Japan’s postwar oc-cupiers, but among Japan’s own early postwar historians as well. It is a view that continues to inform many comparative histories. It sees the policies of the Toyotomi Hideyoshi regime, namely, the cadastral surveys (Taiko- kenchi), the separation of the warrior and peasant classes (heino- bunri), and the peace order forbidding neighboring territories to fi ght over boundaries (sūbuji), as the logi-cal end point of an intensifi cation and deepening of military rule. Furthermore, this view links to an understanding of the Meiji Restoration as another manifes-tation of these historical conditions and Asiatic principles. Japan’s transition to modernity was incomplete, the argument goes, because each successive po liti cal regime, including the U.S. occupation aft er World War II, was an imposition from above that used existing hierarchical structures for support and legitimation.

Community Vitality in Medieval JapanTsuguharu Inaba4

72 Tsuguharu Inaba

Th e revisionist view, that Japa nese history includes a rich tradition of local vigor and re sis tance to control, has developed over the last twenty years through a reexamination of data from village communities in the late medi-eval period. Th is view radically reinterprets the fi ft eenth to the seventeenth centuries as a period in which villages, towns, and local po liti cal regimes were vigorous and active in shaping their environments. Th e Toyotomi Peace Or-der against local confl ict and the separation of warriors and peasants into separate classes are understood as responses to the needs and demands of such communities rather than as symptoms of their repression. Community autonomy is thought to have survived the tumultuous events of medieval life and to have bequeathed to early modern Japan a legacy of self- regulation and self- assertion from society’s lower rungs.

Power and Social Values

Norms, customs, morals, and values in medieval Japan were generated fi rst and foremost in village communities rather than from military overlords, in tandem with the mechanisms of self- government. Village councils (yoriai) of peasant house holds gained decision- making importance from the fi ft eenth century throughout Japan. Th e councils enacted rules (mura- okite) governing local behavior and maintained the right to enforce these rules.

One of the tasks undertaken by the village leadership was to handle land payments and taxes on behalf of the whole village (souson). Peasants paid land taxes to the village council, which in turn paid the lord his due. Under this system (mura- uke- sei), the village functioned as a mutual aid or ga ni za tion for peasant house holds and helped to regularize contractual relationships be-tween village communities and feudal lords.

Medieval self- governing communities also undertook, to some degree, to manage their own military security. Adult males in the village community armed to protect the communal gathering and hunting grounds in neighbor-ing mountains, as well as the water sources for irrigating the fi elds. Peasants carried swords at their sides from the age of 15 or 16. Arms such as spears and bows as well as techniques for using these arms spread widely in villages. Ri-fl es were also introduced in the late sixteenth century.

Th e villages’ military forces, consisting of males between the ages of 15 or 16 and 20, were called wakashu. Th ey were well or ga nized and well equipped, and could sometimes repel a lord’s army.

When territories or water supplies were invaded by neighboring villages, armed adult males in the village community rushed to the scene to protect

Community Vitality in Medieval Japan 73

their property. Wars among villages could become chronic and sometimes expanded to involve other neighboring villages and warriors in the region. It is known, for example, that villages in Fushimi located in the south of Kyoto fought with each other about every fi ve years in the early fi ft eenth century.

Th e most serious concern of the medieval village community was how to keep confl icts over territory from ruining their livelihoods. Village communi-ties enacted laws to improve their mobilizational capacity. Villagers who failed to rush to the scene to help in a boundary dispute were, in some cases, deprived of village membership, whereas acts of bravery were conspicuously rewarded. Continual exposure to territorial struggles no doubt enhanced village solidar-ity and diminished internal confl icts within the community. Th e village com-munity provided its members with economic aid as well as military security. Th eir own rules, more than those of any feudal lord, ordered their daily lives.

The Bargaining Hypothesis of Democracy

Farm communities were prepared to mobilize for defensive war, but they pre-ferred to stay out of the larger confl icts of the region’s warrior lords.

Tax documents called sashidashi reveal a contractual relationship between a village community and a warrior lord. When a warrior lord defeated his op-ponent and conquered his opponent’s territory, he fi rst had to collect the tax documents from villages in the territory that he had newly gained. Th is prac-tice became common in the fi ft eenth century during the Ashikaga shogunate. Th e tax documents showed the kinds and amount of taxes the village paid to the previous lord, and were renegotiated periodically based on cadastral sur-veys of the quantity and quality of land under cultivation. When land changed hands, taxes that the new lord could collect from villages were, by tradition, restricted by the pre ce dents described in the village’s tax documents. Village communities did their best to circumscribe the lord’s powers to raise new re-sources from their productive eff orts, and negotiated hard for minimal tax increases as productivity grew.

Tax documents also laid out the public duties that a village provided for the previous lord, including labor and military ser vice. Th e peasants’ main military ser vice was transporting military equipment and supplies to the battlefi eld. Peasants called up for this duty accounted for more than 30 per-cent of all the personnel of the daimyō’s army in the sixteenth and seven-teenth centuries. In the sixteenth century, as regional lords consolidated their territories and began to fi ght for supremacy, village communities were in con-tinual negotiation over the provision of military ser vice.

74 Tsuguharu Inaba

One exception to the rule that peasants served as carriers rather than as warriors came from the desperate attempts of the Hojo (the regent family of the Kamakura shogunate) to raise large armies of peasant soldiers. In the sec-ond half of the sixteenth century, the Hojo fi elded peasant armies against Takeda Shingen and again against Toyotomi Hideyoshi. Even here, however, the peasants were not expected to fi ght on the front lines of the battles but to stand in defense of castles close to their villages. Th e rule, at fi rst unspoken and later formalized as “separation of warrior and farmer” (heino bunri), re-fl ected the preferences of villagers to protect their livelihood more than being an imposition of class hierarchy.

During war time, local lords allocated logistical ser vice to village com-munities in their domains, depending on village size. Th e village council in turn decided who would be called up for duty and for how long. Th e local lord typically discounted about 10 percent of taxes as long as the village com-munity regularly performed public duties such as logistical support for mili-tary campaigns.

When war damaged crops and otherwise undermined agricultural produc-tivity, local lords were forced to make economic concessions such as reducing or canceling debts owed by the village to the lord. Th e Hojo canceled debts and released debt slaves for all the villages in their territory in 1550 and 1560 when the villages were totally exhausted. Th e lord’s retainers were prohibited from imposing additional public duties on the villages without permission, and peas-ants were given the right to appeal to the lord directly in the event they felt this rule was not honored. Although local lords were militarily powerful, their power rested on the resources and implicit support of peasant villages.

In the busy seasons of planting or harvesting, during a bad crop year, or in a time when villages were struggling to recover from war- related damage, vil-lage communities oft en off ered monetary payments in lieu of labor ser vices in order to keep all hands on the farm. In some cases, they demanded total exemption from duty. Historical documents of the Hojo reveal that local lords oft en had no choice but to accede to these requests.

Th roughout the sixteenth century, village communities became more so-lidifi ed at the same time that military lords consolidated territorial control. Because of negotiated restrictions on the military use of peasants, warrior lords were limited in the range and duration of potential military operations. As a result, battles among warrior families in the sixteenth century tended to be localized territorial confl icts in the border regions of their provinces.


Mobilization and Expansion

Th e peace of the Toyotomi, which facilitated the peace of the Tokugawa, was brought by reor ga niz ing village communities and warrior lord authority. It was impossible in the sixteenth century to expand po liti cal authority simply by marshalling superior military power. Toyotomi Hideyoshi did not resort to arms primarily in order to expand his authority. Aft er occupying Kyoto, Hideyoshi issued a proclamation to the regional lords to stop fi ghting:

I was ordered by the Emperor to pacify the country. You must stop fi ghting for territory immediately and come up to Kyoto. I will listen to your claims on territory thoroughly and make judgments. If you will not follow my judg-ments, you will be punished. You must reply to me immediately and tell me if you agree with this order.

By 1590, most of the regional lords accepted the peace proclamation— or more precisely, the order to accept the existing status quo. Hideyoshi guar-anteed the whole or part of their territories, establishing a lord and vassal relationship with the regional warrior families and expanding authority without further warfare. Th e Hojo was the only daimyō who failed to get a consensus of opinion in favor of Hideyoshi’s peace edict from among their retainers. When they declined to accept the order, Hideyoshi’s forces destroyed them.

In addition to ensuring the military status quo, the Toyotomi regime also guaranteed the autonomy and status of village communities. Village commu-nities secured their role as intermediaries or contractors responsible for col-lecting taxes and or ga niz ing public duties, and enhanced their role as mutual aid organizations to peasant house holds.

Although peasants and warriors were already accustomed to a division of labor in warfare, the Toyotomi regime made the division more explicit. Hide-yoshi proscribed peasants from carry ing arms such as swords and prohibited village communities from resorting to arms for self- protection or confl ict set-tlement. Hideyoshi announced to the peasants:

Peasants will prosper forever, if they carry faming tools only and are engaged in farming. Th is is a policy for peasants and for peace. I will introduce this policy, following that of a legendary emperor of China. Peasants should devote themselves to farming.

76 Tsuguharu Inaba

From the standpoint of village communities, Toyotomi’s authority solved two of their thorniest problems: the need to protect themselves from neigh-boring villages and the need to provide local lords with corvée labor during war time. Both were now illegal.

Th ere is still the question of why Sengoku daimyō and village communi-ties accepted Toyotomi Hideyoshi’s rescripts, which were, aft er all, not unlike rules that had been periodically handed down by preceding governments. Th e traditional explanation, that Hideyoshi was the fi rst with military muscle to back up his words, ignores the demand for compliance.

By the middle of the sixteenth century, many of the regional lords (war-ring states daimyō) had secured their power within their regions. But tension then shift ed from struggles with regions to the border areas between prov-inces. Regional lords found themselves pinched between the expense of bor-der confl icts on the one hand and peasant demands for economic concessions on the other. To avoid fi nancial ruin, regional lords oft en settled territorial con-fl icts through treaties mediated by neighboring lords.

Something similar occurred in territorial confl icts among villages. Con-fl icts were oft en settled by mediators from neighboring villages or by collective decisions of villages in the region. Village communities, which maintained their unity by fi ghting against neighboring villages in the beginning, became more stable as self- governing, mutual aid organizations. By the end of the six-teenth century, villages shift ed their activities from military preparedness to economic risk sharing. Th e peace of Toyotomi was possible because territorial consolidation by regional lords had increased the costs relative to the benefi ts of warfare, for peasants as well as for themselves.

At the end of the sixteenth century, the foundation of the Tokugawa central government was thus established. In the Tokugawa regime, peasant house holds and village communities existed as the foundation of Japa nese society. Th e shogun’s offi cials (kougi) ruled over regional lords, or daimyō, who in turn governed local societies. Although the Tokugawa regime is characterized by its hierarchical or ga ni za tion with the shogun as its apex, it should not be over-looked that more than 60,000 village communities and about 250 daimyō authorities (han- kenryoku) maintained considerable autonomy.

Peasants and Mobilization for the Invasion of Korea

Hideyoshi’s invasion of Korea has always puzzled historians of this period. Was Hideyoshi a megalomaniac hell- bent on taking everything in sight re-


gardless of the cost to his domestic reign? Alternatively, did he think making Korea into an enemy would strengthen his control over Japan?

Successive Chinese dynasties had ruled over East Asia, at least in name, claiming to be the center of the world (chuuka) and requiring tribute from neighboring “barbarian” (iteki) countries. However, China’s regional hege-mony became unstable in the days of Nobunaga and Hideyoshi because of the decline of the Ming Dynasty and the encroachment of Eu ro pe an navies. Th is provided China’s neighbors with the opportunity to upend China’s claim to suzerainty. Although Hideyoshi’s plan failed, the Manchurian descendants of the Jurchens did manage to conquer the Ming Dynasty in the middle of the seventeenth century.

Hideyoshi’s understanding of Korea was poor, and troops were sent to Korea under the same logic as he employed for the internal unifi cation of Japan. He demanded that the local lords in Korea should stop fi ghting and yield to his authority. If they did not comply, he would invade them, as he had in other parts of Japan.

Th e Toyotomi regime’s mobilization of peasants for the invasion of Korea illustrates the importance to peasants of staying out of wars, or being com-pensated when they were drawn in. Kato Kiyomasa, who was the domain lord of Higo, now the prefecture of Kumamoto, bore the heaviest burden of mobi-lizing for war against Korea.

Th e problem of the peasant logistical laborers, which Sengoku daimyō had faced, also bothered Kiyomasa from the preparatory stage of the invasion. As was true in the case of internal wars, peasants did not like to be sent to the battlefi eld. Th ey strongly and frequently demanded substitution of monetary payments for logistical duties to daimyō and their retainers beginning with the mid- sixteenth century. Peasants were aware that, even if their duties remained logistical, war drew down their resources. By confi scating arms from all but the warrior class, on the one hand, the Toyotomi regime responded to the peasants’ demand to be left out of war. On the other hand, however, Toyotomi now needed many logistical peasants to support his Korean invasion. In the province of Bungo or the Ohtomo’s territory, it was assumed from the begin-ning that peasants would refuse mobilization for the invasion, and retainers were instructed to prepare for their refusal with what ever force was necessary.

In 1591, Kiyomasa sent a letter to his territorial province, Kumamoto, aft er six peasants fl ed from the invading base of Nagoya in the province of Hizen. It read:

78 Tsuguharu Inaba

If these peasants are found to return to their village, catch them immediately and put them in prison. If they hide somewhere, imprison their wives, chil-dren, and relatives. If those who did not have clear identities without families or relatives were sent for logistical duties, that was a deceit by a representative of the village. Put the village representative in prison. If the village representa-tive cannot be imprisoned, that is the responsibility of an offi cer who admin-isters my direct domain. Since troops will be fi nally dispatched to Korea next spring, investigate closely.

As during the period of the Warring States that preceded Oda Nobunaga and Toyotomi Hideyoshi, peasants disliked logistical duties and instead sent convicts or men kidnapped by the village for this purpose. Knowing this fact, Kiyomasa could not take any eff ective countermea sures.

In the base camp of the battlefi eld in Korea, Kiyomasa found a letter sent by a peasant in a village in the province of Higo to logistical peasants in the battlefi eld. Th e letter stated, “It is time you should return to the village. Offi -cers will not investigate, even if a group of you fl ee back to the province of Higo.” Th e letter shocked Kiyomasa. Concluding that the villages had planned this act in conspiracy with offi cers in his territory, he became furious. Appar-ently, however, he was unable to handle the continual desertion of peasants to their villages. Kiyomasa ordered punishment for the sender of the letter and a recall of those peasants who had fl ed. Peasant desertion not only refl ected re-sis tance to mobilization by the peasants and their villages in the province of Higo, but also implicated local offi cers who sought to maintain local order. Th e escapes were neither haphazard nor sporadic.

Being shaken by the peasants’ re sis tance to the war, Kiyomasa said in the spring of the second year of the invasion, “It seems that Korean and Chinese soldiers have been prevailing recently. Th is is because Japa nese troops have become exhausted due to their long- term assignments.” In the next spring, in order to secure as many peasant carriers as possible, he raised the peasant levy, requiring that uncultivated lands in villages should count toward their labor obligations. Th is order, which broke with pre ce dent, only intensifi ed the villages’ re sis tance and failed to increase peasant mobilization. In 1596, Kato Kiyomasa was reassigned to Kyoto, ostensibly because he had hindered peace negotiations. In fact, Hideoyoshi was worried because the peasants in the province of Higo formed a negative opinion that Kato had refused to accept his governance.


When the second Korean invasion ended with Hideyoshi’s passing, Kato Kiyomasa ordered his high- ranking retainers to post signs in his territorial province of Higo and appeal to peasants as follows:

For the last few years, I have given you suff ering, imposing many labor duties on you. I have just been ordered by Hideyoshi to return to my domain. When I get back, I will exempt you from all the labor duties except for the land tax for the next two to three years. Th us, I would like you, the peasants, to be pleased with this and await my return.

During the period of the invasion, Kato ordered that peasants be punished at once if they did not pay land taxes. Immediately aft er the failure of the inva-sion, he changed his position and off ered the peasants exemption from labor duties for the next two to three years. However, this off er did not simply ema-nate from Kato’s benevolence. Th e suff ering of the peasants described in the signs he had posted should be considered his recognition that peasants now infl uenced his role as a daimyō, by insisting that the primary duty of a lord is to protect the peasants’ engagement in farming and the productivity of the vil-lages. Kato feared their criticism of his qualifi cations as a lord. Before return-ing to his home ground, the Kumamoto castle, he posted signs guaranteeing the peasants’ engagement in farming. Toyotomi and his vassals failed to con-struct a peasant conscription system for a foreign invasion that had nothing to do with the security of the peasants’ daily life.

Kumamoto: A Case Study of Village Self- Governance After Unifi cation

As part of a joint project of the History Section, the Faculty of Arts and Hu-manities at Kumamoto University, I conducted a fi eld survey of more than 600 stone monuments in Chuo- machi in Mashiki County (today’s Misato- machi). Th is fi eld survey revealed that groups of villagers began erecting stone monuments around the sixteenth century, documenting that village commu-nities were collecting their own taxes by medieval times in northern Kyushu. Th e number of these stone tablets is approximately the same as the number of residential blocks or village clusters. Surveys conducted by freelance research-ers and by municipal governments also suggest that roughly the same number of stone monuments exist in other prefectures. Based on surveys of stones and existing documents, we can safely acknowledge the establishment of village communities in Kumamoto in the sixteenth century.

80 Tsuguharu Inaba

When we consider how the Hosokawa family structured the organs of gov-ernance in the seventeenth century, following the unifi cation of Japan under Tokugawa rule, the dispute settlement pro cess concerning land usage over the Kuju Mountains provides an insightful example of continued involvement in local management by the villagers themselves. Villagers in Kuju in the Hoso-kawa family of Kumamoto and Kutami Go in the Oka Clan of Bungo (today’s Oita) had continually disputed the use of resources harvested from the moun-tains that stretch between these areas. Disagreements between the two vil-lages involved settling borders between the two feudal domains and therefore encompassed the entire hierarchy of po liti cal authority, from villagers up to the top of the clan.

Th e dispute resolution mechanism was multilayered. As the fi rst step, vil-lage headmen or the representatives of the directly impacted population ne-gotiated how to adjudicate the dispute. When this stage failed, the headmen reported the situation to the clan’s county offi ce. Th e chief retainers of the clan directed the headmen to negotiate with his counterpart in the other clan. In the event this second stage of negotiations failed, both of the clans jointly sponsored a court ruling or asked a third party, mostly a feudal lord in nearby clans, to adjudicate the dispute.

Th e system of dispute settlement was an embodiment of complex layers of po liti cal authority in the traditional Japa nese society and consistent with Kogi or the Constitution of the feudal clans as well as the preexisting autonomous village or ga ni za tion. Th e involvement of peasants, headmen, county offi cials, and chief retainers can be summarized in the following manner.

1. Th e information on which the clan government relied in making deci-sions was provided by the headmen. In par tic u lar, existing documents detail the pro cess of disputes, the topology of the disputed areas, and the situation surrounding the opponents’ territory.

2. Th e documents were prepared by the headmen and other village offi cers and the county resident representative passed the information on to the offi cial in charge of local administration in the central clan’s government. Th en the information was pro cessed to the chief retainers. Th e chief retainers almost always accepted the information without questioning it and based their decision on it. Th en the decision was passed down to the village level and was implemented by the headmen of the relevant villages.


3. Even aft er the chief retainers decided what to do with the disputes, the headmen and village clusters were actively involved in the intelligence activities and dispute settlement. Without the information they provided, the clan’s government could not reach a decision.

4. When the headmen were active, peasants held meetings at both the regional and village levels. It appears that headmen of neighboring villages contacted each other on a regular basis.

As early as the seventeenth century, a close relationship already existed between local communities and the clan’s administration. Village headmen were eff ectively territorial lords on a small scale and also were chief adminis-trative staff , and they did not limit their local administrative activities to clan border disputes. An example comes from a stone monument beside the Sa-mata Canal in Kumamoto, which was opened in 1683 as an irrigational canal to convert vegetable fi elds into underwater rice paddies. Because the drilling of rock was a diffi cult task, those who were involved in the project celebrated the opening of the canal when it was completed. Th e roster on the stone lists Mr. Nakayama, the hereditary headman of Nakayama; the offi cial of the clan in charge of the construction work; offi cials in charge of the area; and the rep-resentatives of Samata Village, whose residents were the target benefi ciaries. While the irrigation system specifi cally benefi ted Samata Village alone, offi -cials in charge of the county as well as offi cials in charge of public works were also listed on the stone tablet. Th e village- level project involved the clan’s gov-ernment. Th e inscription shows that the target benefi ciaries coordinated their project and worked out the details of water use rights. Th en the villagers sub-mitted the project request to the village headman. Aft er the headman achieved consensus within the village, he applied for project approval and assistance from the clan’s government. Th en the clan’s government approved of the proj-ect. Although only this stone epigraph documents this project, the names on the stone reveal a great deal about the decision- making pro cess.

Th e mid- to late eigh teenth century marks a historical watershed in the de-velopment of communal governance. During this period, the village headman system transformed itself across the Kumamoto house. First of all, the Reform of Horeki (1751– 1764) marks a transition from an inherited village headman system to an appointment- based system. Aft er the reform, a nontrivial number of village headmen were ousted from their positions, and headmen from peasant families were appointed to approximately 10- year terms. Hereditary

82 Tsuguharu Inaba

headmen apparently were not capable of handling the expanding scope and the complicated nature of local administrative tasks.

Second, headmen started to collect tax revenues on behalf of the feudal domains and developed their own local development funds. As village head-men became the main tax- collecting agents, a portion of the revenues was ex-pended for the benefi t of their local communities. According to Maeda, as of 1843, the fi ft y- two headmen he examined collected a total of 2,538 ryo of gold, 43 kan of silver, 5,270 monme of coins, and 85,585 koku of rice. Th e headmen’s revenue base includes taxes of various kinds levied on the rice harvest, dona-tion and contributions by the peasants, and returns on their own investment projects (e.g., land reclamation). In this sense, local development funds pro-vided a material foundation to the headmen’s governance. Th e feudal domain’s resident representatives in the communities were no longer present, and most local administration tasks were conducted by villagers themselves.

Th e decentralization initiative and the formation of the local development funds resulted in further development and enhancement of headmen as au-tonomous local governance functionaries. Because headmen had their own development funds, they gradually but eff ectively became eco nom ical ly in de-pen dent of the feudal clans and were instead agents of communal planning and regional policymaking. Th e new system of appointing and relocating vil-lage headmen and peasant civil servants provided eff ective administrative ser vice to the community, and they together constituted the cornerstone of regional policy by the Kumamoto house.

Village self- governance continued to develop throughout the nineteenth century. Under the name of village headmen or other representatives, village communities submitted petitions for public work projects and other favors, using predesignated forms of paperwork. In the event that the county repre-sentative accepted the form, the deputy county representative forwarded the proposal to the county representative’s Secretary Bureau, and then the chief county representative decided whether to authorize or veto each proposal.

As the aforementioned example of the Samata irrigation canal might sug-gest, the content of the proposals oft en concern water use as well as other types of public work projects. Th e deputy county representative forwarded the headmen’s proposals to the clan’s secretary, and then the clan offi cials evalu-ated the merit of the proposal and made decisions over project approvals. Th e offi cials reported the result to the deputy county representative, and if the vil-lage headman agreed, the project would be implemented.


Th e form and contents of these memos indicate that the system of consen-sual decision making was developing in the Kumamoto family’s local admin-istration, with village headmen functioning as the basic unit. Th e system of village headmen became institutionalized and grew into a bureaucracy aft er the Horeki Reform period. In this period, the headmen villages and village clusters were administrators who were born of peasant families, and these positions were distinct from formal status and ranks that could be obtained by making fi nancial contributions to the clan. Each of the “meeting places” of the village hired promising young boys as trainees. Th ey were promoted to village headmen (mura shoya) at the age of 20 or so. Th ese headmen were then promoted to the executive board of the headmen council in their late thirties, and the headman was usually selected from these board members. Th us, the village headmen at the hamlet level were eff ectively administrators and bu-reaucrats, and whenever these offi cials could not handle the administrative function in these communities, village headmen collectively petitioned for fi ring incompetent headmen. On the other hand, when the headman was ef-fi cient, the village headmen petitioned against relocation of the headman. Similarly, peasants oft en initiated boycott movements against ineffi cient vil-lage headmen.

Th e bureaucratic nature of the headman system is illuminated when we examine the merit evaluation system of the headmen in the Kumamoto family in the eigh teenth century. Th is is eff ectively a system of per for mance evalu-ation conducted by the clan’s Election Bureau, and the pro cess was based on information contained in reports from the village headmen and the county deputy. Th e bureau was monitoring how well the headman was planning policies and implemented them. Based on this per for mance evaluation, the headmen were awarded, relocated, or fi red. As a result, the headmen from one village were sometimes relocated to other villages.

Th is per for mance evaluation system consequently aff ected personnel af-fairs in each headman council so that the preferences of peasants who be-longed to each headman council were closely refl ected. Th erefore, as long as the headmen were willing to work for the benefi t of the residents, they were serving as administrative representatives of the villages. Whether they could ascend to a higher ranking administrative position also depended on their administrative per for mance and the evaluation by the election bureau.

Increasingly, headmen were those who functioned as representatives of local interests and were born of a peasant family. Th e meeting place was managed by

84 Tsuguharu Inaba

staff who were also from peasant house holds. Th ese offi cials aggregated the preferences of the local residents and arbitrated interests. Th ey together planned and implemented public works that would benefi t the locale on the whole. Th e project proposals were submitted to the clan’s government, and the system of consensual decision making was institutionalized. In addition, this system of decision making preserved an eff ective mechanism of checks and balances be-tween the clan’s government and the local administrators at the village level. When headmen discussed disaster relief mea sures aft er a tsunami hit the vil-lage in the early nineteenth century in Konoura Tenaga in Udo County, the headmen decided to solicit grants from the clan’s government instead of fi nanc-ing the project themselves. Th e clan offi cial in charge of local administration revamped the proposal from the headmen, but the headmen objected to this decision in more than four rounds. Consequently, the clan government’s de-cision was modifi ed to a considerable degree, and a compromise was worked out. In this pro cess, the system of consensual decision making was respected, and the clan’s government was not capable of enforcing their decision in a top- down way. Th e village headmen eff ectively constrained the clan’s government. Village communities, headmen, and the clan’s secretariat were closely involved in the decision- making pro cess, making headmen eff ective veto players in the decision- making procedure.

Modernization aft er the Meiji Restoration was greatly facilitated by this ex-perience with local self governance. Th e popularly elected prefectural assembly in Kumamoto was regarded as the most demo cratically advanced among the Japa nese prefectures of the time, together with Shizuoka and Kochi. First, the election for the Kumamoto Prefectural Assembly was conducted during Gover-nor Yasuoka Ryosuke’s tenure starting in June 1873. Yasuoka was appointed governor by the Meiji government, and his original intention was to appoint village headmen and or ga nize their meeting (Kansen Kuchokai). Instead of a meeting of po liti cally appointed headmen, prefectural assembly members were popularly elected, with near perfect male universal suff rage. Second, the elec-tion was conducted in multiple rounds. Voters fi rst elected members of “small districts” (sho ku). Th en the elected members chose their representatives serv-ing in “large districts” (dai ku), and only one in ten members of the small dis-tricts was selected into the main prefectural assembly members (dai ku). In this multilayered electoral system, shoku roughly corresponded to the boundaries of tenaga in the premodern period, and it appears that the elected members of the Prefectural Assembly were former headmen of tenaga. Th e Assembly eventually


confronted Governor Yasuoka’s mea sure to control villages’ bud getary base gobi kin (village endowments).

Th e village headman system not only incubated a system of repre sen ta-tion but also made a streamlined transition to modern village communities. Th ese villages had their own local revenue base, and villagers fi nanced local public goods that prepared for modernization of the local community in later periods.

Th e Higo Restoration faction of the Jitsugakuto government in Kuma-moto (est. 1871) removed each village’s development funds and placed them at the prefectural level. Th e prefectural government then used these funds to fi nance the establishment of medical schools, hospitals, and to pay headmen’s salaries. In 1873, Governor Yasuoka, a native of Tosa, forwarded part of the village’s revenues to the Ministry of Finance in Tokyo, and the remaining fund was administered by headmen appointed by the governor. However, during massive uprisings in 1877, the protestors requested that the village de-velopment funds be returned to the village communities. Municipal council members also joined forces. In 1895, the protestors appealed to the court for judicial decisions.

Th e fact that those who participated in the riot of 1877 were demanding the return of communal development funds to the village indicates that dur-ing this period local residents perceived that the funds were their own com-mon property. In 1880, the Meiji government made a policy shift . Th at year’s order from the Department of State (Daijokan) stipulated that the village de-velopment fund would belong to the alliance of village trusts (gobi zaisan rengokai).

Th us, the development funds were invested in the Kyushu railroad compa-nies and other newly emerging enterprises. Th e funds enabled new companies to issue bonds, and allowed regional communities to establish hospitals, schools, and other types of infrastructure, as well as to hire school personnel and other civil servants. Th e funds supported municipal government fi nance especially when their fi scal foundations were vulnerable.

Although the Meiji government tried to confi scate the funds accumulated in each local community, it ended up making a policy transition to allow greater local autonomy. Th e engine of modernization in local communities in Japan in fact resided in capital accumulation in the local communities, and these communities protected their common property from confi scation by the state.

86 Tsuguharu Inaba

In 1890 the Meiji government tried to impose a new local entity of coun-ties, which were nothing more than geographic segments for the sake of im-plementing public works projects. But Kumamoto Prefecture initially op-posed the introduction of the modern county system in 1890, and the central government did not have substantive administrative capability to impose the new system. Th e functional administrative units were the villages, and their agents were the headmen serving these communities.

Th e cornerstone of modern Japa nese administration was carried over from local self- governance in the premodern period. Local government fi -nance, consensual decision making, civil servants, and repre sen ta tion were all present in their early form in medieval Japa nese villages.

