War and State Building in Medieval Japan 9780804774314

Citation preview

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 Press Stanford, California

Stanford University Press Stanford, California ©2010 by the Board of Trustees of the Leland Stanford Junior University. All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or in any information storage or retrieval system without the prior written permission of Stanford University Press. Printed in the United States of America on acid-free, archival-quality paper Library of Congress Cataloging-in-Publication Data 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

Acknowledgments

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 greatest thanks to the six historians who were willing to bridge the divide between history and social science to engage with us about the processes of state building in medieval Japan: Thomas 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 Noriko 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 Japanese Studies graciously made facilities available to us for the workshop, and the translators of Simul International did an exceptional job of translating the workshop in real time in both directions. The Council on East Asian Studies at Yale University provided funding and the administrative support for the workshop. Teruo Utsumi, a postdoctoral fellow in Political Science at Yale in 2004–2005, translated the workshop memos from the Japanese participants into English, 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 Hoffmann, 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 entire 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

WE HAVE INCURRED

v

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.

Contents

Contributors 1 War and State Building in Medieval Japan

ix 

John A. Ferejohn and Frances McCall Rosenbluth

2 They Were Soldiers Once: The 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 Political 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: Organizational Technology and the Consolidation of Regional Power in Japan, 1333–1600 Thomas Conlan



viii Contents

Postscript



John A. Ferejohn and Frances McCall Rosenbluth

Glossary



Index



Contributors

John A. Ferejohn is a political economist and democratic theorist who has writ-

ten widely in the areas of political institutions and behavior, judicial politics, and the philosophy of social science. He was the Carolyn S. G. Munro Professor of Political Science at Stanford University before becoming Professor of Law and Political Science at New York University in 2009. Ferejohn is a member 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 political theory. He sits on the editorial boards of Social Choice and Welfare, Democratization, Supreme Court Economic Review, and the Cambridge Press series Philosophy and Law and Economics and Philosophy. Frances McCall Rosenbluth is a political economist with a special interest in Ja-

pan. She is the Damon Wells Professor of International Politics at Yale University. 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 Contemporary Japan (Cornell 1989), Japan’s Political Marketplace (Harvard 1994, with Mark Ramseyer), The Politics of Oligarchy: Institutional Choice in Imperial Japan (Cambridge 1996, with Mark Ramseyer), an edited book on The Political ix

x

Contributors

Economy of Japan’s Low Fertility (Stanford 2007), Women, Work, and Power (Yale University Press 2010, with Torben Iversen), and Japan Transformed: Political Change and Economic Reform (Princeton University Press 2010, with Michael Thies). Thomas Conlan is a historian of Japan at Bowdoin College. Professor Conlan

studied Japanese history at the University of Michigan (B.A., 1986), Kyoto University, and Stanford University (Ph.D., 1998). Conlan’s scholarship focuses on medieval Japanese history and in par ticular on the nature of warfare and the political role of esoteric (Shingon) Buddhism. His books include In Little Need of Divine Intervention (Cornell 2001); and State of War: The Violent Order of Fourteenth Century Japan (Michigan 2003). Karl Friday teaches at the University of Georgia, where he focuses on Japanese

military institutions and traditions. His Ph.D. is from Stanford in history in 1989. His books include Hired Swords: The 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 The 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 history,” which examines political economy from “the bottom up.” Eschewing older models of history that focused on elite behavior or economic determinism, 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 fifteen 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 Japanese history, The World Upside Down: Medieval Japanese Society (Columbia 2001).

Contributors xi

Carol Richmond Tsang is a specialist in the religious movements in medieval

Japan that successfully resisted Japanese territorial consolidation for many decades. 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

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

Introduction The 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 figure. It is generally known that the ninja existed sometime in the mists of Japanese history. Less well understood is that the ninja was but one manifestation of fierce and extensive resistance to encroaching armies in the dying years of medieval Japan. Local farming communities, particularly those in mountain valleys, armed themselves with simple weapons and guerrilla techniques to forestall the trend toward territorial consolidation and centralized taxation. The transformation of “ninja” (the “forbearing ones” or shinobi mono) into warriors with virtually supernatural powers is a recent invention that glorifies the struggle of humble mountain villages for local autonomy in the late sixteenth century. The world is more familiar with similar events in Europe. The legend of William 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 after 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 magnificent cavalry of Duke Leopold I of Austria in an ambush at Morgarten Pass, pushing countless horses and their riders off a steep mountainside, spearing other unfortunates through with pikes, and causing the rest to flee in terror. Swiss pikers from mountain villages managed to protect their land from foreign invaders, thereby assuring Swiss autonomy. Feared and admired the 1

2 Ferejohn and Rosenbluth

world over for their ferocity in battle, Swiss fighters were recruited into mercenary armies throughout Europe. The Roman Catholic pope chose them for his own guards, a role they continue to serve, at least symbolically, to this day. Unlike the Renaissance Italians or the seventeenth-century English, the Swiss did not elaborate an indigenous theory of limited government, though their practices of cantonal government with local referenda have endured. The Swiss mountain warriors were uneducated farmers and woodsmen scrabbling out a living in alpine valleys and were unfamiliar with the classical Greek and Roman texts that inspired Italian and English antimonarchical theorizing. What distinguishes the Swiss in the forest cantons from farmers elsewhere—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 independence. There is little wonder that the great plains of Europe, which sometimes doubled as highways for marauding armies, were populated with seemingly weak-kneed farmers who chose instead to exchange their labor for military protection. Japanese mountain dwellers and Swiss alpine farmers took naturally to fighting for their freedom, not because they were braver than their lowland counterparts, but because their craggy fortresses gave them the possibility of resisting domination. Three other groups in Japan successfully resisted political incorporation for centuries. For seafaring pirates (wako), water provided the functional equivalent of mountain defense. Their ships navigated deft ly through the coastal waters, 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 records testify, pirates plagued coastal commerce around the Japanese archipelago from time immemorial. The political arm of Buddhism constituted a second group in medieval Japanese society that managed for centuries to repel the government’s encroaching territorial and jurisdictional authority. Buddhist temples, monasteries, and farming communities, often heavily armed but also often allied with members of the imperial family, avoided government taxation and regulation until Oda Nobunag, one of the unifiers of Japan, finally 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 necessary. 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. They 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-flung 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. This battle occurred some 15 years after 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. The 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. This early set of successful Hobbesian bargains at the local level would influence 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 affected than others, succumbed 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 political parties were testament to a latent yearning for self-governance. This is not to say that Japan’s freedom-fighting past was a continuing legacy that kept alive the potential for resistance. Resistance or acquiescence in Japan’s early history followed a pattern of opportunity or necessity. The Japanese 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 constraints than by long-standing mental frames. This book relates the tumultuous events of Japan’s medieval and early modern history—roughly 1185 to 1600—to theorizing about war and politics elsewhere. Japanese 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. The 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 battlefield. While Japanese 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 financial

4 Ferejohn and Rosenbluth

and labor resources for military protection. We do not intend to paint a picture of happy peasants bargaining and contracting for a better life. Rather, we seek to underscore the severe circumstances in which the Japanese, 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. Japanese and Swiss fighters offer 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 stiffer, and theorizing about limited government was more prolific in Italy and England, whose indeterminate topography left open a range of political outcomes. War, though miserable for those who fight and for those whose homes and fields are destroyed in the path of battle, can sometimes function as a political leveler. History provides some dramatic examples of political 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 affect how much and to whom rulers need to make concessions in exchange for resources, including whether the people supplying resources for war value the protection of a powerful ruler. If communities are confident of their ability to protect themselves, they will be willing to fight for others only in exchange for something of value such as political freedoms or, if they are already free, for pay. Japanese mountain villagers needed relatively little protection from overlords 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 revenues in exchange for protection. Their fear of violence and instability was greater even than their desire for freedom from domination. Their willingness to supply resources for large armies lies at the root of Japan’s political unification in the sixteenth century. The same logic accounts for the rise of Europe’s large territorial monarchies.

War and State Building in Medieval Japan 5

The Rise and Fall of Decentralized Military Rule in Medieval Japan The debates among social and economic historians over the repressive nature of Japanese 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’ leverage 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. The 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 economic 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 fisher folk from Polynesia who enjoyed land abundance and relatively egalitarian social structures. Then, in about 300 b.c., waves of immigrants from Korea invaded Japan and pushed the earlier inhabitants into the mountains and outlying islands. The new ruling elite organized into political units (uji) that jostled among themselves 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 contested, was usually managed peacefully through negotiations among a coalition of leading clans. Unlike many powerful monarchical dynasties in China or Europe, imperial succession rules were loose, allowing for a large number of potential heirs. A significant part of the ruling class derived benefit from the imperial institution, giving it the structure of an oligarchy rather than an autocracy. The scions of some court families emigrated to the provinces beginning in the late ninth century. They did well for themselves by exploiting their connections to powerful court figures and institutions, and by obtaining sinecures as government officials or managers of private estates. The court, in turn, cultivated 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 internal and external threats.

6 Ferejohn and Rosenbluth

Access to abundant frontier land, followed by the scramble to clear new arable land out of forests and swamps, afforded a modicum of bargaining leverage 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 efforts to claim new land for themselves. Farmers often 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 neighbors. In the centuries that followed until the sixteenth century, the imperial court became overshadowed by military order provided by one group of warriors 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. The 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 often switched sides to join the victors rather than have their land confiscated 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-fifteenth 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 farmers and warriors (heino bunri) that had already begun to emerge under domainal 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 fight in battles. Although taxes increased, so did agricultural productivity and economic growth. Making good on the promise to protect farmers gave these leaders an enormous advantage over their opponents. Leaving farmers on the farm, Oda Nobunaga and Toyotomi Hideyoshi created disciplined and skilled armies. Histories of modern warfare herald Maurice of Nassau (1567–1625) and Gustav Adolf of Sweden (1594–1632) for building regimented and skilled armies, but Oda

War and State Building in Medieval Japan 7

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 decades, Hideyoshi and Nobunaga reversed the centrifugal movement toward smaller political units and created significant 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 coalition big enough to finish the job, Nobunaga and Hideyoshi had broken the back of resistance to central military control. Warrior-farmers in mountain hideaways and armed monks in monasteries 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 afforded came at a dear price: their autonomy. The 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. The organizational and numerical superiority of the conquering army was made possible by the taxes of millions of war-weary farmers.

War and State Building in Japan and Europe 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 Europe. Farmers in the surrounding countryside provided labor, crops, or both in exchange for protection from invaders provided by the nobility’s cavalry. Common to both Japan and Europe was the rather small territorial size of feudal domains after the breakup of empires, given the difficulty of protecting large tracts of open terrain with small bands of warriors on horseback. 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 affected by factors such as modes of economic production and military technology. But economics and technology, whether alone or together, leave unexplained substantial parts of the variation in scale economies. Economic theory might suggest, for example, that Eastern Europe was dominated by larger fiefs and more persistent

8 Ferejohn and Rosenbluth

serfdom because economies of scale in grain production there are greater than for the crops cultivated in the hillier terrain of Western Europe. The fi xed costs of maintaining teams of oxen and other farm equipment would be too great for small landholders, making large manors more economically productive. There is no obvious reason, however, why farmers might not have worked out some cooperative arrangement to share expensive livestock and tools. Another influential economic model of serfdom (and slavery) turns standard bargaining logic on its head. Precisely because abundant land-to-labor ratios favored peasants in Eastern Europe compared to Western Europe, political 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 effort and productivity from peasants. But this model does not explain the origins of coercive power. Military technology is another possible explanation for changes in economies to territorial size. The introduction of stirrups to Europe 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 afford 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 midfifteenth century. With the development of cannon in 1449–1450, Charles VII of France regained much of Normandy by knocking down 60 fortifications— each of which took the English a year to build—at a clip of more than one a week. The 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 shifted the relative productivity between capital- and labor-intensive modes of warfare, with potential consequences 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 military value of foot soldiers before the stirrup and after the cannon. But the exposure of farmland to plundering armies meant that farmers were always exploited, even when they were mobilizationally useful. They were too vulnerable to make use of their relative scarcity as bargaining leverage. Early modern Poland is instructive here because the inability of Polish peasants to assert their natural bargaining advantage conferred by the abundance of labor relative to land has always been something of a mystery to economists. The dominance of the Polish landed nobility makes more sense when one re-

War and State Building in Medieval Japan 9

members the serial invasions by Magyar and other horseback invaders, against whom the Polish cavalry was quite successful. The Polish cavalry’s battle record was so glorious that Poland neglected military innovation and was destroyed by Russian and Prussian 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 benefit of exit options among multiple political units. This was particularly true in Eastern Europe 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. The remaining puzzle is identifying the source of this money. The medieval Japanese experience, like the European one, revolved around an implicit peasant demand for protection. Lords competed for the loyalty of lesser lords, but ultimately the whole edifice 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 confident of their ability to defend themselves did peasants resist taxes and territorial consolidation. Territorial consolidation began in those flat areas that were most vulnerable to military invasion and spread as the armies of those areas gained preeminence. 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 acceptance in the wake of harrowing religious violence in the sixteenth century. Elsewhere in Europe, taxes were higher and armies were larger in the great flatlands of Prussia and Russia, in the pathway of steppeland hordes. Big armies can often conduct an effective 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 nationstate with territorial control, first in the form of the absolutist state. This recognition of the importance of peasant demand for protection differs from the Marxist suggestion that the landed nobility needed an absolutist state to fix their status against the onslaught of the proto-industrial bourgeoisie. Nor were undefended “church lands” available for confiscation in Japan because monasteries tended to be armed to the hilt. In Japan, the unifiers robbed their competitors of their lands militarily, whereupon they divided the

10 Ferejohn and Rosenbluth

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 Japan’s medieval period that by the mid-fifteenth 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 after Octavius’s victories. The “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 Europe, Russian and Prussian “absolutism” 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. This is not to deny that peasants were often miserable, 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. Their choices had significant consequences for the kinds of states that would subsequently emerge.

Conclusions Widespread territorial vulnerability and fear of violence created the territorial state. This book tells that story as it unfolded in medieval and early modern Japan, but it follows a general logic that holds in Europe and elsewhere. There is some irony to the way vulnerability paved the way for strong, hierarchical governments with extensive coercive powers over the subject population. 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 profitably refuse to fight unless the leader is willing to offer better terms of exchange. History gives us a number of examples of political concessions to peasant fighters, including fift h century b.c. Athens when Cleisthenes granted the masses full participatory rights in exchange for their help in ousting the Spartan-installed oligarchy. In Republican Rome, fighting wars for Rome was the ticket to citizenship, first 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 after the war was won. In modern times, World War I ushered in female suffrage 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 political rights in one setting and an abdication to absolutist government in another?

War and State Building in Medieval Japan 11

Japanese 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 bidding 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 afford 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 Japanese people might also have won political concessions in exchange for emergency mobilization. The piecemeal intensification of violence in Japan worked against broader mobilizational concessions. Once widespread destruction reached an intolerable 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 materials to provide a close look at the choices made by lords and peasants throughout the medieval period and across different 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 warriors 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 Kamakura 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 centralized 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, after the decline of the imperial authority and before unification under Oda Nobunaga. Competition 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. The civil war that broke out in 1437 among lords competing for territory forced local villagers to

12 Ferejohn and Rosenbluth

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 wartime and explains the relief with which peasants greeted Oda Nobunaga’s policy of dividing labor between farmers and warriors. Rather than being distressed to have their swords confiscated, many farmers greeted with relief the new leader’s ambition to curb the fighting clans. Throughout 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. The Honganji Temple’s thousands of followers in villages across a sizable area in today’s Ishikawa prefecture created a structure of local self-rule and political order to match that provided by local military leaders. The villagers who joined in this religious league had little need for the protection offered by secular lords because they were well armed and well disciplined. Their belief that fidelity to their cause would be awarded in the afterlife empowered them to fight 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. Lacking the economies of scale that the centralizers had developed, the most passionately independent local resistance 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 different consequences for Japanese political 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, Thomas Conlan addresses the question of how the more successful military lords managed to consolidate territory. Conlan documents the organizational 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-

War and State Building in Medieval Japan 13

dation, Conlon credits these generals’ methodical approach to gaining public acceptance, raising revenues, and building armies, rather than the guns with which they equipped their men. In Japan, as in modern Europe, 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. The variation within Japan over time and place suggests a mechanism for its emergence that was at work in Europe as well. Farmers in the pathway of armies became desperate for protection, even at the cost of their money and freedom. Although less vulnerable 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 resistance of these natural fortresses. In Japan, the entire archipelago became as one.

Notes 1. “Ninja” is a term of modern origin used to describe rustic fighters in ages past shrouded in myth and mystery. Following forcible political unification in 1600, these resistance fighters 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 century, 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 escaped and then did shoot the magistrate after 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 Renaissance theorizing on liberty during a time when liberty—from both foreign and domestic domination—was under threat (Wootton 2007). The English carried on the tradition in their resistance to monarchs who overstepped traditional boundaries. An authoritative source is Pocock 1975. The famous Swiss champion of self-governance, Jean Jacques Rousseau (1712–1778), wrote much later, having the full advantage of earlier Italian and English theorizing. 4. Petrarch (1304–1374) is said to have revived the study of Roman thought, making him one of the fathers of the Italian Renaissance. 5. North and Thomas 1971. 6. Murai 1993. We are grateful to an anonymous reviewer for this parallel. 7. Adolphson 2001; Adolphson and Ramseyer 2009.

14 Ferejohn and Rosenbluth

8. Adolphson 2001; Adolphson and Ramseyer 2009. 9. We are indebted to an anonymous reviewer for this point. 10. In The Leviathan (1660), written after a brutal civil war in England, Thomas 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 effect 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 observed 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 fifteenth centuries the French experienced the ravages of English armies, which were ordered to plunder the countryside and show no mercy to enemy and noncombatant alike. This brutal tactic, known as chevauche, had the dual advantage to the English of relieving the costs of feeding their own troops and depriving the French of supplying theirs (Sumption 2001: 272). The sixteenth century witnessed the vicious religious wars, in which Huguenots and Catholics fought without mercy for doctrinal control. In the seventeenth century, the Thirty Years’ War brought plundering armies into villages over large swaths of Germany and France. 15. Montesquieu, in eighteenth-century France, mused that England’s relative isolation may have given it some advantage in developing a theory and practice of limited government (The 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 misleading (see Henshall 1992; Hoff man and Rosenthal 1997). 17. Japa nese economic history in the 1920s and 1930s centered on a debate between 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. After 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-

War and State Building in Medieval Japan 15

nomic surplus from unfree labor. Tom Scott (1998) documents peasant choices and resourcefulness in the European context. See also Brown 1974. 18. Genetic tests show that modern Japanese inhabitants of the northernmost island of Hokkaido and the southernmost island of Okinawa share more DNA with each other than with Japanese living on the central islands of Kyushu, Shikoku, and Honshu (see Imamura 1996). 19. Kiley 1999. 20. Piggott 1997. 21. The Japanese “estate” was “a group of plots, often scattered, that were bound together under a common proprietor. The proprietor, who might be the head of a powerful local family, a member of the aristocracy or the imperial family, or a religious institution, 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 Tsunetomo, the author of the samurai classic, Hagakure, invented a code of abject loyalty during 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 fields. 26. Roberts 1956; Parker 1988; Lynn 2003; Rogers 1995. 27. Duus 1993: 76–77. 28. Conlan, this volume, establishes the insufficiency of the technology argument. 29. Hideyoshi did finish 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 first to accomplish that. Berry (1986: 242) points out that the unifiers did not establish a strong, interventionist state beyond what was necessary to carry out their mandate of imposing peace and security. 30. The 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. The real lesson of Nagashino is that a large force ensconced behind field fortifications can defeat a small force attacking it. It was an advantage of defense over attack, given those technologies and strategic ideas. That 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 Russian nobles competed for scarce labor in the sixteenth century, the government restricted the peasants’ freedom. When labor is scarce, it has higher marginal value than

16 Ferejohn and Rosenbluth

land, and landlords have an incentive to control labor in order to expropriate the full value of labor’s marginal product. This has become the standard explanation for establishing 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. The stirrup and cannon are meant here as shorthand for a host of reasons why mass armies may or may not be effective. 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 fighters who know how to use terrain in their favor. 35. Frost 2000. 36. Eastern European 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. This was thanks to the natural fortification 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 thoroughfares. See Henneman 1976. 39. Anderson 1974. 40. As in Europe, “there was a persistent moral threat insisting that the lord should grant fiefs 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 treasury, 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 confiscated property. The 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.