Conclusion

Th is chapter examines how wars changed the po liti cal map of medieval Japan. Confl icts between villages regarding agricultural production led to the for-mation of autonomous village communities, while feudal lords consolidated territorial control through armed confl ict. In this pro cess, villages demanded physical protection and limits to their role in military mobilization. Th e Toy-otomi regime generalized the settlement between villages and daimyō to the national level.

Th e failed invasion of Korea renewed the government’s commitment to the peasants’ specialization in farming. Th e Tokugawa shogunate gave up on expansionism toward the external world, at least in part because the peasants were tired of war and demanded a government that would secure their peace and prosperity.

It is interesting to consider how certain characteristics of medieval Japa-nese society infl uenced modern Japa nese norms and practices. First, succes-sive governments of Japan were able to appropriate the loyalty that peasants felt toward their villages and use it for patriotic projects of various kinds. Historically, peasants worked hard for their villages or regions, but disliked mobilization by regional lords or the shogunate, even as nonsoldiers. Peas-ants in feudal society did not envision themselves toiling for the nation. Th ey would still feel this way when the Tokugawa shogunate fell to disenchanted domains in the Meiji Restoration of 1868. Th e mass conscription system es-tablished by the Meiji government did not function well in its beginning. To create an army of the people in modern Japan, it was necessary to implant an idea in the peasants’ minds that to sacrifi ce their lives for the nation- state was the same as doing so for their families or villages. It took the imperial


fascism of the 1930s militarists to capture the imagination of village commu-nities in such a way.

Second, although daimyō under the Tokugawa were guaranteed the right to introduce their own policies to their regions by the shogunate, they were badly criticized by village communities when their policies were quite diff erent from the traditional practice of making economic concessions to peasants in hard times. Th e shogunate maintained its regime by depriving daimyō of the right to rule when they could not ensure agricultural productivity. In the late eigh teenth century, some village communities began to gather beyond the geo graph i cal boundary of a daimyō’s territory to discuss regional problems such as fl ooding and water management. Th ey handled these problems by choosing representa-tives of the region by election, or by proposing policies to a daimyō and receiving fi nancial aid to put their proposals into practice. Such activities became common in many parts of the country. It was relatively easy for the Meiji government to shift taxation from the domain to the village level because villages had already gained considerable experience in managing their own aff airs.

Th ird, the structure of traditional Japa nese society, which consisted of autonomous, mutual aid organizations against the background of central au-thority, is similar to that of Japa nese corporate society in the twentieth and twenty- fi rst centuries. During the economic boom following World War I, many Japa nese corporations competed for scarce labor by guaranteeing the livelihoods of employees and their families through the lifetime employment system and or ga niz ing numerous subcontract companies under big companies. Interfi rm competition provided an incentive to off er employment contracts that managed workers’ lifetime risks. However, the global economic integration of Japan’s economy has now destroyed much of this ethic. Japa nese society has not yet found solutions to numerous problems arising from increased unemployment, homelessness, and income in e qual ity. Village communities were broken up in the 1960s, while corporate organizations were destroyed through the 1990s to the beginning of the twenty- fi rst century. It was the biggest, and perhaps the worst, change that Japa nese society has experienced since the sixteenth century.

Notes1. Downing (1992), for example, implies that Japan failed to develop demo cratically

because its medieval and early modern history lacked experience with or a theory of limited government.

2. Th e territory of the village community included residential areas (mura), agri-cultural fi elds (nora), and lands for gathering (yama). Irrigation water needed for rice

88 Tsuguharu Inaba

farming was oft en gained through channels from rivers. Th e yama was a space of col-lective use for peasants of the village community. Like irrigation water, the yama was thus maintained by the whole village (sou) and had to be protected from neighboring villages. Th e village community was the actor of self- redress.

3. Th e size of cultivated lands in the village, which was determined by daimyō authority and the village community through the pro cess of making a record book, was called mura- daka.

4. Th is practice of making economic concessions was known as tokusei and fol-lowed the Confucian idea of virtuous rule or tokuji- shiso.

5. Th is is the sūbuji- rei.6. Th e taiko- kenchi.7. Th e katanagari- rei was a policy to complete the heino- bunri under the name of

the tokusei.8. Th e kenka- choji- rei.9. Th is is called kai- chitsujyo (literally, to bring order to China and the

bar barians).10. With this policy of the sūbuji, or “peace rescript,” Hideyoshi controlled Ky-

ushu in 1587 (in the fi ft eenth year of Tensho) and then tried to hold Korea, Kanto, and Ohu at the same time.

11. See Chapter 7.12. Inaba 2005.13. Inaba 2005.14. Yoshimura 2001.15. Maeda 1997, 1998.16. Nishimura 2003.17. One of the important collections of documents that delineate the Kumamoto

house’s local administration is a large number of memos that have been stored in ma-terials collected by Yoshimura (2001). Th e memos were chronologically bound, and some of them are as thick as 30 centimeters. Th e originals used high- quality paper sheets of standardized sizes, which were perhaps provided by the clan’s central govern-ment for use by the county resident representative offi ce and the headmen.

18. Yoshimura 2001.19. Fujino 1977.20. Yoshimura 2001.21. Yoshimura 2001.22. Maeda 1997, 1998.23. Maeda 1998. Because of this objection, the implantation of the new system was

postponed until 1896.24. Th is replacement of failed daimyō was called the kaieki.


ReferencesDowning, Brian. 1992. Th e Military Revolution and Po liti cal Change: Th e Origins of

Democracy and Autocracy in Early Modern Eu rope. Prince ton, NJ: Prince ton Uni-versity Press.

Fujiki, Hisashi. 1985. Toyotomi Heiwa- rei to Sengoku Shakai [Th e Toyotomi Peace Or-der and Sengoku Society]. Tokyo: Tokyo Daigaku Shuppan- kai.

Fujiki, Hisashi. 2005. Katana- gari [Th e Sword Confi scation]. Tokyo: Iwanami- shoten.Fujino, Tamotsu, ed. 1977. Kyushu kinseishi kenkyu soso, Kyushu to ikki [Medieval

Kyushu Historical Documents, Kyushu and Ikki Movements]. Kumamoto: Uni-versity Historical Documents Project.

Inaba, Tsuguharu. 2001. Mura no Buryoku- doin to Jinbu- yaku [Mobilization of Peas-ants and Logistical Duties]. In Senso to Heiwa no Chu- Kinsei- shi [War and Peace in the Medieval to Modern Age], edited by the Rekishi o benkyo suru kai [Asso-ciation for the Study of History]. Tokyo Aoki- shoten.

Inaba, Tsuguharu. 2003. Heino- bunri to Shinryaku- doin [Th e Separation of the War-rior and Peasant Classes and Mobilization for an Aggressive War]. In Nihon no Jidai- shi 13: Tenka- toitsu to Chosen- shinryaku [vol. 13 of A History of Japan: Th e Unifi cation of the Country and the Korean Invasion], edited by Susumu Ike. To-kyo: Yoshikawa Kobunkan.

Inaba, Tsuguharu. 2004. Sengoku- ki Hokubu Kyushu ni okeru Ryogoku- shihai to Mura ni kansuru Oboe- gaki [A Note on the Ruling of Territorial Provinces and Villages in Northern Kyushu in the Sengoku Period]. In Shoen to Mura wo aruku 2 [Exploring the Shoen and Village 2]. Tokyo: Azekura- shobo.

Inaba, Tsuguharu. 2005. Sengokuki ni okeru chiiki kosei no fukugenteki kenkyu [Re-search Documents on Local Or ga ni za tion in the Warring States Period]. Kuma-moto University Liberal Arts Division, Historical Research Group working papers.

Maeda, Nobutaka. 1997. “Gobikin no kenkyu oboegaki” [Memos from Local Accounts], in Shishi kenkyu Kumamoto [Th e History of Towns in Kumamoto], vol. 8, pp. 43– 69.

Maeda, Nobutaka. 1998. “Zoku gobikin no kenkyu oboegaki [Memos from Local Ac-counts], Continued,” in Shishi kenkyu Kumamato [Th e History of Towns in Ku-mamoto], vol. 9, pp. 21– 55.

Makihara, Norio. 1998. Kyaku- bun to Kokumin no Aida: Kindai Minshu no Seiji Ishiki [Between Outsiders and a People: A Po liti cal Mind of People in the Modern Era]. Tokyo: Yoshikawa Kobunkan.

Nakano, Hitoshi. 1996. Toyotomi- seiken no Taigai- shinryaku to Taiko- kenchi [Th e Foreign Invasion and the Cadastral Surveys by the Toyotomi Regime]. Tokyo: Azekura- shobo.

Nishimura Haruhiko. 2003. “Horeki kara Tenhoki ni okeru Higo Hosokawa han no nosei to seimensei” [Agricultural management and governance in the Higo

90 Tsuguharu Inaba

Hosokawa Clan During the Eigh teenth Century], in Kumamotoshigaku [Kuma-moto Historical Studies], vol. 82, pp. 1– 82.

Sakata, Satoshi, Masaharu Ebara, and Tsuguharu Inaba. 2002. Nihon no Chusei 12: Mura no Senso to Heiwa [vol. 12 of A History of Medieval Japan: War and Peace in Villages]. Tokyo: Chuo- koron- shinsha.

Yoshimura Toyoo. 2001. Bakuseika no mura to zaicho [Villages and Towns Under the Tokugawa Shogunate]. Ichinomiyacho: Ichinomiya shiyakusho.

91

NESTLED IN THE FOOTHILLS of the Japa nese Alps about an hour’s bus ride from the bustling metropolis of Kanazawa City lies the tranquil farming village of To-rigoe, population 3,086. Now an obscure village, Torigoe was the site of the last stand in Kaga province (modern- day Ishikawa prefecture) of a powerful religious group’s re sis tance to Oda Nobunaga’s relentless march to power. On a 312- meter- high mountain overlooking the Tedori River, together with its sister fortress of Futoge on a hill across the valley, the fortifi ed temple of Tori-goe commanded a view of the valley below and the roads winding along the river from the castle town of Kanazawa by the sea. In 1581, Oda Nobunaga’s general Shibata Katsuie captured both fortresses and set up a garrison of 300 men, but within a month the resisters had recaptured them and killed Oda Nobunaga’s defending troops to a man. Th e fortresses were to change hands two more times in ever more ferocious fi ghting, until Oda Nobunaga with a vast army of tens of thousands fi nally prevailed in 1582 to overwhelm the 300 barricaded resisters, who had continued to fi ght despite their leader’s capitu-lation two years before. Oda Nobunaga’s men did not stop at crucifying the survivors on the riverbed below; they set about slaughtering the population in the surrounding villages over the next three years. Th e backbone of religious re sis tance to po liti cal unifi cation was broken, and within a few years Oda Nobunaga’s successor, Toyotomi Hideyoshi, was undisputed ruler of Japan. Th e last great bastion of warrior- monks at the Mount Kōya temple voluntarily handed over their cache of arms in 1588 without a fi ght.

Th is chapter documents the rise and fall of the fascinating phenomenon of po liti cally mobilized religious movements in medieval Japan that resisted, but

“Advance and Be Reborn in Paradise . . . ”Religious Opposition to Po liti cal Consolidation in Sixteenth- Century Japan

Carol Richmond Tsang

5

92 Carol Richmond Tsang

ultimately failed, to block Oda Nobunaga’s unifi cation campaign. I leave it to others, including the editors of this book in their introduction, to consider why temples and monasteries became so deeply involved in providing physi-cal protection and economic management in the medieval period. My con-cern here is to provide a narrative account of the ikkō ikki, literally translated as leagues of members of the Honganji sect of True Pure Land Buddhism (Jōdo Shinshū) in the sixteenth century, and to evaluate the motivations and success in fi ghting Oda Nobunaga’s forces.

“Advance and be reborn in Paradise, retreat and go instantly to hell” read the banner that led fi ghters to war against Nobunaga under the leadership of Honganji Temple. Honganji, along with its members, is best known for this fi erce stand against Oda Nobunaga’s attempt to unify Japan under his rule. More amazingly, if less dramatically, Honganji adherents joined with other groups in the province of Kaga to establish a more or less autonomous zone of rule from 1488 until they were fi nally vanquished by Oda Nobunaga’s forces in 1574. Th e sect’s re sis tance and history of militancy contributed to some of the most visible changes in Japan’s government and social or ga ni za tion, spe-cifi cally the separation of warriors from the countryside and the proscription of Christianity carried out under Nobunaga’s successors, Toyotomi Hideyoshi and Tokugawa Ieyasu.

Although the actions of Honganji sect members between 1465 and 1580, usually in warfare, were highly visible and therefore widely reported, most of them were ultimately unsuccessful. Nobunaga defeated the sect in the end, and so did most of its other enemies over time. Th e largest single success, when the members took over the province of Kaga with warrior help, resulted in shared rule with a number of other groups. Finally, the Honganji Temple took on the role of military governor approximately forty years aft er the ini-tial rebellion. While some may count this last a success, the ultimate winner was the leader of the Honganji sect more than its members, who fared better than they would have under a daimyō, but much the same as they did under previous military governors.

Th ere were some important successes as well, though on the whole these occurred when Honganji sect members responded to calls for assistance from warrior leaders, such as the deputy shoguns in 1506 and 1532. Th e military prowess that Honganji’s sectarians displayed in all their eff orts, however, im-pressed both warriors and aristocrats, not necessarily favorably, leading to results the sectarians failed to foresee. Some of the warriors saw them as

“Advance and Be Reborn in Paradise . . . ” 93

potential allies, others, along with the aristocrats, saw them as a threat. Th e Honganji members’ activities, and the consequences of those activities, do not easily lend themselves to tidy summation because of the variety of battles and the sect’s relatively infrequent successes. Nevertheless, the ikkō ikki, as the members’ fi ghting leagues were known, were arguably the most powerful and visible expression of commoners’ struggles in the sixteenth century.

Th ere were many other commoners’ struggles in the form of ikki of one kind and another, especially demonstrations demanding tax or debt relief, known as tokusei ikki. Ikki as a whole dominated medieval Japan, including leagues of warriors who joined for mutual assistance in addition to leagues of commoners. Several things made the ikkō ikki stand out, however.

One important diff erence, indeed the defi ning characteristic of ikkō ikki, was the dominance of Honganji sect members. Th is characteristic lent the various struggles a deceptive consistency that they, in fact, lacked. Th e ikko ikki leagues’ aims varied widely, as did the leadership. Honganji Temple some-times did direct and coordinate the ikkō ikki’s activities, but not always. Nev-ertheless, even when the ikki’s ventures occurred for local reasons and had nothing whatsoever to do with Honganji Temple, the imperial and military authorities demanded that Honganji’s leaders take a hand in ending them.

Th e perceived ideology associated with ikkō ikki also diff erentiated them from other leagues and league activities. As will be described in greater detail later in this chapter, True Pure Land Buddhism teaches that reliance on a bud-dha named Amida was the only eff ective means of achieving Buddhahood. As nearly as we can tell, most members of the Honganji sect understood the the-ology to mean that no gods (kami) or buddhas other than Amida had any real power over them. Th is can be seen from repeated charges that they “slandered the gods and buddhas” and destroyed sacred objects. In the 1532 ikki attack on Kōfukuji Temple in Nara, members burned sutras and other religious ob-jects. Th is behavior, naturally, increased the adherents’ notoriety and fur-thered the case for suppression of the sect.

Ikkō ikki must also be understood in the context of the Honganji branch religion, not just as commoner activities. Much if not most of their fi ghting involved some aspect of religion: support of the Honganji branch’s leader, re-sis tance to religious oppression, or defense of local temples. More impor-tantly, what held the members’ fi ghting organizations together was their shared religious identity, an identity that delineated a clear boundary between Honganji followers and those of other Buddhist sects.


True Pure Land Buddhism teaches that the only sure path to Buddhahood is to throw oneself at the mercy of one Buddha, Amida, who promised to save all sentient beings. Th rough his grace, aft er death one is reborn in his Pure Land paradise, where one can attain Buddhahood eff ortlessly. Understand-ably, for most followers being reborn in paradise suffi ced as an incentive to worship Amida, and the ultimate achievement of Buddhahood was men-tioned less and less over time. Rebirth in the Pure Land rested entirely in Amida’s hands, the Honganji branch taught, so one had to abandon all thought that one’s own actions contributed to Amida’s act. Even faith was to be regarded as Amida’s gift , not an attitude chosen by the believer. Th e devo-tion to a single Buddha, and the belief that relying on his grace provided the only chance to achieve rebirth in paradise or the attainment of Buddhahood, earned the Honganji branch the sobriquet of the “one- direction,” or ikkō, sect.

Because Amida bestowed his grace regardless of merit, only on the basis of his own vow to save sentient beings, Honganji and other True Pure Land teachings opened a path whereby one’s occupation no longer barred one from salvation. Previously, those whose livelihoods depended on killing, for exam-ple, whether of human beings for warriors or insects and vermin for farmers, were thought to be unworthy of auspicious reincarnations. With the worship of Amida, these disadvantages no longer proved to be obstacles to salvation. All equally shared the possibility of a Pure Land rebirth, though other sects might still require that some rituals be observed.

Honganji’s teachings diff ered from those of other Amidist sects in that they identifi ed selfl ess faith in Amida as the mechanism by which rebirth in paradise was assured. Traditional methods of achieving merit toward reincar-nation on a better level, such as reading and copying sutras, or paying for reli-gious ser vices, no longer applied. Th us, not only did the commission of sins in pursuit of a livelihood present no obstruction, but also, followers needed no special skills, leisure time, or disposable income to qualify for admission to the Pure Land. An auspicious rebirth was available to all, regardless of social status or rank. Status was multiple in medieval Japan, as it could refer to court rank, occupation or religious affi liation, for example. A peasant who was a Honganji member shared a religious status with a warrior who was also a Honganji member, though aside from religion, the warrior occupied a higher social rank than the peasant. A warrior could be granted court rank, which gave him a higher status than other warriors without court rank, though they both shared warrior status. In short, “status” was fl uid and multiple, and its importance varied depending on context.

Th at did not mean, therefore, that all social rank disappeared for believers. Just as the belief that all were equal before God in Christianity failed to level social distinctions in the world, so did Amidist teachings fail to erase distinc-tions of other kinds of status. Rather, the religious glue that held the ikkō mem-bers together was of a diff erent kind. Loosely speaking, they did share a reli-gious status, which laid the foundation for the formation of ikki, which required that some kind of status be shared by league participants. Th e strength of the glue, however, relied on the singular focus of the Honganji sect faith. Because of the absolute reliance on Amida, and the teaching that all other paths to Bud-dhahood had lost their effi cacy, Honganji propagated an exclusive, and there-fore exclusionary, faith. To most believers, no one outside the Honganji faith was destined for the Pure Land or for Buddhahood. Th is was a radical depar-ture from orthodox Buddhist teachings, which hold that there are many eff ec-tive paths to Buddhahood.

To be sure, other kinds of Buddhism besides the Honganji branch taught exclusivism. For one, the Nichiren sect advocated a single- minded reliance on the Lotus Sutra. One faction of the Nichiren sect, known as the fuju fuse, completely shunned people not of their faith, to the extent of refusing alms from believers of other branches of Buddhism. Interestingly, both Honganji and Nichiren followers formed leagues to advance their interests, and they went to battle on their own. In 1532, Nichiren followers in Kyoto answered the call of the acting deputy shogun, the most powerful warrior in Japan at that time, and attacked Honganji Temple in the city’s outskirts. Th ey acted with warrior allies, not alone, but they acted as members of the Nichiren sect. As townspeople, they lacked warrior status, so they would not have fought as in-dividuals in warrior armies anyway. Following their success in destroying Honganji, the Nichiren followers returned to Kyoto and undertook various duties in administering the city.

In both cases of religious ikki, Nichiren and Honganji, the exclusive nature of the sect’s beliefs provided a cohesive basis for the or ga ni za tion of followers for other purposes. Other religious institutions, such as the Tendai monastery Enryakuji, generally had specialized warriors. For the Honganji and Nichiren sects, the members themselves acted as warriors in both military and admin-istrative capacities, as when the Nichiren adherents ruled over Kyoto.

By the sixteenth century in Japan, villages had little need for protection by any but local warriors. Villagers took care of disputes between themselves and neighboring villages, such as over water rights, resorting to warrior assistance only in diffi culties on a larger scale. Indeed, with the endemic warfare of that



time, warriors used villagers to supplement their armies and off ered rebates or partial exemptions of the taxes they levied on the villages for military protection.

Furthermore, local warriors formed their own leagues. In a region without a strong military governor, local warriors might join each other and recog-nize the others’ territorial boundaries. Th ey would also pledge to come to the others’ assistance in the event of attack from without. None, however, tran-scended regions as the ikkō leagues did. Ikki demanding tax or debt relief did, over time, transcend regions as the ikkō ikki did, though they tended to in-volve adjacent regions. Ikkō ikki might draw participants from distant prov-inces, bringing warriors from Kaga, for instance, over a hundred miles to the Osaka region, as they did in 1506. It will be well to review the most important of the ikkō leagues’ struggles before going on to discuss the ideologies that ap-parently drove them, and the long- term eff ects of their activities. Not all of the leagues’ military actions were rebellions, but the most striking of these strug-gles were, specifi cally those in Kaga and Mikawa provinces, and that against Nobunaga.

Th e ikkō leagues had their greatest success in Kaga province, the southern part of modern- day Ishikawa prefecture. At fi rst, in 1473, an ambitious war-rior seeking to gain control of Kaga recruited Honganji sect members to help defeat his rival, thereby becoming Kaga’s military governor for the next fi ft een years. It is worth noting that his rival had the support of a diff erent branch of True Pure Land supporters, who oft en took what ever side Honganji members opposed, and vice versa. In offi ce this warrior worked to restructure the ad-ministration of the province and to consolidate power in his own hands, just as many of his contemporaries did. In other words, he attempted to form a unifi ed daimyō domain.

Landholding before the daimyō domains was multiple and layered. A landlord, usually a temple or an aristocrat in Kyoto, might have holdings in several diff erent villages, and each village might have parts owned by several diff erent entities. Th e cultivators did not own land as much as cultivation rights, and they might owe rents to more than one landlord for diff erent fi elds. Warriors generally had the right to collect taxes and rents on a par tic u lar landlord’s holdings, so warriors might compete to collect rents from the same cultivators. Daimyō generally overrode these complicated arrangements by taking over all lands. Toyotomi Hideyoshi would eventually grant the cultiva-tors own ership of the lands they tilled.

Local temples and sometimes warriors oft en had holdings over which they were able to exercise largely autonomous authority. Th ey had the right to bar entry to warrior offi cials, a right that amounted to exemptions from numer-ous taxes, and the responsibility for investigating and punishing criminal ac-tivity within the temple precincts. Th e daimyō also overrode these rights, taking direct control of all the aff airs of the province.

Th is became a problem for Honganji sect members, as in Kaga and else-where Honganji temples ruled areas with this kind of autonomy. Honganji temples had become centers of commerce, as towns grew up around them on land owned by the temples, and the right to bar entry to warriors extended to the temple towns. In time some of these towns claimed the rights even when they had not been specifi cally conferred by the governor. As one chronicle put it, “new priests and temples sprang up here and there . . . [declared] them-selves jinai (temple towns) . . . and ignored” warrior authorities, not paying taxes either in goods or labor. Even the earlier, less centralized, rule of the military governor had trouble with towns claiming autonomous authority without offi cial sanction. For a daimyō, it was worse. Th ere was no room in a daimyō domain for autonomous territories governed by people outside his chain of command.

Although there is no specifi c evidence of the new governor abrogating the “no entry” rights of Honganji sect temples in Kaga, we do know that he took over estates owned by other temples, shrines, and aristocrats. Among the many letters sent out by the shogunate in response to landlords’ complaints about nonpayment of rents, some target the new Kaga governor by name. It would be odd, indeed unique, if he ignored Honganji temple towns.

His policies also evidently threatened the rights of some of his own retain-ers. In 1488 many of these retainers joined Honganji sect members and at-tacked the military governor. Aft er a short burst of battles, the governor’s castle fell to the rebels, and he committed suicide. Th e victors installed a less ambitious relative of his as military governor and began a new system of gov-ernment in which warriors and ikkō leagues shared rule over Kaga. One con-temporary author wrote that because of this appointment from below, as it were, Kaga seemed “like a province held by commoners.” Partly because of this oft - quoted phrase, it has become a common impression that the ikkō leagues, that is, villagers, ran the province at that time. According to an anal-ysis of documents, however, it seems they played a role along with numerous other actors on the po liti cal stage of the province.



It appears that there was little eff ective control over warriors and the col-lection of rents in the provinces aft er 1488. Judging from letters sent by the shogunate to adjudicate cases of withheld rents and taxes, the right to collect rents continued to be highly contested among warriors with no direct in-volvement of ikki or Honganji sect temples. In one case, Honganji adherents make an appearance, helping one warrior against another warrior, each intent on rights to collect taxes (and keep some) in a par tic u lar landholding. Over time, the ikki, Honganji sect temples, and Honganji itself seemed to become more active in disputes over rent and tax collection. Eventually, around 1530, the province came under the rule of the main Honganji Temple itself, which acted as military governor. Interestingly, Honganji did not establish a daimyō- like domain but rather maintained the former system of overlapping authori-ties on a local level, where other temples and aristocrats had jurisdiction over estates they held. Indeed, rather than try to replace the old, complicated sys-tem of land own ership, Honganji tended to champion the old temples and aristocrats in their attempts to collect taxes and rents from their holdings.

Where rule by leagues, either on their own or in a kind of co ali tion of leagues and warriors, represented an alternative to daimyō rule, ultimately such a government was unsustainable in the context of widespread warfare and the struggle to unify the country. For any kind of warfare, a unifi ed leader-ship, or at least a pyramidal or ga ni za tion, confers distinct advantages. In the case of invasion, all the diff erent groups— leagues, diff erent warrior bands— might fi ght the invader, but they will be much less eff ective if there is no overall coordination. In addition, diff erent groups within the co ali tion might back dif-ferent national military fi gures, resulting in provincial disorder.

Another factor was that, almost by defi nition, league and local rule re-quired a looser central authority than the new daimyō domains aff orded, whether it was exercised in a single province or writ large over the whole coun-try. A domain made for a more effi cient administration and allotted greater power to the daimyō, so it was an advance in po liti cal or ga ni za tion. Th us, while leagues potentially represented a new form of provincial government, in another sense their interests were more conservative than radical. Th ey repre-sented a new way of governing locally, yet required the old way of governing nationally. Th e daimyō domain, and later the unifi ers, introduced new regimes on both levels.

An uprising similar to that of Kaga occurred in Mikawa province (the eastern half of modern- day Aichi prefecture) in 1563. Th at province, like

Kaga, featured a young military governor seeking to establish centralized daimyō rule. In this case, the warrior was Tokugawa Ieyasu, who ultimately founded the stable government that ruled Japan until 1867. Aft er establishing himself in western Mikawa, he turned east, where three Honganji sect tem-ples had bustling temple towns with rights to bar entry to warriors. Th ese rights confl icted with Ieyasu’s ambitions as well as his ability to protect his retainers when they suff ered at the hands of townspeople, because the temples alone had the authority to punish wrongdoers.

Th ere are two explanations for the proximate cause of the uprising. In one, Ieyasu tried to levy a tax on a temple, and its supporters rebelled, going so far as to kill the warriors Ieyasu sent to demand the tax rice. In the other version, a warrior tore apart the shop of a merchant in a temple town, and the towns-people set upon the warrior. He later complained to Ieyasu, who moved to punish the townspeople, in theory the prerogative of the temple. Regardless of which incident sparked the rebellion, Ieyasu attacked the temples’ rights, and the sectarians came to the temples’ defense, resulting in outright warfare. Some of Ieyasu’s more powerful retainers also joined the rebels, as they op-posed some of his policies, though not necessarily those pertaining to the Honganji sect temples. One opposed a recent move Ieyasu had made to ally with his neighbor to the west, Oda Nobunaga, instead of a longer- standing alliance with the daimyō family to the east, the Imagawa. Others appear to have wanted to take Ieyasu’s place.

Perhaps more ominously for Ieyasu, his own retainer band split, as many were Honganji sectarians, and they joined the temples rather than supporting their military overlord. Some of his generals and closest retainers were also Honganji adherents, but they remained loyal. A few changed their religious affi liation to accommodate their lord, and one refused to renounce his faith but handed over his son as hostage to guarantee his loyalty. Th ose of Ieyasu’s retainers who held lower positions in his or ga ni za tion split cleanly along reli-gious lines. Th is must have come as a shock, because before this, religion had played little role in warrior politics. In an earlier confl ict in Mikawa, members of the same temple congregation fought on diff erent sides in the warfare.

It took six months before Ieyasu could prevail, and even then he had to grant amnesty to the sectarians in his band before the opposition would surrender. In spite of a promise to allow the temples to continue “as they were originally,” he declared that “originally they were fi elds, so fi elds they shall be.” He had them razed and banned the Honganji sect temples in his domain.



Th e experience of the Mikawa ikkō ikki rebellion had repercussions in later years, for members of both sides of the uprising. Ieyasu had learned the dangers of exclusivist religions, and their power to override other types of loyalty, a lesson he would not, apparently, forget. For the Honganji adherents, the destruction of the temples must have underscored the necessity of Hon-ganji’s in de pen dence and military power for its continued existence. Th us, when Oda Nobunaga threatened the main Honganji Temple’s in de pen dence, the members fought all the more vigorously in the sect’s defense.

In 1568, Nobunaga rode into Kyoto triumphantly, in a position to domi-nate the national government and impose his rule, gradually, over the rest of Japan. At fi rst Honganji, among others, welcomed and congratulated him; the Patriarch Kennyo sent him gift s just before Nobunaga entered the capital, and more within a few months, at the new year. Th e cordiality would not last. By this time Honganji itself, the sect’s headquarters on Naniwa Bay (in what is now Osaka), came to be “an elaborate fortress- like complex with magnifi cent temple buildings and a large temple town ( jinai machi) surrounded by walls and moats.” Even though Honganji claimed the tax exemptions and autono-mous rule enjoyed by most temple towns, Nobunaga demanded that the tem-ple pay a special tax of 5,000 coins (kan). Honganji regarded this as an aff ront, an abrogation of its rights, as indeed it was, and reacted much the same way that the Mikawa temples had: it treated Nobunaga’s actions as a threat to its very existence and rallied the sect’s followers to oppose the daimyō.