War and State Building in Medieval Japan 17

References Acemoglu, Daron, and James Robinson. 2006. The 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 Premodern Japan. Honolulu: Hawaii University Press. Adolphson, Mikael, and J. Mark Ramseyer. 2009. “The Competitive Enforcement of Property Rights in Medieval Japan: The Role of Temples and Monasteries.” Journal of Economic Behavior and Organization, 71, 3: 660–668. Anderson, Perry. 1974. Lineages of the Absolutist State. Atlantic Highlands, NJ: Humanities Press. Bartlett, Robert. 1993. The Making of Europe: Conquest, Colonization, and Cultural Change, 950–1350. Princeton, NJ: Princeton University Press. Bean, Richard. 1973. “War and the Birth of the Nation State.” Journal of Economic History, 33, 1: 203–221. Berry, Mary Elizabeth. 1982. Hideyoshi. Cambridge, MA: Harvard University Press. Berry, Mary Elizabeth. 1986. “Public Peace and Private Attachment: The Goals and Conduct of Power in Early Modern Japan.” Journal of Japanese Studies, 12, 2: 237–271. Berry, Mary Elizabeth. 1994. The Culture of Civil War in Kyoto. Berkeley: University of California Press. Berry, Mary Elizabeth. 2005. “Presidential Address—Samurai Trouble: Thoughts 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 Japanese Studies, 11, 2: 369–400. Blum, Jerome. 1957. “The Rise of Serfdom in Eastern Europe.” 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. The Perspective of the World: Civilization and Capitalism. New York: Harper & Row. Brown, Delmer M. 1993. “The Yamato Kingdom.” In Ancient Japan, volume 1 of The Cambridge History of Japan, edited by Delmer M. Brown. New York: Cambridge University Press, pp. 108–162. Brown, Elizabeth R. 1974. “The Tyranny of a Construct: Feudalism and Historians of Medieval Europe.” 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. The Beginnings of Rome: Italy and Rome from the Bronze Age to the Punic Wars. London: Routledge.

18 Ferejohn and Rosenbluth

Dewald, Jonathan, and Liana Vardi. 1998. “The Peasantries of France, 1400–1789.” In Tom Scott, ed., The Peasantries of Europe. New York: Longman Press. Domar, Evsey. 1970. “The 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: The 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 Japanese 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. The Northern Wars: War, State and Society in Northeastern Europe, 1558–1721. London: Longmans Harlow. Henneman, John Bell. 1976. Royal Taxation in Fourteenth Century France: The Captivity and Ransom of John II, 1356–1370. Philadelphia: American Philosophical Society. Henshall, Nicholas. 1992. The Myth of Absolutism: Change and Continuity in Early Modern European 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, Thomas. Leviathan. 1651/1968. London: Penguin Classics. Hoff man, Philip, and Jean-Laurent Rosenthal. 1997. “The Political Economy of Warfare and Taxation in Early Modern Europe.” In John Drobak and John Nye, eds., The Frontiers of New Institutional Economics. New York: Academic Press. Hurst, G. Cameron, III. 1990. “Death, Honor, and Loyalty: The Bushido Ideal.” Philosophy East and West, 40, 4: 511–527. Imamura, Keiji. 1996. Prehistoric Japan: New Perspectives on Insular East Asia. Honolulu: University of Hawaii Press. Israel, Jonathan. 1995. The 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 Historian, 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. The Right to Vote: The 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 The Cambridge History of Japan, edited by Donald H.

War and State Building in Medieval Japan 19

Shively and William H. McCullough. New York: Cambridge University Press, 1999: 236–340. Kliuchevsky, Vladimir. 1937. Kurs russkoi istorii. Moscow: Gosudarstvennoe sotsail’noekonomicheskoe izdatel’stvo. Translated by C. J. Hogarth as A History of Russia (New York: Russell and Russell, 1960). Levi, Margaret. 1997. Consent, Dissent, and Patriotism. New York: Cambridge University Press. Lintott, Andrew. 1999. The Constitution of the Roman Republic. New York: Oxford University Press. Locke, John. 1690/2003. Two Treatises of Government and A Letter Concerning Toleration, 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: Stanford University Press. Machiavelli, Niccolo. 1970/1517. The Discourses. London: Penguin Classics. Martins, Roberto Borges. 1982. “Growing in Silence: The Slave Economy of Nineteenth Century Minas Gerais, Brazil.” Journal of Economic History, 42, 1: 222–223. Millward, Robert. 1982. “An Economic Analysis of the Orga nization of Serfdom in Eastern Europe.” 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 Thomas. 1971. “The Rise and Fall of the Manorial System: A Theoretical Model.” Journal of Economic History, 31, 4: 777–803. Parker, Geoff rey. 1988. The Military Revolution: Military Innovation and the Rise of the West, 1500–1800. New York: Cambridge University Press. Piggott, Joan. 1997. The Emergence of Japanese Kingship. Stanford, CA: Stanford University Press. Pocock, J. G. A. 1975. The Machiavellian Moment: Florentine Political Thought and the Atlantic Republican Tradition. Princeton, NJ: Princeton University Press. Roberts, Michael. 1956. The Military Revolution. Belfast: Queens University Press. Rogers, Clifford, ed. 1995. The Military Revolution Debate: Readings on the Military Transformation of Early Modern Europe. Boulder, CO: Westview Press. Rogowski, Ronald, and Duncan MacRae. 2004. “Inequality and Institutions: What Theory, History, and (Some) Data Tell Us.” Manuscript, UCLA Political Science Department. Samons, Loren J., ed. 1998. Athenian Democracy and Imperialism. New York: Houghton Mifflin. Scott, James. 2009. The Art of Not Being Governed: An Anarchist History of Upland Southeast Asia. New Haven, CT: Yale University Press.

20 Ferejohn and Rosenbluth

Scott, Tom, ed. 1998. The Peasantries of Europe from the Fourteenth to the Eighteenth Centuries. New York: Longman Press. Strauss, Barry. 2004. The Battle of Salamis: The Naval Encounter That Saved Greece— And Western Civilization. New York: Simon and Schuster. Sumption, Jonathan. 2001. The Hundred Years’ War: Trial by Fire. Philadelphia: University 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. “The True Origins of Republicanism or de vera respublic.” In Manuala Albertone, ed., Il repubblicanesimo moderno: L’idea di repubblica nella reflessione storica di Franco Venturi. Naples: Bibliopolis. Yamamura, Kozo. 1974. “The Decline of the Ritsuryo System: Hypotheses on Economic and Institutional Change.” Journal of Japanese Studies, 1, 1: 3–37.

2

They Were Soldiers Once The Early Samurai and the Imperial Court Karl Friday

MEDIEVAL JAPAN WAS TRULY, as Souyri recently styled it, a “world turned upside down.” The prevailing forces of the preceding epoch, the Nara (710–794) and Heian (794–1185) periods, had been centripetal, and the realm was governed politically, economically, 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 decades historians have radically and dramatically revised their thinking on how and when the one epoch gave way to the other. It was once common to cite the creation of the first 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 twelfth, thirteenth, and early fourteenth centuries—as a time of transition between the classical and medieval ages. The 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. “The rise of the samurai” was less a matter of dramatic revolution than one of incremental evolution, occurring in fits 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 delegation of administrative responsibility that distinguished the Heian polity from its Nara predecessor. Its roots sprang from shifts in imperial court military policy that began in the middle decades of the eighth century and picked up momentum in the ninth. 21

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 together by a confederation of regional chieftains, among which one—the royal, or Yamato, house—stood as first 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 autonomous. In theory, these provincial hegemons drew their titles from the court, but in practice, their positions were permanent, hereditary, and only nominally 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 vehicle for cooperation among the great houses in matters of “national” concern. All of this changed rapidly and dramatically during the seventh century, as this polity gave way to a centralized imperial regime. The changeover accelerated after the sixth month of 645, when radicals led by the future Emperor Tenji seized power by hacking their political 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 measures collectively known as the Taika Reforms, after the calendar era in which the first were launched. Over the next several decades, the great regional powers were stripped of their independent bases and converted to true officials of the state, while the Yamato sovereigns were restyled in the image of Chinese emperors, as transcendent repositories of all political authority. The result was what historians have come to refer to as the imperial, or ritsuryō (after the administrative and penal codes that formed its framework), state. The reformers prevailed through cajolery, cooptation, and coercion, aided in no small measure 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 fleets looming over the horizon muted opposition to losses of local or hereditary privilege and promoted support for state-strengthening reforms, as central and provincial noble houses set aside their differences in the face of a perceived common enemy. For it was obvious to all concerned that the Yamato military organization 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 process.

They Were Soldiers Once 23

“National armies” of the Yamato state had been cobbled together from forces raised independently by provincial chieftains, who then led them into battle under the banner of the Yamato sovereign. Recruitment, training, organization, and mobilization varied from place to place and from conflict to conflict. So did command, which was eclectic, and often 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 officers—had been subsumed under the direct control of the newly emergent emperor and his court. To be sure, the ritsuryō military system is distinguished from its predecessor more in terms of principle and formal structure than function or personnel. But there was a crucial difference, for henceforth centrally appointed officers and officials oversaw all military units and activities, and direct conscription by the imperial court replaced enlistment of troops through regional chieftains. Under the new system, all free male subjects between the ages of 20 and 59, other than rank-holding nobles and individuals who “suffered from long-term illness or were otherwise unfit 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 fields. Provincial governors maintained copies of regimental rosters, which they used as master lists from which to select troops for training; for peacetime police, guard, and frontier garrison duties; and for service in wartime armies.

Draftees and Horsemen The model for the ritsuryō military system had been Tang China. Contrary to the images that still dominate many popular histories, however, the new institutions—like the rest of the imperial state structure—were not simply adopted wholesale. The architects of the imperial state carefully adapted Chinese practices to meet Japanese needs and circumstances. At the same time, the system they designed was all too often the product of conflicting priorities, and accordingly, incorporated more than a few rather unhappy compromises. Moreover, the original foibles of the system were exacerbated by changing conditions: by the mid-eighth century the needs and priorities of the Japanese state differed considerably from those of the late seventh. One of the difficulties 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 efforts 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 Japanese imperial armies were mixed weapons-system forces: predominantly infantry, but augmented by heavily armored archers on horseback. This infantryheavy balance was the product of both design and necessity. The ritsuryō military had been contrived in the face of two perceived threats: a Chinese invasion and regional insurrections led by the old provincial chieftains. The architects of the system seized on large-scale, direct mobilization of the peasantry as a key part of the answer to both dangers. The apparatus they created enabled the court to corner the market on military manpower and to create loyalist armies of daunting volume, thereby effectively 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 Japan to muster large-scale fighting 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 technology of the age, constructing a force composed primarily of infantry, while the premier military technology of the day was mounted archery. The state did try to maintain as large a cavalry force as it could, but efforts to that end ran afoul of major logistical difficulties. Foremost among these was the simple truth that fighting from horseback, particularly with bows and arrows, demands complex skills that require years of training and practice to master. It was just not practical to attempt to develop first-rate cavalrymen from shortterm peasant conscripts. The court addressed this problem through the straightforward expedient of staffing its cavalry units only with men who had acquired basic competence at mounted archery on their own, prior to induction. This 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. Thus only a small portion of

They Were Soldiers Once 25

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 decades of the eighth century, however, the political climate—domestic and foreign—had changed enough to render the provincial regiments anachronistic and superfluous in most of the country. The Chinese invasion the Japanese had so feared simply never materialized. Whatever 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. The likelihood of violent challenges to the central polity from the regional nobility had also dwindled rapidly, as former provincial chieftains came to accept the imperial state structure as the arena in which they would compete for power and influence. The passing of these crises all but ended the need to field 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 martial needs of the interior provinces—the vast majority of the country—quickly pared down to the capture of criminals and similar policing functions. Unwieldy 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 officers and provincial officials to misuse the conscripts who manned them; they were borrowed, for example, for free labor on their personal homes and properties. The court responded to these challenges with a series of adjustments and general reforms. The pattern of the reforms, which began as early as the 730s, indicates that the government had concluded that it was more efficient 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 ordinary peasantry played smaller and smaller roles in state military planning, while the role of elites expanded steadily. The provincial regiments were first 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 The institutions and procedures of the early imperial state had their roots in an intriguing blend of idealism, wishful thinking, pragmatism, and compromise. Few survived wholly intact for more than a few decades, although the spirit of the system—particularly the principle of centralized authority culminating in the emperor—remained strong into the fourteenth century and beyond. The ritsuryō provisions for governing the countryside had been calculated, above all, to establish centralized control and to produce revenue for the court. They 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 collectively as lord. This 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 provinces 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. Nevertheless, 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 stronger, not thinner, during the Heian period. This odd mixture of centrifugal and centripetal development transpired because the provincial gentry and the court nobility embraced fundamentally different notions of what constituted wealth and power. The Kyoto aristocracy viewed wealth in terms of rice and other products collected, and power as a function of administrative oversight. The provincials, on the other hand, saw both in terms of hands-on control over productive resources—lands and people. As long as they were allowed a share of the profits from the land, the gentry

They Were Soldiers Once 27

were content to acquiesce to a central authority and to focus not on ultimate ownership but on actual control—management of the agricultural process, collection of taxes and rents, and other domainal sorts of authority. One shortcoming of the ritsuryō landholding and tax structure was that it effectively presumed a countryside populated by a largely undifferentiated mass of peasant households. The peasants were to be assigned—or leased— fields to cultivate, from which they would draw their livelihoods and in exchange for which they would pay taxes and rents to sustain the emperor, the court, and local officials. But the theory behind this system failed to account for differences in individual ability: Some peasants proved to be better—or simply luckier—farmers than others. Some were able to make a profit year after year; others found themselves incapable of making ends meet, let alone meeting their tax obligations. Thus some peasants grew wealthy, while others were forced to become dependents of neighbors, who increasingly organized the populations around their homes under their own control. A gap was developing between the structure 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 difficult. In a remarkably pragmatic effort 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. The 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 whatever revenues were deemed appropriate from the specific locales in which they had influence. Governors and local managers alike welcomed such policy measures as opportunities for increasing their personal wealth and power. In a matter of decades, the new tax structure turned everyone involved—except peasant cultivators—into tax farmers collecting revenues beyond their assigned quotas and pocketing the surplus. Even as the state was redefining 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 influence outside the official government structure. Such efforts became a persistent source of trouble for provincial residents and officials from the middle decades of the ninth century onward. Council of State edicts from this period railed against “temples, shrines, princes and officials behaving like peasants, contesting over peasant paddy fields, stealing moveable property, ignoring governors and district magistrates, and using their prestige and influence” to intimidate provincial residents. “Agents of temples, shrines, government officials, and great houses,” we are told, were “confiscating boats, carts, horses and men by force,” “robbing tax shipments,” and “causing much suffering for the people.” Provincial governors, for their part, demonstrated a less-than-noble inclination to regard their provinces chiefly as revenue-producing resources, and became increasingly indifferent 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 demanding the impeachment of rapacious governors. The most famous of these, filed in 988 by the “district officials 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, often beginning his collection efforts 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 throughout the province and operate it as private, tax-exempt holdings. They also “commandeered” horses and cattle from provincial residents, and then sold them back several days later for three to five times their fair price. Provincial elites, too, were flexing their political and economic muscles. One of the defining characteristics of the imperial state was the monopolization 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 political and social back benches. In practice, however, the old regional chieftains never fully surrendered their wealth or their familial authority in the agricultural 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 resistance, when it did occur, was generally small scale, but it

They Were Soldiers Once 29

did sometimes involve violence. From the turn of the ninth century onward, the court issued a steady stream of complaints about armed marauders “burning people’s homes and using weapons to rob,” “striding about the villages opposing government officials 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 influence among multiple groups, including provincial-resident elites, provincial officials, and the “temples shrines, princes and officials” of the court. At the axis of this competition were the middle-ranked court nobles whose careers centered on appointments to provincial government offices. Such career provincial officials (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 offices, 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 after their terms office expired. Heian society was rigidly stratified, characterized by functionally unbridgeable 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, classified both the central and the provincial nobility. A man’s rank determined his status, his eligibility for government posts, and his place in the complex protocols for official and social events. Each stratum in this hierarchy had access to specific 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 effectively barred from provincial or local posts as provincial elites were from becoming top court officials. This arrangement generated intense competition and rivalry among peers, including siblings, who contested in circumscribed and ever-morecrowded arenas for the same baskets of fruit. But it also formed a basis for cooperation between members of different strata because neither party could challenge 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 office-holding became progressively deeper and more firmly entrenched, and eligibility for

30 Karl Friday

posts became limited to smaller and smaller numbers of houses. As the prospect that descendants of par ticu lar families would hold the same posts generation after generation became increasingly predictable, many offices—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 responsibilities more difficult 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 regents or other senior courtiers, who would become their patrons and sponsor their advancement at court. In exchange, the clients served on the patron’s household staff and vouchsafed his interests in the course of their official duties in the posts he obtained for them. This 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 provincial governors in the administration of their private affairs, and it assured lowerranking nobles of continued appointment to lucrative posts. Alliances were also developing between senior court figures and provincial elites, as the elites learned to use their connections—real or pretended—with the court figures to gain greater autonomy from the provincial government. By the late 800s, court edicts complained that the great houses and religious institutions 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 provinces” disrupting peasant households and government business. The edicts also indicated that “natives and drifters claiming to be housemen of princes and government officials did not fear the authority of the governor and did not obey the injunctions of district magistrates.” The “natives and drifters” cited in these documents included a new group of provincial residents. For members of the career provincial governor class, the heightened competition, factionalism, and atmosphere of hereditary prerogative that characterized 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 finding that they could use the power and perquisites of their

They Were Soldiers Once 31

offices, 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 after their terms of office expired. As former officials 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 influence. Among other things, the newcomers seem to have been continuing to collect taxes—especially back taxes—just as if they were still officials. 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. The 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 “officials who have finished their terms of office, and sons and younger brothers of princes and court officials,” who were, “settling down in their [former] areas of jurisdiction, where they hindered agriculture, gathered up the peasantry like fish, and constructed plans for their own evil gains.” By the mid-tenth century, however, the court had given up this effort and was instead coming to terms with the immigrants. Former provincial officials and their children began to gain appointments to assistant governorships and other provincial offices. Nevertheless, few of the nobles who “settled down” in the countryside actually abandoned life in the capital for a provincial existence. Some individuals 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. They could not afford to do otherwise, for to cut themselves off completely from the court would have meant severing themselves from the source of official appointments and from critical personal connections. This would thereby have ended all hope of maintaining the family’s social and political position— even in provincial society. Typical exurban provincial officials 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 political affairs in the capital and often maintained homes there, to which they shipped most of the profits 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 political and official world of the capital. The 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 officials were taking wives and sons-in-law from provincial elite houses with considerable frequency.

The New Warrior Order The expansive social and political changes taking shape during the Heian period spawned intensifying competition for wealth and influence 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 government 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, increasingly turned to military ser vice as a path to the entourages of powerful aristocrats and lucrative government posts. Compelled by a need to defend themselves and their prerogatives against outlawry and armed resistance, as well as by the desire to maximize the profits that could be squeezed from taxpayers, many provincial governors began to include “warriors of ability” among the personal staffs that accompanied them to their provinces of appointment. Some also took up arms for themselves and established reputations as military troubleshooters. By the tenth century, military ser vice at court and ser vice as a provincial official had become parallel and mutually supportive careers for many zuryō, resulting in the emergence of the group Japanese scholars have dubbed the miyako no musha, or “warriors of the capital.” These were men of the fourth or fift h court rank, who curried the patronage of the higher nobility and recognition 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 efforts to prune the imperial family tree and dispose of extraneous princes and princesses. There were seventeen major lines of Minamoto and four of Taira, all descended from different emperors. Branches of two of these—the Seiwa Genji deriving from Emperor

They Were Soldiers Once 33

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 Minamoto surname. The 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 fathered nine sons, at least three of whom became warriors of some repute. The most important of these was Mitsunaka. It is to Mitsunaka and the progeny of his three eldest sons, Yorimitsu, Yorichika, and Yorinobu, that historians generally refer when they speak of the Seiwa Genji. The Kammu Heishi comprised four main lineages, each descended from a different prince. The line that produced warriors began with Kammu’s eldest son, Katsurahara, through his son Takami and grandson Takamochi. According to tradition, Takamochi, who was presented with the Taira surname and appointed assistant governor of Kazusa in 889, acquired both the appointment 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 twelfth century, warriors of Minamoto descent 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 considerably 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 successive retired emperors (in) similar to the one that existed between the Minamoto and the Fujiwara Regents’ House. By the second decade of the twelft h century, the strength of this new alliance brought the Taira to full parity of prestige with the Minamoto at court. This 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 first, owing to marriage or other ties. Marriages among the court aristocracy were polygamous or serially monogamous, 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 allowed. 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 significantly higher station than the groom’s, the children—and sometimes the new husband—often 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 The history of Japanese military organization is a tale dominated by dialectics between personal and institutional authority, and between localized and centralized sanctions, jurisdictions, and structures of command. By mid-Heian times, warfare and law enforcement had become the preserve of a warrior order that armed, trained, and organized itself. But armies were raised and retained 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 experimented its way toward a system that centered on commissioning professional warriors with new military titles legitimating their use of private martial resources on behalf of the state—a principle that would characterize military affairs in Japan down to the modern era. The evolution of military posts during the Heian period and beyond reflects the emergence of the samurai across the same span of time. The two processes were, in fact, reciprocal: Deputizing provincial elites and members of the middle and lower central aristocracy inevitably had a catalytic effect on the development of private martial resources under these leaders, which in turn led to the introduction of new assignments and the modification of existing ones. The 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

They Were Soldiers Once 35

ever-changing sociopolitical reality it served, the cardinal features persisted well into the medieval era. Standing out among these features was the bifurcation of organizational 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 organization 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. Appointments and compensation alike came from the center, following principles and procedures that closely paralleled those specified for comparable posts under the ritsuryō codes. These essential similarities enabled the court to retain exclusive authority over—and at least general control of—military affairs throughout the Heian period. But the new offices were also fundamentally different from their ritsuryō antecedents on one critical point: They were premised not on the existence of a publicly conscripted pool of manpower over which the officer’s commission gave charge, but on the appointee’s ability to recruit troops for himself. Curiously, the court established no statutory guidelines for drafting or otherwise raising troops immediately after its dismantling of the provincial regiments in 792. For most of the ninth century, troop mobilizations remained grounded in public authority but were conducted through ad hoc means. Responsibility for mustering fighting men as the need arose rested with provincial governors. The specific manner in which this mobilization was to be accomplished varied from case to case, but it generally involved drafting the necessary manpower on the basis of the general corvée obligations required of all imperial subjects. This 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. The phenomenon 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 decade of the tenth century, private military networks of substantial scale had begun to appear, centered on major provincial 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 midtenth century, the government had begun to co-opt private arrangements between warriors, transferring much of the responsibility for mustering and organizing the forces necessary for carry ing out military assignments to warrior leaders. These 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 interactive but diverse entities. During the Heian period, private martial entourages were assembled by retired emperors and top courtiers, by monastic institutions, by career provincial officials, 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 varied 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 patchwork assemblages of several types of forces. Leading warriors in both the provinces and the capital maintained relatively small, core bands of fighting 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 variety of sources. Some troops were simply hired mercenaries; others were sons or close relatives of the organization’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 difficult to ascertain, for few reliable sources record the numbers of followers under a given warrior’s direct command. Those that do, moreover, indicate substantial variation from one

They Were Soldiers Once 37

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, fishermen, and other residents of the lands in and around the estates and districts they administered. Such men were, strictly speaking, not under the warrior’s control, but they often leased land from him, borrowed tools and seed from him, and conducted trade at his compound, making his residence an important economic center for them. By exploiting whatever political and economic 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 different sociopolitical status. This technique made it possible for warriors to assemble forces many times the size of their core organizations. The 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 first place. The 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, however, for the corollary to the factors and principles that encouraged vertical cooperation between members of different sociopolitical strata was that the interests of men of similar station were generally at odds with one another. This tended to exercise a divisive influence, rendering lateral alliances unstable and therefore difficult to maintain for long periods, with the result that, for most of Japanese 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 organizational pattern for warrior cooperation was hierarchical, centering on figures whose prestige enabled them to serve as rallying points for alliances between warriors of lesser pedigree. 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 reflected 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 process by which estates (shōen) were formed—alliances between warriors were not supported by legal contracts. The exchange of obligations that accompanied warrior partnerships during Heian times was far less palpable, and the nature, extent, and duration of these obligations much less precise, than in medieval times. Formal arrangements under which specified benefices were offered in return 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 political circumstances. Even the most powerful warriors of the age occupied only intermediate positions in the sociopolitical hierarchy and were dependent on connections with the higher echelons of the court to maintain their political and economic positions. Their 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, offering 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 often brought perquisites over lands and peoples, and sometimes involved the transfer of properties hitherto administered by warriors on the losing side of a conflict, Heian samurai were rarely, if ever, able to specify the size or the particulars of rewards for themselves. Any transfers of lands were accomplished 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 shortlived. The larger the organization, the more ephemeral it tended to be. On occasion, illustrious warriors like Minamoto Yoshiie or Yoshitomo were able to construct martial networks that extended across multiple provinces, but until the 1180s, no such organization survived the death of its founder. 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

They Were Soldiers Once 39

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 The events that led to the birth of Japan’s first 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 decades, Kiyomori’s prestige and influence 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) ascended the throne as Emperor Antoku. That same year, however, a frustrated claimant to the throne, Prince Mochihito, provided Yoritomo with an opportunity 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 misadventure two decades before had cost him the career as a government official 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 guaranteeing whatever lands and administrative rights an enlisting vassal considered to be rightfully his own. In essence, Yoritomo was proclaiming the existence of an independent state in the east, a polity run by warriors for warriors. But he took pains to portray himself as a righteous 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 justified 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 conservative revolutionary. Rather than maintain his independent 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 courtcentered national polity for formal court recognition of many of the powers he had seized. The regime that this agreement established—the Kamakura shogunate— was a kind of government within a government that was both part of and distinct 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. After the Jōkyū War of 1221, an ill-fated attempt by a retired emperor, Go-Toba, to eliminate the shogunate, the balance of real power shifted 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 affairs. 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”). This already venerable title had hitherto been a temporary, war time commission given to officers who led imperial armies in the east. Yoritomo held it briefly 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 Tokimasa, and her brother, Yoshitoki, deposed Yoriie and replaced him with his brother, Sanetomo. They legitimized this new power arrangement by having Sanetomo appointed seii taishōgun. Thus Sanetomo, who is remembered as the third Kamakura shōgun, was actually the first 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 office were figurehead rulers (many of them were children at the time they took office), while the Hōojō ran things behind the scenes as hereditary directors of the shōgun’s private chancellery. The samurai-centered military and police system of the Heian era had represented a curious mixture of public and private parts: It depended on private

They Were Soldiers Once 41

resources—private training and acquisition of skills, private recruitment and mobilization, and private equipment—for its operations; but it was organized and directed through centralized, public principles. In most respects, the organizing principles of Kamakura period armies and military campaigns remained true to this pattern. The key difference was, of course, the introduction of the shogunate, which, after 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 respect 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. That is, the shōgun—or rather the institution he symbolically headed—became in effect the lone surviving miyako no musha, carrying out the court’s law enforcement and national defense jobs by mobilizing personal retainers, in much the same way that Heian samurai leaders had. Militarily, 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 twelfth-century military officers, whose command authority lasted only for the duration of a specific mission, the thirteenth-century shogunate exercised an ongoing and more or less exclusive jurisdiction over warfare. It also introduced new mechanisms for organizing and directing its housemen, as well as an unprecedented clarity to the reciprocal obligations that bound them. In its relationships to the imperial court and to samurai in the countryside, the first shogunate is perhaps best understood as a kind of warriors’ union. Before its creation, warriors in the provinces were merely local government administrators or caretakers for estates that belonged to court nobles or temples. The court had kept them politically weak by playing them against one another. By insulating an elite subgroup of the country’s provincial warriors from direct court control or employ, however, the shogunate significantly changed the rules of the game. Initially, this merely served to vault Yoritomo (and after him, the shogunate) into prominence, but in the long run, it created a mechanism for unraveling the fabric of centralized authority. The 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 diplomatic efforts over the next five years, extended this organization 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 between them, insisting that all calls to ser vice, all rewards, and all disciplinary matters involving Kamakura vassals pass through him. This arrangement became permanent—hereditary—under the Hōjō after Yoritomo’s death. The existence of the shogunate, therefore, rested on two competing obligations: 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. This is what made the regime legal and formed the basis of its national 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. These competing obligations led the shogunate to adopt a policy of minimal 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 records 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 whatever 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 personal claims to their lands—and the people on them. They did this in much the same way that a persistent dog will sometimes crawl under the blankets and take over most of its owner’s bed. The first step in the process usually involved a warrior exceeding his authority 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 warrior was a Kamakura vassal, the estate owner could not discipline the man himself; he could only complain to the shogunate. The shogunate, however, was less concerned with abstract matters of right and wrong than with keeping order and minimizing bad feelings on either side. After hearing arguments and examining documents submitted, its rulings more often than not involved some kind of compromise between the claims of the two parties. This meant that the warrior got less than he had tried to seize, but also that the estate owner would, from then on, be forced to accept less than he had before

They Were Soldiers Once 43

the dispute began. This 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 after generation. The process worked like a ratchet: Each new settlement represented a net gain for warriors and a net loss for the court. Thus, through a gradual advance by fait accompli, real power over the countryside bled steadily from the center to the hands of local figures, 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 progressed 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. The regime fell in 1333, as the result of events spawned by an imperial succession 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 connections between military and political power for many centuries thereafter. Unlike medieval Europe, where knights and military lordship arose more or less in tandem, Heian Japan remained firmly under civil authority. There 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 sociopolitical hierarchy thought of themselves as warriors in much the same way that modern corporate executives view themselves as shoemakers, automobile manufacturers, or magazine distributors: They identified 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 firms and other industries than with the workers, engineers, or secretaries in their factories, design workshops, and offices. While the descendants—both genealogical and institutional—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 twelfth century, the samurai remained the servants, not the adversaries, of the court and the state.

Notes 1. Pierre François Souyri, The World Turned Upside Down: Medieval Japanese Society (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: Japanese Historians Interpreted in English (Ithaca, NY: Cornell University East Asia Series, 2006); and Jeff rey P. Mass, ed., The Origins of Japan’s Medieval World: Courtiers, Clerics, Warriors, and Peasants in the Fourteenth Century (Stanford, CA: Stanford University Press, 1997). 3. The following account of the origins of the samurai is condensed primarily from Karl Friday, Hired Swords: The Rise of Private Warrior Power in Early Japan (Stanford, CA: Stanford University Press, 1992). For an alternative view, see William Wayne Farris, Heavenly Warriors: The Evolution of Japan’s Military, 500–1300 (Cambridge, MA: Harvard University Press, 1992). 4. For more on the political structure and the events of this period, see Bruce Batten, “Foreign Threat and Domestic Reform: The Emergence of the Ritsuryō State.” Monumenta Nipponica 41, no. 2 (1986): 199–219; Cornelius J. Kiley, “State and Dynasty in Archaic Yamato,” Journal of Japanese Studies 3, no. 1 (1973): 25–49; Inoue Mitsusada with Delmer M. Brown, “The Century of Reform,” in Ancient Japan, vol. 1 of The Cambridge History of Japan, edited by Delmer M. Brown (New York: Cambridge University Press, 1993), 163–220; Joan R. Piggott, The Emergence of Japanese Kingship (Stanford, CA: Stanford University Press, 1997). 5. Batten, “Foreign Threat and Domestic Reform,” 10–14. 6. See, for example, the forces described in Nihon shoki 591 11/4 and 602 2/1 (Shintei 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 The 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. dissertation, University of California, Berkeley, 1980); and Torao Toshiya, “Nara Economic and Social Institutions,” in Ancient Japan, vol. 1 of The Cambridge History of Japan, edited by Delmer M. Brown (Cambridge: Cambridge University Press, 1993), 415–452. 9. Ryō no gige, p. 183.

They Were Soldiers Once 45

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 “pacification campaigns” in the northeast, see Karl Friday, “Pushing Beyond the Pale: The Yamato Conquest of the Emishi and Northern Japan,” Journal of Japanese 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 process 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, Kamens, and Matsumoto, Heian Japan. 16. Cornelius J. Kiley explores this central idea in detail in “Provincial Administration and Land Tenure in Early Heian,” in Heian Japan, vol. 2 of The 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. Morita 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 kobunkan, 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). Ishimoda 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 peasants in revolt against the “misrule” of the governor.

46 Karl Friday

22. This and similar phrases appear in numerous documents from the ninth century onward, lamenting the depredations of central court figures 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 Province (Princeton, NJ: Princeton 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). The court issued repeated prohibitions 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 bungaku 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 English, see William H. McCullough, “The Heian Court, 794–1070,” in Heian Japan, vol. 2 of The Cambridge History of Japan, edited by Donald H. Shively and William H. McCullough (New York: Cambridge University Press, 1999), 39–40; or Friday, The First Samurai: The Life and Legend of the Warrior Rebel, Taira Masakado (New York: John Wiley & Sons, 2008), 35.

They Were Soldiers Once 47

35. Sompi bummyaku, in Shintei zōho kokushi taikei (Tokyo: Yoshikawa kōbunkan, 1983) vols. 3 and 4. The 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. The Seiwa Genji have been the subject of numerous 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 Mitsunaka࣬Yorimitsu: 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 offshoot of the Kammu Heishi. They 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 Morita Tei, Zuryō, 139–140. The definitive study of the early Heishi is Takahashi Masaaki, Kiyomori izen. 38. Friday, Hired Swords, 98–99; Farris, Heavenly Warriors, 188–189; Fukuda Toyohiko, 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 Administration,” 281. For more on Heian marriages and family structure, see William McCullough, “Japa nese Marriage,” Harvard Journal of Asian Studies, no. 27 (1967): 103–167; Peter Nickerson, “The Meaning of Matrilocality: Kinship, Property, and Politics 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, “Marriage and Property from the Perspective of Women’s History,” Journal of Japanese 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 (Tokyo: 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, The First Samurai. 43. Friday, Hired Swords, 70–97. The warrior alliances of the Heian and Kamakura periods are commonly referred to by Japa nese historians as bushidan (literally, “warrior groups”), but, for several reasons, I find 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 Japanese History (Honolulu: University of Hawaii Press, 2007) offers 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 Thomas Conlan, In Little Need of Divine Intervention: Scrolls of the Mongol Invasions 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, offer illuminating discussions of the sizes of early medieval fighting 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. This 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. The so-called seven leagues of Musashi are discussed in detail by Yasuda Motohisa, 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 alliances 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— difference between the two phenomena. 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 between court nobles, a warrior entering the ser vice of another presented his new master with his name placard (myōbu). There is, however, no evidence that the ju nior

They Were Soldiers Once 49

party to the arrangement ever received any written confi rmation in exchange. For examples of warriors offering 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 Minamoto efforts to create lasting warrior networks. In-depth treatments of the factors that defined 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, The Development of Kamakura Rule 1180–1250: A History with Documents (Stanford, CA: Stanford University Press, 1979) and Lordship and Inheritance in Early Medieval Japan: A Study of the Kamakura Sōryō System (Stanford, CA: Stanford University Press, 1989).

References Adolphson, Mikael S. Teeth and Claws of the Buddha: Monastic Warriors and Sōhei in Japanese 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 Threat and Domestic Reform: The Emergence of the Ritsuryō State.” Monumenta Nipponica 41, no. 2 (1986): 199–219. Brown, Delmer M., and Inoue Mitsusada. “The Century of Reform.” In Ancient Japan, vol. 1 of The 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: The Evolution of Japan’s Military, 500–1300. Cambridge, MA: Harvard University Press, 1992. Friday, Karl. The First Samurai: The Life and Legend of the Warrior Rebel, Taira Masakado. New York: John Wiley & Sons, 2008. Friday, Karl. Hired Swords: The Rise of Private Warrior Power in Early Japan. Stanford, CA: Stanford University Press, 1992. Friday, Karl. “Pushing Beyond the Pale: The Yamato Conquest of the Emishi and Northern Japan.” Journal of Japanese Studies 23, no. 1 (1997): 1–24. Friday, Karl. Samurai, Warfare and the State in Early Medieval Japan. London: Routledge, 2004.

50 Karl Friday

Hall, John W. Government and Local Power in Japan, 500–1700: A Study Based on Bizen Province. Princeton, NJ: Princeton 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 The 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 Japanese 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. The Development of Kamakura Rule, 1180–1250: A History with Documents. Stanford, CA: Stanford University Press, 1979. Mass, Jeff rey P. Lordship and Inheritance in Early Medieval Japan: A Study of the Kamakura Sōryō System. Stanford, CA: Stanford University Press, 1989. Mass, Jeff rey P., ed. The Origins of Japan’s Medieval World: Courtiers, Clerics, Warriors, and Peasants in the Fourteenth Century. Stanford, CA: Stanford University Press, 1997. Mass, Jeffrey P. Warrior Government in Medieval Japan: A Study of the Kamakura Bakufu, 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. “The Heian Court, 794–1070.” In Heian Japan, vol. 2 of The Cambridge History of Japan, edited by Donald H. Shively and William H. McCullough. New York: Cambridge University Press, 1999, 20–96.