Th is struggle, which would continue off and on for ten years, constituted the largest of the ikkō leagues’ fi ghts and one of the most per sis tent obstacles to Nobunaga’s ambitions. Oda Nobunaga laid siege to Honganji headquarters, seeking to isolate it from outside help, but the navy of the Mōri family of war-riors helped to provision the temple from the sea. Th e defenders inside its walls, for their part, included “thousands of arquebus gunners” who drove back repeated assaults by Oda Nobunaga’s forces. In 1576, Oda Nobunaga himself was wounded in the leg in one of these battles in spite of possessing a preponderance of force. Honganji ultimately capitulated in 1580, starved of food and ammunition in part because of the defeat of the Mōri, but even then it was in a position strong enough to guarantee the sect’s continued presence, unlike the temples in the parallel case of Mikawa. To be sure, the Honganji Temple and its town were handed over to Nobunaga. Th e temple and town were also burned to the ground, though it has never been clear whether the priests or the townspeople set the fi re to deny Nobunaga its assets, or whether

Nobunaga ordered that it be burned. Nevertheless, he did not outlaw the sect as Ieyasu had in Mikawa.

If the in de pen dence of the three temples in Mikawa threatened Ieyasu’s ability to rule his province, then Honganji’s power made it impossible for Nobunaga to exercise the kind of authority he needed to achieve his goal of unifying Japan under his rule. Honganji had members in strong numbers throughout the Hokuriku provinces of Echizen, Kaga, Noto, and Etchū, in the provinces of Ōmi to the east of Kyoto, Kii to the south of Osaka, and Owari, the western half of modern- day Aichi prefecture. It had close ties to many daimyō who opposed Nobunaga. Perhaps most importantly, it controlled a signifi cant part of the country’s economy through its temple towns and the province of Kaga. Finally, Nobunaga had no authority within Osaka or the temple towns. Osaka took its right to bar entry to warriors seriously. Once, a daimyō sent his vassal to Osaka with a gift for the Patriarch, who sent a mes-senger to meet the vassal and pick up the gift outside the town because he would not permit the vassal to enter. It seems extremely unlikely that Nobu-naga could have allowed this system of autonomous towns to continue, and Honganji’s in de pen dence, with its military past and many members through-out central Japan, had to be destroyed.

Th e geographic spread of Honganji’s power meant that it took Nobunaga years to fi nally conclude the war decisively. He attacked and defeated Honganji sectarians methodically, moving from one geographic region to another. Some places put up spirited re sis tance. For two years, Nobunaga underestimated the amount of force necessary to defeat the adherents in Ise province, at Na-gashima. Nagashima was a community built on a swampy delta at the confl u-ence of three great rivers, the Kiso, Nagara, and Ina, and many smaller ones. Living on islands in the marshes, the inhabitants used the dubious terrain as an asset for self- defense. It took two unsuccessful assaults for Nobunaga to re-alize that he needed a navy to mount his attack if he intended to be victorious. Legend has it that Honganji adherents booby- trapped the reed beds with old pots buried up to the necks in the sand, which could break the ankles of invad-ing foot soldiers or battle horses, and tied ropes to stakes just below the water level to serve as tripwires. Nobunaga had to make three desperate forays be-tween 1571 and 1574 fi nally to subdue— and massacre— the members. Massacre was Nobunaga’s preferred method for dealing with Honganji fi ghters, whom he regarded as worthless at best. It is said that as many as 40,000 residents of Nagashima, including women and children, died as a result of Nobunaga’s last



assault, although one must note that numbers in premodern sources tend to be exaggerated. If he meant to demoralize the other Honganji adherents, he seems to have had scant success. Each time they met in battle, the ikkō ikki participants fought tenaciously— and Nobunaga slaughtered them.

Of the three rebellions, only the one opposed to Nobunaga left any signifi -cant documentation from the Honganji side. Th e events in Kaga must be gleaned primarily from sources outside the province, and in Mikawa, very little remains from the members’ side. So far, only one document explaining the outbreak from one temple’s perspective, and one narrative about the fi ght-ing written by a temple’s supporter, have emerged. In the case of the opposi-tion to Nobunaga, by contrast, numerous letters written by the Honganji leader have been assembled in one collection, which includes correspondence from the leader to members and also to the sect’s daimyō allies.

Th e fi rst call to arms that the leader, named Kennyo, sent out to his follow-ers in 1570 said that Nobunaga had been causing trouble for the sect. He urged his followers to exert themselves in defense of “our line,” that is, the line of teaching set down by the found er, from whom the leaders also descended. In no extant document did Kennyo explicitly promise rebirth in paradise to those who fought for the sect, nor did he threaten damnation to those who failed to do so. His father and pre de ces sor once wrote a note to console the families of some members who fell in defense of the sect, assuring them that “Th ere is no doubt that the people who died in battle achieved rebirth in the Pure Land” because they had fought as the found er’s allies. Kennyo, how-ever, did not go so far. Once, he told followers that their willingness to fi ght for the temple demonstrated their fundamental desire for rebirth in the Pure Land, implying that this desire constituted the requisite faith to achieve that rebirth. Otherwise, he restricted himself to urging them to “settle their faith,” in the standard language of the sect’s teachings— that is, to abandon themselves to Amida’s grace and his promise to save all sentient beings. Th is was all the more important in time of war, as death came so frequently. If one died without faith, one would regret it forever. He did, however, expel mem-bers who failed to support Honganji and go to war.

Ordinary members tended to believe that expulsion from the sect neces-sarily meant that the expelled member was condemned to hell. Strictly speak-ing, Honganji’s teachings held that faith in Amida’s grace alone would result in rebirth in his paradise. One must recognize that no power of one’s own, not even the will to have faith, would avail. Only the deep understanding that

even faith was called forth by Amida’s grace, without regard to any will or merit adhering to the individual, would work.

Compared to other teachings, even within the True Pure Land teachings, this was abstract and complicated thinking. Other sects tended to rely on more tangible means to assure members of their rebirth in the Pure Land. Some used registers in which the believers’ names were inscribed. By that act, resulting in a physical artifact, the believer was certain to achieve salvation. Another distributed amulets that established a karmic bond with Amida that would bring the believer to his land aft er death. Only Honganji depended on Amida’s own “other” power, denying the effi cacy of amulets, registers, priests, or even one’s will to faith. It should be no surprise that many followers appar-ently had trouble grasping all the implications of Honganji’s teaching and placed their faith in Honganji’s leaders as much as they did in Amida. If the leader, in this case Kennyo, expelled a member, that meant that he judged the member’s behavior to be inconsistent with faith; therefore the member lacked faith and would fail to be reborn in paradise. Although in general Buddhism off ers multiple possibilities for reincarnation, on several diff erent planes of existence, in the sixteenth century Pure Land and True Pure Land followers seem to have believed in two alternatives: reincarnation in paradise or in one of several possible hells.

Th us, Honganji followers believed the leader, in this case Kennyo, had the power to condemn people to hell. According to one of the sect’s sixteenth- century historians, they also believed the leader had the power to guarantee rebirth in paradise. So even if Kennyo did not explicitly claim these powers, we must assume that they formed part of the mindset of those who fought on Honganji’s behalf. Furthermore, since Kennyo assured his followers that No-bunaga threatened the very existence of the sect, it would have been hard to conceive that anyone with faith in Amida and Honganji’s teachings would fail to respond to the call to arms. Th is, then, was the context in which the “Ad-vance and be reborn in paradise . . . ” banner appeared.

Th is simple ideology of the Honganji leader guaranteeing either salvation or damnation did not prevail over the entire era of ikkō leagues and their fi ghting. In most cases, including many that have not been discussed here, the motivat-ing factor was survival, either of specifi c temples or of the sect itself. Although the Honganji sect taught obedience to the worldly authorities, no one could ex-pect members to acquiesce to the destruction of their temples or the proscrip-tion of their religion by those authorities. In those cases, with or without the



encouragement of the sect leaders, the followers took up arms. In the case of Kaga, the justifi cation for fi ghting is more obscure because we do not know the proximate cause of the rebellion. We only know the general condition of his consolidation of rule, and that the ikkō members and his retainers took an op-portunity he gave them by temporarily absenting himself from the province.

Whether or not the leaders of the sect sanctioned it, and for the most part they did not, the teachings allowed members to feel less threatened by the kinds of authority to which they were subject, outside the sect. Th ey had little need of warrior protection, since they demonstrated time and again their own military abilities. Th ey had little fear of repercussions because they believed their future lay in paradise. So much of Japa nese religious belief at the time involved this- world benefi ts to be derived from worship that it is likely that Honganji sect members, too, expected some protection from Amida in the present world, even though such beliefs had no place in the sect’s teachings. Most importantly, however, they believed that they were saved and that others outside the sect were not. Th is last tenet, combined later with the beliefs in the power of the leader, proved to be extremely potent as motivation and cohe-siveness for the leagues. Th at the cohesiveness could override other ties, as they did in the case of Ieyasu’s retainer band, made the sect seem especially dangerous to warriors.

Ikkō leagues, as administrative units overseeing government within a re-gion, rarely, if ever, appeared outside of Kaga. Other leagues did, and the ikkō leagues presented no more nor less of an alternative to daimyō government. Under assault, however— and most regions found themselves under assault at one time or another in the sixteenth century— the benefi ts of a unifi ed com-mand trumped any benefi ts of league government. Because of policies ad-opted by Oda Nobunaga, Toyotomi Hideyoshi, and Tokugawa Ieyasu, the ikkō leagues had no real descendants or imitators. Nevertheless, they played an important role in the formation of those policies, albeit a negative one from their perspective.

Nobunaga began, and Hideyoshi extended nationally, a policy of separat-ing warriors from villagers, oft en identifi ed as farmers. Th ey would bring the warriors into castle towns and make them dependent on what were essentially stipends rather than on lands they worked or oversaw directly. Th e villagers were then deprived of some of their weapons, though they were never wholly disarmed. Th is policy was designed, in part, to pacify the countryside and to prevent feuds, or private wars, from breaking out in the provinces. It also

brought the warriors more closely under the supervision of their overlords. For Honganji members, this had the eff ect of removing the most able leaders of their numbers and may, indeed, have been intended to. Even without the experience of ikkō leagues and their uprisings, the national leaders may have pursued the same or a similar policy because of the debt cancellation and tax relief ikki that also occurred frequently at that time. It is nonetheless diffi cult to imagine that the Honganji members’ activities played no role in making such a policy desirable.

From the beginning of their rule, daimyō recognized that temples had to be separated from their in de pen dent economic and military power. Th e uni-fi ers Oda Nobunaga, Toyotomi Hideyoshi, and Tokugawa Ieyasu in par tic u lar went about making sure that all temples relied on them for income and pro-tection. Honganji’s re sis tance to Nobunaga formed part of that struggle, but Nobunaga’s policy of subduing temples did not begin with Honganji. One of his most notorious attacks leveled an 800- year- old temple on top of Mount Hiei, just to the north of Kyoto in 1571. He encircled it with his army and com-manded his men to take no prisoners but rather to kill any people they came across. Th e attack horrifi ed Kyoto aristocrats, who saw in Nobunaga the en-emy of all Buddhism.

More interesting, however, is the Tokugawa regime’s policy that imposed Honganji’s main- branch temple system on all Buddhist sects. Honganji was by no means the only religious or ga ni za tion to be so constituted, but one can argue that its authority over its followers demonstrated forcefully the advan-tages of the main- branch pyramidal structure. It was through this pyramidal structure that Honganji managed to mobilize so many of its members, in so many distant provinces, to come to its aid in several ikki episodes.

Most directly, the ikkō leagues proved decisively how a religious or ga ni za-tion could provide motivation, cohesiveness, and a structure to match that of the warrior or ga ni za tion. Ties of a common religion could and did overcome those of warrior loyalty, most vividly in the case of Mikawa and Ieyasu’s re-tainer band. As noted earlier, the ikkō leagues were not the only religious or-ga ni za tion that displayed this tendency. Th e Nichiren leagues and the highly exclusive fuju fuse branch of Nichiren also did their share to draw warriors’ attention to the danger posed by religious solidarity. Th e Nichiren followers, however, came primarily from the urban classes, not from the village elite and warriors that Honganji drew. Th ey did not, then, threaten daimyō nearly so much as Honganji did. No other Buddhist sect cost the warriors anywhere



near the number of lives that ikkō leagues claimed. Nobunaga himself lost several close relatives to the Honganji adherents. None showed so clearly how religion could override retainer fealty as Honganji members did in Mikawa.

A policy begun by Hideyoshi, one that ultimately became one of the benchmark policies of the Tokugawa shogunate, was the exclusion of Chris-tianity and its teachings from Japan. Th e hegemons feared the allegiance of Christians to a foreign leader, the Pope. Perhaps even more, they feared the automatic alliance of Japa nese Christians with foreign Christians, and the threat that would pose should a Christian country choose to invade Japan. Given the alacrity with which Christian nations fought with each other, one would be forgiven for thinking that the hegemons worried a bit too much. Nevertheless, Ieyasu had seen his own trusted followers split into two camps based on shared religion, and the Mikawa uprising was well known among his descendants. It is unclear to what extent Ieyasu and his contemporaries shared a common Japa nese identity strong enough to make that more impor-tant than a common religion. Although it would no doubt be too much to state that Christianity was banned in Japan simply because of Honganji and the ikkō leagues, surely Nobunaga’s and Ieyasu’s experiences with them weighed heavily as Nobunaga’s successors estimated the potential dangers of the new religion.

In the end, the most visible of the commoner movements, if such we can call the ikkō ikki, died without successors. Leagues as such continued to be important or ga niz ing mechanisms for villagers, especially in their opposition to overtaxation, certainly, but ikkō ikki were only one manifestation of the league, and leagues did not originate with Honganji. Th e ikkō leagues’ great-est impact was the policies that guaranteed they could not occur again.

Notes 1. Onuma Tsutomu, 2001; Kikuchi Masanori 2008: 185ff . 2. Fujiki Hisashi, 2005. 3. See also Adolphson and Ramseyer, 2009. 4. More precisely, Honganji and warriors together put in place a weak military

governor (shugo). Honganji itself began to act as shugo offi cially in 1530. See Tsang, 2007: 94ff . One must note that, as described above, the fi ghters themselves remained active until 1582; it was Honganji’s temporal power in Kaga that succumbed in 1574.

5. Nijōji shukaki, entry dated Tenbun 1 [1532] July 17; Nisuiki, same date. 6. See Imatani Akira, 1989. 7. Fujiki Hisashi, 1998.

8. Th ese taxes were known as hanzei. See Tanaka Katsuyuki, 1993. 9. For a more detailed discussion of the right to bar entry to warrior offi cials, also

known as shugo funyū, see Tsang, 2007: 24– 26.10. Hogo no uragaki, 1977: 747– 748.11. Muromachi bakufu monjo shūsei, vol. 1, 1986: 347, letter dated Bunmei 12 [1480]

(October 2).12. Jitsugo, 1977a: 416.13. Bugyōnin hōsho, vol. 2, 1986: 59– 60, letter dated Eishō 2 [1505] (May 23).14. Th e year of the uprising is not entirely clear. It may have been 1562, though 1563

is more likely. See Tsang, 2007: 209.15. Shingyō Norikazu, 1975: 225– 231.16. Ōkubo Hikozaemon Tadataka, 1974: 95.17. Kennyo shōnin bun’an, 1979: 1139, 1143.18. Kuroda, 2006: 46.19. Shinchō- kō ki, 1965: 193.20. Lamers, 2000: 165.21. Tsang, 2007: 223.22. Suitō Makoto, 1992: 95– 96.23. Shinchō- kō ki mentions nine rivers, 159.24. Turnbull, 2005. Shinchō- kō ki, the most reliable primary source on the battles,

does not mention them. Th en again, as Nobunaga’s offi cial biographer, author Ōta Gyūichi may not have wanted to make any of the ikki fi ghters sound clever or capable.

25. Shinchō- kō ki, 1965: 179, says they are “mononokazu ni mo sezu,” essentially “worthless beings.”

26. Lamers, 2000: 103. He states that 20,000 died in battle, and another 20,000 had died of starvation. Also, Tōji kōmyō kakochō mentions 40,000, including those killed by the sword and drowned. Shinchō- kō ki, 1965: 163, says that Nobunaga killed 20,000 men, women, and children alone by herding them into a fortress, then setting fi re to it.

27. Eiroku ikki yurai, 1989: 666– 667; Watanabe Chūuemon oboegaki, 1977: 83– 94.28. Kennyo shōnin bun’an, 1979: 1133– 1178.29. Th e Honganji line of teaching and leaders traced their descent from Shinran

(1173– 1262), a priest active in the thirteenth century. For the call to arms, see Honganji monjo, 1986: document 23 [unnumbered page].

30. Letter quoted in full in Kinryū Shizuka, 1989: 100– 101.31. Kennyo shōnin bun’an, 1979: 116632. Jitsugo, 1977a: 586. He dates the beginning of this belief to the reign of Kenn-

yo’s pre de ces sor, Shōnyo, and criticizes it as heretical.33. It was a sign of Oda Nobunaga’s growing momentum that in the years between

1571 and 1573 he went to war against fi ve military houses (Asakura, Asai, Takeda,



Rokkaku, Miyoshi, and Matsunaga) in addition to two powerful religious institutions, the temples of Enryakuji on Mount Hiei and Honganji in Osaka. See Berry, 1982: 44.

ReferencesAdolphson, Mikael, and J. Mark Ramseyer. 2009. “Th e Competitive Enforcement of

Property Rights in Medieval Japan: Th e Role of Temples and Monasteries.” Jour-nal of Economic Behavior and Or ga ni za tion 71 (3): 660– 668.

Berry, Elizabeth. 1982. Hideyoshi. Cambridge, MA: Harvard University Press.Bugyōnin hōsho, vol. 2, 1986: 59– 60, letter dated Eishō 2 [1505] (May 23).Eiroku ikki yurai. 1989. In Kodai, Chūsei shiryō, vol. 6 of Shinpen Okazaki- shi. Oka-

zaki: Okazaki shiyakusho.Fujiki Hisashi. 1998. Sengoku no sahō: mura no funsō kaiketsu [Rules of the Warring

Provinces: Solving Village Disputes]. Tokyo: Heibonsha.Fujiki Hisashi. 2005. Katanagari: buki o fūinshita minshū [Th e Sword Hunt: Th e Com-

mon People Who Sealed Up Th eir Weapons]. Tokyo: Iwanami Shoten.Hogo no uragaki. 1977. In Shinshū shiryō shūsei, vol. 2 of Rennyo to sono kyōdan, ed.

Katada Osamu. Kyoto: Dōhōsha.Honganji monjo. 1986. Edited by Chiba Jōryū and Kitanishi Hiromu. Tokyo: Kashiwa

shobō.Imatani Akira. 1989. Tenbun hokke no ran: busōsuru machishū [Th e Lotus Disturbance

of the Tenbun Era: Th e Townspeople Who Took Up Arms]. Tokyo: Heibonsha.Jitsugo. 1977a. Honganji sahō no shidai in Rennyo to sono kyōdan, vol. 2 of Shinshū

shiryō shūsei, edited by Katada Osamu. Kyoto: Dōhōsha.Jitsugo. 1977b. Tenshō sannen ki in Rennyo to sono kyōdan, vol. 2 of Shinshū shiryō

shūsei, edited by Katada Osamu. Kyoto: Dōhōsha.Kennyo shōnin bun’an. 1979. In Ikkō ikki, vol. 3 of Shinshū shiryō shūsei, edited by Ki-

tanishi Hiromu. Kyoto: Dōhōsha, 1979, 1133– 1178.Kikuchi Masanori. 2008. Sengokushi [History of the Warring States Period]. Tokyo:

Saitosha.Kinryū Shizuka. 1989. “Kinai Tenbun no ikkō ikki” [Th e Ikkō Ikki of the Tenbun

Years in the Region Around the Capital], in Sengoku, Shokuhō, vol. 5 of Komonjo no kataru Nihonshi, edited by Minegishi Sumio. Tokyo: Chikuma shobo.

Kuroda, Toshio. 2006. “Leaders in an Age of Transition,” translated by Th omas Kirch-ner. In Rennyo and the Roots of Modern Japa nese Buddhism, edited by Mark L. Blum and Shin’ya Yasutomi. New York: Oxford University Press.

Lamers, Jeroen. 2000. Japonius Tyrannus: Th e Japa nese Warlord Oda Nobunaga Re-considered. Leiden: Hotei Publishing.

Muromachi bakufu monjo shūsei, Bugyōnin hōsho. 1986. 2 vols. Kyoto: Shibunkaku shuppan.

Nijōji shukaki. 1532. In Zoku nengyō zatsuroku. Tokyo University Historiographical Institute, manuscript facsimile, entry dated Tenbun 1 [1532] (July 17).

Nisuiki. 1532. Tokyo University, Historiographical Institute, manuscript photocopy.Ōkubo Hikozaemon Tadataka. 1974. Mikawa monogatari in Nihon shisō taikei, vol. 26,

edited by Saiki Kazuma et al. Tokyo: Iwanami shoten.Onuma Tsutomu. 2001. Ikkō ikki to iu monogatari [Th e Story of an Ikkō Ikki]. Kanazawa:

Konishi Publishers.Shinchō- kō ki. 1965. In Sengoku shiryō sōsho, vol. 2, edited by Kuwada Tadakichi. To-

kyo: Jinbutsu Ōraisha.Shingyō Norikazu. 1975. Ikkō ikki no kisō kōzō: Mikawa ikki to Matsudaira- shi [Th e

Basis of the Ikkō Ikki: Th e Mikawa Ikki and the Matsudaira Family]. Tokyo: Yoshi-kawa kōbunkan.

Suitō Makoto. 1992. “Ōsaka jinaichō no hibi: Tenbun nikki kara” [Daily Life in the Osaka Temple Town: From the Tenbun Era Diary (of the Honganji Patriarch)]. Kokuritsu rekishi minzoku hakubutsukan kenkyū hōkoku 39.

Tanaka Katsuyuki. 1993. “Mura no ‘hanzei’ to senran/tokusei ikki— sengokuki Kyōto kinkō sonraku no rentai to buryoku dōin” [Villages’ “Military Tax” and Warfare/Tax Cancellation Demonstrations— Associations and Military Activities of Vil-lages Near Warring Provinces- Era Kyoto], in Shigaku zasshi 102:6.

Tsang, Carol Richmond. 2007. War and Faith: Ikkō Ikki in Late Muromachi Japan. Cambridge, MA: Harvard University Asia Center.

Turnbull, Stephen. 2005. Japa nese Fortifi ed Temples and Monasteries, ad 710– 1602. Oxford: Osprey Publishing.

Watanabe Chūuemon oboegaki. 1977. In Shingyō Norikazu, “Mikawa ikkō ikki ron ho- i (sono ichi).” Chūsei shi kenkyū 2: 83– 94.


110

LOCATED EAST OF YAMATO (about 50 miles southeast of Kyoto), the old province of Iga covers a basin perfectly surrounded by mountains with summits towering from 1,600 to 3,200 feet. Th e main river crossing the province, the Nabarigawa, snakes down between impressive gorges. Th e region forms a circle about 50 miles from north to south and from east to west.

Just north of Iga lies the district of Kōga with a landscape of hills descend-ing down to the Biwa Lake. Between Iga and Kōga, there are mountains whose highest peaks can reach up to about 3,200 feet. Kōga was historically a part of Ōmi province.

Southwards of Iga, on the other side of mountains that demarcate the bor-der between the Iga and Ise provinces, there lies a landscape of hills and plains that presently skirt the Kintetsu railroad line between Ōsaka and Matsuzaka. At the end of the medieval period, this very narrow stretch of country was a manor estate called Oyamato Shōen.

Th ose regions, which were only a two or three days’ walk from Nara and Kyoto, the former Japa nese capitals (and today, less than an hour from Osaka by express train), were not located on main trade routes and thus never con-stituted a particularly important po liti cal or strategic stake during medieval and ancient times. In the absence of centralized po liti cal control the arable lands in the region quickly fell under the infl uence of the great monasteries of Nara like Tōdaiji or of big sanctuaries like that of Ise which controlled those areas (then called shōen or mikuriya). In the fi ft eenth century, the region was held under the offi cial domination of two families, the Nikis and the Tsutsuis, warlords whose names did not become famous. Sometimes pilgrims and im-

Autonomy and War in the Sixteenth- Century Iga Region and the Birth of the Ninja PhenomenonPierre Souyri

6

Autonomy and War 111

perial messengers en route to the great shrine would take a secondary road to go from Kyōto to Ise. Th at road cut through the basin of Iga and the land of Oyamato but avoided the district of Kōga, which lay further north.

From afar, the basin of Iga seems tranquil and easy to cross, but once the traveler takes a closer look at the spurs surrounding the region, the picture looks quite diff erent. Abrupt summits, slopes covered with pine trees, narrow, winding paths, streams turning into torrents in deep gorges would give the impression of a mountain if it were not for the humble height of the surround-ing hills. Th e people from the region called those extremely narrow paths ko-guchi or tigers’ mouths (only one horse could go through them at a time), and it was thought that only the most intrepid would dare to cross those passes. Needless to say, in medieval times these passes were strategic spots: A mere handful of armed men, well- retrenched and knowledgeable of the terrain, could easily control convoys trying to make their way through the hills. Th e problems were the same for military troops, who saw their own incursions into the province frustrated by the terrain and the locals’ skilled use of it. Th ere was no crossing the Iga region without the approval of local authorities, for huge bands of ruffi ans and various bandits (akutō) were scouring the country at the end of the thirteenth century.

It was becoming ever clearer to the people of the region that without a means to ensure local stability and curtail the raids of the akutō, local com-munities would not survive. In 1494, in the anarchic early years of the War-ring States (sengoku) period, the male inhabitants of the region of Oyamato wrote down two sworn declarations addressed to a certain monk named Shinsei, the abbey of the Jōganji, a temple recently created in the region. Th e documents were of an explicitly po liti cal nature, but given the overwhelming po liti cal infl uence of the Buddhist temples on the inhabitants of the land, this is not surprising.

Th e fi rst of the Oyamoto declarations represented a kind of local constitu-tion establishing new rules for local life. Th e fi ve- article document, established on the fi ft eenth day of the eighth moon of Meiō, was signed by 350 common people (hyakushō) gathered in the villages. One of its provisions, designed to safeguard local institutions of rural life and stem violence in the region, de-clared, “Th e hyakushō shall not strife for paddy fi elds, mountains or forests. Th ey shall not seize cultivation rights or steal.” Th is declaration might be thought of as a code of conduct for the farmers who signed it, rules each per-son pledged to obey in the name of the collective security.

112 Pierre Souyri

Th e second document, dated a month later, on the twenty- fi rst day of the ninth moon, was sworn to by forty- six people who called themselves ikkachū, or “families.” In this document, the families sought to ensure the regional peace, vowing not to fi ght over taxes on peasants and to try to prevent insur-gent acts by subordinated people who might “cause trouble.” Th e charter’s in-tention was stated unequivocally. One of its articles read, “If anyone acts badly, inside or outside Oyamoto, he will be judged and sentenced.”

Th ese forty- six citizens of Oyamoto were the leading class in the country-side. Th is class of people is well known by historians of Japan, who call them the jizamurai or samurai of the soil— that is, local warriors. On this occasion, the Oyamoto jizamurai were laying down the foundations of an ikki, a pledged or ga ni za tion with the clear aim of maintaining order within the domain.

Th e league of forty- six families was created fi ve weeks aft er the league of the “common people,” aft er both had gathered and sworn their pledges. Th e warriors’ league seized administrative and judicial control over the estate, its authority based on the charter signed by the 350 peasants. However, most im-portant to the success of the endeavor was the understanding that these two groups had formed a united front. Although the local warriors had their own system of cooperation, they were compelled to respect their agreement with the peasants, without whose cooperation the region’s autonomy might be im-periled. Both groups of pledges had many shared priorities, but the most im-portant, clearly, was the common objective of maintaining local order and peace and limiting or preventing violence. Violence here seems an endemic feature of the local society, the result of social confl icts over land and access and not that of external aggression. Th ere were no wars, as such, at the time. Th ese leagues appear to have been designed to establish expectations of local cooperation and stability.

Th rough these organizations, a double structure of power began to de-velop: the assembly of the forty- six local warriors and the general peasant as-sembly. Th e “families,” or petty samurai, led society but only on the basis of the agreement— one might want to call it a consensus— of the “common people,” who were the fi rst to or ga nize into this sort of social arrangement, as far as we know. Th e 1 to 7 size ratio between the two groups suggests that the pro cess of social diff erentiation was limited, although, on the other hand, the very exis-tence of the two groups suggests that the social identity of the two groups was clear for all parties concerned. Th e two groups had a relationship predicated on power and domination, but without either a suzerain or absolute power. Th is

double structure was created within a community facing the threat of internal violence, whose concrete origins are still unclear but are most likely to have involved land and water rights.

For several de cades at least, the system seems to have worked. From 1494 on, the domain of Oyamato, once a minor estate of patrons in Kyoto, became an autonomous society operated in a markedly in de pen dent fashion from outside hierarchical control, at least at the po liti cal and judiciary levels. Th e local communities had seized the right to govern themselves without seeking the permission of any outside power. At the level of the old domain, a collec-tive local or ga ni za tion had appeared to supplant the waning powers of the ancient lords, and taxes were now to be collected and distributed within the domain. In Oyamato, there had once been a collective or ga ni za tion ruling over a relatively vast rural community (not an urban one like those in Sugaura or Imai) that was then called gō. By this point, however, all feudal seigniorial power had been swept away, and yearly taxes were for the most part kept within the community.