They Were Soldiers Once 51

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 Lordship 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. Morita Tei. Zuryō. Tokyo: Kyōikusha, 1978, 50–52, 107–117, 136–138. Morris, Dana Robert. “Land and Society.” In Heian Japan, vol. 2 of The 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. “The 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: Japanese Historians Interpreted in English. Ithaca, NY: Cornell University, East Asia Series, 2006. Piggott, Joan R. The Emergence of Japanese Kingship. Stanford, CA: Stanford University Press, 1997. Sekiguchi Hiroko. “Kodai kazoku to kon’in keitai.” In Koza Nihonshi 2, edited by Rekishigaku kenkyūkai and Nihonshi kenkyūkai. Tokyo: Tokyo daigaku shuppankai, 1984: 287–326. Souyri, Pierre François. The World Turned Upside Down: Medieval Japanese 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 The 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 Japanese 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 shuppankai, 1970, 105–130.

3

Competence over Loyalty Lords and Retainers in Medieval Japan Susumu Ike

Introduction It is hard for modern Japanese, accustomed to political and social stability, to imagine the chaos that characterized life in medieval Japan. This is particularly 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-fledged civil war broke out as well. The increasing economic self-sufficiency and autonomy of the peasants eroded control by local lords and eventually led to an amalgamation of estates into larger territorial units throughout the Japanese archipelago. In par ticular, the archetypal models of the daimyō domains were first established primarily in the Tokai and Chugoku areas, which are not too far from Kyoto. In more distant areas, the rivalry of local estate owners (kokunin) persisted for a long time, probably because peasants were more deeply reliant on materials supplied by estate owners. In contrast, in areas surrounding Kyoto, there were robust network grassroots pacts among local residents (kokunin ikki) that resisted intrusion by centralized political leadership. Interestingly enough, the consolidation into territorial domains took longer to develop in both the hinterlands and in Kyoto’s vicinity. The territorial extension of daimyō rule intensified rivalry among themselves. But fighting at every opportunity was not always the best course of action for staying in the game. Daimyō often shifted 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 53

54 Susumu Ike

process of winnowing out the weak led to persistent rivalry among those who survived. In due course, rivalry among powerful domains gradually engulfed the hinterlands as well. This chapter explores the dramatic disruptions in political power that rocked the world of medieval inhabitants, and how those shifts shaped the relationships between lords and their retainers. The 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. The third section documents the weak bonds of loyalty between the local estate owners and their retainers, contrary to more romanticized views of medieval Japan. This is followed by an analysis of emerging local territorial states in the late medieval period, as exemplified by Uesugis in Echigo. The fift h section examines the process of territorial consolidation during the Warring States (sengoku) period. As local domains were gradually consolidated, the nature of military conflicts shifted from internal strife to battles among domains. The sixth section explores conditions that eventually contributed to Toyotomi Hideyoshi’s unification of Japan. The 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 example 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 government did not intervene in the battle while Masakado fought against his uncles, Kunika and Yoshikane. However, Masakado attacked a kokuga (the government’s provincial office) of the province of Hitachi (now, the prefecture of Ibaragi) and began to call himself a new emperor. Only then did the government 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 fourteenth centuries, warrior clans fought each other for de facto control of Japan.

Competence over Loyalty 55

The Minamoto’s defeat of the Taira clan in 1185 established the Minamoto family as the official protectors of the court. Although the emperor hardly had a choice, he granted Minamoto no Yoritomo the right to appoint his retainers, or gokenin (“housemen”) as military governors (shugo) in the provinces and military stewards ( jito) in both public and private landed estates. Thus 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 decades after Yoritomo’s death, the imperial court raised an army to challenge the Hojo regents who were ruling in the Minamoto name. The Hojo successfully defended their control of the Kamakura shogunate in a military contest against the imperial troops. Upon achieving military 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 conflicts peacefully, the Kamakura shogunate established a judicial system. Because of limited capacity, however, it relied entirely on selfreporting. This is the source of the proverb to the effect that “Even if a dead body lies down in front of the police office or the court, the police will not take any action unless someone specifies a criminal and accuses him to the police” ( gokuzen no shinin uttaenakunba kendannashi). Since criminal cases were treated in this rather offhand way, it is not difficult to imagine how unsatisfactorily many civil cases were handled. According 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. The defendant then submitted a letter of petition for defense. Th is process was repeated three times. Then, the two confronted each other face to face and, after that, a sentence (saikyo) was fi nally handed down. The system seems reasonable if these procedures were actually followed. 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 defendant referred to laws enacted by the shogunate in order to justify his position, he had to prove that those laws were really enacted. More problematically still, even if a sentence was pronounced, the shogunate did not have a system of enforcement. Thus, 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. The shogunate took seriously threats to its regime, but played a limited role in maintaining social order in the Kamakura period, doing little to intervene in conflicts 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 often failed. The Soga brothers, Goro and Juro, for example, murdered Kudo Suketsune in retaliation for killing 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 Europe. The Kamakura shogun recognized as his retainers (gokenin) the lords of local estates who pledged fealty to him. The shogun guaranteed the local lord’s land rights either by assigning him the position of steward of an estate ( jito-shiki) or by ratifying his position of estate officer already assigned by the estate owners (sokan-shiki). The Kamakura shogunate typically preserved 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 officially part of the master’s household rather than owning their own estates and formed a kinship-like band for self-protection. The master’s household directly managed dozens of hectares of lands, using not only slave-like direct laborers (genin) but also neighboring peasants of very small operating units (menka). Furthermore, households had workshops for food, clothes, and agricultural tools, and were highly self-sufficient. In the unit of the household, kin members and retainers controlled tenant workers and protected landed estates. Local social order was maintained through the coexistence of these large operating units, which also placed neighboring peasants under their control and protection. A master controlled his household, and even the government could not easily intervene in its internal affairs. For example, Article 26 of the Goseibaishikimoku (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 children. 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

Competence over Loyalty 57

enormous resources. Central tax revenues were minimal, forcing the shogunate 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 destroyed 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 effort that in the end Ashikaga Takauji, a former Kamakura vassal, emerged from the shadows and established a new shogunate under his own name. The Ashikaga government gave local jurisdiction to military generals (shugo), who became daimyō or domain lords. These local lords were given authority to punish illegal activities such as harvesting someone else’s rice plants (karida-rozeki). The government also established a law called kosenbosen-ho to restrict private wars. The term kosen means initiating a fight, 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 conflicts peacefully. Local residents might form a pact (kokujin-ikki) to solve territorial conflicts, particularly regarding territorial boundaries, and troubles in markets and towns, through peaceful discussions rather than with force. These local pacts emerged as a consequence of the breakdown of the household system. Early pacts were often negotiated at the instigation of local lords attempting to reunite a clan against the growth and independence of branch families in the clan. But for several reasons, local social order became more urgent, leading to regionwide pacts. The independence of branch families entailed the division of landed estates, disrupting sharecropping arrangements. Moreover, with the development of new lands peasants became increasingly self-supporting and dissatisfied with labor duties on a lord’s estate. As the estate system loosened, stipends became attached to the lands themselves rather than to the official positions (shiki) associated with protecting the land. Estate owners 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 development of commerce increased the stake in smoothly functioning markets. The 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 destabilized succession plans that kin and retainers feared would fail to put the strongest heir in place as leader. The result was full-scale civil war (the Ōnin War) between 1467 and 1477, and continued jostling among domains for another hundred years after that.

The Warring States Period After 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 mobilization, 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 securing the political leadership in local society or securing the right to trade with China. In the Kyoto region and its neighboring provinces, the powerful Hosokawa 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 ownership. An even more important change was the breakdown of the household (soryo) system of bequeathing landed estates to kin members and retainers. With the chaos of civil war, relatives and warrior retainers remained loyal to a household head only as long as he could protect their lands, rather than risk all through unconditional loyalty. As retainers began to intervene in the selection of a new master at the time of household succession, either through council meetings or by resorting to force, the lord-retainer relationship became more mutual. Fights for household 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

Competence over Loyalty 59

States daimyō. A master’s ability to protect land rights was an important condition 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 frequently, close retainers died together with their masters in battles between Sengoku daimyō. There 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 after avenging their dispossessed lord by killing his rival. The story has lasting appeal, perhaps because a show of this kind of loyalty was so rare. Far more often, retainers 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 political 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 independent, 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, territorial conflict occurred between bordering lords, and one party would invade the other’s territory. Alternatively, a neighboring lord might intervene in a dispute over fishery rights, but his mediation could fail if it was favorable to the one party over the other. Conflicts among neighboring lords were sometimes unyielding to local mediation altogether. Other evidence of local disharmony comes from the province of Suruga, now the prefecture of Shizuoka. The 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 rapidly came to ruin after 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 includes many articles regarding the settlement of conflicts through judicial process. Article 1 concerns the rights of ownership and management associated with cultivated lands, and it is important for studying social structure during that historical period. Article 2 deals with conflicts on boundaries among fields and mountains and prescribes severe punishment to the party losing a  lawsuit. The punishment was confiscation of a third of the party’s land. Later, retainers claimed that the punishment was too harsh, and it was softened. But the prescription of such severe punishment shows how seriously boundary disputes were taken. Article 3 focuses on redevelopment of barren land destroyed by flooding and other natural disasters (kawanari, uminari no chi). This indicates that the extension of development of barren land was one of the major causes of boundary conflicts. The 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 conflict and mentions that development of fields and mountains from two sides, by extending to the boundary, could cause a boundary conflict (Article 58). Boundary conflicts of the day were closely related to the entrepreneurial cultivation of new lands by ordinary peasants. Where property rights or common 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). The Mori household 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 independent lords with their own retainers swore allegiance to Mori Motonari. Conflicts among them were to be solved cooperatively, and the authority of the master was limited to punishing offenders. 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 chubatsujiken). Following the incident, 237 retainers submitted a written oath to Mori, in which they agreed that conflicts among them would be subject to Mori’s jurisdiction. They forswore territorial expansion and conflict resolu-

Competence over Loyalty 61

tion by self-redress. The expulsion of the Inoue clan succeeded because many of the retainers hoped that Mori’s firm 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 effective in providing social order than the prevailing system of selfhelp and vigilante justice. Consolidation became a trend as in region after region, a dominant family arose. The Uesugi (later, the Nagao) managed Echigo, the Imagawa managed Suruga, and this pattern was replicated in many provinces. The Imagawa, for example, punished both sides of a territorial squabble with capital punishment. The Imagawa code also included detailed rules regarding the judicial system and clearly demonstrated a trend toward firm rule by the regional public authority.

War in Medieval Japan The proximate causes of the Ōnin War (1467–1477) were disputes over household succession in the Shiba and Hatakeyama families, both powerful military governors. By the fifteenth century, the Ashikaga shogunate was internally divided and could not intervene effectively in the succession conflicts of retainer 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 shogunate, were at loggerheads. The political system of the shogunate was dysfunctional for all practical purposes, and conflicts could be solved only by force. The Ōnin War ended when Hosokawa Katsumoto and Yamana Sozen died and their respective heirs chose to settle up, at least temporarily. But the underlying structural problem of widespread incentives to grab whatever 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’ household succession. Furthermore, the positions they took were often complicated by their own stakes in various territorial disputes. This situation often made succession problems intractable, and the recognition of a household heir by the shogunate authority could no longer break logjams. After the Ōnin War, the Hosokawa and Hatakeyama continued to fight each other for lordship over the Kinai region. Amazingly, in 1493, twenty years after the end of the Ōnin War, Hosokawa Masamoto, a powerful regent

62 Susumu Ike

of the shogunate, discharged the shogun himself. In the process, Hosokawa began to consolidate his control over the Kinai region. The same process— regional consolidation—continued in the rest of Japan, with the Hojo moving in on the southern Kanto region. The 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, resistance to such authority emerged, and such resistance was forcefully quelled. The 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. The primary locus of this struggle was among a master, his sons, and his brothers. The 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 household control provided 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 winning 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 conflicts by force and could not expand their lands by invading their neighboring lands within their domains. The next stage in Japan’s Warring States period began in the second half of the sixteenth century when the newly consolidated domains began attacking each other.

Competence over Loyalty 63

The written oath that Mori’s band of retainers submitted after the purge of the Inoue clan contained many articles concerning war. They prescribed war readiness and a pledge to serve bravely in battle as needed. They 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 insufficient. The oath was quite explicit about the connection between ser vice and reward. It was a contract between Mori and his band of retainers to unite toward external expansion 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 battlefield without permission. In 1555, five years after the purge of the Inoue clan, Mori fought in the Island 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 neighboring 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 organized army with high morale, giving Mori the platform to become a powerful daimyō with great influence 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, deplored their change in character. As territory became consolidated, the nature of war also changed from internal struggles to interdomainal war. The Battle of Kawanakajima, the most famous military struggle in the entire Warring States period, was such a case. The 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. The 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 regional lords began to disappear and domainal boundaries became well demarcated. 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, political and military alliances 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 coalitions 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 conflicts between the daimyō and arranged temporary truces between Mori and Amako of the province 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 political 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 province of Omi (now, the prefecture of Shiga), and the Asakura of the province of Echizen. The 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 popular 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 following a practice that daimyō lords before him had initiated. More original to Oda Nobunaga was his logistical move of controlling an economically advanced

Competence over Loyalty 65

region like Kinai. In fact, immediately after 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 officers. Merchants supported the unifiers because they valued political stability in a broad region since it increased their profit 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 relationship with local lords in their ruling regions. Instead, he organized 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. The unification of the nation was taken over by Hashiba (Toyotomi) Hideyoshi, who became Oda Nobunaga’s successor. Hideyoshi conquered Chugoku, Shikoku, Kyushu, Kanto, and Ou and completed the unification of the nation. He was successful in this effort largely because of his ability to mobilize enormous 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. The Hojo, whom Hideyoshi vanquished, were good at the conventional strategy of holding 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 thousand soldiers for more than three months. The Hojo finally 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 conflicts between them, and he ordered them to obey the judgment of judicial officials. 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 conflicts at that time, justifying his continued use of force against them. When he defeated them, Toyotomi rewarded his warriors with the newly acquired lands, creating the so-called Toyotomi daimyō including Ishida, Fukushima, and Kato. The 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 sufficient to require compliance. Toyotomi Hideyoshi failed to conquer Korea and died of illness, after which the government became unstable. But Tokugawa Iyeyasu’s conquest in 1600 and his regime’s consolidation in 1615 finished a job that Oda Nobunaga and Toyotomi Hideyoshi had nearly completed themselves.

Conclusion Competition among local lords resulted in the territorial consolidation of Japan. 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 heterogeneous forms rule and authority coexisted fail to endure? Why did the struggle among warlords eventually lead to unification and the establishment of the Tokugawa shogunate? Some historians have argued that the microcosm of a nation already established at the domain level provided foundations of legitimacy for national unification 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, ruthlessness, and singleness of purpose—is unlikely to have divided Toyotomi from other men. The 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 owners and peasants, creating fewer but larger territories. Even as Hideyoshi’s sword hunt of 1588 helped to cement his primacy as a warlord, it also managed to ramp down the levels of violence to which the Japanese farmers had long been subjected.

Notes 1. Susumu Ike, Daimyō Ryokokusei no kenkyu [Research on the Territorial System of Daimyō], in Daimyō-ryogoku-sei no Kenkyu [Studies on the Daimyō Domain System] (Tokyo: Azekura shobo, 1995), pp. 66–68.

Competence over Loyalty 67

2. After Yoritomo died in 1199, effective power in the shogunate passed into the hands of the Hojo family, who were the family of Yoritomo’s wife, Masako. 3. This episode is known as the Jokyu Disturbance. 4. This 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. This right was technically possessed by the king of the nation, that is, the shogun. 7. Soryo-sei teki bushi-dan, in which kin members and retainers (ie-no-ko, roto) were organized under a master called soryo. 8. The 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. The Meiji government also put its ex post stamp of approval on Emperor Go Daigo, who refused to acquiesce to Ashikaga’s rule. In defiance 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, Ashikaga moved the capital back to Kyoto and built for himself an opulent castle that dwarfed the emperor’s. 9. The 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 objective. The kokujin means a person who resides in the province. Putting these terms together, the kokujin ikki was an equal-basis-unification of local lords based on the regional relations. 10. See Jeffrey 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 ᐓᮮ 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 between 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. The story was memorialized in the kabuki play, Kanedehon Chūshingura, and later in the classic 1941 Mizoguchi movie, The Forty-Seven Samurai. The war time government must also have liked the emphasis on self-sacrifice. 14. Susumu Ike, 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 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 Political Thoughts in the Medieval Period, No. 2], vol. 21 of Nihon Shiso Taikei [A History of Japanese Thoughts] (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 essential for rice fields, and this is also a conflict related to development. 16. Dai-nihon Kobunsho Mori-ke Bunsho [Japa nese Old Documents: The Documents 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: The Documents of the Mori Family], No. 398. 19. Dai-nihon Kobunsho Mori-ke Bunsho [Japa nese Old Documents: The Documents of the Mori Family], No. 401. 20. The original statement is “rihi wo ronzezu,” meaning whether or not there is a reason does not matter. 21. This is called the Bunsei Political 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 household, which created a conflict with Yoshimi. 23. This episode is called the Meio Political Disturbance. 24. Also at about that time, Ise Nagauji (Hojo Soun) caused a stir when he removed an official, the Horikoshi Kubo. The Horikoshi Kubo, a member of the Ashikaga clan, was a replacement of the Kamakura Kubo, who was an original administrator of the Kanto region of the Ashikaga shogunate but revolted against the shogunate. The 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 ambitious daimyō found it difficult to extend his political 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 Dainihon Kobunsho Mori-ke Bunsho [Japa nese Old Documents: The Documents of the Mori Family], No. 418.

Competence over Loyalty 69

29. Specifically, gekokujo included the overthrow of a master by a powerful retainer and that of a shugo by a lieutenant shugo. Nagao’s failed usurpation of Uesugi’s position was an example of the latter. 30. Dai-nihon Kobunsho Mori-ke Bunsho [Japa nese Old Documents: The Documents of the Mori Family], No. 613. 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. This 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. The Miyoshi family ruled the Kinai region after the Hosokawa. 33. The ikko-ikki mobilized common believers and thus had very large numbers of people in arms. Furthermore, their morale was strong due to their religious motivation. 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 territorially. They surrendered to Nobunaga once he promised the continuation of their religious orga nization. The 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 [The Creation of Shogun Authority] (Tokyo: Iwanami shoten, 1994). 35. The term erizeni referred to an act of denying par ticu 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: The 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. The weakening of the Ming Empire inspired Toyotomi to imagine a different 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 [The Unification of the Nation and the Invasion of Korea], edited by Susumu Ike (Tokyo: Yoshikawa Kobunkan, 2003). 38. See, for example, Katsumata Shizuo, Sengoku-ho [The Law of the Warlords] (Tokyo: Iwanami, 1976), p. 208. 39. See, for example, Fujiki Hisashi’s discussion of the so-buji in Toyotomi Heiwarei to Sengoku Shakai [The Peace of Hideyoshi and the Japanese Society During the Warring States Period] (Tokyo: Tokyo Daigaku Shuppankai, 1985).

70 Susumu Ike

References Asao, Naohiro. 1994. Shogun Kenryoku no Soshutsu [The Creation of Shogun Authority]. Tokyo: Iwanami shoten. Hisashi, Fujiki. Toyotomi Heiwarei to Sengoku Shakai [The Peace of Hideyoshi and the Japanese Society during the Warring States Period]. Tokyo: Tokyo Daigaku Shuppankai, 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 Daimyōryogoku-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 System]. Tokyo: Azekura shobo, 1995, pp. 66–68. Ike, Susumu. Nihon ShiTenka Toitsu to Chosen Shinryaku [The Unification of the Nation 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: The Currency and Nation in Pre-modern Japan]. Tokyo: Aoki shoten. Mass, Jeff rey P. Lordship and Inheritance in Early Medieval Japan: A Study of the Kamakura Soryo System. Stanford, CA: Stanford University Press. 1989. Shizuo, Katsumata. Sengoku-ho [The Law of the Warlords]. Tokyo: Iwanami, 1976.

4

Community Vitality in Medieval Japan Tsuguharu Inaba

Introduction Autonomous communities are known to have developed in medieval Japan. There 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 political 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 communities managed to maintain or even expand their autonomy, forming the basis of a robust modern society capable of democratic self-governance. The characterization of Japan as a typical “Asiatic nation” governed by a centralized authority was not only a common view among Japan’s postwar occupiers, 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 fight over boundaries (sūbuji), as the logical end point of an intensification and deepening of military rule. Furthermore, this view links to an understanding of the Meiji Restoration as another manifestation of these historical conditions and Asiatic principles. Japan’s transition to modernity was incomplete, the argument goes, because each successive political regime, including the U.S. occupation after World War II, was an imposition from above that used existing hierarchical structures for support and legitimation. 71

72 Tsuguharu Inaba

The revisionist view, that Japanese history includes a rich tradition of local vigor and resistance to control, has developed over the last twenty years through a reexamination of data from village communities in the late medieval period. This view radically reinterprets the fifteenth to the seventeenth centuries as a period in which villages, towns, and local political regimes were vigorous and active in shaping their environments. The Toyotomi Peace Order against local conflict 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 first and foremost in village communities rather than from military overlords, in tandem with the mechanisms of self-government. Village councils (yoriai) of peasant households gained decision-making importance from the fifteenth century throughout Japan. The 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 organization for peasant households and helped to regularize contractual relationships between 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 neighboring mountains, as well as the water sources for irrigating the fields. 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. Rifles were also introduced in the late sixteenth century. The villages’ military forces, consisting of males between the ages of 15 or 16 and 20, were called wakashu. They were well organized 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 five years in the early fifteenth century. The most serious concern of the medieval village community was how to keep conflicts over territory from ruining their livelihoods. Village communities 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 solidarity and diminished internal conflicts within the community. The village community provided its members with economic aid as well as military security. Their 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 preferred to stay out of the larger conflicts 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 opponent and conquered his opponent’s territory, he first had to collect the tax documents from villages in the territory that he had newly gained. Th is practice became common in the fifteenth century during the Ashikaga shogunate. The tax documents showed the kinds and amount of taxes the village paid to the previous lord, and were renegotiated periodically based on cadastral surveys 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 precedents described in the village’s tax documents. Village communities did their best to circumscribe the lord’s powers to raise new resources from their productive efforts, 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. The peasants’ main military ser vice was transporting military equipment and supplies to the battlefield. Peasants called up for this duty accounted for more than 30 percent of all the personnel of the daimyō’s army in the sixteenth and seventeenth centuries. In the sixteenth century, as regional lords consolidated their territories and began to fight for supremacy, village communities were in continual 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 second half of the sixteenth century, the Hojo fielded peasant armies against Takeda Shingen and again against Toyotomi Hideyoshi. Even here, however, the peasants were not expected to fight on the front lines of the battles but to stand in defense of castles close to their villages. The rule, at first unspoken and later formalized as “separation of warrior and farmer” (heino bunri), reflected 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 communities in their domains, depending on village size. The village council in turn decided who would be called up for duty and for how long. The local lord typically discounted about 10 percent of taxes as long as the village community regularly performed public duties such as logistical support for military campaigns. When war damaged crops and otherwise undermined agricultural productivity, local lords were forced to make economic concessions such as reducing or canceling debts owed by the village to the lord. The Hojo canceled debts and released debt slaves for all the villages in their territory in 1550 and 1560 when the villages were totally exhausted. The lord’s retainers were prohibited from imposing additional public duties on the villages without permission, and peasants 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, village communities often offered 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 often had no choice but to accede to these requests. Throughout the sixteenth century, village communities became more solidified 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 conflicts in the border regions of their provinces.

Community Vitality in Medieval Japan 75

Mobilization and Expansion The peace of the Toyotomi, which facilitated the peace of the Tokugawa, was brought by reorga nizing village communities and warrior lord authority. It was impossible in the sixteenth century to expand political authority simply by marshalling superior military power. Toyotomi Hideyoshi did not resort to arms primarily in order to expand his authority. After occupying Kyoto, Hideyoshi issued a proclamation to the regional lords to stop fighting: I was ordered by the Emperor to pacify the country. You must stop fighting 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 judgments, 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 guaranteed the whole or part of their territories, establishing a lord and vassal relationship with the regional warrior families and expanding authority without further warfare. The 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 communities secured their role as intermediaries or contractors responsible for collecting taxes and organizing public duties, and enhanced their role as mutual aid organizations to peasant households. Although peasants and warriors were already accustomed to a division of labor in warfare, the Toyotomi regime made the division more explicit. Hideyoshi proscribed peasants from carrying arms such as swords and prohibited village communities from resorting to arms for self-protection or conflict settlement. Hideyoshi announced to the peasants: Peasants will prosper forever, if they carry faming tools only and are engaged in farming. This 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 neighboring villages and the need to provide local lords with corvée labor during war time. Both were now illegal. There is still the question of why Sengoku daimyō and village communities accepted Toyotomi Hideyoshi’s rescripts, which were, after all, not unlike rules that had been periodically handed down by preceding governments. The traditional explanation, that Hideyoshi was the first 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 (warring states daimyō) had secured their power within their regions. But tension then shifted from struggles with regions to the border areas between provinces. Regional lords found themselves pinched between the expense of border conflicts on the one hand and peasant demands for economic concessions on the other. To avoid financial ruin, regional lords often settled territorial conflicts through treaties mediated by neighboring lords. Something similar occurred in territorial conflicts among villages. Conflicts were often settled by mediators from neighboring villages or by collective decisions of villages in the region. Village communities, which maintained their unity by fighting against neighboring villages in the beginning, became more stable as self-governing, mutual aid organizations. By the end of the sixteenth century, villages shifted their activities from military preparedness to economic risk sharing. The peace of Toyotomi was possible because territorial consolidation by regional lords had increased the costs relative to the benefits 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 households and village communities existed as the foundation of Japanese society. The shogun’s officials (kougi) ruled over regional lords, or daimyō, who in turn governed local societies. Although the Tokugawa regime is characterized by its hierarchical organization with the shogun as its apex, it should not be overlooked 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-

Community Vitality in Medieval Japan 77

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 hegemony became unstable in the days of Nobunaga and Hideyoshi because of the decline of the Ming Dynasty and the encroachment of European navies. This 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 unification of Japan. He demanded that the local lords in Korea should stop fighting and yield to his authority. If they did not comply, he would invade them, as he had in other parts of Japan. The Toyotomi regime’s mobilization of peasants for the invasion of Korea illustrates the importance to peasants of staying out of wars, or being compensated when they were drawn in. Kato Kiyomasa, who was the domain lord of Higo, now the prefecture of Kumamoto, bore the heaviest burden of mobilizing for war against Korea. The 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 battlefield. They 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 confiscating 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 beginning that peasants would refuse mobilization for the invasion, and retainers were instructed to prepare for their refusal with whatever force was necessary. In 1591, Kiyomasa sent a letter to his territorial province, Kumamoto, after six peasants fled 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, children, 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 representative cannot be imprisoned, that is the responsibility of an officer who administers my direct domain. Since troops will be finally 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 effective countermeasures. In the base camp of the battlefield in Korea, Kiyomasa found a letter sent by a peasant in a village in the province of Higo to logistical peasants in the battlefield. The letter stated, “It is time you should return to the village. Officers will not investigate, even if a group of you flee back to the province of Higo.” The letter shocked Kiyomasa. Concluding that the villages had planned this act in conspiracy with officers in his territory, he became furious. Apparently, 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 fled. Peasant desertion not only reflected resistance to mobilization by the peasants and their villages in the province of Higo, but also implicated local officers who sought to maintain local order. The escapes were neither haphazard nor sporadic. Being shaken by the peasants’ resistance 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. This is because Japanese 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. This order, which broke with precedent, only intensified the villages’ resistance 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.

Community Vitality in Medieval Japan 79

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 suffering, 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. Thus, 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 after the failure of the invasion, he changed his position and offered the peasants exemption from labor duties for the next two to three years. However, this offer did not simply emanate from Kato’s benevolence. The suffering of the peasants described in the signs he had posted should be considered his recognition that peasants now influenced 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 villages. Kato feared their criticism of his qualifications as a lord. Before returning to his home ground, the Kumamoto castle, he posted signs guaranteeing the peasants’ engagement in farming. Toyotomi and his vassals failed to construct 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 Unification As part of a joint project of the History Section, the Faculty of Arts and Humanities at Kumamoto University, I conducted a field survey of more than 600 stone monuments in Chuo-machi in Mashiki County (today’s Misatomachi). This field survey revealed that groups of villagers began erecting stone monuments around the sixteenth century, documenting that village communities were collecting their own taxes by medieval times in northern Kyushu. The number of these stone tablets is approximately the same as the number of residential blocks or village clusters. Surveys conducted by freelance researchers 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 governance in the seventeenth century, following the unification of Japan under Tokugawa rule, the dispute settlement process 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 Hosokawa 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 mountains that stretch between these areas. Disagreements between the two villages involved settling borders between the two feudal domains and therefore encompassed the entire hierarchy of political authority, from villagers up to the top of the clan. The dispute resolution mechanism was multilayered. As the first step, village headmen or the representatives of the directly impacted population negotiated how to adjudicate the dispute. When this stage failed, the headmen reported the situation to the clan’s county office. The 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. The system of dispute settlement was an embodiment of complex layers of political authority in the traditional Japanese society and consistent with Kogi or the Constitution of the feudal clans as well as the preexisting autonomous village organization. The involvement of peasants, headmen, county officials, and chief retainers can be summarized in the following manner. 1. The information on which the clan government relied in making decisions was provided by the headmen. In particular, existing documents detail the process of disputes, the topology of the disputed areas, and the situation surrounding the opponents’ territory. 2. The documents were prepared by the headmen and other village officers and the county resident representative passed the information on to the official in charge of local administration in the central clan’s government. Then the information was processed to the chief retainers. The chief retainers almost always accepted the information without questioning it and based their decision on it. Then the decision was passed down to the village level and was implemented by the headmen of the relevant villages.

Community Vitality in Medieval Japan 81

3. Even after 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 effectively territorial lords on a small scale and also were chief administrative staff, and they did not limit their local administrative activities to clan border disputes. An example comes from a stone monument beside the Samata Canal in Kumamoto, which was opened in 1683 as an irrigational canal to convert vegetable fields into underwater rice paddies. Because the drilling of rock was a difficult task, those who were involved in the project celebrated the opening of the canal when it was completed. The roster on the stone lists Mr. Nakayama, the hereditary headman of Nakayama; the official of the clan in charge of the construction work; officials in charge of the area; and the representatives of Samata Village, whose residents were the target beneficiaries. While the irrigation system specifically benefited Samata Village alone, officials in charge of the county as well as officials in charge of public works were also listed on the stone tablet. The village-level project involved the clan’s government. The inscription shows that the target beneficiaries coordinated their project and worked out the details of water use rights. Then the villagers submitted the project request to the village headman. After the headman achieved consensus within the village, he applied for project approval and assistance from the clan’s government. Then the clan’s government approved of the project. Although only this stone epigraph documents this project, the names on the stone reveal a great deal about the decision-making process. The mid- to late eighteenth century marks a historical watershed in the development 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. After 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 headmen became the main tax-collecting agents, a portion of the revenues was expended for the benefit of their local communities. According to Maeda, as of 1843, the fift 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. The headmen’s revenue base includes taxes of various kinds levied on the rice harvest, donation and contributions by the peasants, and returns on their own investment projects (e.g., land reclamation). In this sense, local development funds provided a material foundation to the headmen’s governance. The feudal domain’s resident representatives in the communities were no longer present, and most local administration tasks were conducted by villagers themselves. The decentralization initiative and the formation of the local development funds resulted in further development and enhancement of headmen as autonomous local governance functionaries. Because headmen had their own development funds, they gradually but effectively became economically independent of the feudal clans and were instead agents of communal planning and regional policymaking. The new system of appointing and relocating village headmen and peasant civil servants provided effective 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 representative 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 suggest, the content of the proposals often concern water use as well as other types of public work projects. The deputy county representative forwarded the headmen’s proposals to the clan’s secretary, and then the clan officials evaluated the merit of the proposal and made decisions over project approvals. The officials reported the result to the deputy county representative, and if the village headman agreed, the project would be implemented.

Community Vitality in Medieval Japan 83

The form and contents of these memos indicate that the system of consensual decision making was developing in the Kumamoto family’s local administration, with village headmen functioning as the basic unit. The system of village headmen became institutionalized and grew into a bureaucracy after 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 financial contributions to the clan. Each of the “meeting places” of the village hired promising young boys as trainees. They were promoted to village headmen (mura shoya) at the age of 20 or so. These 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. Thus, the village headmen at the hamlet level were effectively administrators and bureaucrats, and whenever these officials could not handle the administrative function in these communities, village headmen collectively petitioned for firing incompetent headmen. On the other hand, when the headman was efficient, the village headmen petitioned against relocation of the headman. Similarly, peasants often initiated boycott movements against inefficient village headmen. The bureaucratic nature of the headman system is illuminated when we examine the merit evaluation system of the headmen in the Kumamoto family in the eighteenth century. This is effectively a system of performance evaluation conducted by the clan’s Election Bureau, and the process was based on information contained in reports from the village headmen and the county deputy. The bureau was monitoring how well the headman was planning policies and implemented them. Based on this performance evaluation, the headmen were awarded, relocated, or fired. As a result, the headmen from one village were sometimes relocated to other villages. This performance evaluation system consequently affected personnel affairs in each headman council so that the preferences of peasants who belonged to each headman council were closely reflected. Therefore, as long as the headmen were willing to work for the benefit 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 performance 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. The meeting place was managed by

84 Tsuguharu Inaba

staff who were also from peasant households. These officials aggregated the preferences of the local residents and arbitrated interests. They together planned and implemented public works that would benefit the locale on the whole. The 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 effective mechanism of checks and balances between the clan’s government and the local administrators at the village level. When headmen discussed disaster relief measures after a tsunami hit the village in the early nineteenth century in Konoura Tenaga in Udo County, the headmen decided to solicit grants from the clan’s government instead of financing the project themselves. The clan official 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 decision was modified to a considerable degree, and a compromise was worked out. In this process, the system of consensual decision making was respected, and the clan’s government was not capable of enforcing their decision in a topdown way. The village headmen effectively constrained the clan’s government. Village communities, headmen, and the clan’s secretariat were closely involved in the decision-making process, making headmen effective veto players in the decision-making procedure. Modernization after the Meiji Restoration was greatly facilitated by this experience with local self governance. The popularly elected prefectural assembly in Kumamoto was regarded as the most democratically advanced among the Japanese prefectures of the time, together with Shizuoka and Kochi. First, the election for the Kumamoto Prefectural Assembly was conducted during Governor 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 organize their meeting (Kansen Kuchokai). Instead of a meeting of politically appointed headmen, prefectural assembly members were popularly elected, with near perfect male universal suffrage. Second, the election was conducted in multiple rounds. Voters first elected members of “small districts” (sho ku). Then the elected members chose their representatives serving in “large districts” (dai ku), and only one in ten members of the small districts was selected into the main prefectural assembly members (dai ku). In this multilayered electoral system, sho-ku 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. The Assembly eventually

Community Vitality in Medieval Japan 85

confronted Governor Yasuoka’s measure to control villages’ budgetary base gobi kin (village endowments). The village headman system not only incubated a system of representation but also made a streamlined transition to modern village communities. These 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. The Higo Restoration faction of the Jitsugakuto government in Kumamoto (est. 1871) removed each village’s development funds and placed them at the prefectural level. The prefectural government then used these funds to finance 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 development funds be returned to the village communities. Municipal council members also joined forces. In 1895, the protestors appealed to the court for judicial decisions. The fact that those who participated in the riot of 1877 were demanding the return of communal development funds to the village indicates that during this period local residents perceived that the funds were their own common property. In 1880, the Meiji government made a policy shift. That year’s order from the Department of State (Daijokan) stipulated that the village development fund would belong to the alliance of village trusts (gobi zaisan rengokai). Thus, the development funds were invested in the Kyushu railroad companies and other newly emerging enterprises. The 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. The funds supported municipal government finance especially when their fiscal foundations were vulnerable. Although the Meiji government tried to confiscate the funds accumulated in each local community, it ended up making a policy transition to allow greater local autonomy. The 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 confiscation by the state.

86 Tsuguharu Inaba

In 1890 the Meiji government tried to impose a new local entity of counties, which were nothing more than geographic segments for the sake of implementing public works projects. But Kumamoto Prefecture initially opposed the introduction of the modern county system in 1890, and the central government did not have substantive administrative capability to impose the new system. The functional administrative units were the villages, and their agents were the headmen serving these communities. The cornerstone of modern Japanese administration was carried over from local self-governance in the premodern period. Local government finance, consensual decision making, civil servants, and representation were all present in their early form in medieval Japanese villages.

Conclusion This chapter examines how wars changed the political map of medieval Japan. Conflicts between villages regarding agricultural production led to the formation of autonomous village communities, while feudal lords consolidated territorial control through armed conflict. In this process, villages demanded physical protection and limits to their role in military mobilization. The Toyotomi regime generalized the settlement between villages and daimyō to the national level. The failed invasion of Korea renewed the government’s commitment to the peasants’ specialization in farming. The 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 Japanese society influenced modern Japa nese norms and practices. First, successive 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. Peasants in feudal society did not envision themselves toiling for the nation. They would still feel this way when the Tokugawa shogunate fell to disenchanted domains in the Meiji Restoration of 1868. The mass conscription system established 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 sacrifice their lives for the nation-state was the same as doing so for their families or villages. It took the imperial

Community Vitality in Medieval Japan 87

fascism of the 1930s militarists to capture the imagination of village communities 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 different from the traditional practice of making economic concessions to peasants in hard times. The shogunate maintained its regime by depriving daimyō of the right to rule when they could not ensure agricultural productivity. In the late eighteenth century, some village communities began to gather beyond the geographical boundary of a daimyō’s territory to discuss regional problems such as flooding and water management. They handled these problems by choosing representatives of the region by election, or by proposing policies to a daimyō and receiving financial 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 affairs. Third, the structure of traditional Japanese society, which consisted of autonomous, mutual aid organizations against the background of central authority, is similar to that of Japanese corporate society in the twentieth and twenty-first centuries. During the economic boom following World War I, many Japanese corporations competed for scarce labor by guaranteeing the livelihoods of employees and their families through the lifetime employment system and organizing numerous subcontract companies under big companies. Interfirm competition provided an incentive to offer employment contracts that managed workers’ lifetime risks. However, the global economic integration of Japan’s economy has now destroyed much of this ethic. Japanese society has not yet found solutions to numerous problems arising from increased unemployment, homelessness, and income inequality. Village communities were broken up in the 1960s, while corporate organizations were destroyed through the 1990s to the beginning of the twenty-first century. It was the biggest, and perhaps the worst, change that Japanese society has experienced since the sixteenth century.