Around the same time in nearby Iga, an or ga ni za tion similar to that in Oyamato appeared. We do not know exactly when this or ga ni za tion was created, in part because its name and “constitution” emerged only when the or ga ni za tion was directly threatened by the power of strong lords of the neighborhood. Th e communities of the Iga River Basin had been defending themselves since around the beginning of the sixteenth century. Th e neigh-boring region of Kōga in Ōmi, similarly or ga nized, had no fewer than 230 fortifi cation works. Residents of Koga had a long history of collectively man-aging the water of the Yasugawa, a river that was oft en diffi cult to tame and whose control ensured the safety of the valley’s crops. It was because the river needed to be controlled from the source down to the valley that strong local or ga ni za tion was indispensable.

Formal consolidation of the Iga communities seems to date from around 1560, when the locals drew up a formal twelve- article constitution. Th e basis of the Iga commune was a federation of village communes, but local power was exercised by the sixty- six local warrior families who made laws from the safety of their local fortresses and held important discussions and decision- making sessions in the nearby Buddhist temple of Heirakuji. Th ese local war-riors collectively administered the territory.

Th e Iga commune seems to have been a sort of geographic extension of Oyamato’s po liti cal and social structure, although it sprang up on a larger


114 Pierre Souyri

scale than that of its neighbor. Th e ruling body of the Iga was, in eff ect, no longer that of the domain but that of the province. Th is or ga ni za tion was called the Iga sōkoku ikki (the league of all the commons of Iga). It is known to us by a single text (the rules of the league), which now belongs to the Ya-manaka archives at the library of the Ise sanctuary. It was probably written between 1552 and 1568, when Oda Nobunaga and his army advanced into the southern part of Ōmi, the domain of the Rokkaku warrior clan, and very near to the Kōga district. Oda Nobunaga would prove to be an insurmountable threat a few de cades later, but the community was prepared to put up a fi erce collective re sis tance. We can surmise that the par tic u lar social system it lays out was created long before, around the beginning of the sixteenth century, almost at the same time as the neighboring league of the inhabitants of Oya-mato. Specialists believe that the league (or the written rules of the league) was created because of an urgent military situation, the looming threat of invasion of the province by Miyoshi Nagayoshi of nearby Yamato in the early 1500s. For example, the rules of the league explicitly forbade serving the Miyoshi. Th is league was led by local warrior families under the rubric of a two- level structure somewhat resembling that of Oyamato. Th e military strength of the league was strongly tied to local warriors’ capacity to train common people and to mobilize them for battle.

Th e league could confer the status of samurai upon draft ed peasants (called ashigaru, “light feet”) if they fought well on the battlefi eld. Th is status was possibly connected to tax relief, but becoming samurai was also, of course, a symbolic reward. We do not know if this “ennoblement” meant the acquisi-tion of a fi ef or a land— for one thing, one might well inquire as to whom this fi ef or land would be taken from. What is certain is that the samurai were proud of their names (ordinary peasants did not have surnames), their ideo-grams and blazons, and the prestigious collective authority they enjoyed. Be-ing a samurai did not necessarily involve an economic advantage, but it did constitute a strong element of prestige. In the Iga region the samurai were called general- warriors (musa taishō), and they led peasants whom they had trained for combat. Leading the others to the battlefi eld was considered a su-preme honor. Conversely, the league could also dismiss members of the or ga-ni za tion who refused to obey or to fi ght.

Point 9 of the league’s rules states that “acts of violence in the villages where troops are staying in case of draft are rigorously banned.” Peacekeeping thus appears to have been a mission of the utmost importance, one whose

success depended on the league members’ sense of common purpose. Th is sense of common purpose and collective discipline of the ikki represented a phenomenon totally diff erent from what occurred in the seigniorial armies of warlords, which were commonly involved in looting and thuggery. Such acts guaranteed them rewards; indeed, their very income depended on these acts. Th e experience of villages occupied by invading warlords was oft en nightmar-ish: as a result, their people oft en fl ed to fortifi ed places on the heights, what Fujiki Hisashi calls “castles of the village” (mura no shiro). Sometimes daimyō would declare a ban on acts of violence against civilians (rōzeki teishi) but troops had minimal respect for these laws. Sometimes local people would seek to buy peace from the warlords too, a practice that would oft en lead to extortion and rent- seeking. It is thus very clear why in local collective organi-zations, including in Iga, the peace proclamation and the ban on violence were essential. Freedom was important, but the freedom to live remained the most crucial of rights.

For Iga, the problem was that the agitation and unrest had been perma-nent since the end of the thirteenth century. Aft er the Ōnin War (1467– 1477), with the authority of the weakened Ashikaga shogunate in tatters, Japan en-tered a new period of anarchy and disorder, the Sengoku era (literally, “coun-try at war”). Th is era spawned two seemingly contradictory phenomena: on the one hand, the inversion of social hierarchies and gradual acceleration of social mobility, known as gekokujō (“the lower commanding the upper”), and on the other, the consolidation of feudal estates at the hands of major war-lords, the daimyō. As sovereigns over their lands, these warlords battled one another to maintain and expand their fi efs. By the late sixteenth century, the daimyō would be contesting the right to extend their authority over the whole country. Rebellion was a threat to the existing order from below, and the ex-pansion of daimyō power was a threat from above, but both contributed to the widespread violence and mayhem that ruled the day.

Th e Ōnin War represented the end of the old estate system and of the hey-day of the major lords (shugo), who emerged from the fi ghting greatly weak-ened. Most of the existing great warrior clans dispersed, torn apart by inner confl ict and unresolved issues of succession. In the resulting chaos, power over the provinces was gradually seized by the vassals of these once- formidable leaders: the shugodai, or governor’s deputies, the kinishū, or local barons, and importantly, the local warriors whom we have already met, the jizamurai. With the upper strata of feudal authorities paralyzed, bands of highwaymen


116 Pierre Souyri

(akutō) began to roam the land, conducting countless raids against monaster-ies such as Tōdaiji. Disorder and unrest reigned, and while religious leaders at the monasteries kept up their vociferous complaints about the bandits, they remained powerless to stop the attacks. Eventually, local warriors began to take matters of security into their own hands, forming a regional alliance of samurai known as the gunnai ichizoku. Th is alliance, which in the sixteenth century would eventually give way to a local league of fi ghters, the Iga ikki, had one main stated objective: restoring peace and order in the province.

In order to consolidate their authority, the Iga jizamurai reached across territorial lines to secure local allies. Th e last point of the rules of Iga’s league said that “the forces of Iga had to join those of Kōga.” With this intent, some common meetings of the leagues took place outdoors on the border between the two local organizations of the two territories, Iga and Kōga. A similar en-tity to the league of Iga, the Koga or ga ni za tion was called Kōgagun chūsō, “Gen-eral Assembly of Kōga district,” and was a small but well- organized league founded on common principles of solidarity and understanding and led by mi-nor warriors and their clans. Each of these warrior clans— the Yamanaka, the Ban, and the Minobe, among others— ruled over villages that were themselves gathered into (districts) sō. Here one fi nds another example of the same two- level system prevailing in Oyamato and Iga, as these villages banded together to form a “league of equal villages” (dōmyōsō).

A district (sō), which most Japa nese historians have compared with the medieval communes of the West, was a federated and allied cluster of villages that ruled territory together. Between the fourteenth and sixteenth centuries, these unique, de facto institutions of local government multiplied in central Japan. At the top level, a structure gathered leading families from the district, the so- called council of three families (sanpōsō). At the time, then, there ex-isted in Kōga three levels of social and po liti cal or ga ni za tion: the now well- known two- level leagues and above them, this additional structure at the district level. One can imagine from this that the lower ranking warriors, the jizamurai, were particularly powerful in Kōga because they were socially well or ga nized.

Or gan i za tion al units below the sō were the “village communes,” or sōson, which arose to replace large estates as the dominant units of authority. Early on, the rules of the commune were discussed and adopted in meetings led by village leaders, discussed in Chapter 4 of the present volume. Th ese communities gained legal recognition and became their own autonomous units, overseeing

irrigation systems and the management of the water supply, clearing land and forests to expand agricultural production, and importantly, collecting large sums of money to be set aside to cover the costs of collective self- defense. Th ey formed protection leagues with other regional communes and, by reaching across old territorial boundaries, further contributed to the weakening of land-owners’ domination over the land.

Th ese and other such alliances also had a precise military aim: that of pre-venting the military forces of outside lords— especially those of the daimyō— from entering the district and destroying the social or ga ni za tion the local warriors had created. To this end, diff erent communes pursued diff erent strategies for maintaining their autonomy. In Sugaura, for example, a sign placed at the entrance to the village read, “It is forbidden for local warlords (shugo) to enter this place, which is under autonomous judicial administra-tion.” In other regional leagues, the main objective was to force warlords to enter into seigneurial pacts that would provide checks on their authority. Th ese pacts, called kashindan, provided for the existence of groups of vassals under the rule of a lord and were highly respected. Local warriors, in fact, oft en called themselves kokujin, “people of the country,” a fact which signals that they saw the powerful lords of Kyōto and elsewhere as outsiders— in other words, as foreign people. Nonetheless, we can see that on some notable occasions, class trumped geography. Th e Iga warriors, in binding themselves together with the neighboring warriors of Kōga, clearly did not consider the Kōga warriors as foreigners but rather as allied forces in the struggle. Th is military alliance between neighboring groups of local warriors, which pro-tected local autonomy and in de pen dence from outside rule, was more impor-tant than narrow geo graph i cal loyalties and refl ects the fact that, when econo-mies of scale called for larger organizations, warriors reached laterally before leveling po liti cal status in their own communities.

Th e relationship between local warriors and peasants was complex, and when confl icts arose between the two, village unions could fall apart. Some-times the warriors might abandon the community and instead join the vassal organizations of major warlords, actually helping to expand the authority of these new feudal barons when it served their material interests. At other times, however, they might choose to emphasize their ties to local community organizations, leading rural uprisings against taxation and debt collection, thereby promoting local autonomy. Because of the existence of a common enemy threat— the territorial encroachment of powerful warlords (sengoku


118 Pierre Souyri

daimyō)—local warrior organizations gained the widespread support of com-moners in the middle of the sixteenth century and were able to extend their control over their territory. In doing so, they slowed the pace of territorial consolidation in Japan. Th ese local leagues, and the institutions they left be-hind, would survive at least until the 1570s, when the threat of outside inva-sion by warlords resurfaced.

Aft er 1568, Oda Nobunaga, with the help of a very powerful and well- organized army, emerged in Kinai as a new type of ruler. His military tactics were unmatched in their savagery and ruthlessness, although the regional commune of Iga tenaciously resisted his attacks for more than ten years. Eventually, however, the federations of local warriors, social and po liti cal structures that took root primarily in backward and mountainous areas, would prove no match for Oda Nobunaga’s armies, whose guns and cannon fi re crushed the small wood- and- earth castles of the Iga warriors. In 1579, Oda’s son, Oda Nobukatsu, launched a major attack against the in de pen dent Iga republic. To the dismay of the great warlord, he was repulsed by the people of Iga (Iga no mono), whose forces consisted of an outnumbered band of peas-ant fi ghters and local warriors. Oda Nobukatsu’s army had greatly underesti-mated the effi cacy of the Iga people’s military tactics, the strategic advantage their local knowledge aff orded them, and their capacity for or ga ni za tion and military mobilization.

Oda Nobunaga would not let his territorial conquest remain unfi nished, however, and in 1581 he launched a second invasion of Iga with a force of 40,000 to 60,000 men attacking from all seven sides of the basin. Th is time, the warlord set the province awash in fi re and blood to forestall any notion of further revolt. With a ferociousness reserved for the fi ercest resisters, Iga was destroyed. Oda Nobunaga and his army put an end to the po liti cal and social structure of Iga and the surrounding areas that had lasted for about a century.

In spite of their crushing defeat at the hands of Oda Nobunaga’s modern-ized military, the historical and po liti cal signifi cance of these autonomous communes is not to be minimized. Regional communes of various sizes ex-isted for diff erent periods of time in Kinai, Omi, Settsu, Izumi, Tanba, and other provinces. Th e existence of these regional and provincial federations eff ectively frustrated, for a time, the attempts of any centralized power to con-trol the provinces of central Japan. Th e Iga league of communes, in par tic u lar, lasted much longer than did the one in neighboring Yamashiro province. Th e reason may be the par tic u lar confi guration of the area, a mountain basin iso-

lated by its geography and situated relatively distant from the major routes. Geographic factors may also have contributed to the strong communal senti-ment and military leadership of the peasants by low- ranking warriors. Al-though diff erences in social rank between peasants and warriors were clear and well- established, they were not insurmountable, for the Iga commune also promoted heroic fi ghters, regardless of their background.

Organizations similar to the Iga community structure had emerged in other plains regions of Japan like Echizen or Mikawa, as well as in the Kansai urban areas of Honganji, Kyōto, and Sakai, but in these cases, they were deeply connected to the growing role of religions, especially Amidism (the Ikkō leagues that Carol Tsang describes in her chapter) and the Lotus sutra creed (the Hokke leagues). Everywhere, local warriors and wealthy merchants created in de pen dent power structures that trained commoners, thrilled by their new faith, for combat against outside forces. Th is created a very strong sense of identity that was tied, moreover, to a religious dimension. In sharp contrast, what sprang up in the Iga and Kōga regions was something akin to a local patriotism at the root of their common identity.

Loyalty to the federations of Iga and Kōga did not spring from a special religious feeling, as was the case for the religious leagues in other parts of Japan. To be sure, the mountainous region was an area where followers of shugendō esoteric asceticism were present, and “mountain monks” (yama-bushi) practiced asceticism in the mountains under the infl uence of Shingon esoterism and Buddhism. Although this movement hardly constituted their shared common identity, the people of Iga or Kōga had no doubt learned valu-able secret traditions and practices from the yamabushi. Some of these prac-tices were connected to the art of fi ghting and of war and may have infl uenced local warriors in their practices. Even the bandits (akutō) who ran raids on the land seemed to have copied the signature yellow scarves they wore over their faces from the yamabushi priests. Nonetheless, much of the Iga commune’s solidarity and longevity derived principally from its geo graph i cal isolation. Historically, Iga province remained inaccessible due to extremely poor road conditions and the inhospitable mountains that surrounded it. Th e subse-quent geographic isolation engendered by these conditions most probably gave the people of Iga and Kōga a stronger feeling of identity than was shared by the people of the plains in the absence, even, of a religious referential.

By way of comparison, we may note that in mountainous areas of medi-eval Eu rope, local wealthy individuals formed organizations that federated


120 Pierre Souyri

into armies designed to defend the integrity of the local autonomy. In par tic-u lar, the inhabitants of the Waldstetten area in the Alps created the concept of Landfrieden (territory in peace) to prevent violence and war against their land. Inhabitants of a village, town, or region swore to respect and maintain peace. Th e territories sworn unto these pledges created alliances among them-selves against outside forces. Th e Treaty of Waldstetten, signed on August 1, 1291, and long considered the foundational act of Switzerland, is simply one of these pledges. Th is is the Landsgemeinde, or Association of Free People of the Land. Subsequently the “Swiss Guards” would furnish mercenary troops to the rest of Eu rope during the modern period, very much like the heirs of the jizamurai who ultimately formed shock troops serving external seigniorial powers. One might suggest that this kind of social collectivization born to prevent war was not so diff erent from what existed in the Sengoku- era Iga province.

In general, the local warriors’ sense of collective identity and identifi cation as a specifi c group does not appear to be directly linked to the uniqueness of their devised solution to the security problem, to the specifi city of the or ga ni-za tion they had created, but rather to the suff ering, violence, and repression they were forced to sustain. Th e people of Iga and Kōga fought desperately to maintain their po liti cal autonomy, their way of life, and their social status. Trained by defeated jizamurai, the inhabitants of Iga and Kōga used tech-niques of warfare and fi ghting that before them the bandits and ruffi ans of past centuries had used. Th ey had good knowledge of the ground, they were mobile, and they acted quickly. For a long time, the people of Iga resisted the “occupation armies” of Nobunaga and Hideyoshi, fi ghting continuously and relentlessly, though their cause would ultimately prove futile. Th e new leaders of Japan, seeking to establish a hegemony over the entire country, felt no small ire at the autonomous actions of the individuals who banded together in ikki. Th e order that they were trying to impose relied, on the contrary, on central-ized vertical power structures. Th e violence they infl icted on local popula-tions matched the severity of the perceived insult of the countryside’s re sis-tance. Generally, local autonomous po liti cal constructions refused the new monarchical order, although this was even more so in the case of the Iga, who far from merely rejecting the new order put forth their own po liti cal system to supplant it. Still, it is worth observing that the Iga people acted not with a par-tic u lar ideology in mind, but rather for the purpose of creating structures that would effi ciently ensure order and freedom at a local level.

Th e people of Iga and Kōga essentially transformed the practice of espio-nage and hiding into a military strategy not only because they had learned to fi ght this way, but also because they had no other choice but to adopt guerrilla tactics. Th e emergence of the ninja tradition was in fact the result of a harsh defeat at the hands of Oda Nobunaga’s army, aft er which the Iga and Kōga warriors were forced to fi ght on against the new order in secrecy, using nin-jutsu techniques. Th e social system had been destroyed by the new Japa nese leaders, who moved quickly to suppress re sis tance movements. Such move-ments had begun to spring up at the hands of local jizamurai, who, far from being incorporated into their new system of vassalage, were punished under it. Th e reluctance of the new authority to integrate the local jizamurai into the new system stemmed from the inability of these in de pen dent but newly de-feated jizamurai to accept and integrate themselves into the hierarchical structures imposed on them. Moreover, their exclusion was exacerbated by their re sis tance: Th e more the jizamurai fought, the more they were denied integration into the feudal order and the status of samurai. As Hayashiya Tatsusaburō explains, the people of Iga and Kōga were “the defeated waste of a big territorial community or ga ni za tion,” holdovers from a sort of micro- republic that had been subverted by the armies of the new monarchy. As to the real or ga ni za tion of the ninja, we can only speculate. It is likely that there were detachments of foot soldiers who specialized in espionage, provocation, and even arson, but it is not clear whether they originally were professionals. Most probably, they were peaceful peasants who took up arms and mobilized at the borders when their republic was under attack. Th ey can be said to have been militias of warrior- peasants.

From then on, the heirs of the defeated jizamurai of these areas were edu-cated in the memory of a happy and free history. At the beginning of the sev-enth century, the people of Iga and Kōga reappeared as auxiliary troops of the Tokugawa shōgun. Tokugawa Ieyasu employed members of the Iga ninja to guard his great castle, relocating a force of 200 men from Iga province to the Yotsuya area of Edo. Th e Iga ninjas were also employed as intelligence agents for the shogun, and during the uprising of Christian kaimin (people of the sea) in Shimabara in 1637, the shogun deployed Kōga people to suppress the upris-ings or used them as spies. In collaborating as auxiliary troops to the repres-sion of the great Shimabara’s uprising, they might have hoped to reacquire the warrior status that they had lost in the wars of the 1570s and that the victorious feudal lords presently still denied to them.


122 Pierre Souyri

From accounts in the chronicles of the age, kept and preserved by the old, wealthy families of Kōga, we learn that the ninja secretly penetrated castles, observed the situation inside, and provided reports of it to attackers. Th e most abundant source that has come down to us concerning these practices is a se-ries of lists of objects detailing the purpose of their missions and the expenses they incurred: scissors, daggers, saltpeter, infl ammable arrows, powder, salted and boiled fi sh, and so on. In oral stories told in these chronicles, one can fi nd almost implausible tales of the special powers of the ninja. Th e families who created these stories and thus invented a tradition were descendants of the jizamurai who had ruled the country until the 1570s.

Although the indomitable Iga people proved a diffi cult ally for the shogunate— in 1606, the Iga guards rebelled against their commander, Hattori Masanari, at Edo Castle due to harsh treatment— their skills were unques-tioned. Iga ninja continued to be employed by the Tokugawa shogunate as guards and intelligence operatives until about 1745, when Tokugawa Yoshimune dismissed all ninja from intelligence work and replaced them with loyal sub-jects from his hometown, Kii province.

. . .

So thoroughgoing has been the romanticization of the ninja tradition that few modern Japa nese, let alone people outside of Japan, know of their origins as lo-cal warriors and farmers of a mountainous region in central Japan who fought for autonomy from outside rule during the sixteenth century. Once they were vanquished by Oda Nobunaga’s overwhelming military force, some of the sur-vivors (and not a few imitators, it must be imagined) in Iga and Koga sold their ser vices to the new Tokugawa overlords as mountain guides, spies, guards, and assassins— representing a sort of early modern SWAT team. Iga and Koga “schools of ninja training” sprang up around Japan to capitalize on their reputa-tion for prowess, and to help them sell it, the schools invented a tradition of ninja wisdom with supposedly ancient roots. In some cases they manufactured false documents to “prove” a lineage that stretched back into the mists of time. But from the vantage point of modern scholars of constitutional history, the plucky medieval villagers who fought for their communities against overwhelm-ing odds hold fascination enough, unvarnished by mythology.

It is intriguing to speculate about how Japan might have turned out diff er-ently had pockets of autonomy such as Iga and Koga held out longer against territorial centralization. Might separate states, some with remarkably fl at

social structures of the sort that governed Iga and Koga, have competed with each other and pushed each other toward an earlier adoption of constitutional rule, as some constitutional scholars suggest? How much does it matter that the outlying domains were the last to succumb to centralized rule and re-mained powerful enough to force the Tokugawa to grant de facto federalism to local warlords rather than to smaller, self- governing communities? Geog-raphy will only get us so far. However mountainous Japan’s topography might be, and however shielded from foreign invasion by surrounding oceans, com-petition for farmland within Japan alone generated violence and insecurity enough to force larger economies of territorial scale than the small mountain communities could match. In this respect, Japan was little diff erent than the lands of Eu rope on the other side of the globe.

Notes1. Nonetheless, the people of Iga and Koga managed to invent a tradition of much

longer duration than their own po liti cal order, that of the ninja (literally, “hidden people” or “forbearing ones,” shinobimono), whose origins were once thought to be even older. Some ninja, in fact, would actively propagate the erroneous account of their past, creating false old documents to “prove” their old tradition and lineage.

2. See, for example, Downing 1992.

ReferencesArai, Takashige. Kuroda akutōtachi no chūseishi [Th e Medieval History of the Kuroda

Ruffi ans]. Tokyo: NHK Books, 2005.Downing, Brian. 1992. Th e Military Revolution and Po liti cal Change: Th e Origins of

Democracy and Autocracy in Early Modern Eu rope. Prince ton, NJ: Prince ton Uni-versity Press.

Fujiki, Hisashi. Mura to ryōshu no sengoku sekai [Th e World of Village and Lords During the Sengoku Period]. Tokyo: Tokyo Daigaku Shuppankai, 1997.

Ishimoda Shō. Chūsei teki sekai no keisei [Formation of the Medieval World]. Tokyo: Iwanami shoten, 1946.

Saeki Shin’ichi. Senjo no seishinshi [A Spiritual History of the Battlefi eld: Th e Illusions of Bushidō]. Tokyo: NHK Books, 2004.

Seta Katsuya. “Chūsei makki no zaichi tokusei” [Debt(s) Moratories at the End of Medieval Period]. Shigaku zasshi, 1968.

Souyri, Pierre. Th e World Turned Upside Down: Medieval Japa nese Society. New York: Columbia University Press, 2001.

Yamaguchi Masayuki. Ninja no seikatsu [Th e Life of the Ninja]. Yūzankaku, 1996.


124

Prologue: The Limitations of Sixteenth- Century Firearms

Shortly aft er a chance landing in 1543 on the Island of Tanegashima, a small island south of Kyushu, Portuguese merchants parted with three of their fi rearms (teppō). A Negoroji priest visiting the region took one of them to his temple, located in central Japan. Th e priests of Negoroji and their affi liated metalworkers soon established a forge of gunsmiths and produced enough weapons to form a force of 300 marksmen (teppō shū) in the 1570s. In spite of Negoroji’s profi ciency in using and producing these weapons, their role in disseminating fi rearms has been ignored. Standard narratives of Japan’s sixteenth- century history portray regional “lords” or daimyō as being the most cognizant of the power of these new weapons and most able to use them eff ectively. Oda Nobunaga, the fi rst of the “three unifi ers” of Japan, has been characterized as a military genius whose concentrated use of fi repower al-lowed him to “revolutionize” warfare, crush his most potent rival, the Takeda of Kai province, and consolidate power from 1570 until his assassination in 1582.

Th e priests of Negoroji realized the importance of these new weapons earlier than any daimyō. In contrast to Nobunaga, who hastily assembled a squadron of gunners for the battle of Nagashino, they fi elded a formidable squad of gunners through mastery of production and training. And when Nobunaga dispatched his brother to attack Negoroji, his army suff ered a convincing defeat. Unlike the “epochal” encounter at Nagashino, this battle has been consigned to oblivion, largely because Nobunaga’s successor, Toyotomi Hideyoshi, inciner-ated most of the temple complex in 1585.

Instruments of ChangeOr gan i za tion al Technology and the Consolidation of Regional Power in Japan, 1333– 1600

Thomas Conlan

7

Instruments of Change 125

Th ose most profi cient in manufacturing and using guns were not destined to achieve po liti cal and military success. As the priests of Negoroji discovered to their detriment, reliance on powerful weapons could not provide security from opposing armies. Negoroji was destroyed because its priests believed that their marksmen alone could defend their temple and its constituent lands. Preoccupied with the manufacture of guns, over time Negoroji lacked suffi cient manpower to defend their territory or, for that matter, aid a belea-guered ally, whose castle fell to the Oda before their gunners could arrive. Negoroji’s 300 marksmen were incapable of defeating an or ga nized and deter-mined adversary, which accordingly suggests that the innovation or adoption of new weapons did not determine po liti cal and military success. Instead, the ability to mobilize, sustain, and supply armies proved to be of paramount importance.

Th is chapter will explore the transformational power of technology and show that, like the priests of Negoroji, historians have tended to overemphasize the signifi cance of new weapons. Th e introduction of fi rearms did not unleash a pro cess whereby fragmented authority was centralized (or “unifi ed”) during the fi nal de cades of the sixteenth century. Instead, or gan i za tion al changes in fourteenth- and fi ft eenth- century Japan provided the impetus for the consoli-dation of regional po liti cal and military power.

Technology is most signifi cant, and best understood, as a technique as op-posed to an instrument or weapon. To date, technology has oft en been con-ceived in material rather than or gan i za tion al terms. Such a view is under-standable, for it is easier to point to improvements in par tic u lar objects than to uncover the pro cess through which they came to be eff ectively used. Neverthe-less, the notion of material objects (“technology”) as being capable of infl uenc-ing historical pro cesses arose in relatively recent times. Th e oldest (1615) use of the word in En glish designated a treatise on arts or skills, and by the mid- nineteenth century it came to represent a par tic u lar practical or industrial art. Even as late as the mid- nineteenth century few conceived of technology— the adoption of new “industrial arts” or materials— as a discrete phenomenon, let alone as an agent of historical change.

Th e impact of technology, conceived of as the creation or use of new mate-rials, was not fully perceived until the carnage of the First World War, when military historians fi rst discussed the importance of weapons such as fi rearms and pikes in the “development of modern warfare.” Since the 1920s, this no-tion of technology has been regarded as an important factor, and the trope of

126 Thomas Conlan

a “technological revolution” remains vibrant, although recent scholars have preferred the meta phor of a “military revolution” to describe the changes wrought by improvements in fi repower. Nevertheless, the conception that “technological advances” such as the adoption of new weapons profoundly infl uenced the waging of war is absent from pre- twentieth- century writings, which instead concentrate on issues of military or ga ni za tion and supply. Th is earlier understanding remains germane, for changing patterns of weapons usage refl ected historical pro cesses rather than caused them.

As we shall see, Japan witnessed a shift toward pike usage, although this happened during the Ōnin War (1467– 1477) and not during the thirteenth and fourteenth centuries as some historians have asserted. In addition, guns gradually supplanted arrows during the course of the fi ft eenth and sixteenth centuries. Th e eff ective use of these new materials hinged upon improvements in military and po liti cal or ga ni za tion. In this sense, the invention of weapons proved less important than improvements in techniques for mobilizing, train-ing, and supplying armies in the fi eld— techniques that will hereaft er be re-ferred to as improvements in or gan i za tion al technology. Men who rose to power in the sixteenth century, such as Nobunaga and Hideyoshi, fully mas-tered these arts, while the priests of Negoroji did not.

Sources

When charting how new weapons came to be adopted and used, one can pos-tulate how they infl uenced the pro cesses of social, military, and institutional change. Unusually precise rec ords describing how wounds were infl icted in battle from 1333 through 1600 enable us to trace the dissemination of weap-ons. One can precisely chart how weapons were used from 1333 onward be-cause battle reports (kassen chūmon) and “petitions for reward” (gunchūjō) record how wounds were infl icted. Th ese documents fi rst appear late in the thirteenth century, in the aft ermath of the Mongol invasions, and continued to be produced through the battles of the early seventeenth century. Written shortly aft er every skirmish, each document mentions the damages incurred by warriors so as to ensure compensation for their actions.

Warriors submitted reports of arrival (chakutōjō) and battle reports to administrators (kassen bugyō), a provisionally appointed body of warriors who inspected these documents. Fourteenth- century battle reports are rela-tively rare, for they were generally discarded aft er they had been summarized in petitions for rewards. Once a warrior completed his battle report, he used it


as evidence to prove his military ser vice and thereupon submitted a petition for rewards. Th ese documents recount all wounds, deaths, and damages in-fl icted, as well as the date and location of battle. Petitions for reward provide a more comprehensive narrative of battle than battle reports because they men-tion how a sequence of such skirmishes unfolded through time. Aft er each document was inspected, it was signed and then returned to the petitioner. One example is as follows.

Izumi Sugi Saburō Nyūdō Dōkaku, a gokenin of Satsuma, respectfully requests to receive rewards and a record of his battle exploits (onchūmon) be-cause of his military ser vice.