Notes 1. Downing (1992), for example, implies that Japan failed to develop democratically because its medieval and early modern history lacked experience with or a theory of limited government. 2. The territory of the village community included residential areas (mura), agricultural fields (nora), and lands for gathering (yama). Irrigation water needed for rice

88 Tsuguharu Inaba

farming was often gained through channels from rivers. The yama was a space of collective 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. The village community was the actor of self-redress. 3. The size of cultivated lands in the village, which was determined by daimyō authority and the village community through the process of making a record book, was called mura-daka. 4. This practice of making economic concessions was known as tokusei and followed the Confucian idea of virtuous rule or tokuji-shiso. 5. This is the sūbuji-rei. 6. The taiko-kenchi. 7. The katana-gari-rei was a policy to complete the heino-bunri under the name of the tokusei. 8. The kenka-choji-rei. 9. This is called kai-chitsujyo (literally, to bring order to China and the barbarians). 10. With this policy of the sūbuji, or “peace rescript,” Hideyoshi controlled Kyushu in 1587 (in the fi fteenth 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 materials collected by Yoshimura (2001). The memos were chronologically bound, and some of them are as thick as 30 centimeters. The originals used high-quality paper sheets of standardized sizes, which were perhaps provided by the clan’s central government for use by the county resident representative office 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. This replacement of failed daimyō was called the kaieki.

Community Vitality in Medieval Japan 89

References Downing, Brian. 1992. The Military Revolution and Political Change: The Origins of Democracy and Autocracy in Early Modern Europe. Princeton, NJ: Princeton University Press. Fujiki, Hisashi. 1985. Toyotomi Heiwa-rei to Sengoku Shakai [The Toyotomi Peace Order and Sengoku Society]. Tokyo: Tokyo Daigaku Shuppan-kai. Fujiki, Hisashi. 2005. Katana-gari [The Sword Confiscation]. Tokyo: Iwanami-shoten. Fujino, Tamotsu, ed. 1977. Kyushu kinseishi kenkyu soso, Kyushu to ikki [Medieval Kyushu Historical Documents, Kyushu and Ikki Movements]. Kumamoto: University Historical Documents Project. Inaba, Tsuguharu. 2001. Mura no Buryoku-doin to Jinbu-yaku [Mobilization of Peasants 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 [Association for the Study of History]. Tokyo Aoki-shoten. Inaba, Tsuguharu. 2003. Heino-bunri to Shinryaku-doin [The Separation of the Warrior 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: The Unification of the Country and the Korean Invasion], edited by Susumu Ike. Tokyo: 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 [Research Documents on Local Organization in the Warring States Period]. Kumamoto University Liberal Arts Division, Historical Research Group working papers. Maeda, Nobutaka. 1997. “Gobikin no kenkyu oboegaki” [Memos from Local Accounts], in Shishi kenkyu Kumamoto [The History of Towns in Kumamoto], vol. 8, pp. 43–69. Maeda, Nobutaka. 1998. “Zoku gobikin no kenkyu oboegaki [Memos from Local Accounts], Continued,” in Shishi kenkyu Kumamato [The History of Towns in Kumamoto], 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 Political Mind of People in the Modern Era]. Tokyo: Yoshikawa Kobunkan. Nakano, Hitoshi. 1996. Toyotomi-seiken no Taigai-shinryaku to Taiko-kenchi [The 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 Eighteenth Century], in Kumamotoshigaku [Kumamoto 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.

5

“Advance and Be Reborn in Paradise . . . ” Religious Opposition to Political Consolidation in Sixteenth- Century Japan Carol Richmond Tsang

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 Torigoe, 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 resistance 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 fortified temple of Torigoe 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. The fortresses were to change hands two more times in ever more ferocious fighting, until Oda Nobunaga with a vast army of tens of thousands finally prevailed in 1582 to overwhelm the 300 barricaded resisters, who had continued to fight despite their leader’s capitulation 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. The backbone of religious resistance to political unification was broken, and within a few years Oda Nobunaga’s successor, Toyotomi Hideyoshi, was undisputed ruler of Japan. The last great bastion of warrior-monks at the Mount Kōya temple voluntarily handed over their cache of arms in 1588 without a fight. This chapter documents the rise and fall of the fascinating phenomenon of politically mobilized religious movements in medieval Japan that resisted, but 91

92 Carol Richmond Tsang

ultimately failed, to block Oda Nobunaga’s unification 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 physical protection and economic management in the medieval period. My concern 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 fighting Oda Nobunaga’s forces. “Advance and be reborn in Paradise, retreat and go instantly to hell” read the banner that led fighters to war against Nobunaga under the leadership of Honganji Temple. Honganji, along with its members, is best known for this fierce 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 finally vanquished by Oda Nobunaga’s forces in 1574. The sect’s resistance and history of militancy contributed to some of the most visible changes in Japan’s government and social organization, specifically 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. The 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 after the initial 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. There 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. The military prowess that Honganji’s sectarians displayed in all their efforts, however, impressed 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. The 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’ fighting leagues were known, were arguably the most powerful and visible expression of commoners’ struggles in the sixteenth century. There 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 difference, indeed the defining characteristic of ikkō ikki, was the dominance of Honganji sect members. This characteristic lent the various struggles a deceptive consistency that they, in fact, lacked. The ikko ikki leagues’ aims varied widely, as did the leadership. Honganji Temple sometimes did direct and coordinate the ikkō ikki’s activities, but not always. Nevertheless, 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. The perceived ideology associated with ikkō ikki also differentiated 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 buddha named Amida was the only effective means of achieving Buddhahood. As nearly as we can tell, most members of the Honganji sect understood the theology to mean that no gods (kami) or buddhas other than Amida had any real power over them. This 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 objects. This behavior, naturally, increased the adherents’ notoriety and furthered 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 fighting involved some aspect of religion: support of the Honganji branch’s leader, resistance to religious oppression, or defense of local temples. More importantly, what held the members’ fighting organizations together was their shared religious identity, an identity that delineated a clear boundary between Honganji followers and those of other Buddhist sects.

94 Carol Richmond Tsang

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. Through his grace, after death one is reborn in his Pure Land paradise, where one can attain Buddhahood effortlessly. Understandably, for most followers being reborn in paradise sufficed as an incentive to worship Amida, and the ultimate achievement of Buddhahood was mentioned 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. The devotion 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 example, 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 differed from those of other Amidist sects in that they identified selfless faith in Amida as the mechanism by which rebirth in paradise was assured. Traditional methods of achieving merit toward reincarnation on a better level, such as reading and copying sutras, or paying for religious ser vices, no longer applied. Thus, 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 fluid and multiple, and its importance varied depending on context.

“Advance and Be Reborn in Paradise . . . ” 95

That 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 distinctions of other kinds of status. Rather, the religious glue that held the ikkō members together was of a different kind. Loosely speaking, they did share a religious status, which laid the foundation for the formation of ikki, which required that some kind of status be shared by league participants. The 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 Buddhahood had lost their efficacy, Honganji propagated an exclusive, and therefore exclusionary, faith. To most believers, no one outside the Honganji faith was destined for the Pure Land or for Buddhahood. This was a radical departure from orthodox Buddhist teachings, which hold that there are many effective 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. They 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 individuals 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 organization 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 administrative 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 difficulties on a larger scale. Indeed, with the endemic warfare of that

96 Carol Richmond Tsang

time, warriors used villagers to supplement their armies and offered 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 recognize the others’ territorial boundaries. They would also pledge to come to the others’ assistance in the event of attack from without. None, however, transcended 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 involve adjacent regions. Ikkō ikki might draw participants from distant provinces, 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 apparently drove them, and the long-term effects of their activities. Not all of the leagues’ military actions were rebellions, but the most striking of these struggles were, specifically those in Kaga and Mikawa provinces, and that against Nobunaga. The ikkō leagues had their greatest success in Kaga province, the southern part of modern-day Ishikawa prefecture. At first, in 1473, an ambitious warrior 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 fteen years. It is worth noting that his rival had the support of a different branch of True Pure Land supporters, who often took whatever side Honganji members opposed, and vice versa. In office this warrior worked to restructure the administration 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 unified 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 different villages, and each village might have parts owned by several different entities. The cultivators did not own land as much as cultivation rights, and they might owe rents to more than one landlord for different fields. Warriors generally had the right to collect taxes and rents on a particular 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 cultivators ownership of the lands they tilled.

“Advance and Be Reborn in Paradise . . . ” 97

Local temples and sometimes warriors often had holdings over which they were able to exercise largely autonomous authority. They had the right to bar entry to warrior officials, a right that amounted to exemptions from numerous taxes, and the responsibility for investigating and punishing criminal activity within the temple precincts. The daimyō also overrode these rights, taking direct control of all the affairs of the province. This became a problem for Honganji sect members, as in Kaga and elsewhere 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 specifically conferred by the governor. As one chronicle put it, “new priests and temples sprang up here and there . . . [declared] themselves 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 official sanction. For a daimyō, it was worse. There was no room in a daimyō domain for autonomous territories governed by people outside his chain of command. Although there is no specific 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 retainers. In 1488 many of these retainers joined Honganji sect members and attacked the military governor. After a short burst of battles, the governor’s castle fell to the rebels, and he committed suicide. The victors installed a less ambitious relative of his as military governor and began a new system of government in which warriors and ikkō leagues shared rule over Kaga. One contemporary 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 analysis of documents, however, it seems they played a role along with numerous other actors on the political stage of the province.

98 Carol Richmond Tsang

It appears that there was little effective control over warriors and the collection of rents in the provinces after 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 involvement 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 ticular 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 daimyōlike domain but rather maintained the former system of overlapping authorities on a local level, where other temples and aristocrats had jurisdiction over estates they held. Indeed, rather than try to replace the old, complicated system of land ownership, 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 coalition 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 unified leadership, or at least a pyramidal organization, confers distinct advantages. In the case of invasion, all the different groups—leagues, different warrior bands— might fight the invader, but they will be much less effective if there is no overall coordination. In addition, different groups within the coalition might back different national military figures, resulting in provincial disorder. Another factor was that, almost by definition, league and local rule required a looser central authority than the new daimyō domains afforded, whether it was exercised in a single province or writ large over the whole country. A domain made for a more efficient administration and allotted greater power to the daimyō, so it was an advance in political organization. Thus, while leagues potentially represented a new form of provincial government, in another sense their interests were more conservative than radical. They represented a new way of governing locally, yet required the old way of governing nationally. The daimyō domain, and later the unifiers, 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. That province, like

“Advance and Be Reborn in Paradise . . . ” 99

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. After establishing himself in western Mikawa, he turned east, where three Honganji sect temples had bustling temple towns with rights to bar entry to warriors. These rights conflicted with Ieyasu’s ambitions as well as his ability to protect his retainers when they suffered at the hands of townspeople, because the temples alone had the authority to punish wrongdoers. There 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 townspeople 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 opposed 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. Those of Ieyasu’s retainers who held lower positions in his organization split cleanly along religious lines. This must have come as a shock, because before this, religion had played little role in warrior politics. In an earlier conflict in Mikawa, members of the same temple congregation fought on different 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 fields, so fields they shall be.”  He had them razed and banned the Honganji sect temples in his domain.

100 Carol Richmond Tsang

The 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 Honganji’s independence and military power for its continued existence. Thus, when Oda Nobunaga threatened the main Honganji Temple’s independence, the members fought all the more vigorously in the sect’s defense. In 1568, Nobunaga rode into Kyoto triumphantly, in a position to dominate the national government and impose his rule, gradually, over the rest of Japan. At first Honganji, among others, welcomed and congratulated him; the Patriarch Kennyo sent him gifts just before Nobunaga entered the capital, and more within a few months, at the new year. The 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 magnificent temple buildings and a large temple town ( jinai machi) surrounded by walls and moats.”  Even though Honganji claimed the tax exemptions and autonomous rule enjoyed by most temple towns, Nobunaga demanded that the temple 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. This struggle, which would continue off and on for ten years, constituted the largest of the ikkō leagues’ fights and one of the most persistent 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 warriors helped to provision the temple from the sea. The 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. The temple and town were also burned to the ground, though it has never been clear whether the priests or the townspeople set the fire to deny Nobunaga its assets, or whether

“Advance and Be Reborn in Paradise . . . ” 101

Nobunaga ordered that it be burned. Nevertheless, he did not outlaw the sect as Ieyasu had in Mikawa. If the independence 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 significant 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 messenger 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 Nobunaga could have allowed this system of autonomous towns to continue, and Honganji’s independence, with its military past and many members throughout central Japan, had to be destroyed. The geographic spread of Honganji’s power meant that it took Nobunaga years to finally conclude the war decisively. He attacked and defeated Honganji sectarians methodically, moving from one geographic region to another. Some places put up spirited resistance. For two years, Nobunaga underestimated the amount of force necessary to defeat the adherents in Ise province, at Nagashima. Nagashima was a community built on a swampy delta at the confluence 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 realize 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 invading 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 between 1571 and 1574 finally to subdue—and massacre—the members. Massacre was Nobunaga’s preferred method for dealing with Honganji fighters, 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

102 Carol Richmond Tsang

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 significant documentation from the Honganji side. The 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 fighting written by a temple’s supporter, have emerged. In the case of the opposition 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. The first call to arms that the leader, named Kennyo, sent out to his followers 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 founder, 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 predecessor once wrote a note to console the families of some members who fell in defense of the sect, assuring them that “There is no doubt that the people who died in battle achieved rebirth in the Pure Land” because they had fought as the founder’s allies. Kennyo, however, did not go so far. Once, he told followers that their willingness to fight 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. This 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 members who failed to support Honganji and go to war. Ordinary members tended to believe that expulsion from the sect necessarily meant that the expelled member was condemned to hell. Strictly speaking, 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

“Advance and Be Reborn in Paradise . . . ” 103

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 after death. Only Honganji depended on Amida’s own “other” power, denying the efficacy of amulets, registers, priests, or even one’s will to faith. It should be no surprise that many followers apparently 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 offers multiple possibilities for reincarnation, on several different 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. Thus, Honganji followers believed the leader, in this case Kennyo, had the power to condemn people to hell. According to one of the sect’s sixteenthcentury 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 Nobunaga 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. This, then, was the context in which the “Advance and be reborn in paradise . . . ” banner appeared. This simple ideology of the Honganji leader guaranteeing either salvation or damnation did not prevail over the entire era of ikkō leagues and their fighting. In most cases, including many that have not been discussed here, the motivating factor was survival, either of specific temples or of the sect itself. Although the Honganji sect taught obedience to the worldly authorities, no one could expect members to acquiesce to the destruction of their temples or the proscription of their religion by those authorities. In those cases, with or without the

104 Carol Richmond Tsang

encouragement of the sect leaders, the followers took up arms. In the case of Kaga, the justification for fighting 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 opportunity 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. They had little need of warrior protection, since they demonstrated time and again their own military abilities. They had little fear of repercussions because they believed their future lay in paradise. So much of Japanese religious belief at the time involved this-world benefits 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. This last tenet, combined later with the beliefs in the power of the leader, proved to be extremely potent as motivation and cohesiveness for the leagues. That 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 region, 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 benefits of a unified command trumped any benefits of league government. Because of policies adopted 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 separating warriors from villagers, often identified as farmers. They 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. The villagers were then deprived of some of their weapons, though they were never wholly disarmed. This policy was designed, in part, to pacify the countryside and to prevent feuds, or private wars, from breaking out in the provinces. It also

“Advance and Be Reborn in Paradise . . . ” 105

brought the warriors more closely under the supervision of their overlords. For Honganji members, this had the effect 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 difficult 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 independent economic and military power. The unifiers Oda Nobunaga, Toyotomi Hideyoshi, and Tokugawa Ieyasu in particular went about making sure that all temples relied on them for income and protection. Honganji’s resistance 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 commanded his men to take no prisoners but rather to kill any people they came across. The attack horrified Kyoto aristocrats, who saw in Nobunaga the enemy 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 organization to be so constituted, but one can argue that its authority over its followers demonstrated forcefully the advantages 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 organization could provide motivation, cohesiveness, and a structure to match that of the warrior organization. Ties of a common religion could and did overcome those of warrior loyalty, most vividly in the case of Mikawa and Ieyasu’s retainer band. As noted earlier, the ikkō leagues were not the only religious organization that displayed this tendency. The 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. The Nichiren followers, however, came primarily from the urban classes, not from the village elite and warriors that Honganji drew. They did not, then, threaten daimyō nearly so much as Honganji did. No other Buddhist sect cost the warriors anywhere

106 Carol Richmond Tsang

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 Christianity and its teachings from Japan. The 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 important 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 organizing 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. The ikkō leagues’ greatest 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 officially in 1530. See Tsang, 2007: 94ff. One must note that, as described above, the fighters 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.

“Advance and Be Reborn in Paradise . . . ” 107

8. These taxes were known as hanzei. See Tanaka Katsuyuki, 1993. 9. For a more detailed discussion of the right to bar entry to warrior officials, 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. The 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. Then again, as Nobunaga’s official biographer, author Ōta Gyūichi may not have wanted to make any of the ikki fighters 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 fire to it. 27. Eiroku ikki yurai, 1989: 666–667; Watanabe Chūuemon oboegaki, 1977: 83–94. 28. Kennyo shōnin bun’an, 1979: 1133–1178. 29. The 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: 1166 32. Jitsugo, 1977a: 586. He dates the beginning of this belief to the reign of Kennyo’s predecessor, 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 five military houses (Asakura, Asai, Takeda,

108 Carol Richmond Tsang

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.

References Adolphson, Mikael, and J. Mark Ramseyer. 2009. “The Competitive Enforcement of Property Rights in Medieval Japan: The Role of Temples and Monasteries.” Journal of Economic Behavior and Organization 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. Okazaki: 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 [The Sword Hunt: The Common People Who Sealed Up Their 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 [The Lotus Disturbance of the Tenbun Era: The 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 Kitanishi 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” [The 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 Thomas Kirchner. In Rennyo and the Roots of Modern Japanese Buddhism, edited by Mark L. Blum and Shin’ya Yasutomi. New York: Oxford University Press. Lamers, Jeroen. 2000. Japonius Tyrannus: The Japanese Warlord Oda Nobunaga Reconsidered. Leiden: Hotei Publishing. Muromachi bakufu monjo shūsei, Bugyōnin hōsho. 1986. 2 vols. Kyoto: Shibunkaku shuppan.

“Advance and Be Reborn in Paradise . . . ” 109

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 [The Story of an Ikkō Ikki]. Kanazawa: Konishi Publishers. Shinchō-kō ki. 1965. In Sengoku shiryō sōsho, vol. 2, edited by Kuwada Tadakichi. Tokyo: Jinbutsu Ōraisha. Shingyō Norikazu. 1975. Ikkō ikki no kisō kōzō: Mikawa ikki to Matsudaira-shi [The Basis of the Ikkō Ikki: The Mikawa Ikki and the Matsudaira Family]. Tokyo: Yoshikawa 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 Villages 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. Japanese Fortified 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.

6

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

(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. The main river crossing the province, the Nabarigawa, snakes down between impressive gorges. The 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 descending 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 border 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. Those regions, which were only a two or three days’ walk from Nara and Kyoto, the former Japanese capitals (and today, less than an hour from Osaka by express train), were not located on main trade routes and thus never constituted a particularly important political or strategic stake during medieval and ancient times. In the absence of centralized political control the arable lands in the region quickly fell under the influence 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 fifteenth century, the region was held under the official domination of two families, the Nikis and the Tsutsuis, warlords whose names did not become famous. Sometimes pilgrims and im-

LOCATED EAST OF YAMATO

110

Autonomy and War 111

perial messengers en route to the great shrine would take a secondary road to go from Kyōto to Ise. That 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 different. 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 surrounding hills. The people from the region called those extremely narrow paths koguchi 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. The 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. There was no crossing the Iga region without the approval of local authorities, for huge bands of ruffians 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 communities would not survive. In 1494, in the anarchic early years of the Warring 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. The documents were of an explicitly political nature, but given the overwhelming political influence of the Buddhist temples on the inhabitants of the land, this is not surprising. The first of the Oyamoto declarations represented a kind of local constitution establishing new rules for local life. The five-article document, established on the fifteenth 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, declared, “The hyakushō shall not strife for paddy fields, mountains or forests. They shall not seize cultivation rights or steal.” This declaration might be thought of as a code of conduct for the farmers who signed it, rules each person pledged to obey in the name of the collective security.

112 Pierre Souyri

The second document, dated a month later, on the twenty-first 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 fight over taxes on peasants and to try to prevent insurgent acts by subordinated people who might “cause trouble.” The charter’s intention was stated unequivocally. One of its articles read, “If anyone acts badly, inside or outside Oyamoto, he will be judged and sentenced.” These forty-six citizens of Oyamoto were the leading class in the countryside. This 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 organization with the clear aim of maintaining order within the domain. The league of forty-six families was created five weeks after the league of the “common people,” after both had gathered and sworn their pledges. The warriors’ league seized administrative and judicial control over the estate, its authority based on the charter signed by the 350 peasants. However, most important 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 imperiled. Both groups of pledges had many shared priorities, but the most important, 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 conflicts over land and access and not that of external aggression. There were no wars, as such, at the time. These leagues appear to have been designed to establish expectations of local cooperation and stability. Through these organizations, a double structure of power began to develop: the assembly of the forty-six local warriors and the general peasant assembly. The “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 first to organize into this sort of social arrangement, as far as we know. The 1 to 7 size ratio between the two groups suggests that the process of social differentiation was limited, although, on the other hand, the very existence of the two groups suggests that the social identity of the two groups was clear for all parties concerned. The two groups had a relationship predicated on power and domination, but without either a suzerain or absolute power. This

Autonomy and War 113

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 decades 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 independent fashion from outside hierarchical control, at least at the political and judiciary levels. The 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 collective local organization 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 organization 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 orga nization similar to that in Oyamato appeared. We do not know exactly when this orga nization was created, in part because its name and “constitution” emerged only when the orga nization was directly threatened by the power of strong lords of the neighborhood. The communities of the Iga River Basin had been defending themselves since around the beginning of the sixteenth century. The neighboring region of Kōga in Ōmi, similarly organized, had no fewer than 230 fortification works. Residents of Koga had a long history of collectively managing the water of the Yasugawa, a river that was often difficult 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 organization was indispensable. Formal consolidation of the Iga communities seems to date from around 1560, when the locals drew up a formal twelve-article constitution. The 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 decisionmaking sessions in the nearby Buddhist temple of Heirakuji. These local warriors collectively administered the territory. The Iga commune seems to have been a sort of geographic extension of Oyamato’s political and social structure, although it sprang up on a larger

114 Pierre Souyri

scale than that of its neighbor. The ruling body of the Iga was, in effect, no longer that of the domain but that of the province. This organization 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 Yamanaka 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 decades later, but the community was prepared to put up a fierce collective resistance. We can surmise that the particular 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 Oyamato. 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. This league was led by local warrior families under the rubric of a two-level structure somewhat resembling that of Oyamato. The military strength of the league was strongly tied to local warriors’ capacity to train common people and to mobilize them for battle. The league could confer the status of samurai upon drafted peasants (called ashigaru, “light feet”) if they fought well on the battlefield. This 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 acquisition of a fief or a land—for one thing, one might well inquire as to whom this fief 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 ideograms and blazons, and the prestigious collective authority they enjoyed. Being 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 battlefield was considered a supreme honor. Conversely, the league could also dismiss members of the organization who refused to obey or to fight. 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

Autonomy and War 115

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 different 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. The experience of villages occupied by invading warlords was often nightmarish: as a result, their people often fled to fortified 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 often lead to extortion and rent-seeking. It is thus very clear why in local collective organizations, 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 permanent since the end of the thirteenth century. After the Ōnin War (1467–1477), with the authority of the weakened Ashikaga shogunate in tatters, Japan entered a new period of anarchy and disorder, the Sengoku era (literally, “country at war”). This 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 warlords, the daimyō. As sovereigns over their lands, these warlords battled one another to maintain and expand their fiefs. 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 expansion of daimyō power was a threat from above, but both contributed to the widespread violence and mayhem that ruled the day. The Ōnin War represented the end of the old estate system and of the heyday of the major lords (shugo), who emerged from the fighting greatly weakened. Most of the existing great warrior clans dispersed, torn apart by inner conflict 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 shugo-dai, 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 monasteries 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. This alliance, which in the sixteenth century would eventually give way to a local league of fighters, 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. The 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 entity to the league of Iga, the Koga organization was called Kōgagun chūsō, “General Assembly of Kōga district,” and was a small but well-organized league founded on common principles of solidarity and understanding and led by minor 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 finds another example of the same twolevel system prevailing in Oyamato and Iga, as these villages banded together to form a “league of equal villages” (dōmyōsō). A district (sō), which most Japanese 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 existed in Kōga three levels of social and political organization: the now wellknown 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 organized. Organizational 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. These communities gained legal recognition and became their own autonomous units, overseeing

Autonomy and War 117

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. They formed protection leagues with other regional communes and, by reaching across old territorial boundaries, further contributed to the weakening of landowners’ domination over the land. These and other such alliances also had a precise military aim: that of preventing the military forces of outside lords—especially those of the daimyō— from entering the district and destroying the social organization the local warriors had created. To this end, different communes pursued different 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 administration.” In other regional leagues, the main objective was to force warlords to enter into seigneurial pacts that would provide checks on their authority. These 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, often 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. The 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. This military alliance between neighboring groups of local warriors, which protected local autonomy and independence from outside rule, was more important than narrow geographical loyalties and reflects the fact that, when economies of scale called for larger organizations, warriors reached laterally before leveling political status in their own communities. The relationship between local warriors and peasants was complex, and when conflicts arose between the two, village unions could fall apart. Sometimes 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 commoners 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. These local leagues, and the institutions they left behind, would survive at least until the 1570s, when the threat of outside invasion by warlords resurfaced. After 1568, Oda Nobunaga, with the help of a very powerful and wellorganized 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 political structures that took root primarily in backward and mountainous areas, would prove no match for Oda Nobunaga’s armies, whose guns and cannon fire crushed the small wood-and-earth castles of the Iga warriors. In 1579, Oda’s son, Oda Nobukatsu, launched a major attack against the independent 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 peasant fighters and local warriors. Oda Nobukatsu’s army had greatly underestimated the efficacy of the Iga people’s military tactics, the strategic advantage their local knowledge afforded them, and their capacity for organization and military mobilization. Oda Nobunaga would not let his territorial conquest remain unfinished, 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. This time, the warlord set the province awash in fire and blood to forestall any notion of further revolt. With a ferociousness reserved for the fiercest resisters, Iga was destroyed. Oda Nobunaga and his army put an end to the political 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 modernized military, the historical and political significance of these autonomous communes is not to be minimized. Regional communes of various sizes existed for different periods of time in Kinai, Omi, Settsu, Izumi, Tanba, and other provinces. The existence of these regional and provincial federations effectively frustrated, for a time, the attempts of any centralized power to control the provinces of central Japan. The Iga league of communes, in par ticular, lasted much longer than did the one in neighboring Yamashiro province. The reason may be the par ticular configuration of the area, a mountain basin iso-

Autonomy and War 119

lated by its geography and situated relatively distant from the major routes. Geographic factors may also have contributed to the strong communal sentiment and military leadership of the peasants by low-ranking warriors. Although differences in social rank between peasants and warriors were clear and well-established, they were not insurmountable, for the Iga commune also promoted heroic fighters, 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 independent power structures that trained commoners, thrilled by their new faith, for combat against outside forces. This 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” (yamabushi) practiced asceticism in the mountains under the influence of Shingon esoterism and Buddhism. Although this movement hardly constituted their shared common identity, the people of Iga or Kōga had no doubt learned valuable secret traditions and practices from the yamabushi. Some of these practices were connected to the art of fighting and of war and may have influenced 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 geographical isolation. Historically, Iga province remained inaccessible due to extremely poor road conditions and the inhospitable mountains that surrounded it. The subsequent 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 medieval Europe, local wealthy individuals formed organizations that federated

120 Pierre Souyri

into armies designed to defend the integrity of the local autonomy. In par ticular, 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. The territories sworn unto these pledges created alliances among themselves against outside forces. The Treaty of Waldstetten, signed on August 1, 1291, and long considered the foundational act of Switzerland, is simply one of these pledges. This is the Landsgemeinde, or Association of Free People of the Land. Subsequently the “Swiss Guards” would furnish mercenary troops to the rest of Europe 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 different from what existed in the Sengoku-era Iga province. In general, the local warriors’ sense of collective identity and identification as a specific group does not appear to be directly linked to the uniqueness of their devised solution to the security problem, to the specificity of the organization they had created, but rather to the suffering, violence, and repression they were forced to sustain. The people of Iga and Kōga fought desperately to maintain their political autonomy, their way of life, and their social status. Trained by defeated jizamurai, the inhabitants of Iga and Kōga used techniques of warfare and fighting that before them the bandits and ruffians of past centuries had used. They 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, fighting continuously and relentlessly, though their cause would ultimately prove futile. The 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. The order that they were trying to impose relied, on the contrary, on centralized vertical power structures. The violence they inflicted on local populations matched the severity of the perceived insult of the countryside’s resistance. Generally, local autonomous political 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 political system to supplant it. Still, it is worth observing that the Iga people acted not with a particular ideology in mind, but rather for the purpose of creating structures that would efficiently ensure order and freedom at a local level.

Autonomy and War 121

The people of Iga and Kōga essentially transformed the practice of espionage and hiding into a military strategy not only because they had learned to fight this way, but also because they had no other choice but to adopt guerrilla tactics. The emergence of the ninja tradition was in fact the result of a harsh defeat at the hands of Oda Nobunaga’s army, after which the Iga and Kōga warriors were forced to fight on against the new order in secrecy, using ninjutsu techniques. The social system had been destroyed by the new Japanese leaders, who moved quickly to suppress resistance movements. Such movements 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. The reluctance of the new authority to integrate the local jizamurai into the new system stemmed from the inability of these independent but newly defeated jizamurai to accept and integrate themselves into the hierarchical structures imposed on them. Moreover, their exclusion was exacerbated by their resistance: The 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 organization,” holdovers from a sort of microrepublic that had been subverted by the armies of the new monarchy. As to the real organization 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. They can be said to have been militias of warrior-peasants. From then on, the heirs of the defeated jizamurai of these areas were educated in the memory of a happy and free history. At the beginning of the seventh 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. The 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 uprisings or used them as spies. In collaborating as auxiliary troops to the repression 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. The most abundant source that has come down to us concerning these practices is a series of lists of objects detailing the purpose of their missions and the expenses they incurred: scissors, daggers, saltpeter, inflammable arrows, powder, salted and boiled fish, and so on. In oral stories told in these chronicles, one can find almost implausible tales of the special powers of the ninja. The 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 difficult 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 unquestioned. 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 subjects from his hometown, Kii province. .

.

.

So thoroughgoing has been the romanticization of the ninja tradition that few modern Japanese, let alone people outside of Japan, know of their origins as local 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 survivors (and not a few imitators, it must be imagined) in Iga and Koga sold their services 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 reputation 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 overwhelming odds hold fascination enough, unvarnished by mythology. It is intriguing to speculate about how Japan might have turned out differently had pockets of autonomy such as Iga and Koga held out longer against territorial centralization. Might separate states, some with remarkably flat

Autonomy and War 123

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 remained powerful enough to force the Tokugawa to grant de facto federalism to local warlords rather than to smaller, self-governing communities? Geography will only get us so far. However mountainous Japan’s topography might be, and however shielded from foreign invasion by surrounding oceans, competition 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 different than the lands of Europe on the other side of the globe.

Notes 1. Nonetheless, the people of Iga and Koga managed to invent a tradition of much longer duration than their own political 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.

References Arai, Takashige. Kuroda akutōtachi no chūseishi [The Medieval History of the Kuroda Ruffians]. Tokyo: NHK Books, 2005. Downing, Brian. 1992. The Military Revolution and Political Change: The Origins of Democracy and Autocracy in Early Modern Europe. Princeton, NJ: Princeton University Press. Fujiki, Hisashi. Mura to ryōshu no sengoku sekai [The 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 Battlefield: The 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. The World Turned Upside Down: Medieval Japanese Society. New York: Columbia University Press, 2001. Yamaguchi Masayuki. Ninja no seikatsu [The Life of the Ninja]. Yūzankaku, 1996.

7

Instruments of Change Organizational Technology and the Consolidation of Regional Power in Japan, 1333–1600 Thomas Conlan

Prologue: The Limitations of Sixteenth-Century Firearms Shortly after a chance landing in 1543 on the Island of Tanegashima, a small island south of Kyushu, Portuguese merchants parted with three of their firearms (teppō). A Negoroji priest visiting the region took one of them to his temple, located in central Japan. The 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 proficiency in using and producing these weapons, their role in disseminating firearms 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 effectively. Oda Nobunaga, the first of the “three unifiers” of Japan, has been characterized as a military genius whose concentrated use of firepower allowed him to “revolutionize” warfare, crush his most potent rival, the Takeda of Kai province, and consolidate power from 1570 until his assassination in 1582. The 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 fielded a formidable squad of gunners through mastery of production and training. And when Nobunaga dispatched his brother to attack Negoroji, his army suffered a convincing defeat. Unlike the “epochal” encounter at Nagashino, this battle has been consigned to oblivion, largely because Nobunaga’s successor, Toyotomi Hideyoshi, incinerated most of the temple complex in 1585. 124

Instruments of Change 125

Those most proficient in manufacturing and using guns were not destined to achieve political 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 sufficient manpower to defend their territory or, for that matter, aid a beleaguered ally, whose castle fell to the Oda before their gunners could arrive. Negoroji’s 300 marksmen were incapable of defeating an organized and determined adversary, which accordingly suggests that the innovation or adoption of new weapons did not determine political and military success. Instead, the ability to mobilize, sustain, and supply armies proved to be of paramount importance. This chapter will explore the transformational power of technology and show that, like the priests of Negoroji, historians have tended to overemphasize the significance of new weapons. The introduction of firearms did not unleash a process whereby fragmented authority was centralized (or “unified”) during the final decades of the sixteenth century. Instead, organizational changes in fourteenth- and fifteenth-century Japan provided the impetus for the consolidation of regional political and military power. Technology is most significant, and best understood, as a technique as opposed to an instrument or weapon. To date, technology has often been conceived in material rather than organizational terms. Such a view is understandable, for it is easier to point to improvements in particular objects than to uncover the process through which they came to be effectively used. Nevertheless, the notion of material objects (“technology”) as being capable of influencing historical processes arose in relatively recent times. The oldest (1615) use of the word in English designated a treatise on arts or skills, and by the midnineteenth century it came to represent a particular 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. The impact of technology, conceived of as the creation or use of new materials, was not fully perceived until the carnage of the First World War, when military historians first discussed the importance of weapons such as firearms and pikes in the “development of modern warfare.”  Since the 1920s, this notion 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 metaphor of a “military revolution” to describe the changes wrought by improvements in firepower. Nevertheless, the conception that “technological advances” such as the adoption of new weapons profoundly influenced the waging of war is absent from pre-twentieth-century writings, which instead concentrate on issues of military organization and supply. This earlier understanding remains germane, for changing patterns of weapons usage reflected historical processes 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 fifteenth and sixteenth centuries. The effective use of these new materials hinged upon improvements in military and political organization. In this sense, the invention of weapons proved less important than improvements in techniques for mobilizing, training, and supplying armies in the field—techniques that will hereafter be referred to as improvements in organizational technology. Men who rose to power in the sixteenth century, such as Nobunaga and Hideyoshi, fully mastered these arts, while the priests of Negoroji did not. Sources When charting how new weapons came to be adopted and used, one can postulate how they influenced the processes of social, military, and institutional change. Unusually precise records describing how wounds were inflicted in battle from 1333 through 1600 enable us to trace the dissemination of weapons. One can precisely chart how weapons were used from 1333 onward because battle reports (kassen chūmon) and “petitions for reward” (gunchūjō) record how wounds were inflicted. These documents first appear late in the thirteenth century, in the aftermath of the Mongol invasions, and continued to be produced through the battles of the early seventeenth century. Written shortly after 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 relatively rare, for they were generally discarded after they had been summarized in petitions for rewards. Once a warrior completed his battle report, he used it

Instruments of Change 127

as evidence to prove his military ser vice and thereupon submitted a petition for rewards. These documents recount all wounds, deaths, and damages inflicted, as well as the date and location of battle. 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. After 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) because of his military ser vice. On the seventh day of the past fift h month, when the gate of Kaseda castle in Kimotsuki district, Satsuma province, was stormed, [Dōkaku and his] son were first to attack. [They] surmounted the moat and cut through the barricades. As [they] performed military valor with fearless abandon, [Dōkaku’s] son Yasaburō Tamotsu was shot through the left thigh [and later had the] arrow 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 service in order to promote the honor of [a practitioner of] the bow and arrow. So humbly stated. Sixth month 1336 Received (copy of the monogram of Shimazu Sadahisa)

This document provides a brief narrative of battle. Dōkaku described the wounds of his son and bannerman—all caused by arrows—and named witnesses for his deeds. He must have originally submitted a battle report that mentioned these casualties in greater detail, but this document no longer survives. 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 services, 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 paucity of military records reflects changes in social and military organization more than the prevalence of warfare per se. As petitions for reward functioned 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 retainers would no longer be able to demand rewards for their military ser vice, and so stopped submitting these documents. Only a dwindling band of warriors maintained autonomy and continued writing battle reports throughout the fifteenth 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 documents 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 ticular, than in the rest of Japan. Similarly, more western warriors managed to preserve their autonomy throughout the political turmoil of the fifteenth and sixteenth centuries than did their brethren in central and eastern Japan. The surviving military reports of the fifteenth and sixteenth centuries reveal less about the nature of warfare than fourteenth-century records. In contrast to earlier petitions, fifteenth-century records rarely mention the names of witnesses or where a particular battle was fought. As armies became more cohesive 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 particular battle. Thus, the very processes that led to the improved ability of armies to mobilize troops and secure supplies caused the historical records 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 fight being randomly distributed, such invitations became unnecessary as the fifteenth century progressed, and armies came to be composed of most, if not all, warriors from a par ticu lar region. Once this process 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 fifteenth-

Instruments of Change 129

century troops. This is unfortunate, for as we shall see, the period of the greatest innovation in military organization is one in which virtually no organizational records survive. The following document, submitted by Kikkawa Mototsune of Aki province in the aftermath of a Kyoto battle, typifies fifteenth-century reports: The 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 killed same name Asaeda Magotarō same name Asaeda Matajirō same name Asaeda Magogorō Yamagata Mago Saemon no Jō: a pike (yari) wound Wada Saburō Saemon no Jō: a pike wound [and] an arrow wound Ono Yaroku same as the above Miyoshi Saemon Tarō Saeki Shoroku a pike wound Tahara Tosho no Suke a pike wound Kawayoshi 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 Katsumoto, 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 received a report that your retainer (hikan) Yuasa Yajirō was killed, and, in addition, 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 after receiving several petitions for reward, fifteenth-century generals issued documents of praise shortly after 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 recognized 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 characterizes 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 fifteenth- and sixteenth-century warfare. A Statistical Survey of War Because fifteenth- 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. These 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 fifteenth- 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 earlier documents. Fourteenth-century data on wounds by weapon are presented 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 important 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 inflicted the remaining 27 percent of fourteenth-century and 25 percent of fifteenth- and sixteenth-

Instruments of Change 131

Table 7.1 Fourteenth-Century Wounds by Weapon Period

One (1333–1338)

Two (1339–1349)

Three (1350–1355)

Four (1356–1394)

Total

Arrow

 ()

 ()

 ()

 ()

 ()

Sword

 ()

 ()

 ()

 ()

 ()

Pike

 ()

 ()

 ()

 ()

 ()

Rock

 ()

 ()

 ()

 ()

 ()

Total

 ()

 ()

 ()

 ()



Source: The 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 records in comparison with other time blocks. Figures below 1 percent were rounded up.

century wounds. The 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 inflict 58 percent of all such wounds through 1600, even though firearms were introduced to Japan in 1466. Guns did not displace bows until 1600, when they inflicted 80 percent of all skirmishing casualties. Warriors tended to fight in close quarters after 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 figures may not seem particularly remarkable, they reveal that handheld weapons inflicted 30 percent more wounds during the years 1333–1338 than was typical for the rest of the fourteenth century (35 percent compared to 27 percent). Not surprisingly, this same period exhibited a disproportionate share of fatalities. The 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 inflicted 28 percent more wounds in 1600 than was typical for the years 1467–1600 (32 percent to 25 percent). (For a list of casualties 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 outcome of wars. The 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

132 Thomas Conlan

Table 7.2 Casualty Lists, 1467–1600 Document date

Collection

..