On the seventh day of the past fi ft h month, when the gate of Kaseda castle in Kimotsuki district, Satsuma province, was stormed, [Dōkaku and his] son were fi rst to attack. [Th ey] surmounted the moat and cut through the barri-cades. As [they] performed military valor with fearless abandon, [Dōkaku’s] son Yasaburō Tamotsu was shot through the left thigh [and later had the] ar-row removed. On the battle of the eighth, the general of the main forces, Shimazu Rokurō, saw an arrow pierce the forearm of [Dōkaku’s] bannerman Rokurōmaru. Furthermore, at that battle, both Ushibari Yamano Hikoshirō nyūdō and Isakuda Hyōbu no suke of Satsuma province witnessed this [as well]. Next at the pitched battle (kakeai kassen) at Hinozaki on the twenty third, [Dōkaku] also performed military ser vice. [Dōkaku] requests that he receive rewards and a record of his battle exploits because of [his] military ser-vice in order to promote the honor of [a practitioner of] the bow and arrow. So humbly stated.

Sixth month 1336Received (copy of the monogram of Shimazu Sadahisa)

Th is document provides a brief narrative of battle. Dōkaku described the wounds of his son and bannerman— all caused by arrows— and named wit-nesses for his deeds. He must have originally submitted a battle report that mentioned these casualties in greater detail, but this document no longer sur-vives. Shimazu Sadahisa, a commander of Satsuma forces, accepted the veracity of Dōkaku’s later petition, for he wrote “received” and signed his monogram on this document. If Sadahisa were suitably impressed, he might write a kanjō, a document praising Dōkaku’s ser vices, and recommend that he be rewarded.

Well over 1,300 petitions and battle reports survive from the fourteenth century, while only ninety- four documents describe the nature of wounds

128 Thomas Conlan

from 1467 to 1600. Such a situation might seem to be paradoxical, for these later centuries are known as the Warring States era (1467– 1600), but this pau-city of military rec ords refl ects changes in social and military or ga ni za tion more than the prevalence of warfare per se. As petitions for reward func-tioned as a means of ensuring that a warrior would be remunerated for his deeds, those who were incorporated into a regional magnate’s network of re-tainers would no longer be able to demand rewards for their military ser vice, and so stopped submitting these documents. Only a dwindling band of war-riors maintained autonomy and continued writing battle reports throughout the fi ft eenth and sixteenth centuries.

Unlike the documents of the fourteenth century, which were uniformly distributed throughout the land, nearly all battle reports created between 1467 and 1600 come from western Japan. Older styles of documents, namely docu-ments of arrival (chakutōjō) and petitions for reward (gunchūjō) survived longer in western Honshu and Kyushu, and in the domains of the Ōuchi and the Ōtomo in par tic u lar, than in the rest of Japan. Similarly, more western warriors managed to preserve their autonomy throughout the po liti cal tur-moil of the fi ft eenth and sixteenth centuries than did their brethren in central and eastern Japan.

Th e surviving military reports of the fi ft eenth and sixteenth centuries re-veal less about the nature of warfare than fourteenth- century rec ords. In con-trast to earlier petitions, fi ft eenth- century rec ords rarely mention the names of witnesses or where a par tic u lar battle was fought. As armies became more co-hesive and the same troops fought together over time, the need to record the names of witnesses and the locations of battles diminished. Documents came to simply mention who was wounded at a par tic u lar battle. Th us, the very pro-cesses that led to the improved ability of armies to mobilize troops and secure supplies caused the historical rec ords pertaining to war to decline.

Indeed, greater military cohesion meant that some types of documents disappeared entirely as fourteenth- century patterns of mobilization became anachronistic. Although the armies of the fourteenth century were mobilized on an ad hoc basis, with “invitations” calling warriors to fi ght being randomly distributed, such invitations became unnecessary as the fi ft eenth century progressed, and armies came to be composed of most, if not all, warriors from a par tic u lar region. Once this pro cess of mobilization became standardized, warriors no longer submitted documents of arrival when reporting to camp, which makes it nearly impossible to reconstruct the movement of fi ft eenth-


century troops. Th is is unfortunate, for as we shall see, the period of the great-est innovation in military or ga ni za tion is one in which virtually no or gan i za-tion al rec ords survive.

Th e following document, submitted by Kikkawa Mototsune of Aki prov-ince in the aft ermath of a Kyoto battle, typifi es fi ft eenth- century reports:

Th e following were killed or wounded during the battle at [the crossroads of] Ichijō and Takakura on the thirteenth day of the ninth month of the fi rst year of Ōnin (1467).

Yuasa Yajirō killedsame name Asaeda Magotarōsame name Asaeda Matajirōsame name Asaeda MagogorōYamagata Mago Saemon no Jō: a pike (yari) woundWada Saburō Saemon no Jō: a pike wound [and] an arrow woundOno Yaroku same as the aboveMiyoshi Saemon TarōSaeki Shoroku a pike woundTahara Tosho no Suke a pike woundKawayoshi Shinsaemon no Jō same as the above

Kikkawa Jirōsaburō Mototsune (monogram)Received (monogram) [Hosokawa Katsumoto]

Unlike the earlier petitions, this report tells us little about the movement of Mototsune and his men, save that four were killed and seven wounded 9. 13, 1467. Mototsune submitted this report to his commander, Hosokawa Katsu-moto, who responded with the following document of praise (kanjō) within ten days of the encounter:

During the battle of the past thirteenth, you exchanged sword blows. I re-ceived a report that your retainer (hikan) Yuasa Yajirō was killed, and, in ad-dition, that many others were wounded. I am extremely pleased and moved [by your battle ser vice]. It is my sincerest desire that you shall continue [such outstanding] military ser vice. Respectfully.

Ninth month, twenty- third day [1467] Katsumoto (monogram)

Katsumoto’s quick dispatch of a document of praise made the later creation of a separate petition for rewards irrelevant. Unlike the fourteenth century,

130 Thomas Conlan

where commanders issued documents of praise aft er receiving several petitions for reward, fi ft eenth- century generals issued documents of praise shortly aft er reading these more informal battle reports.

Mototsune was unusual in that he preserved his battle report. Most of his compatriots tended to discard battle reports once their battle ser vice was rec-ognized with documents of praise (kanjō). Unfortunately, these laconic kanjō reveal little about the nature of warfare. Katsumoto only mentions one out of eleven Kikkawa casualties by name, and then inaccurately character-izes the encounter as one where troops “came to blows with swords” (tachi uchi) when most were actually wounded by pikes! Although the kanjō was the most common type of military document in the Warring States era, it proved inadequate for reconstructing the nature of fi ft eenth- and sixteenth- century warfare.

A Statistical Survey of War

Because fi ft eenth- and sixteenth- century documents record little more than the names of various warriors and how they were wounded, their deepest insights can best be derived through statistical analysis. Th ese rosters of wounded soldiers allow for a survey of how men were wounded in battle and, by extension, how war was fought. As the average number of people and wounds mentioned in fi ft eenth- and sixteenth- century battle reports increases over time, a smaller sample of military reports nevertheless provides adequate data. Fourteenth- century documents mention 8,634 individuals, but they mostly record only the names of participating warriors, describing how wounds were incurred in a mere 721 cases. By contrast, the ninety- four documents dating from 1467 to 1600 describe 1,208 wounds, 487 more than all of the ear-lier documents. Fourteenth- century data on wounds by weapon are pre-sented in Table 7.1.

Of course, statistics have their limitations, for they provide a false sense of precision not matched by the sources. Still, they represent the only means of holistically comprehending the battle data. Statistical analysis suggests impor-tant trends that cannot be comprehended through anecdotal evidence alone.

Surviving data reveals that warfare consisted primarily of skirmishing throughout the years 1333– 1600. During the fourteenth century projectiles caused 73 percent of all wounds, while this percentage increased slightly, to 75 percent, during the years 1467– 1600. Handheld weapons infl icted the remain-ing 27 percent of fourteenth- century and 25 percent of fi ft eenth- and sixteenth-


century wounds. Th e bow remained the favored weapon for skirmishing, but its dominance eroded over time. Arrows caused 99 percent of all projectile wounds during the 1300s and continued to infl ict 58 percent of all such wounds through 1600, even though fi rearms were introduced to Japan in 1466. Guns did not displace bows until 1600, when they infl icted 80 percent of all skir-mishing casualties.

Warriors tended to fi ght in close quarters aft er the outbreak of large- scale war. During the “Genkō and Kenmu Disturbance” of 1333– 1338, and the Ōnin War of 1467– 1477, pikes and swords caused 35 percent of all wounds, while projectiles were responsible for the remaining 65 percent. Although these fi g-ures may not seem particularly remarkable, they reveal that handheld weap-ons infl icted 30 percent more wounds during the years 1333– 1338 than was typical for the rest of the fourteenth century (35 percent compared to 27 per-cent). Not surprisingly, this same period exhibited a disproportionate share of fatalities. Th e Ōnin War of 1467– 1477 also witnessed a 40 percent increase in pike and sword wounds (35 percent compared to an average of 25 percent), while handheld weapons infl icted 28 percent more wounds in 1600 than was typical for the years 1467– 1600 (32 percent to 25 percent). (For a list of casual-ties recorded during this period, see Table 7.2.)

Even though anywhere from two- thirds to three- fourths of all casualties stemmed from skirmishing, the skirmishes did not decisively impact the out-come of wars. Th e opening years of war were fought with the greatest intensity, and accounted for most casualties. Suzuki Masaya has argued that skirmishes remained the mainstay of battles and that close- quartered clashes were of

Table 7.1 Fourteenth- Century Wounds by Weapon

PeriodOne

(1333– 1338)Two

(1339– 1349)Th ree

(1350– 1355)Four

(1356– 1394) Total

Arrow () () () () ()

Sword () () () () ()

Pike () () () () ()

Rock () () () () ()

Total () () () ()

Source: Th e data are reproduced from Conlan, State of War, p. 58.Note: Numbers in parentheses indicate percentage in each time block, save for those in the total column,

which designate the relative percentage of these rec ords in comparison with other time blocks. Figures below 1 percent were rounded up.

132 Thomas Conlan

Document date Collection Arrow Gun Rock Pike Sword Killed Total

.. Kikkawa

.. Kikkawa

.. Kikkawa

..– Kikkawa

.. Kikkawa

.. Mōri

.. Mōri

.. Kobayakawa

.. Miura

.. Kobayakawa

.. Kikkawa

.. Kikkawa

.. Migita Mōri

.. Tagaya

.. Miura

.. Ura

.. Masuda

.. Mita

.. Tagaya

.. Miura

.. Miura

.. Kutsunoya

.. Migita Mōri

.. Migita Mōri

.. Migita Mōri

.. Masuda

.. Migita Mōri

.. Migita Mōri

.. Miura

.. Reisen

.. Shidō

.. Migita Mōri

.. Miura

.. Ōtomo

Table 7.2 Casualty Lists, 1467– 1600



.. Ōtomo

.. Yano

.. Kobayakawa

.. Kobayakawa

.. Migita Mōri

.. Migita Mōri

.. Dewa

.. Katsumata

.. Iwami Kikkawa

.. Amano

.. Hiraga

.. Iwami Kikkawa

.. Miyoshi

.. Mōri

.. Asonuma

.. Dewa

.. Kikkawa

.. Yuasa

.. Ōtomo

.. Ōtomo

.. Ura

.. Irie

.. Ōtomo

.. Kusakari

.. Sugi

.. Kikkawa

.. Irie

.. Ōtomo

.. Ōtomo

.. Ōtomo

.. Ōtomo

.. Sugi

.. Tōshima

.. Ōtomo

Table 7.2 (continued)

(continued)

134 Thomas Conlan


.. Ōtomo

.. Ōtomo

.. Ōtomo

.. Ōtomo

.. Ōtomo

.. Tōshima

.. Ōtomo

.. Ōtomo

.. Ōtomo

.. Kodama

.. Ōtomo

.. Ōtomo

.. Ōtomo

.. Ōtomo

.. Ōtomo

.. Ōtomo *

.. Ōtomo

.. Ōtomo

.. Ōtomo

.. Ōtomo

.. Ōtomo

.. Ōtomo

.. Ōtomo

.. Ōtomo

.. Ōtomo

.. Kikkawa

Total


Sources: For the Kikkawa documents, see Dai Nihon Komonjo Iewake, Series 9, Kikkawa ke monjo, vol. 1, docs. 320– 324, pp. 272– 277, docs. 328, 329, pp. 280– 281, doc. 509, pp. 453– 456, doc. 511, pp. 457– 462, doc. 513, pp. 463– 469, and doc. 728, pp. 674– 701. For the Iwami Kikkawa, see ibid., vol. 3 (1932), the Iwami Kikkawa ke monjo appendix, doc. 55, pp. 59– 60 and doc. 57, p. 61. For the Mōri, see Dai Nihon Komonjo Iewake, Series 8, Mōri ke monjo (Tōkyō teikoku daigaku shiryō hensanjo, 1920), vol. 1, docs. 123– 125, pp. 116– 118, and doc. 293, pp. 305– 319. Dai Nihon Komonjo Iewake, Series 11, Kobayakawa ke monjo (Tōkyō teikoku daigaku shiryō hensanjo, 1918), vol. 2, doc. 153, pp. 60– 62, doc. 197, pp. 84– 86, doc. 429, pp. 271– 273 and doc. 432, p. 275– 277 reproduce the Kobayakawa rec ords. Th e Ura documents also appear in ibid., vol. 2, Ura ke monjo appendix; see doc. 4, p. 3 and doc. 11, pp. 7– 9. For the Miura rec ords, see Dai Nihon Komonjo Iewake, Series 14, Kumagai ke monjo– Miura ke monj– Hiraga ke monjo (Tōkyō teikoku daigaku shiryō hensanjo, 1937), doc. 67, pp. 370– 371, doc. 87, pp. 384– 389 and docs. 95– 98, pp. 393– 398. Th e 1549 Hiraga record also appears in ibid., Hiraga ke monjo doc. 169, pp. 628– 633.



marginal importance. But his generalization cannot account for why war-riors fought so fi ercely during the years 1333– 1338, 1467– 1477, and 1600. Rather, one can surmise that a distinction existed between skirmishing, aimed at har-rying enemy forces, and pitched battles, where the ability to physically control strategic areas assumed crucial importance. With the outbreak of war, supply lines had to be secured, which necessitated the occupation of contested grounds by infantry forces.

During the fi ft eenth century, a marked and surprising shift in the nature of hand- to- hand combat took place. Contrary to common assumptions, swords, the so- called souls of the samurai, were rarely used aft er 1467. While swords generated 92 percent of all wounds stemming from close-quarters combat in the fourteenth century, they were responsible for only 20 percent of these wounds from 1467 onward. Pikes, which had infl icted 7 percent of all such wounds in the fourteenth century, caused 80 percent of them from 1467 until 1600. Th is preference intensifi ed over time, for pikes caused 74 percent of all nonprojectile wounds from 1467 to 1477 and 98 percent of all such wounds by 1600.

Even though pikes assumed paramount importance in close combat aft er 1467, swords maintained a cultural and linguistic signifi cance long aft er they ceased to be militarily important. Commanders continued to deploy the

Yamaguchi kenshi shiryōhen, Chūsei, no. 2 (Yamaguchi, 2001), pp. 330, 386– 387, 732, 943, and 945 contains the Kutsunoya, Reisen, Katsumata, and Kodama rec ords. For the Masuda rec ords, see Dai Nihon Komonjo Iewake, Series 22, Masuda ke monjo (Tōkyō daigaku shiryō hensanjo, 2000), vol. 1, doc. 208 (8.24.1511), pp. 159– 162 and doc. 278 (3.18.1527), pp. 243– 245. See also Hagi, vol. 1, maki 16, pp. 456– 457 (the Shidō), maki 34, p. 823 (Kusakari), and maki 35, pp. 829– 830 (Asonuma), vol. 2, maki 43, pp. 138, 148 (Dewa), maki 79, pp. 774– 775 (Sugi), vol. 3, maki 109, p. 351 (Miyoshi), and vol. 4, maki 164, p. 361 (Mita) and maki 161, p. 309 (Tōshima). A Yano reference to a man who was twice wounded by pikes can be found in Niigata kenshi shiryōhen, vol. 5, Chūsei, no. 3, docs. 3658– 3660, p. 663 and doc. 3483, p. 599. For the Amano on 9.5.1547, see Shizuoka kenshi shiryōhen, vol. 7, Chūsei, no. 3 (Shizuoka ken, comp. Shizuoka, 1994), doc. 1860, pp. 644– 645. For the other Amano documents, see the Migita Mōri ke monjo, Yamaguchi kenshi shiryōhen, Chūsei, no. 3, pp. 438, 440– 443, 449. See also ibid., pp. 786– 787 for the Tagaya documents, and p. 999 for the Yuasa ke monjo. For the Irie, see Irie monjo (Ueda Jun’ichi, ed. Zoku gunsho ruijū kansekai, 1986), doc. 107, pp. 171– 173 and ibid., doc. 108, pp. 174– 175. Ōtomo documents are found in Zōho teisei Hennen Ōtomo shiryō, vol. 16, docs. 93– 94, pp. 42– 45; vol. 19, docs. 371– 372, pp. 210– 213; vol. 20, doc. 113, pp. 55– 56; vol. 21, docs. 395, 415, pp. 190, 198– 199; vol. 22, docs. 225, 229, 369– 374, 438– 439, 451, 470, pp. 101– 103, 162– 172, 205– 207, 210– 211, 220– 221; vol. 23, docs. 65, 170– 171, pp. 24, 73– 74; vol. 24, doc. 312, pp. 165– 166. For the 1581 battles, see vol. 25, docs. 490, 492, pp. 215– 217. Next, see vol. 26, docs. 37, 72, 75, 91, 93, 137, 335, 338, 357, 486, 489, 583, pp. 14– 18, 31– 32, 44– 45, 65– 66, 153– 156, 161– 162, 218– 220, and 260– 261 and vol. 27, doc. 83, p. 36.

Note: Only wounded and killed warriors were counted. Due to the same person occasionally suff ering multiple wounds by diff erent weapons, the totals do not invariably add up with the number of wounds by weapon. Finally, multiple wounds by the same weapon were not separately counted. Th us one man shot four times by a gun would only be counted as one person shot, but one hit by both an arrow and a bullet would be counted in each category separately.

136 Thomas Conlan

phrase “coming to blows with swords (tachi uchi)” to describe combat between groups of pike- wielding soldiers. Although the military signifi cance of swords proved negligible, their cultural value remained. Indeed, a certain mystique arose concerning them with the seventeenth- century establishment of an en-during peace in Japan.

Japan’s Fifteenth- Century Transformations

Th e adoption of new weapons did not cause tactical change. Th e pike initially appeared in 1333 but was hardly used throughout the fourteenth century. Only 15 men were wounded with pikes, in contrast to 523 who can be documented as being wounded by arrows and 178 by swords. Pikes could only be used ef-fectively, and widely adopted, when soldiers could be mobilized in cohesive units that could withstand charging cavalry. Surprisingly, this transformation appears to have arisen during seventy years of relative peace, from 1392 until 1467. Chronicles mention how, in 1454, warriors were killed during a confron-tation with pikes (yari awase senshi) in central Japan. And as we shall see, improvements in military or ga ni za tion ensured that pikes would become the favored weapon for foot soldiers from 1467 onward.

Changes in Military Or ga ni za tion, 1392– 1467

During the seventy- fi ve years separating the fi nal battles of Japan’s fourteenth- century confl ict and the onset of the Ōnin War, Japan witnessed a shift from loosely or ga nized armies to semipermanent regional units. As only a few spo-radic skirmishes were fought between the years of 1392 and 1467, one cannot readily discern transformations in tactics. Surviving documents suggest, however, that war was waged as it had been in the fourteenth century, with the exception that military units were becoming increasingly regionally based and cohesive.

Th e armies of Japan’s fourteenth and early fi ft eenth century represented little more than clusters of bow- wielding skirmishers scattered among small bands of horse men. Th ese foot soldiers were vulnerable in open spaces, for horse men could easily charge and shoot them with arrows, and instead pre-ferred fi ghting in inaccessible terrain, or in towns and villages, where they could easily hide and fi re arrows at enemy cavalry. When encountering such a force of skirmishers, horse men would burn dwellings and obstacles in order to create enough room for their horses to roam. As their dominance remained unchallenged in open spaces, the signifi cance of cavalry proved greater than


aggregate numbers would imply. A few of these highly trained horse men could decisively defeat a larger number of scattered skirmishers. Armies were cob-bled together from these warrior houses, and the ability to entice these men into alliances underpinned military power in fourteenth- century Japan.

Early fi ft eenth- century petitions, drawn from eastern Japan, are stylisti-cally indistinguishable from fourteenth- century petitions, except for infer-ences to warriors from a single region fi ghting together. For example, during the years 1417– 1418 warriors from Musashi province fought as squads, or ikki, based on geographic origins rather than kinship ties. As the fi ft eenth cen-tury progressed, units tended to be identifi ed by their provincial origins. As evidence of this we see warriors from the provinces of Musashi, Kōzuke, and Shinano fi ghting as cohesive forces in 1423, and by 1440, generals command-ing troops drawn from a single province— again Musashi and Kōzuke— were perceived as being normative.

Surviving rec ords suggest that no tactical transformations arose during the years 1392– 1467. Battles continued to be waged as they had been in the fourteenth century: men shot their opponents with arrows, bludgeoned them with swords, or hacked their way into fortifi cations. Tellingly, horses were also used conspicuously, and even as late as the tenth month of 1455, some were slashed with swords. Swords caused all recorded examples of fi ft eenth- century horse wounds, a trend consistent with the latter de cades of the four-teenth century. Th at horses continued to be slashed reveals that horse men continued to charge through infantry formations, just as they had in the four-teenth century. Nevertheless, forces drawn from the central provinces of Yamato and Kii used pikes as early as 1454, and they would meet with sudden and unexpected success a dozen years later on the Ōnin battlefi elds.

Improvements in Provincial Po liti cal Or ga ni za tion, 1392– 1467

Improvements in the ability to collect revenue allowed regional magnates to forge and sustain provincial armies. Once semipermanent forces were established, pikes could be adopted and used eff ectively. Th e 1351 promulga-tion of the hanzei edict allowed provincial constables (shugo) to use half of their province’s “public” tax revenues for military supplies. As time passed, shugo managed to assess increasingly burdensome levies from their prov-inces. Th ese taxes sometimes assumed extraordinary proportions, with the Ōyama estate in Tanba province supplying 755 laborers for their shugo Ya-mana Ujikiyo in 1390.

138 Thomas Conlan

Th e establishment of the hanzei enabled shugo to outstrip their non-shugo rivals in wealth and military power, for it provided a mechanism for them to mobilize and sustain provincial military forces. Unlike many of the battles of the early fourteenth century, where warriors fought heedless of their re-gional origins, the armies of the late fourteenth and early fi ft eenth centuries were better or ga nized and less dispersed. Already as early as 1355 armies tended to be based more on warriors drawn from a specifi c area, and fi ght-ing under their shugo’s command, than had been typical during the years 1333– 1338.

Shugo taxation remained onerous even aft er the wars of the fourteenth century drew to a close, as levies for horses, messengers, and workmen con-tinued unabated. Th rough their deputies, shugo used their judicial powers and taxing authority to forge the bonds of regional lordship. Th is ability to collect funds also enabled shugo to maintain their provincial military power.

Th e powers of the shugo post and its limitations are both evident in the case of Tanba, where Hosokawa Yorimoto attempted to dominate this prov-ince aft er being appointed shugo in 1392, just as the wars of the Northern and Southern Courts were ending. Although a newcomer to Tanba province, with no historical ties, Yorimoto managed to wield infl uence through his deputies, who legitimated their actions under the broad powers of the hanzei. Th ese men attempted to extract wealth from their province and its nominally im-mune estates by conscripting laborers and levying taxes for shrine repairs and enthronement ceremonies, but they met with limited success.

Shugo had diffi culties establishing powerful provincial lordships despite the advantages off ered by the hanzei. Tanba warriors such as the Nakazawa resisted Hosokawa authority and continued occupying lands that had been nominally confi scated. Th e Nakazawa’s struggle continued unabated from the waning years of the fourteenth century through the mid- fi ft eenth century. Disputes erupted sporadically, and even as late as 1445 the Nakazawa still oc-cupied these contested lands in spite of repeated orders to desist.

Shugo powers coalesced when local warriors became incorporated into their regional network. Th e contrast between the Kasai (fi rst Jōken and then his close relative, Motosuke), powerful retainers (miuchi) from Shikoku who were appointed Tanba’s deputy shugo in 1414, and Naitō Motosada, a Tanba warrior and former deputy shugo of Settsu province, who was appointed aft er the Kasai in 1431, proves illuminating. Th e Kasai’s tenure lasted little more than fi ft een years amid simmering re sis tance from other Tanba warriors and


estate residents. On July 24, 1431, Kasai Motosuke was criticized as being “un-principled” (in eff ec tive?) and divested of his deputy shugo post in favor of Naitō Motosada. Motosada collected more taxes in his fi rst year than Kasai Motosuke had during the fi nal year of his appointment. Motosada even ap-pears to have secured funds from the recalcitrant Nakazawa, and by 1435 Motosada had doubled some levies. To no avail, some Ōyama estate residents absconded in protest of the Naitō’s tax increases, but only succeeded in having their lands confi scated by Hosokawa retainers who monopolized the position of estate manager.

Once shugo delegated their authority to local warriors, their network be-came better able to procure revenue. Th is ability to govern locally proved inte-grally related to the shugo’s ability to forge a regional army supplied through taxation. In other words, although the hanzei caused considerable hardship among Tanba residents, it also allowed for most of the area’s warriors to be melded into a semipermanent provincial army. For example, provisions were levied, and porters— sometimes hundreds of people— were conscripted so that the Naitō could easily travel to and from Kyoto.

Th e Hosokawa were able to control and mobilize most of Tanba’s wealth, and nearly all its landed income, by leveraging the appointment of their re-tainers to both “public” offi ces, such as the post of Tanba deputy shugo, and as “private” managers of major estates. By 1457, a portion of Tanba hanzei were directly given to the Naitō and their retainers, while by 1460, a local warrior gained control of the Ōyama estate, with a promise to provide merely 20 kanmon of cash to its proprietor. Th is trend continued as the fi ft eenth century progressed. In 1482, the Nakazawa came to “manage” the Ōyama es-tate, aft er which the absentee proprietor no longer received income from these lands.

Th e example of Tanba reveals that shugo and their deputies became increas-ingly skilled at administratively dominating their provinces. With their consoli-dation of power over all major Tanba offi ces by 1460, the Hosokawa were ideally suited to mobilize and sustain a provincial army drawn from Tanba. Once the Ōnin War erupted in 1467, this army would play a key role in the Hosokawa’s strategy of opening supply lines to the west of the capital. Although not all of Japan’s sixty- six provinces would be so tightly controlled by their shugo, the trends evident in Tanba prove illustrative. Th e shugo’s monopolization of provincial ad-ministrative offi ces facilitated the development of regional warrior networks, which in turn formed the nucleus of fi ft eenth- century armies.

140 Thomas Conlan

Improvements in po liti cal or ga ni za tion and the ability to extract revenue manifested themselves in the ability to forge and sustain regional forces. Th ey did not, however, immediately translate into a shift of tactics. Several de cades were required for shugo, or their deputies, to realize that formations of foot soldiers, rather than squadrons of cavalry, proved capable of dominating the battlefi eld. Only with the onset of an indeterminate civil war during the years 1467– 1477 could commanders make pike- wielding soldiers the mainstay of their armies and transform the nature of battle in Japan.

The Onin War: Tactical and Or gan i za tion al Transformations, 1467– 1477

For the fi rst time in centuries, cohesive units of infantry, armed with pikes and capable of occupying contested ground indefi nitely, were mobilized in 1467. However, these important or gan i za tion al innovations have been over-looked by scholars, who have tended to focus on the rationales for the out-break of the Ōnin War and its social and po liti cal consequences. Most mono-graphs ignore both how the Ōnin War was fought and how its armies were supplied. Recently, Paul Varley has emphasized the importance of ashigaru, a fl eet- footed force of light infantry, but posits the most signifi cant changes as occurring during the Warring States era, where a “continuing technological revolution, refl ected in the greater use of guns” transformed Japan.

Or gan i za tion al technology, rather than the adoption of the gun, proved critical in instigating change. Improvements in the ability to provision armies enabled armies to occupy regions indefi nitely. Once troops trained together and mastered formations, they became profi cient in using pikes. Th is allowed them to defeat horse men on the open battlefi eld, a task that scattered bands of sword- wielding men could never accomplish.

Th e rapid and comprehensive mobilization of provincial forces under the aegis of the deputy shugo propelled the Ōnin War, and their ability to supply troops in the capital helped prolong the confl ict. In 1467, the Naitō entered the capital leading most of Tanba’s warriors, including the Nakazawa, where they fought on behalf of Hosokawa Katsumoto. It is a mea sure of the Naitō’s suc-cess in mobilization that they denuded their province of warriors. When en-emy Yamana forces invaded Tanba during the sixth month of 1467, they met with no re sis tance. Indeed, only one prominent Tanba family seems to have actively opposed the Naitō. Other rec ords also mention the regional cohe-sion of the Naitō’s forces, naming them “Tanba armies (tanshū no onzei),” as military units were typically described in the mid- fi ft eenth century.


Provincial armies were maintained through a steady stream of supplies, which tended to be collected and dispersed by their deputies since most shugo were absentee fi gures. No tax rec ords survive during the years 1467– 1477 for the Ōyama estate, but this gap in the sources refl ects a desperation regarding the procurement of surpluses that did not allow for complaints to be issued or depredations to be recorded. Documents from the Niimi estate reveal how the outbreak of the Ōnin War in 1467 caused “public levies” to increase dramati-cally as funds and produce were transported, sold, or disrupted by enemy forces during the course of the war.

Shugo powers were thus both the cause of the outbreak of war and the mechanism for its prolongation. Th e post of shugo became the fulcrum of pro-vincial power and the basis for regional or ga ni za tion, and yet the concentra-tion of power inherent in this offi ce served to destabilize Japan. Competition for the post exacerbated tensions that already existed both within and among shugo houses. Some provinces, such as Tanba, shift ed from the control of one house to another. At the same time, the indivisible nature of the shugo post served to intensify inheritance disputes, as only one candidate could inherit each offi ce. Th us, the inheritance— or lack thereof— of shugo offi ces focused rivalries both within and among houses. By magnifying warrior dissent and dissatisfaction, competition for the post of shugo ultimately contributed to the intense and protracted nature of the Ōnin confl ict.

A violent dispute between two Hatakeyama candidates vying for a portfo-lio of shugo positions provided the spark for war during the fi ft h month of 1467. Rival shugo houses, such as the Hosokawa and the Yamana, came to sup-port diff erent Hatakeyama factions, thereby causing the fi ghting to spread throughout the capital. As all shugo lived in Kyoto, their residences served as nascent encampments for provincial troops and supplies. Each dwelling func-tioned as a base that had to be consolidated, or destroyed, in order for one co ali tion of shugo to dominate the capital.

Hosokawa Katsumoto, the shugo of several provinces, operated from a position of superiority, for he had been preparing for battle since early 1467. Accordingly, his co ali tion, which became known as the eastern army, man-aged to control the northeastern areas of the capital, where the imperial and shogunal palaces were located. Th e forces of his opponents, led by Yamana Sōzen, occupied northwestern Kyoto and hence were called the western army. On May 26, 1467, Hosokawa Katsumoto reduced the strategic dwelling of Is-shiki Yoshitada to ashes. Yoshitada’s abode, located adjacent to the shogunal

142 Thomas Conlan

palace, had been the eastern outpost of the Yamana’s army and their only means of contacting the Ashikaga shogun. Th e Yamana responded by de-molishing the dwellings of Hosokawa partisans residing in western Kyoto, such as the home of Hosokawa Katsuhisa. Th ereupon two large armies co-alesced, dominating the northeastern and northwestern quadrants of the capital, respectively.

Arson and fi erce fi ghting characterized the onset of hostilities as each army attempted to create space for their horse men to roam. During the initial two days of battle, the residences of three eastern and three western shugo, located in indefensible positions, were burned, along with seven temples and countless other dwellings. Th e Hosokawa controlled all the strategic palaces in the north-east but could not occupy southeastern Kyoto, nor could they make signifi cant inroads to the northwest in spite of repeated sharp and bloody exchanges.

A tactical stalemate arose, and both sides started digging trenches and constructing barricades. Katsumoto proved unable to crush the outnum-bered western forces. In order to press his advantage, he ordered the deputy shugo of neighboring provinces to reinforce the capital. Th e Naitō led almost all of Tanba’s men to Kyoto, which further bolstered Hosokawa strength. But by overly concentrating the eastern army’s forces, Katsumoto blundered, for he allowed Yamana Sōzen, the commander of the opposing “western army,” to smash through depopulated Tanba on June 8, 1467, and to strengthen his position in the capital. Naitō Sadamasa, the deputy shugo of Tanba, perished with dozens of family members and retainers while defending the border be-tween Tanba and Kyoto. Th ereupon the “eastern” and “western” armies for-tifi ed their positions and, for lack of a better option, peppered their opponents with projectiles. In skirmishes on June 25, 1467, three more shugo residences, two nobles’ abodes, and numerous other structures were burned. Neverthe-less, as western reinforcements continued arriving via Tanba roads, the initia-tive lay with them. Ōuchi Masahiro led a strong contingent of warriors to Kyoto during the eighth month of 1467, which allowed western commanders to contemplate seizing the off ensive for the fi rst time.

Surviving Kikkawa battle reports reveal that pikes were widely used dur-ing the autumn of 1467, where the brunt of fi ghting revolved around several large watchtowers. On September 13, 1467, six members of the Kikkawa fam-ily were stabbed by pikes, while on October 2– 3, 1467, six more were so wounded in street fi ghting. Of course, not all of the Kikkawa wounds stemmed from pikes— eight more were caused by arrows, one by a sword, and


fi ve by rocks— but these twelve pike wounds, infl icted over a period of three weeks, suggests the rise of new tactics, as tightly or ga nized forces of pikemen came to dominate the battlefi eld.

Th is tactical transformation becomes evident when recounting the western army’s off ensive of the tenth month. On October 3, 1467, the western armies launched a rolling attack on the eastern army’s positions. Th eir off ensive, cen-tered on Shōkokuji, the lynchpin of eastern defenses located near the shogunal and imperial palaces, would be suddenly and dramatically checked in spite of their numerical superiority.

Th e battle at Shōkokuji reveals that pikes had become the preferred weapon for hand- to- hand combat. According to the “Chronicle of Ōnin,” Hatakeyama Masanaga, a commander of the eastern army and a veteran of the 1454 Yam-ato and Kii campaigns, routed a large force of western Rokkaku cavalry on this day with a compact squad of pikemen. Aft er closing in on the Rokkaku horse men at the burned- out grounds of Shōkokuji temple, an arena that fa-vored cavalry by allowing for mobility, Masanaga’s pikemen surged into the enemy. Th e Rokkaku were broken, and sixty- seven were killed before their cavalrymen fl ed in defeat.

Th e Hatakeyama played a crucial role in perfecting how to use pikes in formation. Th e “Chronicle of Ōnin” suggests that contemporaries were shocked that Masanaga’s 2,000 men on foot chose to attack cavalry forces three times larger (6,000– 7,000) at Shōkokuji, and they were even more surprised by his stunning victory over the Rokkaku. Th e battle- hardened Masanaga was confi -dent of success, however, for he purportedly boasted that “I will defeat even an enemy of a million” as his troops advanced in tight formation behind shields. Th e “Chronicle of Ōnin” attributes Masanaga’s success to the fact that western pikemen proved unable to establish formations in the confusion of the Rokkaku fl ight. Other shugo or their deputies lacked the training to respond immediately to the Hatakeyama’s tactics. Although the logistical prowess of shugo made standing armies possible, they could not master the use of pikes in tight forma-tion without training.

During the opening months of the Ōnin War, broad areas of the capital were burned in order to provide cavalry with space to roam, but the innova-tive use of pikes checked the mobility of horse riders even in these open areas. Although horse men remained eff ective in supporting infantry formations, or harassing the enemy, their role became peripheral once they could no longer dislodge a force of pike- wielding infantry. Cavalry, formerly the mainstay of

144 Thomas Conlan

battle, became relegated to reconnaissance and skirmishing, while trained units of pikemen came to constitute the backbone of military forces from the mid- fi ft eenth century onward.

War continued to be waged through skirmishing, but pitched battles could now only be won when men armed with pikes physically occupied contested grounds. Although pikes infl icted relatively few casualties, units of pikemen were indispensable. Th ese formations were so potent that even the threat of their use proved adequate cause for armies to withdraw. Hatakeyama Masanaga well understood this principle, for he retreated from Shōkokuji rather than face a force of enemy pikemen led by the western commander Hatakeyama Yoshi-nari, another veteran of the 1454 Yamato confl ict, who entered the fray in the aft ermath of the Rokkaku defeat. Th ereupon each army hunkered down in its respective fortifi cations, leaving the blackened grounds of Shōkukuji a desolate no- man’s-land.

Ōnin generals soon realized the futility of frontal attacks on entrenched units of infantry. Early in 1468, the eastern army began digging elaborate trenches and constructing earthworks of unpre ce dented height and depth at strategic locales. Both Kyoto armies burrowed trenches 3 meters deep and 6 meters wide, which caused some sections of the capital to resemble the West-ern Front. Locked in a stalemate, huddled in trenches with few opportunities for off ensive action, the warriors of Ōnin alleviated the intense tedium by writ-ing Japa nese poems (waka) or poetic phrases on small silk crimson fl ags, which they attached to themselves.

Each army, unable to seize the off ensive, viewed the other’s troop move-ments from a cluster of watchtowers, which became the focus of pitched bat-tles. Th e western army eventually constructed a structure 21 meters high. Not to be outdone, the eastern army answered with a tower that loomed 30 meters above the burned out grounds of Shōkokuji. Th ereupon each army used fl aming arrows and rocks to support their off ensives and defend against attacks.

In order to compensate for their tactical stalemate, commanders relied on daring raids by small groups of foot soldiers to infi ltrate and disrupt enemy lines. Some of these patrols managed to burn enemy fortifi cations, while other mobile squadrons of lightly armored soldiers (ashigaru) demolished lodgings harboring enemy troops in night raids. Th ese fl eet- footed ashigaru are best characterized as irregular units of skirmishers. Contemporary sources variously describe them as carry ing shields, strong bows, and pikes


and occasionally wearing helmets or, to the contrary, “holding no pikes wear-ing no armor, and carry ing only swords.” Ashigaru were guerrilla fi ghters who excelled in sudden attacks and skirmishes but could not occupy con-tested grounds. Th ey did not constitute phalanxes of pikemen, and indeed, they were punished in off ensive battles against entrenched forces. During one such encounter on 9. 7, 1468, Koma Tarō, the leader of the eastern army’s ashi-garu, was killed and his forces were decimated.

Not well equipped to engage in pitched battles, commanders directed their energies toward cutting enemy supply lines. Cavalry were used to raid villages so as to constrict the fl ow of goods to enemy camps, but they no longer deci-sively infl uenced the outcome of what had become a war of attrition. Except for a few skirmishes in the capital, most military actions seem to have focused on the hinterlands in 1468. Both the eastern and western armies struggled to gain control of Yamashina, a village located to the southeast of Kyoto that provided the main conduit for supplying the increasingly beleaguered eastern army. Th e western army briefl y managed to occupy Yamashina, which re-duced the eastern army to illness- plagued malnutrition, but they could not hold the position, thereby allowing the stalemate to continue. Th e warriors of Tanba once again aided the eastern army by launching a probing attack from the west in an ultimately unsuccessful attempt to open another supply line to Katsumoto’s forces.

Th e need to dislodge entrenched enemies encouraged innovations in weaponry. One fi nds reference to a craft sman from Izumi province construct-ing a catapult— known as a hō— that could launch 3- kilogram projectiles for over 300 yards! Unsei Daigoku, the author of the Hekizan Nichiroku, be-lieved that this catapult was a device of great antiquity— modeled perhaps on third- century Chinese devices— and insisted that it did not represent an in-novation. Th is emphasis on the “oldness” of such catapults is all the more ironic because at the same time a relatively new weapon, the gun, appeared in Japan.

The Introduction of Firearms to Japan

Unsei Daigoku remarked, with remarkably little surprise or wonder, how on November 6, 1468, a hihō hisō, or literally a “fl ying projectile fi re spear,” was discharged from a besieged tower. Unsei Daigoku’s nonplussed reaction stemmed in part from the antiquity of such devices and (in contrast to con-temporary catapults) from their unimposing nature. Explosive shells (teppō)

146 Thomas Conlan

had been known in Asia since the thirteenth century, when the Mongol in-vaders of Japan used primitive bombs, composed of ceramic projectiles fi lled with gunpowder, in tandem with rounded rocks to terrorize the Japa nese de-fenders. Th ese bombs should not, however, be confused with later fi rearms. Chinese sources refer to a primitive gun, known as a “fi re dragon spear” (hiryūsō) being created in 1355, and the oldest surviving specimens are thought to date from the 1370s.

Surviving sources indicate that the Ryūkyū Kingdom, which comprises modern Okinawa, fi rst introduced fi rearms to Japan some eighty years before the Portuguese visited Tanegashima. A fi ft eenth- century diary rec ords how an offi cial from the Ryūkyū Kingdom surprised many bystanders in Kyoto with the report of his fi rearm (teppō) on 7. 28, 1466. Archaeological evidence from Akenajō and Katsurenjō, two Okinawan castles (gusuku), reveals that guns were used prior to the mid- fi ft eenth century. Defenders supplemented the weakest point in Akenajō’s defenses with a portal especially designed for use by snipers, placed low in its stone walls. Furthermore, stone, earthen-ware, and on occasion, iron bullets have been uncovered within the battle-ments of Katsurenjō, which was destroyed in 1458 and never rebuilt.

Primitive fi rearms composed of three metal tubes welded together, known as fi re arrows (hiya), were widely disseminated in Eu rope and Asia through-out the late fourteenth and early fi ft eenth centuries. Remarkably, some of these fi ft eenth- century hiya were fi red as late as the early twentieth century, and they were reportedly capable of blasting projectiles for 200 yards.

Most primitive fi rearms (hiya) discharged rounded stones, and intrigu-ingly, data drawn from sixteenth- century military petitions reveals a sudden upswing in rock wounds, particularly in western Japan, which suggests the dissemination of the primitive fi rearms. Even though only a handful of cases where soldiers were wounded by rocks can be documented in the fourteenth century, rocks injured eighty- two men during the years 1524– 1552, with over half (forty- four) occurring during the seventh month of 1552.

Th e fi rst documentary evidence describing wounds caused by fi rearms ap-pears in Amano Okisada’s kassen chūmon of 11. 27, 1527, where one man is listed as being “shot wounded” in the right foot. Documents submitted by Okisada six months earlier, on 5. 13, 1527, refer to “arrow wounds” but refrain from the elocution of “shot wounded.” Th ese suggestive sources can be cor-roborated with letters dating from 1569, which use the same verb (iru) to de-scribe wounds infl icted by guns (teppō).


Th e earliest documented example of a man being explicitly wounded by “fi rearms” occurs on 1. 27, 1563, when Hara Rokurō, a retainer of Sugi Mat-suchiyo, was shot near his left armpit by a “hand fi re arrow,” or tebiya, by sup-porters of the Ōtomo in Northern Kyushu. Weapons known as as teppō can be fi rst verifi ed on 11. 13, 1563, when the Amako of Izumo province mauled the Kikkawa, wounding thirty- three by teppō, six by arrows, fi ve by rocks, and one by a sword.

As a result of Amako prowess, the number of gun casualties supersedes those of bows by a fi gure of 88 to 64 during the 1560s. Nevertheless, the data are not comprehensive enough to hypothesize about patterns of gun dissemi-nation. Although twice as many bullet wounds (17 to 8) were recorded as ar-rows in the 1570s, both were infl icted at roughly analogous rates (19 to 16) in the 1580s. Indeed, from 1467 until 1600, arrows caused 58 percent of all projec-tile wounds, while bullets were responsible for 28 percent and rocks the re-maining 13 percent.

Guns required nearly half a century to supplant bows because they were only incrementally more eff ective in range and penetration. Gun wounds were oft en infl icted in close proximity to enemy forces, as were arrows. Sev-eral examples exist of warriors wounded with both bullets and arrows, or being shot and stabbed with a pike in the same encounter. Instances where the same warrior was shot repeatedly and yet survived also attest to the limita-tions of sixteenth- century aim and fi repower.

Negoroji appears to have fi rst recognized the power of a squad of gunners. Th eir marksmen gained fame for their prowess during the 1570s and mid- 1580s. In fact, they proved to be well ahead of their time. Not until 1600 do surviving documents reveal a pronounced preference for guns (teppō), which caused 80 percent of all projectile wounds on the plains of central Japan.

Th e adoption of guns, be they the primitive hiya or, for that matter, Portu-guese teppō, seemed to cause no discernible change in tactics. Guns did not stop pikemen from fi ghting in close quarters even aft er they largely supplanted bows. Although the battles of 1600 substantiate the marked dissemination of guns, pikes increased in use as well and were responsible for 32 percent of all wounds. Such a fi gure is consistent with fourteenth- and fi ft eenth- century mili-tary patterns.

Th e 1600 data might lead one to assume incorrectly that the introduction of the more powerful teppō from the Portuguese led to a “military revolution” that paved the way for the po liti cal “unifi cation” of sixteenth- century Japan.

148 Thomas Conlan

Teppō were able to achieve parity with arrows by 1580, but they could only be-come an eff ective weapon when gunners were or ga nized into cohesive groups, a pro cess that only began during the 1570s. Th e infl uence of guns on changes in tactics is seen in the fi rst rosters recording military units and the par tic u lar weapon used by each, which fi rst appeared in 1575. Firearms infl uenced the outcome of battle when marksmen were or ga nized much as pikemen had been during the course of the Ōnin War.

The Onin Legacy, 1477– 1600

Even though the Ōnin War witnessed the use of bows, catapults, and fi rearms, these weapons proved inadequate in shaping the course of battle. In the end, victory could only be achieved when forces of pikemen were in a position to thwart enemy forces and blockade their supplies. Improvements in military or ga ni za tion proved to be decisive and caused battles to shift from mobile encounters of horse men to entrenched formations of infantry. Commanders focused their energies on disrupting enemy supply lines and, when this proved ineff ectual, attempted to destroy their opponents’ regional support by foment-ing the rebellion of deputy shugo. Or ga ni za tion became crucially important in maintaining and sustaining po liti cal and military power, while the ability to eff ectively use weapons, be they “old” pikes or “new” guns, merely refl ected this more fundamental pro cess.

In the confusion of the Ōnin War, when all authority became contested, depu-ties were encouraged to rebel against their shugo. Delegated powers proved ripe for usurpation, as deputy shugo were best placed to or ga nize, supply, and com-mand regional military forces aft er 1467. Th ese men were intimately aware of their lands and personally led men to battle. Asakura Takakage, the deputy of Echizen province, astutely abandoned his shugo, Shiba Yoshikane, and de-fected from the western army in 1471. Takakage ensured the continued survival of the eastern army and more importantly, from his point of view, managed to gain eff ective control over Echizen a mere three years later. Similar be-trayals became endemic over the ensuing century. Th e Oda of Owari prov-ince, another deputy shugo family of the Shiba, would also overthrow their shugo and come to control most of central Japan by 1582 under the steward-ship of Oda Nobunaga.

Th e powers inherent in the offi ce of the shugo did not invariably accrue to the men appointed to this post. Indeed, most shugo were unable to capitalize on the strength of their armies, as until 1467 they were forced to reside in the


capital and entrust de facto command to their deputies. Th at an individual might be simultaneously appointed as the shugo of several provinces miti-gated against direct po liti cal supervision of any single province. Aft er 1471, deputy shugo increasingly displaced their shugo in violent fashion, thereby giv-ing rise to the moniker gekokujō, or “lower overcoming higher,” which aptly characterizes the turbulent Warring States era.

Gekokujō did not depend on the adoption of pikes, or for that matter, guns. Instead, those most profi cient in or ga niz ing their troops were best able to amass po liti cal authority. Although oft en overlooked in the pro cess of sixteenth- century “unifi cation,” the July 21, 1547, triumph of the deputy shugo Miyoshi Nagayoshi over his Hosokawa rivals epitomizes how the ability to mobilize a formidable force, train them, and equip them with pikes proved essential. Nagayoshi’s success was directly related to the eff ective use of units of pikemen, for the Ashikaga Kiseiki recounts how his army of 900 pikemen clashed with a similar force of Hosokawa troops, infl icting hundreds of casualties before en-suring Hosokawa Harumoto’s defeat. Aft er Miyoshi Nagayoshi defeated and destroyed Harumoto, his “lord,” he expelled the Ashikaga shogun Yoshiteru from the capital in 1549. Nagayoshi scorned accepted titles as sources of legiti-macy and preferred instead to base his authority on military prowess.

Miyoshi Nagayoshi’s victory, and his subsequent actions, reveal a new atti-tude regarding authority that emphasized coercive force over all other systems of legitimacy. Military power, based on the ability to mobilize large formations of pike- wielding soldiers, had become established as the basis for all po liti cal endeavors by 1547, the year of Nagayoshi’s victory, and a mere four years aft er Portuguese fi rearms were initially introduced to the Japa nese archipelago. In short, most of the innovations thought to have arisen from the introduction of Portuguese fi rearms were fully established before these weapons ever arrived.

Conclusion

Ambiguities inherent in the word “technology” allow two interpretations of its role in Japa nese history to be true, namely, that technology was instrumen-tal in determining Japan’s fi ft eenth- and sixteenth- century transformations and was a refl ection of these pro cesses. When understood as a technique or method of or ga niz ing armies, technology— here characterized as or gan i za-tion al technology— proved decisive. When conceived as the innovation or adoption of new weapons, its most common defi nition, technology func-tioned as a barometer of historical and, for that matter, or gan i za tion al change.

150 Thomas Conlan

Japa nese warfare provides an ideal arena for assessing the importance of new weapons and technologies because during the years 1333– 1600 pikes re-placed swords as the preferred weapon for hand- to- hand combat, while guns also gradually displaced bows. Of these transformations, the hitherto over-looked shift from swords to pikes proved dramatic and signifi cant, while the dissemination of guns was gradual and less consequential.

Th ese changing preferences were not based on the innovation of new weap-ons. In fact, the lag between a weapon’s initial appearance and its widespread adoption undermines materialist explanations of tactical change. References to pikes fi rst appear in documents dating from 1333, but they were not widely used until 1467. Similarly, guns fi rst arrived in Japan by 1466, but they only marginally infl uenced tactics for the ensuing century. Most of the transforma-tions attributed to guns, or for that matter, pikes, stemmed from fourteenth- and fi ft eenth- century improvements in the ability to supply and maintain armies.

Although guns have attracted considerable attention, they did not alter the nature of battle. Instead, most of the changes that have been attributed to them, such as their checking of cavalry, had already occurred at the time of the Ōnin War. Th e infl uence of guns was limited to giving men from the most prosperous regions— the capital and western Japan— a slight military advan-tage over their more remote rivals as the sixteenth century progressed. Guns represented a minor improvement in range and stopping power over arrows, but no force of gunners, no matter how well trained, could break advancing pikemen or horse riders, as the priests of Negoroji discovered to their detri-ment in 1585.

Th e widespread adoption of pikes signifi ed a more important development than the gradual dissemination of guns. Troops wielded pikes only aft er re-gional authorities (shugo and their deputies) possessed the ability to sustain large armies in the fi eld. In other words, the practical ability to transform signifi cant numbers of foot soldiers into a cohesive fi ghting block proved to be vitally important. Only when troops could be mobilized, fed, and trained over time could they fi ght in formation, which proved indispensable for both pikes and, ultimately, fi rearms to be used eff ectively.

Improvement in the ability to extract surplus revenue, supplies, and man-power enabled shugo to strengthen their po liti cal control over Japan’s prov-inces during the course of the fourteenth through sixteenth centuries. Instead of relying on autonomous warrior houses to forge fi ssile armies, as had been


typical through 1350, shugo created a network of warriors who could be read-ily mobilized, thereby enabling them to overshadow provincial rivals, and project power into neighboring provinces.

Th is improved or ga ni za tion is refl ected by the increasingly well- defi ned structure of Japa nese provincial armies. As the fi ft eenth century progressed, or gan i za tion al techniques became increasingly sophisticated as shugo relied on their economic and po liti cal resources to break down the autonomy of Japan’s warriors and forge them into cohesive organizations. Once armies achieved a modicum of coherence, troops could train and master sophisticated forma-tions, resembling the phalanx, where a massed body of men wielded pikes in unison and thus could withstand cavalry charges on open ground. Th ereupon, the need to conscript as many men as possible and forge them into a unifi ed army superseded the need to entice fi ckle warriors into one’s camp. Th ose most skilled in governance ultimately proved most skilled at or ga niz ing, training, and maintaining provincial armies.

Deputy shugo managed to train and or ga nize troops regardless of the presence of fi rearms. During the latter half of the sixteenth century, a scion of a deputy shugo family named Oda Nobunaga managed to assert control (“uni-fi cation”) over the capital and central provinces, but he could have accom-plished this task even if he had possessed no guns. His military powers, and those of his compatriots, were predicated on the improved or gan i za tion al abilities and logistical prowess of fi ft eenth- and sixteenth- century armies, which in turn hinged on increased po liti cal control over economic surpluses. Improvements in or gan i za tion al technology determined the nature and pace of mechanical, tactical, and ultimately po liti cal change.

Notes1. Hamada Toshiyasu, “Teppō denrai no keii ni tsuite ni san kōsatsu,” Reimeikan

chōsa kenkyū hōkoku 14 (Kagoshima ken rekishi shiryō sentaa Reimeikan, 2001), pp. 85– 100 and Hora Tomio, Teppō- denrai to sono eikyō (3rd printing; Kyoto, Shi-bunkaku, 2001), pp. 1– 35, 154– 160.

2. Asao Naohiro, “Th e Sixteenth Century Unifi cation,” in John Hall, ed., Th e Cambridge History of Japan, vol. 4 (Cambridge: Cambridge University Press, 1991), pp. 40– 95, particularly p. 54, claims that guns “revolutionized” warfare, as does Paul Var-ley in his “Warfare in Japan, 1467– 1600,” in Jeremy Black, ed., War in the Early Mod-ern World, 1450– 1815 (London: UCL Press, 1999), pp. 67– 73, and Geoff rey Parker in Th e Military Revolution (New York: Cambridge University Press, 1988), pp. 140– 142.

152 Thomas Conlan

3. Wakayama kenshi chūsei shiryō, vol. 2 (Wakayama ken, comp. Wakayama, 1973), Honganji Saginomori betsuin monjo, doc. 14, 10. 11 [1577] Rairen shojō, pp. 427– 428, and doc. 7, pp. 423– 424.

4. I would like to thank Michael Como for this insight.5. See “technology” in Th e Complete Oxford En glish Dictionary (New Edition).6. See Hans Delbrück, History of the Art of War, particularly vol. 3, Medieval War-

fare, pp. 649– 656, and vol. 4, Th e Dawn of Modern Warfare, pp. 23– 57 (Walter Renfroe, Jr., trans., Lincoln: University of Nebraska Press, 1990).

7. One of the fi rst formulations of a “technological revolution” appears in Marc Bloch’s 1931 classic, French Rural History: An Essay on Its Basic Characteristics (Berkeley: University of California Press, 1966). For the concept of a “military revolu-tion” stemming from improvements in fi repower, see Parker, Th e Military Revolu-tion, pp. 24ff .

8. A concern for military or ga ni za tion pervades the writings of Clausewitz, who ignores the role of new weapons on battle. See Michael Howard and Peter Paret, trans., On War (Prince ton, NJ: Prince ton University Press, 1976).

9. For the Mongol invasions and the creation of petitions for reward, see Th omas Conlan, In Little Need of Divine Intervention (Ithaca, NY: Cornell University East Asia Series, 2001), pp. 207– 210, 217– 222.

10. Nanbokuchō ibun, Kyūshū hen (7 vols.; Seno Sei’ichirō, comp. Tōkyōdō shuppan, 1980– 1992), vol. 1, doc. 662, 6.1336 (Kenmu 3) Izumi Dōkaku gunchūjō utsushi, p. 206.

11. Gokenin, or “honorable house men,” were warriors who submitted petitions for reward demanding compensation for their military ser vice. Conlan, State of War: Th e Violent Order of Fourteenth Century Japan (Ann Arbor, MI: Center for Japa nese Stud-ies, 2003), pp. 107– 140.

12. Conlan alludes to this pro cess in State of War, pp. 104– 106, 162– 164, 222– 229.13. Dai Nihon Komonjo Iewake, Series 9, Kikkawa ke monjo (Tokyo: Tōkyō tei-

koku daigaku shiryō hensanjo, 1928), vol. 1, doc. 320, pp. 272– 274.14. Ibid., doc. 327, p. 279.15. For one example of a reference to a casualty list that no longer survives, see

Niigata kenshi shiryōhen, vol. 4, Chūsei, no. 2 (Niigata, 1983), doc. 2039, 4.21(1460) Ashikaga Yoshimasu kanjō utsushi, p. 523.

16. See Table 7.2. Th e fi gure of 1,208 wounds was reached through the addition of all sword, rock, pike, gun, and arrow wounds, whereas the fourteenth- century data is reproduced in Table 7.1. Th e fourteenth- century data also appears in Conlan, State of War, pp. 53– 69.

17. Table 7.1 reveals that the highest percentage of hand- to- hand combat (33 percent) occurred during 1333– 1338. Th ese years concurrently experienced the greatest number of deaths. Approximately 60 percent of all those killed during the wars of the fourteenth century (704 out of 1,173) perished at this time. See Conlan, State of War, pp. 53– 69.


18. See Table 7.2, where projectiles (arrows and rocks) caused 35 of 54 casualties during the Ōnin War, while bows and guns infl icted 165 of 241 wounds in 1600. For the 1600 fi gures, look at the penultimate row of the table, dating from 1600.8.26 (a document from the Kikkawa monjo).

19. Suzuki Masaya, Nihonjin to teppō (Chikuma gakugei bunko, 2000), pp. 208– 238. See also pp. 167– 207.

20. See Table 7.2. From 1467 to 1475, pikes caused 14 of 19 nonprojectile wounds, while in 1600, they infl icted 75 of 76 such wounds.

21. See G. Cameron Hurst, Armed Martial Arts of Japan (New Haven, CT: Yale University Press, 1998).

22. Conlan, State of War, pp. 58– 69.23. See the 5.1454 reference in the Kōya Shunshū Hennen Shūroku, in Dainihon

Bukkyō Zensho, vol. 131 (Bussho kankōkai, 1912), maki 11, p. 240.24. Taiheiki (Jingū chōkōkanbon) (Hasegawa Tadashi, Kami Hiroshi, Ōmori

Kitayoshi, Nagasaka Shigeyuki, eds. Ōsaka: Izumi shoin, 1994), maki 35, “Shodaimyō kasanete Tennōji ni mukau koto,” p. 1040.

25. Th e Northern White Flag ikki and the Southern ikki, both of Musashi, fought separately and autonomously. See Saitama kenshi shiryōhen, vol. 5, Chūsei, no. 1 (Saitama, 1982), doc. 689, 1.1417 Beppu Owari nyūdō dai Uchimura Katsuhisa chakutōjō, pp. 466– 467 and doc. 707, 7, 24, 1419 Ashikaga Mochiuji gunzei saisokujō utsushi, p. 475.

26. See ibid., doc. 739, 8.1423 Beppu Michitada gunchūjō, doc. 740, October 10 Hatakeyama Michiie shojō, pp. 490– 491, and doc. 798, 10. 15, 1440, Senba Hitachi no suke shojō, pp. 522– 525.

27. Saitama kenshi shiryōhen, vol. 5, Chūsei, no. 1, doc. 690, 1.1417 Toyoshima Noriyasu gunchūjō, p. 467. See also doc. 739, 8.1423 Beppu Michitada gunchūjō, pp. 490– 491 and doc. 848, 10. 14 [1455], Uesugi Tatsuwaka kanjō, p. 553.

28. An equal number of horses were wounded with swords and arrows from 1331 through 1338, but a majority (75 percent) were wounded by swords thereaft er. Conlan, State of War, pp. 67– 69.

29. Hyōgo kenshi shiryōhen chūsei (9 vols.; Kobe, 1983– 1997), vol. 6 (1991), doc. 320, 1390 Ōyama no shō shugo’eki ninpu mokuroku, pp. 237– 238 (hereaft er HKSC). For more on the hanzei, see Conlan, State of War, pp. 95– 98, 225– 229.

30. Cohesive, regionally based units were forged aft er 1350, some twenty years af-ter the onset of hostilities. See Conlan, State of War, pp. 72– 76.

31. HKSC, vol. 6, p. 299, doc. 413 and doc. 410, p. 298, doc. 338, 2. 10, 1398, Ogas-awara Masamoto uchiwatashijō an, p. 252, doc. 366, the 8.7 Tanba no kuni shugodai kakikudashi an, pp. 275– 276 and doc. 389, 9. 29, 1409, Tanba no kuni shugo dai Hoso-kawa Tōtōmi no kami kakikudashi an, p. 288.

32. Th e Nakazawa disobeyed repeated injunctions by the Ashikaga shogun, the Tanba shugo, and his deputies while defending their lands. See HKSC, vol. 6, doc. 334,

154 Thomas Conlan

11. 14, 1397 Muromachi shōgunke migyōsho, pp. 249– 250, and doc. 335, 12. 27, 1397 Tanba no kuni shugodai Ichinomiya Eishō jungyōjō, p. 250. For their 1411 defense of their lands, see doc. 393, 12. 17, 1410 Tanba no kuni shugo dai kakikudashi, p. 292 and doc. 400, 10.1411 Nakazawa Masamoto shojō an, pp. 294– 295.

33. For the appointment of the Kasai, see ibid., doc. 412, p. 299; and doc. 422, p. 303. Miuchi were retainers of a “lord,” who lacked autonomy in action. For Naitō Motosada’s earlier career as the deputy shugo of Settsu province, and a good survey of the situation in Tanba, see Ogawa Makoto, Yamana Sōzen to Hosokawa Katsumoto (Shinjinbutsu Ōraisha, 1994), particularly p. 91.

34. Manzei junkō nikki (2 vols.; Hanawa Hokinoichi, comp. Zoku gunsho ruijū hoi, no. 1, Zoku gunsho ruijū kanseikai, 1928), vol. 2, 7. 24, 1431, p. 270.

35. Kasai Motosuke seems to have had diffi culty in collecting funds to rebuild Shinomura shrine, whereas Naitō Motosada succeeded. Compare HKSC, vol. 6, doc. 519 (of 1430), Ōyama no shō daikan Tsuchiya Sōgen shojō an, p. 370, Shinomura tansentō irime chūmon with doc. 531, Ōyama no shō Shinomura Hachimangu zōei tansen irime chūmon, pp. 375– 376.

36. Ibid., doc. 531, Ōyama no shō Shinomura Hachimangu zōei tansen irime chūmon, p. 376 for securing Nakazawa funds, and doc. 553, Ōyama no shō hyakushōra moshijō an, pp. 395– 396 for the doubling of tax revenue.

37. Ibid., doc. 560, Ōyama no shō Ichii no tani hyakushōra moshijō, pp. 398– 399.38. Ibid., doc. 615, Ōyama shōmu kakukudashi an, pp. 449– 450 for military provi-

sions being shipped to the capital and doc. 619, Ōyama no shō Ichii no tani hyakushōra mōshijō narabi ni renshokishōmon, pp. 451– 452, and doc. 638, Ōyama no shō ichii no tani shugo’eki fusen chūmon, p. 475 for shugo and their deputies conscripting hun-dreds of porters.

39. Ibid., doc. 750, Ōyama no shō tansen iriashi haitō chūmon, p. 581 and doc. 768, September 6, 1460, Shindō Toshisada Ōyama no shō ryōkegata Ōgatabun daikan-shiki, pp. 591– 592.

40. Ibid., doc. 803, Daikanshiki buninjō an, and doc. 804, Nakazawa Motoki dai-kanshiki ukebumi, p. 613.

41. For the pioneering study on the origins of the Ōnin War, see Paul Varley, Th e Ōnin War (New York: Columbia University Press, 1967). For a more recent analysis, see Elizabeth Berry, Th e Culture of Civil War in Kyoto (Berkeley: University of Califor-nia Press, 1994), particularly pp. 14– 34.

42. Katsumata Shizuo’s “Jugo- roku seiki no Nihon,” Iwanami kōza Nihon tsūshi, vol. 10, Chūsei, no. 4 (Iwanami shoten, 1994) and Nagahara Keiji’s Gekokujō no jidai, Nihon no rekishi, vol. 10 (Chūō kōronsha, 1974) ignore Ōnin tactics, save for a brief description of lightly armored foot soldiers (ashigaru).

43. Varley, “Warfare in Japan, 1467– 1600,” pp. 53– 86.


44. Th e Kuge, Nakazawa, Ogino, Honjo, Adachi, Yaku, and Ashida can be docu-mented as follows: the Naitō in battle two years later, in 1469. See the “Chronicle of Ōnin” (“Ōninki”) found in Gunsho ruijū, vol. 20, Kassenbu, no. 1 (Hanawa Hokinoi-chi, comp. Zoku gunsho ruijū kanseikai, 1931, pp. 355– 419), p. 403.

45. See the actions recorded in 6. 29, 1467, Kyōkaku shiyōshō found in Mukō shishi shiryōhen, p. 275 and Dai Nihon shiryō Series 8, vol. 1 (hereaft er 8.1) (Tōkyō teikoku daigaku shiryō hensanjo, 1913), pp. 283– 287. Miyada Bingo no kami was the only Tanba warrior who fought against the Naitō. See Hyōgo kenshi, vol. 3, p. 99.

46. For an informative petition mentioning military forces from Kii, Settsu, Harima, and Tanba provinces, see Mukō shishi shiryōhen, March 1474, Noda Yasutada gunchūjō, pp. 272– 274. Th e document is located most con ve niently in Dainihon shiryō 8.1, pp. 33– 37.

47. Okayama kenshi iewake shiryō, vol. 20 (Okayama, 1985), docs. 426– 433, pp. 787– 791.

48. Dainihon shiryō 8.1, pp. 238– 269 for the battles of May 26, 1467. For the attack on the Isshiki, see pp. 238– 242.

49. Ibid., p. 266. Much of the ensuing Ōnin narrative has been drawn from the varied chronicles, diaries, and documents appearing in Dainihon shiryō 8.1 for the fi ft h and sixth months of 1467, pp. 201– 325.

50. Dainihon shiryō 8.1, pp. 238– 270.51. Ibid., p. 275, for the actions of 5. 30, 1467.52. “Ōninki,” in Dainihon shiryō 8.1, p. 285.53. Dainihon shiryō 8.1, pp. 320– 323.54. Ōuchi Masahiro smashed into the capital and built an encampment at Tōji on

8. 24, 1467. See Dainihon shiryō 8.1, pp. 343– 347, and 357– 359.55. A Kikkawa Mototsune jihitsu kassen tachiuchi chūmon mentions fi ghting in

the vicinity of watchtowers 10. 4, 1467. See Dai Nihon Komonjo Iewake, Series 9, vol. 1, doc. 324, p. 277.

56. Kikkawa ke monjo, vol. 1, docs. 320– 324, pp. 272– 277.57. “Ōninki” in Dainihon shiryō 8.1, pp. 454– 459. See also Dainihon shiryō 8.1, pp.

441– 464.58. Dainihon shiryō 8.1, pp. 454– 459. See also Ogawa, Yamana Sōzen to Hosokawa

Katsumoto, pp. 184– 185.59. Dainihon shiryō 8.1, pp. 454– 459.60. “Ōninki” in Dainihon shiryō 8.1, pp. 454– 459.61. See the priest Unsei Daigoku’s diary, Hekisan Nichiroku (Zōho Zoku Shiryō

Taisei, vol. 20. Rinsen shoten, 1982), 1. 29, 1468, p. 181, and May 3, 1468, p. 202 for de-scriptions of deep trenches and high walls.

62. Ogawa Makoto estimated in his Yamana Sōzen to Hosokawa Katsumoto, p. 188.63. Hekisan Nichiroku, 1. 29, 1468, p. 181.

156 Thomas Conlan

64. Ibid., 11. 6, 1468, p. 231. For fi ghting around these towers, see Kikkawa ke monjo, vol. 1, doc. 324, Kikkawa Mototsune jihitsu kassen tachiuchi chūmon, p. 277.

65. Hekisan Nichiroku, 4. 14, 1468, p. 199. Ōuchi Masahiro erected another tower, named Daiseirō, to the southwest of this edifi ce 4. 25, 1468. See ibid., p. 200.

66. Ibid., 5. 27, 1468, p. 206.67. Hekizan Nichiroku, 1. 5, 1468, p. 177, 4. 26, 1448, p. 200, and 6. 21, 1468, p. 210.68. Ibid., 5. 17, 1468, p. 191.69. Ibid., 6. 15, 1468, p. 209, 6. 21, 1468, p. 210, and 8. 2, 1468, p. 215.70. For the former, see ibid., 11. 3, 1468, p. 234; for the latter, 6. 15, 1468, p. 209.

Th ese passages are diffi cult to decipher because of ambiguous terms. Pikes were anachronistically called hoko, designating a spear- like weapon that had not been used for centuries. I would like to thank Karl Friday for bringing this to my attention.

71. Varley, “Warfare in Japan, 1467– 1600,” pp. 59– 60.72. Hekizan Nichiroku, 9. 7, 1468, p. 221.73. Th at villages were burned to disrupt supply routes is evident from the Hekizan

Nichiroku, 11. 5, 1468, pp. 234– 235.74. Hekizan Nichiroku, 7. 21– 25, 1468, p. 213 and 8. 9, 1468, p. 215. For the western

army’s fi nal defeat on intercalary 10. 17, 1468, see p. 231.75. For more on this contingent of 1,500 Tanba warriors, see Hekizan Nichiroku, 9.

21, 1468, p. 223, and Noda Tadayasu’s petition for reward (Dainihon shiryō 8.1, pp. 34– 35), 8. 24– 9. 7, 1468.

76. Hekisan Nichiroku, 1. 29, 1468, p. 181.77. Ibid.78. Ibid., 11. 6, 1468, p. 235.79. See the 10. 20, 2001 eve ning edition of the Asahi shinbun, p. 12, for the discov-

ery of three of these ceramic projectiles in the wreckage of the Mongol fl eet. See also Conlan, In Little Need of Divine Intervention, pp. 12, 73.

80. Th e 1972 cata logue, Nihonshi ni mieru teppō ten contains pictures of these early weapons on p. 2. For a reference to fi rearms (hiryūsō) being created in China by 1355, see the Bubi Hiryūkyō, in ibid., p. 2. For the best coverage of the early guns, see Hora Tomio, Teppō- denrai to sono eikyō, pp. 1– 12; 36– 56.

81. Inryōken Nichiroku (5 vols.; Dainihon bukkyō zensho, no. 133– 137. Bussho kankōkai hensan, 1912– 1913), vol. 2 (134), July 28, 1466, p. 670.

82. Toma Shi’ichi, “Hiya ni tsuite,” Nantō Kōko, no. 14 (December 1994), pp. 123– 152. I visited these structures on December 20, 2001. Portals in the walls of two castles, Akenajō and Nakagusukujō, are located within 18 inches of the ground, and are thus too low to allow for arrows to be fi red. For more on Akenajō, Katsurenjō, and Nakagusukujō, see ibid., pp. 66– 73, and Toma, “Hiya ni tsuite,” pp. 129– 139.

83. In the case of Katsurenjō, eleven stone bullets were uncovered, while one was made of fi red earth and another of iron. See Toma, “Hiya ni tsuite,” pp. 135– 136, and


p. 152. Limestone, coral, and sandstone were used to make stone bullets, while metal bullets were cast from either copper or iron. See ibid., pp. 129, 134– 135.

84. See Toma Shi’ichi, “Hiya ni tsuite,” particularly pp. 123– 129, 140– 141. Chigira Yoshinori of the Okinawa Prefectural Museum generously helped in securing a pho-tograph of one of the museum’s hiya.

85. See Toma, “Hiya ni tsuite,” pp. 127– 128. Archaeological excavations reveal that numerous projectiles peppered the walls of Okinawan castles, the largest being the size of a soft ball. See Toma, ibid., pp. 129– 136.

86. See Table 7.2. Toma Shiichi’s discoveries suggest that most early bullets were made from rocks rather than lead. See his “Hiya ni tsuite.”

87. Yamaguchi kenshi shiryōhen, Chūsei, no. 3 (Yamaguchi, 2004), p. 443 for the 5. 13 and 11. 27 documents.

88. Yamaguchi kenshi shiryōhen, Chūsei, no. 3, p. 971, for a 2. 7 and a 5. 30, 1569 letter by Tagayama Michisada. Th e former document explicitly mentions teppō. See also p. 972 and p. 1077 for 5. 7 and 5. 20, 1569, letters by Fukuhara Sadatoshi describing a total of twenty- three enemies being “shot.”

89. Hagi han batsu’etsuroku (5 vols.; Yamaguchi ken monjokan, 1967) (hereaft er Hagi), vol. 2, maki 79, p. 774.

90. See Table 7.2.91. Of the 907 projectile wounds recorded in Table 7.2, 528 were by arrows, 257 by

guns, and 122 by rocks. Percentages do not add up to 100 because of rounding.92. Th e hapless Otsumaru Sakyō no suke was shot with bullets and arrows. See

Zōho teisei Hennen Ōtomo shiryō (33 vols.; Takita Manabu, comp. Ōita, 1962– 1971), vol. 22, doc. 29, p. 103. Among the Kikkawa casualties of 1600, one fi nds references to some men who were shot with bullets and arrows, or shot and stabbed at the same encounter. Each wound has been counted separately in Table 7.2. Multiple gunshots were also fairly common.

93. Guns caused 132 out of 165 projectile wounds in 1600, with arrows causing the remaining 33 wounds. Two distinct words are used for guns in the 1600 Kikkawa document. Both are homonyms for teppō.

94. For a list of armies or ga nized according to units of pikes, guns, and horse men, see the 2. 16, 1575 Uesugi ke gun’eki chō, Dai Nihon Komonjo Iewake, Series 12, Uesugi ke monjo (Tōkyō teikoku daigaku shiryō hensanjo, 1935), vol. 2, docs. 639– 640, pp. 1– 58. Hōjō Ujimasa likewise refers to formations of pikemen, bowmen, and gunners in a 1587 report. See Sengoku ibun (6 vols.; Sugiyama Hiroshi, comp. Tōkyōdō shuppan, 1989– 1995), vol. 4, doc. 3229, 12. 9, 1587, Hōjō Ujimasa chakutō kakidashi utsushi, pp. 197– 202.

95. For Takakage’s actions, see Kurushima Noriko, Ikki to Sengoku Daimyō (Kōdansha, 2001), pp. 14– 17, and Suitō Makoto, Asakura Yoshikage (Yoshikawa kōbunkan, 1981), pp. 5– 22.

96. Th e Miyoshi led a revolt (do- ikki) against the shugo of Awa province in 1487. Despite this disobedience, they were appointed as the deputy shugo of Sesshū early in

158 Thomas Conlan

the sixteenth century. See Shōzui jidai Miyoshi Nagayoshi Tenka o seisu (Shō zuijō kan-seki Kokushiseki shiteikinen tobubetsu ten, Tokushimajō hakubutsukan, October 2001), and Nagae Shō’ichi, Miyoshi Nagayoshi (Yoshikawa kōbunkan, 1968).

97. See the “Ashikaga Kiseiki,” in Kaitei Shiseki Shūran, vol. 13 betsukirui vol. 2 (Kondō katsuhanjo, 1902), pp. 132– 264, for the description of the Battle of Busshariji on 7. 21, 1547, pp. 192– 193. See also Nagae, Miyoshi Nagayoshi, pp. 90– 95.

98. Miyoshi Yoshitsugu, Nagayoshi’s heir, ultimately killed Ashikaga Yoshiteru on 5. 19, 1565. See Niigata kenshi shiryōhen, vol. 5, Chūsei, no. 3 (Niigata, 1984), doc. 3740, 6. 24 (1565?) Yasumi Munefusa shojō. Th e most accurate transcription of this document appears in Jōetsu shishi sōsho 6, Uesugi- ke gosho shūsei, vol. 1 (Jōetsu shi, 2001), doc. 288, p. 182. Nagayoshi had already died on 7. 4, 1564. See Nagae, Miyoshi Nagayoshi, p. 275.

99. Th is is particularly true for the years 1553 to 1558. See Imatani Akira, Sengoku Daimyō to Tennō (Kōdansha gakujutsu bunko, 2001), p. 194, and Imatani, Sengoku ji-dai no kizoku (Kōdansha gakujutsu bunko, 2002), pp. 188– 210.

159

WAR IS CREDITED WITH THE CREATION of the modern nation- state as we know it. Po-liti cal sociologists beginning with Max Weber but more recently including Tilly (1975, 1978), Tilly and Tarrow (2003), Downing (1992), and Ertman (1997), have argued that the mobilization of big armies and the execution of long wars in the sixteenth century transformed loosely or ga nized feudal entities or isolated merchant city- states into large and centralized territorial states. In meeting external threats, governments had to fi nd better ways to squeeze re-sources out of their populations and in the pro cess developed greater internal capacity to govern, administer, and control their own territories.

Th e macro- sociological accounts may be right that the nation- state was a side eff ect of the fi scal needs of early modern governments waging wars. In Japan, as in Eu rope, wars eventually led to centralization, though as we saw in the example of the Swiss alpine villages and for a time even in mountain or monastic redoubts in Japan, the threat of violence can under some conditions reinforce smaller scale and more horizontal forms of po liti cal or ga ni za tion. Th e explanation off ered in this book for territorial consolidation boils down to economies of scale in the provision of security, which is a factor that varies, among other things, with terrain. In high mountain valleys or in small island communities where natural defenses reduce the usefulness of large armies and where the eff ort of each fi ghting man is valued, territorial scale can be small without losing strength. Scale economies in defense explain why terri-torial consolidation originated in fl at, fertile regions where farmers were will-ing to trade taxes for protection out of dire necessity. Eventually, national boundaries radiated outward until more remote areas were swallowed up

PostscriptJohn A. Ferejohn and Frances McCall Rosenbluth


under centralized control but not until communities living in prosperous lowlands supported enormous and opportunistic armies. In Japan, as in early modern Eu rope, farmers living in fertile plains accepted— oft en under dire circumstances— strong, centralized government as the solution to their phys-ical vulnerability.

Scale economies account for considerable variation across cases in the ef-fects of war on territorial consolidation, but an explanation centering on the relative costs of expansion is as incomplete as the macro- sociological conven-tional wisdom in accounting for the onset of wars in the fi rst place. Aft er all, Japan enjoyed freedom from foreign invasion, and a decentralized system of overlapping jurisdictions lasted for centuries aft er imperial rule from Kyoto collapsed at the end of the twelft h century. As Susumu Ike asks at the end of Chapter 3, what broke the equilibrium, engulfi ng Japan in internal wars of conquest?

One explanation suggested in the literature on the military revolution in Eu rope is that fi rearms changed the balance of power in favor of aggression over defense by making castles more vulnerable to attack. Th omas Conlan, in Chapter 7, warns against the technical change argument, pointing out that guns were not always decisive in battle. Th e timing of fi rearms and territorial consolidation, moreover, is not quite right. Conlan argues instead that military success rested on superiority in administering territories, raising revenues, and or ga niz ing large armies. Th is still leaves open the question, of course, of why some domain lords became more successful than others in fi scal, military, and logistical administration.

Ike’s and Inaba’s analyses of the relationship among warlords, estate own-ers, and farmers, in Chapters 3 and 4, supply some missing pieces. For centu-ries, Japa nese farmers were enmeshed in clientelistic relationships with local estate own ers for their livelihood. Gradually, farmers became more eco nom-ical ly self- suffi cient, particularly in the Tokai region along the thoroughfare between Kyoto and Kamakura where proximity to resources and markets gave farmers alternatives to clientelistic de pen den cy. Once more in de pen dent, farmers were free to shift alliances for self- protection and advancement in a way that shook the entire ladder of de pen den cy, from villages at the bottom to warlords at the top.

Patrilineal bequests of landed estates gave way to struggles among rival groups, creating complex possibilities of power displacement and land confi s-cation. Insecurity and instability intensifi ed, which increased the farmers’

Postscript 161

willingness to pay taxes to whichever co ali tion of leaders was perceived as be-ing best able to secure life and property. Once po liti cal authority and military power lined up territorially, relatively small perturbations in force sizes un-raveled a delicate balance of power among warlords, unleashing the wide-spread warfare of the Warring States period. Successful military leaders and territorial administrators expanded the size and scope of their acquisitions, while weaker leaders were winnowed out. Although fi rearms were an instru-ment of conquest, the reasons for upheaval and the ensuing ac cep tance of a repressive peace lay deeper in changing societal structure. Ironic though it is that growing agricultural self- suffi ciency should have triggered the chain of events leading to more repressive governance, the episode showed that eco-nomic autonomy is not as empowering as mobility or defensibility. Th e speci-fi city of agricultural assets in land meant that farmers— even those with mul-tiple suppliers and markets and who could therefore sell their products at market prices— still had to contract with someone for physical security.

Th e pro cess of social turmoil and ensuing territorial consolidation in me-dieval Japan displays some striking similarities and contrasts with what hap-pened in Eu rope. Had Charles Tilly and Sidney Tarrow (2003) turned their attention to Japan, they would have noted that in sixteenth- and seventeenth- century Japan, as in Prus sia where there were relatively few own ers of mobile assets, own ers and tillers of land were the main players in the struggle for po-liti cal control. Although by the end of the sixth century merchants of the bustling city of Sakai (near Osaka) helped to bankroll Oda Nobunaga’s forces, the vast bulk of money supplying armies in the preceding de cades had come from farming villages desperate to make deals with one overlord or another. By contrast, in the Italian city- states or En gland, or even France where mon-archs sold infl uence to increasingly wealthy merchants, the groups bargaining with the monarch were more mobile and less vulnerable than farmers, tem-pering the nature of centralized rule. In medieval Italian city- states as well as in En gland following its Glorious Revolution of 1688 and the Netherlands af-ter freeing itself from the Habsburg Empire, merchants can even be said to have gotten the upper hand. Italian republicanism was displaced by foreign domination by the sixth century, so we can only imagine how an uninter-rupted history would have played out, but in En gland and the Netherlands the republican checks on monarchical rule were never entirely reversed.

An additional question that this book alludes to but addresses only obliquely concerns how far territorial consolidation goes. What are the conditions under


which it marches all the way to full, nationwide absolutism, as in seventeenth- and eighteenth- century Prus sia or France, and when instead does it stop short at federal bargains in which provinces retain some autonomy as in seven-teenth- and eighteenth- century Japan or even Italy and the United Kingdom (if we consider Scotland, Ireland, and Wales)? Th e Tokugawa shogunate ruled over the entire archipelago of modern Japan, but domain lords— even those who had fought on the “wrong” side and were defeated by Tokugawa at Sekigahara— controlled most of the ordinary administration of their lands. Japan was governed by a federated landed aristocracy for over two and a half centuries (1600– 1868). Where there is no external threat to force the state to become strongly command- and- control oriented, permitting a mea sure of local economy can be a low- cost way to govern.

Aristocratic rule is likely to be a weak form of po liti cal or ga ni za tion for the purposes of war- readiness against external threats. In the fi rst place, as Machiavelli noted of his native Florence, aristocracies (of the merchant vari-ety in the case of Florence) fail to excite the loyalty and morale of the potential citizen base in ser vice of military need. Second, the peasantry is not likely to be in strong physical or economic health in the event they are conscripted because landed nobles want to extract rents from peasants, and if they are lo-cal monopolists, they have the power to do so.

Th e absolutist monarchies that arose in early modern Eu rope can be inter-preted as a broader, rather than narrower, co ali tion than the feudal govern-ments they replaced in response to perceived defensive need. Unlike the ear-lier rule by landed aristocracy, kings tended to force the nobility to reduce their demands on the peasantry so that the peasants could better contribute taxes and manpower for national defense. In Prus sia a strong monarchy bal-anced the interests of the Junker nobility and the peasants who made up the bulk of the army. Poland provides a curious case that resisted this logic, for the nobility that made up one- tenth of the population was reluctant to arm the peasants against the Prus sians and Rus sians in the eighth century for fear that a victorious peasant army might turn on the Polish nobility. Poland sur-vived for many years as a narrowly based oligarchy, but was caught unpre-pared by the large armies from Prus sia and Rus sia that dismembered the proud Polish republic in 1797.

Th e Polish anomaly aside, the contrast between Eu ro pe an absolutism and Tokugawa’s loose federation of domains seems to suggest that tighter central-ization follows higher levels of military threat. For centuries between the

Postscript 163

failed Mongolian invasions of the thirteenth century and the successful American one of the nineteenth, Japan did not have external threats of the sort France and Prus sia faced that would have pressed territorial domains into common cause. Without real fear of foreign invasion, there was no pres-sure from the center to elicit extensive mobilization, nor did the domains have any interest in providing it. Th e domains paid the Tokugawa nominal alle-giance and taxes in exchange for retaining de facto local control and dynastic rule, while the Tokugawa shogunate enacted policies to keep the domains weak rather than to make the nation strong. But the domains were barred from investing in military preparedness, and for over two and a half centuries samurai warriors were, by virtue of inheritance, veterans on a stipend without ever having fought a war.

Insuffi ciently paid as war- less warriors, many in the military class became bureaucrats. As in France where the monarchy lured aristocrats into becom-ing functionaries of a nascent modern state by ensconcing them in Versailles and cutting them off from their local support bases, the Tokugawa required daimyō families to maintain sumptuous estates in Edo at great cost to do-mainal trea suries. But unlike in France where the nobility was incorporated into a strong central government, the Tokugawa regime opted for a stunted level of national integration that kept members of former “enemy” domains outside the circles of power in Edo. When Japan fi nally faced a foreign threat in the form of the “Black Ships” of the U.S. Navy steaming into Edo Bay be-ginning in 1854, samurai- bureaucrats in the enemy domains launched a revo-lution that toppled the Tokugawa in short order, given the lack of national preparedness. Th e Meiji government that they established in 1868 abandoned federalism in favor of radical centralization, abandoning the federal structure that had prevented the strongest possible marshalling of national resources. Tokugawa Ieyasu had not needed it in 1600 and was shrewd enough to know that trying to achieve it might have brought to power another constellation of forces in his place.

Notes1. Of course, not all states landed on effi cient solutions to their need for funds.

Beik (1985) argues that the French crown raised war funding by making offi ce into secure property that could be sold, but in so doing, increased the group of nobility exempt from taxation. Th e French monarchy’s short- term solution to its fi nancial problem created long- term shortfalls.


2. Bean 1973.3. On the Tokugawa regime, see, for example, Banno and Mitani 2006; Maruyama

1974; Totman 1967; Najita 1987.4. Th e Tokugawa won contingent victories over rival domain lords in 1600 and

again in 1615: as long as domains were largely autonomous with respect to internal af-fairs, they would settle for nominal Tokugawa overlordship. Th e Tokugawa tried to keep the domains from gaining strength by soaking them for ceremonial appearances in Edo, and restricted internal and external trade to limit the possibilities of uneven growth. Th is produced a static, but relative to the outside world, weakening economy. Few realized how weak it was until foreigners arrived with gunships in 1854, at which time support for the Tokugawa regime evaporated in favor of Meiji centralizers.

5. Contrary to Downing (1992), full mobilization need not lead to absolutism. Machiavelli writing in Re nais sance Florence had in mind a diff erent response to the weakness of aristocracy: expand the franchise in order to harness the power of the public to military valor and conquest. Machiavelli was inspired by the Roman deci-sion in about 450 b.c. to give fi ghting men the right to vote, but Cleisthenes’s “demo-cratization” of Athens in 508 b.c. in order to overthrow Spartan- backed aristocrats was an even more ancient pre ce dent. In Japan, as in Switzerland, only the isolated mountain or monastic communities came close to this model.

6. En gland, which enjoyed some protection from the Channel, also avoided the po liti cal consequences of full- scale military mobilization, but if we are correct, devel-oped a form of mixed government rather than territorial federalism because of the power of merchants to protect themselves from the arbitrary rule of monarchs.

ReferencesBanno, Junji, and Ichiro Mitani. 2006. Nihon no kindaishi jukko: Rekishi o tsukuro

mono [Essays in Modern Japa nese History: What History Is Made Of]. Tokyo: Iwanami shoten.

Bean, Richard. 1973. “War and the Birth of the Nation State.” Journal of Economic His-tory, 33, 1: 203– 221.

Beik, William. 1985. Absolutism and Society in Seventeenth Century France: State Power and Provincial Aristocracy in Languedoc. New York: Cambridge University Press.

Downing, Brian M. 1992. Th e Military Revolution and Po liti cal Change: Th e Origins of Democracy and Autocracy in Early Modern Eu rope. Prince ton, NJ: Prince ton Uni-versity Press.

Ertman, Th omas. 1997. Birth of the Leviathan: Building States and Regimes in Medieval and Early Modern Eu rope. New York: Cambridge University Press.

Maruyama, Masao. 1974. Studies in the Intellectual History of Tokugawa Japan. Trans-lated by Mikiso Hane. Tokyo: University of Tokyo Press.

Postscript 165

Najita, Tetsuo. 1987. Visions of Virtue in Tokugawa Japan: Th e Kaitokudo Merchant Academy of Osaka. Chicago: University of Chicago Press.

Tilly, Charles. 1975. “Revolutions and Collective Violence.” In F. Greenstein and N. Polsby, eds., Th e Handbook of Po liti cal Science, vol. 3. Reading, MA: Addison- Wesley.

Tilly, Charles. 1978. From Mobilization to Revolution. Reading, MA: Addison- Wesley.Tilly, Charles, and Sidney Tarrow, eds. 2003. Th e Politics of Collective Violence. New

York: Cambridge University Press.Totman, Conrad. 1967, 1988 (paperback edition). Politics in the Tokugawa Bakufu,

1600– 1843. Berkeley: University of California Press.

Glossary and Index

169

Akutō , “evil bands”; groups of peasants and minor warriors involved in loot-ing, raiding, and other local crimes during the late thirteenth and fourteenth centuries

ashigaru , literally, “light feet”; low- ranked infantry; draft ed peasantsbōsen , defensive warfarechakutōjō , reports of arrivalchūka , literally, “center of civilization,” oft en used to refer to Chinachūnin- sei , mediation by powerful individualsDaijōkan , Council of Statedaimyō , domain lordsgekokujō , literally, “lower overcoming higher”; proverb that aptly character-

izes the turbulent Warring States era which experienced the widespread phenom-enon of subordinates rebelling against their lords

genin , shoju , direct laborersgō , village or districtgokenin , literally, “house men” of the Kamakura bakufu who were obligated to

perform guard duty but were otherwise elite members of provincial society. Th e term also refers to a low- ranking warrior in the Tokugawa era.

Goseibai- shikimoku (Jōei Formulary) , the law code of the Kamakura bakufu, promulgated in 1232 and appended frequently thereaft er through the re-mainder of the Kamakura age

gunchūjō , literally, “petitions for reward”; petitions for reward provide a more comprehensive narrative of battle than battle reports because they mention how a sequence of such skirmishes unfolded through time

gundan , provincial regiments

Glossary

170 Glossary

gunnai ichizoku , regional alliance of samurai that in the sixteenth century would eventually give way to a local league of fi ghters, the Iga ikki, which had one main stated objective: restoring peace and order in the province

gusuku , castleshan- kenryoku , daimyō authoritieshanzei , literally, “half the tax”; provision by which shugo under the Muromachi

shogunate were allowed to retain for their own purposes half the taxes collected from public lands and centrally owned estates

heinō- bunri , separating farmers and warriors into two classes. Toyotomi Hideyoshi’s Sword Hunt (1588) formalized a pro cess of specialization that had al-ready begun under military governors

heishi , a conscript soldier enrolled in one of the provincial regiments under the ritsuryō codes

hihō hisō , literally, a “fl ying projectile fi re spear”; that is, fi rearmhikan , retainerhiryūsō , Chinese sources refer to a primitive gun, known as a “fi re dragon

spear,” being created in 1355hiya , primitive fi rearms composed of three metal tubes welded together, known

as fi re arrowshō , a catapulthōkōshū , direct retainers of the shogunhonryō , right to the landhyakushō , common people; that is, persons without court rank and subject to

taxationIga ikki , local league of fi ghters in Iga with one main stated objective: re-

storing peace and order in the provinceIga sōkoku ikki , the league of all the commons of Iga. Specialists be-

lieve that the league (or the written rules of the league) was created because of an urgent military situation, the looming threat of invasion of the province by Miyo-shi Nagayoshi of nearby Yamato in the early 1500s

Ikkachū , “families”; name for the forty- six signatories who sought to ensure the regional peace, vowing not to fi ght over taxes on peasants and to try to prevent insurgent acts by subordinated people who might “cause trouble”

ikki , a pledged or ga ni za tion with the clear aim of maintaining order within the domain; squads

ikkō ikki , literally translated as leagues of members of the Honganji sect of True Pure Land Buddhism (Jōdo Shinshū) in the sixteenth century

Imagawa kanamokuroku , code of laws, enacted by the Imagawa, Sen-goku daimyō of the province of Suruga in 1526

Glossary 171

Inoue- shū chūbatsu- jiken , clan from the pact for invading a neigh-bor’s territory

iru , to shoot; used for both bows and fi rearmsiteki , term for neighboring “barbarian” countriesjinai , temple groundsjinai mach , temple townjitō , estate stewards or managers appointed by the Kamakura shogunate; chiefl y

concerned with tax collection and local police dutiesjitō- shiki , the rights, duties, and perquisites associated with the position of

jitōjizamurai , literally, “samurai on the land”; yeoman farmers who also functioned

as warriorsJōdo shinshū , literally, “True Pure Land” Buddhism; this was the sect that

sponsored a large self- governing league in Kaga provincekaimin, , people of the seakami , Shintō divinitieskanjō , document of praise, which could include a recommendation that a mili-

tary leader be rewarded for his actions in battlekarita- rōzeki , rice plantskashindan , seigneurial pacts that would provide checks on the authority of

warlords and provided for the existence of groups of vassals under the rule of a lord and were highly respected

kassen bugyō , battle reports to administratorskassen chūmon , battle reports, written shortly aft er every skirmish, with

each document mentioning the damages incurred by warriors so as to ensure compensation for their actions

kebiishi , literally, “Investigators of Oddities”; offi cers of the Offi ce of Impe-rial Police (kebiishi- chō) in the capital; also a provincial military/police post in the provinces established in the mid- ninth century

kenka- ryōseibai- hō , rule under which both aggressors and defenders were regarded with equal severity and private wars were prohibited altogether

kinjo- no- gi , mediationkōgi , shogun’s offi cialskoguchi , literally, “tiger’s mouths”; referred to extremely narrow pathskokuga , the provincial government offi cekokujin or kokunin , literally, “people of the country” or local warriors; during

the fourteenth and early fi ft eenth centuries, principally used to distinguish pow-erful provincial warriors whose activities were confi ned to a single province from those who also held posts as shugo; also a person who resides in the province

kokujin- ikki , local social pact of equal- basis unifi cation of local lords based on regional relations

kokushū , emerging warrior leaders known as Warring States daimyōkosen , off ensive warfarekosenbōsen- hō , law that restricted private warskugyō , high- ranking aristocratsmenka , peasantsmiuchi , powerful retainersmiyako no musha , literally, “warriors of the capital,” men of the fourth or

fi ft h court rank who utilized the profession of arms as a vehicle for more gen-eral career advancement; sometimes called the samurai in English- language sources

mura no shiro , fortifi ed places on the heights, what Fujiki Hisashi calls “cas-tles of the village”

mura- okite , rules governing local behaviormura- uke- sei , system in which peasants paid land taxes to the village council,

which in turn paid the lord his duemusha taishō , literally, “warrior general”; high- ranking warriorsnengu , land taxesninja , literally, “stealthy ones”; or more commonly read as shinobi mono; lightly

armed warriors who fought with homemade implements and used tactics of sur-prise, stealth, and intrigue

ōryōshi , literally, “Envoy to Subdue the Territory”; one of the two main titles granted provincial warrior leaders under the Heian provincial military/police system

Ritsuryō , the legal codes that defi ned the structure and operations of the impe-rial state

saikyo , legal judgmentsaikyo- jō , letter of sentencesashidashi , tax documents that reveal a contractual relationship between a vil-

lage community and a warrior lord. When a warrior lord defeated his opponent and conquered his opponent’s territory, he fi rst had to collect the tax documents from villages in the territory that he newly gained

seii taishōgun , literally, “Field Marshal for Subduing the Barbarians”; originally a temporary title for the offi cers appointed to command of ritsuryō- era expeditionary forces operating in the east and northeast; later the title for the head of the Kamakura and Muromachi shogunates

sekkanke , Regents’ House; the sublineage of the northern branch of the Fuji-wara house that came to hold in heredity the post of sesshō (regent for an emperor still in his or her minority) or kampaku (“regent” for an adult emperor)

172 Glossary

sengoku , warring statessengoku daimyō , powerful warlords of the Warring States periodshakizu , a wound from gunshotshiki , originally a function or offi ce with attached perquisites; later the right to

designated income under the title of an offi ce; oft en hereditaryshisen , private wars; that is, confl icts fought without the sanction of government

authoritysho- ku, small districtsshōen , a landed estate, normally held in proprietorship by a high- ranking mem-

ber of the central nobility or by one of the great religious institutions; the domi-nant form of landholding and land administration from the eleventh through the fourteenth centuries, its full history runs from the eighth through the sixteenth

shugo , province- level “constables” appointed by the Kamakura shogunate; pro-vincial military governors under the Muromachi shogunate

shugodai , shugo deputiessō , district; most Japa nese historians have compared them to the medieval com-

munes of the Westsōbuji , peace order forbidding neighboring territories to fi ght over boundariessōkan- shiki , estate own erssōryō , the head of extended warrior houses during the Kamakura period; in

spite of a general pattern of divided inheritance, the house head retained general leadership of and responsibilities to and for cadet houses

sōson , village; or gan i za tion al units below the sō, which arose to replace large estates as the dominant units of authority

tachi uchi , literally, “clash of swords”; a general appellation for battletaikō- kenchi , cadastral surveys, part of the policies of the Toyotomi Hidey-

oshi regimetanshū no onzei , a typical way to describe mid- fi ft eenth- century military

unitstebiya , “hand fi re arrow”teppō , explosive shells; fi rearmsteppōshū , marksmenTsuibushi , literally, “Envoy to Pursue and Capture”; one of the two main titles

granted provincial warrior leaders under the Heian provincial military/police system

Tsuitōshi , literally, “Envoy to Pursue and Strike Down” ; a temporary deputa-tion used by the court from the mid- Heian period to grant special military or po-lice powers to warrior leaders for specifi c missions

uji , po liti cal units of the pre- imperial era

Glossary 173

waka , Japa nese poemswakashū , villages’ military forceswakō , seafaring piratesyakizu , “arrow wounds”yamabushi , “mountain monks”: practiced asceticism in the mountains under

the infl uence of Shingon esoterism and Buddhismyari awase senshi , confrontation with pikesyoriai , village councilszuryō , career provincial offi cials of the mid- to late Heian period

174 Glossary

175

Absolutism, 14n16Akutō. See Bandits (akutō)Alliances, 30, 116–117; warrior bands and,

36–39, 48n49, 49n51Amako clan, 64Amida, 93–95, 102–104Archery. See Arrows, use of; Mounted

archeryAristocratic rule, 162Aristocrats, court, 26, 29, 34. See also

Nobility, courtArrival, reports of (chakutōjō), 126Arrows, use of, 126–127, 131, 131t, 132–135t,

147–148, 153n28, 157n93Asai house, 64Asakura house, 64Asceticism, 119Ashigaru (light infantry), 140, 144–145Ashikaga shogunate, 11, 57, 61, 64Ashikaga Takakage, 148Ashikaga Takauji, 57, 67n8Ashikaga Yoshiaki, 64, 69n32Ashikaga Yoshimasa, 68n22Ashikaga Yoshimi, 68n22Ashikaga Yoshiteru, 149, 158n98Autonomy, 71, 76, 162, 164n4; background of

local, 1–2; economies of scale and, 3, 160–161; in Iga region, 110–123; peasants and, 53, 57–58, 60

Bandits (akutō), 111, 115–116, 119Bargaining leverage, farmers’, 5–6, 8–11Battle reports (kassen chūmon), 126–130

Battles. See Pitched battles; Skirmishes. See also under names of specifi c battles

Beik, William, 163n1Bodin, Jean, 14n13Buddhahood, 93–95Buddhism: Honganji sect of, 69n33, 92–106;

Jōdo Shinshū (True Pure Land), 2–3, 92–95, 103; Nichiren sect of, 69n33

Bullets: stone, 146, 156–157n83; wounds by, 147

Cadastral surveys (Taiko-kenchi), 71, 73Cannon, 8, 16n34Career provincial offi cials (zuryō), 29–30, 32,

34, 46n28Castles, 8; of the village (mura no shiro), 115Catapult (hō), 145Cavalry: defense and, 8–9; ritsuryō military

system and, 24–25; technology and, 136–137, 140, 143–145, 150–151

Chakutōjō. See Arrival, reports of (chakutōjō)

Change, organizational, 125, 149. See also Technology, organizational

Chevauche, 14n14Chieft ains, regional, 22–23, 25, 28China, 5, 77; coins from, 69n35; Tang,

22–23, 25Christianity, 92, 106“Chronicle of Ōnin,” 143Client-patron relationships, 30, 33,

48–49n49, 66, 160Commoners’ struggles, 93, 106. See also

Ikkō-ikki

Index

176 Index

Community vitality. See Village communities

Competence over loyalty, 53–70Conscription: Meiji government and, 86;

privatized, 35–36; ritsuryō military system and, 23–26

Consolidation. See Territorial consolidationContracts: oaths and, 63; village

communities and, 73–74; warrior alliances and, 38–39, 48n49. See also Oaths

Court politics, 29–31, 46n28Currencies, circulation of, 65, 69n35

Daimyō (domain lords), 3, 13–14, 15n25, 15n29; Honganji sect and, 96–98, 105; Sengoku, 59–60, 76, 115, 117–118; territorial expansion of, 53–54, 57; Toyotomi, 65–66; Warring States (kokushu), 58–59. See also Lords and retainers

Date Uemune, 62Decision-making, consensual, 83–84Democracy, bargaining hypothesis of, 73–74Discourses (Machiavelli), 13n3Dispute resolution system, 80–81Districts (sō), 116Downing, Brian M., 159, 164n5

Economies of scale: autonomy and, 3; changes in, 7–9; consolidation and, 160–161

Electoral system, multilayered, 84–85Elites, provincial: alliances of, 30; military

service and, 32–34; wealth and power of, 26–29

England, 14nn14–15, 161, 164n6Erizeni, 69n35Ertman, Th omas, 159Estate system, 15n21, 115, 160; Niimi, 141;

Ōyama, 137–139, 141Europe, 5, 8–9, 14nn14–15, 16n36, 161–163,

164nn5–6Evaluation, merit system of, 83

Farmers’ bargaining leverage, 5–6, 8–11Federal system, Tokugawa’s de facto, 3Feudalism, 5Firearms: introduction of, to Japan, 124,

145–148, 157n93; limitation of 16th century, 124–136, 132–135t;

organizational technology and, 124, 160–161

Flatlands. See Plains (fl atlands)Foot soldiers: as leverage, 8, 10, 16n34. See

also InfantryForeign invasion, 1–2, 24, 79, 160, 163Forty-Seven Samurai, Th e (fi lm), 67n13France, 9, 14n14, 161–163Fujiwara regents, 30, 33

Gekokujō (“the low overthrow the high”), 62, 69n29, 69n31, 115, 149

Gempei War, 40Genji preeminence, 32–33Gentry, provincial, wealth and power of,

26–29Go Daigo, (emperor), 57, 67n8Gokenin (“shogunal housemen”), 55–56,

152n11Governance. See Imperial court; Judicial

systems; Laws, code of; Lords and retainers; Shogunates; Social order

Governors, provincial: military service and, 32, 35; ritsuryō system and, 27–28

Gunnai ichizoku (alliance), 116Guns. See Firearms

Hand-to-hand combat, 131t, 152n17Hanzei edict, 137–139Hashiba (Toyotomi) Hideyoshi. See

Toyotomi HideyoshiHatakeyama Masanaga, 143–144Headmen systems, 80–85Heian period, 21, 28–29, 34–39Heiji Incident, 39Heinō-bunri. See Separation of warrior and

peasant classes (heinō-bunri)Hino Tomiko, 68n22Hiya (fi re arrows), 146Hobbes, Th omas, Th e Leviathan, 14n10Hojo family, 40, 55–57, 62, 64, 74–75Hojo Soun. See Ise Nagauji (Hojo Soun)Hokkaido, 15n18Honganji sect, 69n33, 92–109, 106n4;

ideology of, 93–95, 102–104, 107n32; justifi cation for fi ghting by, 102–104; power of, 101; social order of, 95–98; uprisings and, 97–102, 107n26

Honshu, 3Horeki, Reform of, 81Horikoshi Kubo, 68n24

Index 177

Hosokawa house, 58, 61–62, 79, 139, 141, 149Hosokawa Katsumoto, 61, 140–142Hosokawa Masamoto, 61–62Hosokawa Yorimoto, 138Household system, 56–57; breakdown of,

58–64

Iga region, 110–123, 123n1; attacks on, 118; ban on violence in, 114–115; districts (so) and, 116; espionage and, 121–122; loyalty in, 119–120; two-level leagues and, 111–116; village communes and, 116–120

Iga sōkoku ikki, 114Ikki, 67n9, 112, 116, 120, 153n25. See also

Ikkō-ikki; Kokujin-ikki (territorial pacts)Ikkō-ikki, 64, 69n33, 119; Honganji sect

resistance and, 92–106Imagawa code of law, 60–61, 68n15, 68n20Imagawa house, 60–64Imagawa Yoshimoto, 63Imperial court, 5–7, 21–52; court politics and,

29–31; dying years of, 54–55; early succession and, 5; emigration to countryside and, 30–32, 46n28; fi rst shogunate and, 39–43; institutions and procedures of early, 26–29; new warrior order and, 32–39; ritsuryō state/army and, 22–26; stratifi cation and, 29–30

Infantry, 148; light (ashigaru), 140, 144–145; ritsuryō military system and, 24–26

Inoue house, 60–63Ise Nagauji (Hojo Soun), 68n24Isshiki Yoshitada, 141–142Italian city-states, 161

Japan, settlement of, 5Japanese corporate society, 87Jizamurai, 112, 115–116, 120–121Jōdo Shinshū (Pure Land) Buddhism, 2–3,

92–95, 103Judicial systems: of Kamakura shogunate,

55–56; Kumanoto dispute resolution system as, 80–81. See also Laws, code of

Kaga province, Honganji sect in, 92, 96–97, 101

Kagemusha (fi lm), 15n30Kamakura shogunate, 21, 40–43, 54–58Kammu Heishi, 33, 47n37Kasai house, 138–139Kasai Motosuke, 138–139, 154n35

Kashindan (pacts), 117Kato Kiyomasa, 77–79Kawanakajima, Battle of, 63Kennyo (Honganji patriarch), 100–103Kikkawa house, 142–143Kitabatake Akiie, 67n8Kōga district, 110–111, 113, 116, 119–121, 123n1Kōgagun chūsō, 116Koguchi (narrow paths), 111Kokujin-ikki (territorial pacts), 53, 57, 62,

67n9Kokushu, 58–59Korean invasion, 76–79Kosenbosen-ho (law), 57Koza-ha (associated with Communist

Party), 14n17Kumamoto, self-governance study and,

79–86Kumamoto house, 81–82, 88n17Kumamoto Prefectural Assembly, 84–86Kuniwake, 64Kusunoki Masashige, 67n8Kyoto, 26, 31–32, 53

Labor, scarce, 16–17n31Landholding, 1–13, 160–161; boundary

confl icts and, 60; Honganji sect and, 96–98; Imagawa code of law and, 60, 68n15; ritsuryō system and, 27–29

Laws, code of: Imagawa, 59–61, 68n15, 68n20; Toyotomi Peace Order and, 75–76; village communities and, 73. See also Judicial systems

Leagues: districts (sō) and, 116; two-level, 111–116

Leviathan, Th e (Hobbes), 14n10Lords and retainers, 53–70, 160; Kamakura

shogunate and, 54–58; loyalty and, 54, 57–59; national unity and, 64–66; Toyotomi Peace Order and, 75–76; village communities and, 72–74; village headmen systems and, 80–85; Warring States period and, 58–64. See also Shugo (provincial constables or military governors)

Loyalty, 53–70

Machiavelli, Niccolo, 162, 164n5; Th e Discourses, 13n3

Marriage, among aristocrats, 34Marxism, 9, 14–15n17

178 Index

Meiji government/Restoration, 67n8, 71, 84–86, 163

Merchants, 65, 160, 161, 164n6Mikawa province, Honganji sect in, 98–99Military organization. See Technology,

organizationalMilitary protection: economies of scale and,

8–9; private market and, 32–39, 43; 16th century unifi cation and, 4–12; village communities and, 72–74

Military revolution, 126, 147, 160Military rule/systems: centralization and

restructuring of, 22–26; decentralized, 5–6; fi rst shogunate and, 41; new order in, 32–34; private warrior groups in, 34–39; village communities and, 72–74

Military technology. See Technology, organizational

Minamoto, 32–34, 55Minamoto Sanetomo, 40Minamoto Yoriie, 40Minamoto Yoritomo, 38–42, 55Minamoto Yoshitomo, 39Ming Dynasty, 77Miuchi (retainers), 138, 154n33. See also Lords

and retainersMiyako no musha (“warriors of the capital”),

32–33, 41Miyoshi house, 64, 157–158n96Miyoshi Nagayoshi, 114, 149Miyoshi Yoshitsugu, 158n98Mochihito, Prince, 39Monasteries, 2, 7, 9, 92, 110, 116Montesquieu, 14n15Morgarten, Battle of, 1Mori house, 60–64, 100Mountain regions: defenses and, 2, 68n25,

118–120, 159; taxes and, 16n38Mounted archery, 24–25Mount Kōya temple, 91

Nagao Tamekage, 62, 68n26, 69n29Nagashino, Battle of, 15n30Naitō house, 139–140, 142Naitō Motosada, 138–139, 154n35Naitō Sadamasa, 142Nakazawa house, 138–140, 153n32Nara period, 21Negoroji marksmen, 124–125, 147Networks of warrior groups. See Alliances,

warrior bands and

Nichiren sect, 95, 105Niimi estate, 141Ninja, 1, 11, 13n1; birth of tradition of,

110–123, 123n1Nitta Yoshisada, 57, 67n8Nobility, court: emigration to the

countryside of, 30–32; military service and, 32–34; wealth and power of, 26–29

Oaths, 63Oda Nobukatsu, 118Oda Nobunaga, 2, 6–7, 15n30, 63–65, 69n33,

124; Honganji sect and, 91, 100–102, 104–106; Iga region and, 114, 118

Okehazama, Battle of, 63Okinawa, 15n18Ōnin War, 58, 61, 115; organizational

innovation and, 140–145, 148–149Organizational change, 125, 149. See also

Technology, organizationalOrganizational technology. See Technology,

organizationalŌtomo house, 64Ōuchi Masahiro, 142, 155n54Ōyama estate, 137, 139, 141Oyamoto declarations and domain, 111–113

Pacts: kashindan, 117; territorial (kokujin-ikki), 53, 57, 62, 67n9

Patron-client relationships, 30, 33, 48–49n49, 66, 160

Peace order (sōbuji), 71–72, 75–76, 88n10Peasants, 3–4, 8–12, 15–16n31, 160; class,

separation of warriors from (heinō-bunri), 71, 74–75, 104–105; farmers’ bargaining leverage and, 5–6, 8–11; household systems and, 56; increasing autonomy of, 53, 57–58, 60; Korean invasion and, 76–79; and lga local warriors, 117–118; ritsuryō system and, 27; village communities and, 72–74; village headmen systems and, 80–85

Petitions for reward (gunchūjō), 126–129, 137Petrarch, 13n4Pikes, use of, 126, 131, 131t, 132–135t, 135–137,

147, 148–150; in Ōnin War, 140, 142–144, 156n70

Pirates, seafaring (wako), 2Pitched battles, 135, 144–145Plains (fl atlands), vulnerability and, 3–4, 9Poland, 8–9, 16n36, 162

Index 179

Power: communities and, 5, 72–74; of Honganji sect, 101; of nobility and provincial elites, 26–29; organizational technology and regional, 124–158; war, protection and, 4–12

Projectiles, use of, 131, 131t, 153n18; guns and, 157n93

Protection. See Military protectionProvincial elites. See Career provincial

offi cials (zuryō); Elites, provincial; Governors, provincial; Lords and retainers

Provisions. See Supplies and supply routesPrussia, 9, 162

Rebirth, belief in, 94, 102Record book (mura-daka), 88n3Reincarnations, 94Religious opposition to consolidation,

91–109; justifi cation for fi ghting in, 102–104; organization and, 105; right to bar warriors and, 97, 101; uprisings in, 97–102, 107n26

Religious wars, in 16th century Europe, 14n14

Reports: of arrival (chakutōjō), 126; battle (kassen chūmon), 126–130; petitions for reward (gunchūjō) as, 126–129

Retainers, lords and. See Lords and retainersRevenue. See TaxationRitsuryō state/army, 23–26; institutions and

procedures of, 26–29; role of elites in, 25–26; tactical limitations of, 24–25

Rocks, use of as weapons, 131t, 132–135t, 147Rokkaku house, 64, 143Romans, 10, 16nn41–42Rono-ha (associated with socialists), 14n17Rousseau, Jean Jacques, 13n3Ryūkyū Kingdom, 146

Samurai, 6–7, 21–52, 32–43, 163; fi rst shogunate and, 39–43; Iga region and, 112, 114, 116; new warrior order and, 32–34; private warrior groups and, 34–39; rise of, 21

Sashidashi. See Taxation, documents (sashidashi)

Scale economies. See Economies of scaleSecurity, 3, 9, 13, 14n10, 160–162. See also

Military protectionSeiwa Genji, 32–33

Sekigahara, battle of, 3Self-governance study, 79–87Sengoku era, 115; daimyō, 59–60, 76, 117–118Separation of warrior and peasant classes

(heinō-bunri), 71, 74–75; Honganji sect and, 104–105

Shibata Katsuie, 91Shimazu house, 65–66Shogunates, 38–39; Ashikaga, 11, 57, 61;

Kamakura (fi rst), 21, 39–43, 54–58; relationship of, to court, 42–43; Tokugawa, 121–122; as warriors’ union, 41–43

Shohei-Tenkei War, 54Shōkokuji, battle at, 143Shugo (provincial constables or military

governors): Ōnin War and, 140–141, 143; political organization and, 55, 57, 115, 137–140, 150–151; rebellion against, 148–149. See also Daimyō (domain lords); Lords and retainers

Skirmishes, 131, 135Social order: of Honganji sect, 95–98; in

Iga region, 110–118; Imagawa code of law and, 60–61; of Kamakura shogunate, 54–58; village community and, 72–74

Sōbuji. See Peace order (sōbuji)Status, multiple, 94–95Stirrups, cavalry and, 8, 16n34Stone bullets, 146, 156–157n83Stone monuments, 79, 81Stratifi cation, in Heian society, 29–30Sue house, 63, 69n31Supplies and supply routes, 140–141, 144–145,

148, 150, 156n73, 161Suzuki Masaya, 131, 135Swiss mountain warriors, local autonomy

and, 1–2, 120, 159Swords, use of, 131, 131t, 132–135t, 135–137,

153n28

Taika Reforms, 22Taiko-kenchi. See Cadastral surveys

(Taiko-kenchi)Taira, 32–34, 47n37, 54–55Taira Kiyomori, 39Taira Masakado, 36, 54Taira Sadamori, 54Takeda house, 62–64Takeda Shingen, 15n30, 62, 64

180 Index

Tanba province, 138, 140Tang China, 22–23, 25Taxation, 1–3, 161, 163; documents

(sashidashi), 73–74; Honganji sect and, 97–100; Kumamoto self-governance and, 79, 82, 85; military protection and, 6–7, 9–10, 16nn37–38; ritsuryō system and, 24, 27–29, 31; shugo and, 137–140; village communities and, 72–74

Technology, meaning of, 125–126Technology, organizational, 124–158,

160–161; defi ned, 125; military organization and, 136–137, 148–151; in Ōnin War, 140–145, 148–149; political organization and, 137–140; weapons and, 126–136, 131t, 132–135t

Tell, William, 1, 13n2Temples, 2, 41–42, 69n33, 91–93, 95–103, 105;

towns of ( jinai machi), 97, 100–101Tenji (emperor), 22Teppō (fi rearms), 145–148, 157n93Territorial consolidation, 6–12, 159–163;

demand for peace and, 66; Iga region resistance to, 110–123; national unity and, 64–66; organizational technology and, 124–158; religious opposition to, 91–109; Toyotomi Peace Order and, 76; Warring States period and, 58–64

Th irty Years’ War, 14n14Tilly, Charles, 159Tokugawa Ieyasu, 99–100, 105, 121, 163Tokugawa regime, 3, 87, 121–122, 162–163,

164n4; Toyotomi Peace Order and, 75–76Torigoe, temple of, 91Toyotomi Hideyoshi, 6–7, 15n29, 65–66,

69n37; Honganji sect and, 91, 104–106; Korean invasion by, 76–79; peace order of, 75–76; regime of, 71–72

Trench warfare, 144

Uesugi house, 59, 62–64, 68n26, 69n29Uji (political units), 5Unifi cation, 58–66; demand for peace and,

66; military protection and, 4–12; Warring States period and, 58–64. See also Territorial consolidation

Unsei Daigoku, 145

Varley, Paul, 140Village communities, 71–90; Iga region

village communes and, 116–120;

Kumamoto self-governance study and, 79–86; military forces (wakushu) of, 72; mobilization and expansion and, 75–79; power and social values in, 72–74; revisionist view of, 72; territory of, 87n2

Village councils (yoriai), 72Village headmen systems, 80–85

Wakushu (village’s military forces), 72Waldstetten, Treaty of, 120War: consolidation and, 1–12, 159–163; as

political leveler, 4. See also under names of specifi c wars

Warring States period, 58–64, 67n12, 161Warriors, Japanese, 1, 6–7, 13n1; alliances

and, 36–38; early samurai, 21–52; household systems and, 56; land and, 10; local, in Iga region, 112–123; local, shugo and, 138–139; names, 32–34; “of the capital,” 32–33, 41; and peasant classes, separation of (heinō-bunri), 71, 74–75, 104–105; privatized recruitment of, 32–36, 43; shogunate as union of, 41–43; temple towns and, 95–98, 101; Warring States period and, 58–64; warrior bands and, 36–39, 48n43, 48n49, 49n51, 67n7. See also Lords and retainers

Warriors, Swiss, 1–2Watchtowers, 144Weapons, use of, 126–136; battle reports

(kassen chūmon) and, 126–130; fi rearms (teppō) and, 145–148, 150; innovation and, 145; in Ōnin War, 140–145; petitions for reward (gunchūjō) and, 126–129; reports of arrival (chakutōjō) and, 126; wounds by, 131t, 132–135t, 137, 146–147, 153n28, 157n93

Weber, Max, 159

Yama, 87–88n2Yamabushi (mountain monks), 119Yamamoto Tsunetomo, 15n23Yamana house, 140–142Yamana Sōzen, 61, 141–142Yamato house, 22–23Yasuoka Ryosuke, 84–85Yoriai. See Village councils (yoriai)

Zuryō. See Career provincial offi cials (zuryō)