Kikkawa



..

Kikkawa



..

Kikkawa

..–

Kikkawa

..

Kikkawa

..

Mōri

..

Mōri

..

Kobayakawa

..

Miura

..

Kobayakawa

Arrow

Gun

Rock

Pike

Sword Killed

 



 

















 

 

Kikkawa

..

Kikkawa

..

Migita Mōri



..

Tagaya

 

 



..

Total











  















 

..

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



 



..







 

 



 

Instruments of Change 133

Table 7.2 (continued) Document date

Collection

..

Ōtomo

Arrow

Gun

Rock

Pike

Sword Killed



Total 

..

Yano

..

Kobayakawa





..

Kobayakawa



..

Migita Mōri



..

Migita Mōri

..

Dewa

..

Katsumata



..

Iwami Kikkawa



..

Amano



..

Hiraga



..

Iwami Kikkawa

..

Miyoshi

..

Mōri

.. .. ..

Kikkawa



..

Yuasa



..

Ōtomo







..

Ōtomo







..

Ura







Kusakari

.. ..



 









 





 

 

Kikkawa







Irie







..

Ōtomo





..

Ōtomo





..

Ōtomo

..

Ōtomo





..

Sugi



..

Tōshima

..

Ōtomo

 



 



 









  



  

























 











 





 

 

Sugi







..

 







..





Dewa

Irie

 

Asonuma

Ōtomo

 

 



..

 



..

 

 



(continued)

134 Thomas Conlan

Table 7.2 (continued) Document date

Collection

..

Ōtomo

.. .. ..

Ōtomo

..

Ōtomo

..

Tōshima





..

Ōtomo







..

Ōtomo







..

Ōtomo







..

Kodama

Arrow

Gun

Rock

Pike









Ōtomo









Ōtomo



















..

Ōtomo

..

Ōtomo







  











 

Ōtomo





..

Ōtomo





..

Ōtomo





..

Ōtomo

..

Ōtomo

..

Ōtomo

 



 *





 

Ōtomo

..

Ōtomo

..

Ōtomo



..

Ōtomo



..

Ōtomo









..

Total





..





 





 

..

Ōtomo



..

Ōtomo



..

Kikkawa

Total

Sword Killed



























 



















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 records. The 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 records, 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. The 1549 Hiraga record also appears in ibid., Hiraga ke monjo doc. 169, pp. 628–633.

Instruments of Change 135

Table 7.2 (continued) Yamaguchi kenshi shiryōhen, Chūsei, no. 2 (Yamaguchi, 2001), pp. 330, 386–387, 732, 943, and 945 contains the Kutsunoya, Reisen, Katsumata, and Kodama records. For the Masuda records, 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 suffering multiple wounds by different 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. Thus 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.

marginal importance. But his generalization cannot account for why warriors fought so fiercely during the years 1333–1338, 1467–1477, and 1600. Rather, one can surmise that a distinction existed between skirmishing, aimed at harrying 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 fifteenth 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 after 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 inflicted 7 percent of all such wounds in the fourteenth century, caused 80 percent of them from 1467 until 1600. This preference intensified 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 after 1467, swords maintained a cultural and linguistic significance long after they ceased to be militarily important. Commanders continued to deploy the

136 Thomas Conlan

phrase “coming to blows with swords (tachi uchi)” to describe combat between groups of pike-wielding soldiers. Although the military significance of swords proved negligible, their cultural value remained. Indeed, a certain mystique arose concerning them with the seventeenth-century establishment of an enduring peace in Japan.

Japan’s Fifteenth-Century Transformations The adoption of new weapons did not cause tactical change. The 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 effectively, 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 confrontation with pikes (yari awase senshi) in central Japan. And as we shall see, improvements in military organization ensured that pikes would become the favored weapon for foot soldiers from 1467 onward. Changes in Military Organization, 1392–1467 During the seventy-five years separating the final battles of Japan’s fourteenthcentury conflict and the onset of the Ōnin War, Japan witnessed a shift from loosely organized armies to semipermanent regional units. As only a few sporadic 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. The armies of Japan’s fourteenth and early fifteenth century represented little more than clusters of bow-wielding skirmishers scattered among small bands of horsemen. These foot soldiers were vulnerable in open spaces, for horsemen could easily charge and shoot them with arrows, and instead preferred fighting in inaccessible terrain, or in towns and villages, where they could easily hide and fire arrows at enemy cavalry. When encountering such a force of skirmishers, horsemen 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 significance of cavalry proved greater than

Instruments of Change 137

aggregate numbers would imply. A few of these highly trained horsemen could decisively defeat a larger number of scattered skirmishers. Armies were cobbled together from these warrior houses, and the ability to entice these men into alliances underpinned military power in fourteenth-century Japan. Early fifteenth-century petitions, drawn from eastern Japan, are stylistically indistinguishable from fourteenth-century petitions, except for inferences to warriors from a single region fighting 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 fifteenth century progressed, units tended to be identified by their provincial origins. As evidence of this we see warriors from the provinces of Musashi, Kōzuke, and Shinano fighting as cohesive forces in 1423, and by 1440, generals commanding troops drawn from a single province—again Musashi and Kōzuke—were perceived as being normative. Surviving records 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 fortifications. 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 fifteenthcentury horse wounds, a trend consistent with the latter decades of the fourteenth century. That horses continued to be slashed reveals that horsemen continued to charge through infantry formations, just as they had in the fourteenth 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 battlefields. Improvements in Provincial Political Organization, 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 effectively. The 1351 promulgation 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 provinces. These taxes sometimes assumed extraordinary proportions, with the Ōyama estate in Tanba province supplying 755 laborers for their shugo Yamana Ujikiyo in 1390.

138 Thomas Conlan

The 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 regional origins, the armies of the late fourteenth and early fi fteenth centuries were better orga nized and less dispersed. Already as early as 1355 armies tended to be based more on warriors drawn from a specific area, and fighting under their shugo’s command, than had been typical during the years 1333–1338. Shugo taxation remained onerous even after the wars of the fourteenth century drew to a close, as levies for horses, messengers, and workmen continued 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. The powers of the shugo post and its limitations are both evident in the case of Tanba, where Hosokawa Yorimoto attempted to dominate this province after 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 influence through his deputies, who legitimated their actions under the broad powers of the hanzei. These men attempted to extract wealth from their province and its nominally immune estates by conscripting laborers and levying taxes for shrine repairs and enthronement ceremonies, but they met with limited success. Shugo had difficulties establishing powerful provincial lordships despite the advantages offered by the hanzei. Tanba warriors such as the Nakazawa resisted Hosokawa authority and continued occupying lands that had been nominally confiscated. The Nakazawa’s struggle continued unabated from the waning years of the fourteenth century through the mid-fifteenth century. Disputes erupted sporadically, and even as late as 1445 the Nakazawa still occupied these contested lands in spite of repeated orders to desist. Shugo powers coalesced when local warriors became incorporated into their regional network. The contrast between the Kasai (first 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 after the Kasai in 1431, proves illuminating. The Kasai’s tenure lasted little more than fifteen years amid simmering resistance from other Tanba warriors and

Instruments of Change 139

estate residents. On July 24, 1431, Kasai Motosuke was criticized as being “unprincipled” (ineffective?) and divested of his deputy shugo post in favor of Naitō Motosada. Motosada collected more taxes in his first year than Kasai Motosuke had during the final year of his appointment. Motosada even appears 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 confiscated by Hosokawa retainers who monopolized the position of estate manager. Once shugo delegated their authority to local warriors, their network became better able to procure revenue. This ability to govern locally proved integrally 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. The Hosokawa were able to control and mobilize most of Tanba’s wealth, and nearly all its landed income, by leveraging the appointment of their retainers to both “public” offices, 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. This trend continued as the fifteenth century progressed. In 1482, the Nakazawa came to “manage” the Ōyama estate, after which the absentee proprietor no longer received income from these lands. The example of Tanba reveals that shugo and their deputies became increasingly skilled at administratively dominating their provinces. With their consolidation of power over all major Tanba offices 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. The shugo’s monopolization of provincial administrative offices facilitated the development of regional warrior networks, which in turn formed the nucleus of fifteenth-century armies.

140 Thomas Conlan

Improvements in political organization and the ability to extract revenue manifested themselves in the ability to forge and sustain regional forces. They did not, however, immediately translate into a shift of tactics. Several decades were required for shugo, or their deputies, to realize that formations of foot soldiers, rather than squadrons of cavalry, proved capable of dominating the battlefield. 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 O¯nin War: Tactical and Organizational Transformations, 1467–1477 For the first time in centuries, cohesive units of infantry, armed with pikes and capable of occupying contested ground indefinitely, were mobilized in 1467. However, these important organizational innovations have been overlooked by scholars, who have tended to focus on the rationales for the outbreak of the Ōnin War and its social and political consequences. Most monographs ignore both how the Ōnin War was fought and how its armies were supplied. Recently, Paul Varley has emphasized the importance of ashigaru, a fleet-footed force of light infantry, but posits the most significant changes as occurring during the Warring States era, where a “continuing technological revolution, reflected in the greater use of guns” transformed Japan. Organizational 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 indefinitely. Once troops trained together and mastered formations, they became proficient in using pikes. This allowed them to defeat horsemen on the open battlefield, a task that scattered bands of sword-wielding men could never accomplish. The 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 conflict. 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 measure of the Naitō’s success in mobilization that they denuded their province of warriors. When enemy Yamana forces invaded Tanba during the sixth month of 1467, they met with no resistance. Indeed, only one prominent Tanba family seems to have actively opposed the Naitō. Other records also mention the regional cohesion of the Naitō’s forces, naming them “Tanba armies (tanshū no onzei),” as military units were typically described in the mid-fifteenth century.

Instruments of Change 141

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 figures. No tax records survive during the years 1467–1477 for the Ōyama estate, but this gap in the sources reflects 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 dramatically 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. The post of shugo became the fulcrum of provincial power and the basis for regional organization, and yet the concentration of power inherent in this office served to destabilize Japan. Competition for the post exacerbated tensions that already existed both within and among shugo houses. Some provinces, such as Tanba, shifted 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 office. Thus, the inheritance—or lack thereof—of shugo offices 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 conflict. A violent dispute between two Hatakeyama candidates vying for a portfolio of shugo positions provided the spark for war during the fift h month of 1467. Rival shugo houses, such as the Hosokawa and the Yamana, came to support different Hatakeyama factions, thereby causing the fighting to spread throughout the capital. As all shugo lived in Kyoto, their residences served as nascent encampments for provincial troops and supplies. Each dwelling functioned as a base that had to be consolidated, or destroyed, in order for one coalition 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 coalition, which became known as the eastern army, managed to control the northeastern areas of the capital, where the imperial and shogunal palaces were located. The 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 Isshiki 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. The Yamana responded by demolishing the dwellings of Hosokawa partisans residing in western Kyoto, such as the home of Hosokawa Katsuhisa. Thereupon two large armies coalesced, dominating the northeastern and northwestern quadrants of the capital, respectively. Arson and fierce fighting characterized the onset of hostilities as each army attempted to create space for their horsemen 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. The Hosokawa controlled all the strategic palaces in the northeast but could not occupy southeastern Kyoto, nor could they make significant 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 outnumbered western forces. In order to press his advantage, he ordered the deputy shugo of neighboring provinces to reinforce the capital. The 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 between Tanba and Kyoto. Thereupon the “eastern” and “western” armies fortified 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. Nevertheless, as western reinforcements continued arriving via Tanba roads, the initiative 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 offensive for the first time. Surviving Kikkawa battle reports reveal that pikes were widely used during the autumn of 1467, where the brunt of fighting revolved around several large watchtowers. On September 13, 1467, six members of the Kikkawa family were stabbed by pikes, while on October 2–3, 1467, six more were so wounded in street fighting. Of course, not all of the Kikkawa wounds stemmed from pikes—eight more were caused by arrows, one by a sword, and

Instruments of Change 143

five by rocks—but these twelve pike wounds, inflicted over a period of three weeks, suggests the rise of new tactics, as tightly organized forces of pikemen came to dominate the battlefield. This tactical transformation becomes evident when recounting the western army’s offensive of the tenth month. On October 3, 1467, the western armies launched a rolling attack on the eastern army’s positions. Their offensive, centered 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. The 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 Yamato and Kii campaigns, routed a large force of western Rokkaku cavalry on this day with a compact squad of pikemen. After closing in on the Rokkaku horsemen at the burned-out grounds of Shōkokuji temple, an arena that favored cavalry by allowing for mobility, Masanaga’s pikemen surged into the enemy. The Rokkaku were broken, and sixty-seven were killed before their cavalrymen fled in defeat. The Hatakeyama played a crucial role in perfecting how to use pikes in formation. The “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. The battle-hardened Masanaga was confident 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. The “Chronicle of Ōnin” attributes Masanaga’s success to the fact that western pikemen proved unable to establish formations in the confusion of the Rokkaku flight. 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 formation 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 innovative use of pikes checked the mobility of horse riders even in these open areas. Although horsemen remained effective 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-fifteenth 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 inflicted relatively few casualties, units of pikemen were indispensable. These 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 Yoshinari, another veteran of the 1454 Yamato conflict, who entered the fray in the aftermath of the Rokkaku defeat. Thereupon each army hunkered down in its respective fortifications, 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 unprecedented 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 Western Front. Locked in a stalemate, huddled in trenches with few opportunities for offensive action, the warriors of Ōnin alleviated the intense tedium by writing Japanese poems (waka) or poetic phrases on small silk crimson flags, which they attached to themselves. Each army, unable to seize the offensive, viewed the other’s troop movements from a cluster of watchtowers, which became the focus of pitched battles. The 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. Thereupon each army used flaming arrows and rocks to support their offensives 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 fortifications, while other mobile squadrons of lightly armored soldiers (ashigaru) demolished lodgings harboring enemy troops in night raids. These fleet-footed ashigaru are best characterized as irregular units of skirmishers. Contemporary sources variously describe them as carry ing shields, strong bows, and pikes

Instruments of Change 145

and occasionally wearing helmets or, to the contrary, “holding no pikes wearing no armor, and carry ing only swords.” Ashigaru were guerrilla fighters who excelled in sudden attacks and skirmishes but could not occupy contested grounds. They did not constitute phalanxes of pikemen, and indeed, they were punished in offensive battles against entrenched forces. During one such encounter on 9. 7, 1468, Koma Tarō, the leader of the eastern army’s ashigaru, 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 flow of goods to enemy camps, but they no longer decisively influenced 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. The western army briefly managed to occupy Yamashina, which reduced the eastern army to illness-plagued malnutrition, but they could not hold the position, thereby allowing the stalemate to continue. The 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. The need to dislodge entrenched enemies encouraged innovations in weaponry. One finds reference to a craftsman from Izumi province constructing a catapult—known as a hō—that could launch 3-kilogram projectiles for over 300 yards! Unsei Daigoku, the author of the Hekizan Nichiroku, believed that this catapult was a device of great antiquity—modeled perhaps on third-century Chinese devices—and insisted that it did not represent an innovation. This 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 “flying projectile fire 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 contemporary catapults) from their unimposing nature. Explosive shells (teppō)

146 Thomas Conlan

had been known in Asia since the thirteenth century, when the Mongol invaders of Japan used primitive bombs, composed of ceramic projectiles fi lled with gunpowder, in tandem with rounded rocks to terrorize the Japanese defenders. These bombs should not, however, be confused with later firearms. Chinese sources refer to a primitive gun, known as a “fire 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, first introduced firearms to Japan some eighty years before the Portuguese visited Tanegashima. A fifteenth-century diary records how an official from the Ryūkyū Kingdom surprised many bystanders in Kyoto with the report of his firearm (teppō) on 7. 28, 1466. Archaeological evidence from Akenajō and Katsurenjō, two Okinawan castles (gusuku), reveals that guns were used prior to the mid-fifteenth 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, earthenware, and on occasion, iron bullets have been uncovered within the battlements of Katsurenjō, which was destroyed in 1458 and never rebuilt. Primitive firearms composed of three metal tubes welded together, known as fire arrows (hiya), were widely disseminated in Europe and Asia throughout the late fourteenth and early fifteenth centuries. Remarkably, some of these fifteenth-century hiya were fired as late as the early twentieth century, and they were reportedly capable of blasting projectiles for 200 yards. Most primitive firearms (hiya) discharged rounded stones, and intriguingly, 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 firearms. 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. The first documentary evidence describing wounds caused by firearms appears 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.” These suggestive sources can be corroborated with letters dating from 1569, which use the same verb (iru) to describe wounds inflicted by guns (teppō).

Instruments of Change 147

The earliest documented example of a man being explicitly wounded by “firearms” occurs on 1. 27, 1563, when Hara Rokurō, a retainer of Sugi Matsuchiyo, was shot near his left armpit by a “hand fire arrow,” or tebiya, by supporters of the Ōtomo in Northern Kyushu. Weapons known as as teppō can be first verified on 11. 13, 1563, when the Amako of Izumo province mauled the Kikkawa, wounding thirty-three by teppō, six by arrows, five by rocks, and one by a sword. As a result of Amako prowess, the number of gun casualties supersedes those of bows by a figure of 88 to 64 during the 1560s. Nevertheless, the data are not comprehensive enough to hypothesize about patterns of gun dissemination. Although twice as many bullet wounds (17 to 8) were recorded as arrows in the 1570s, both were inflicted at roughly analogous rates (19 to 16) in the 1580s. Indeed, from 1467 until 1600, arrows caused 58 percent of all projectile wounds, while bullets were responsible for 28 percent and rocks the remaining 13 percent. Guns required nearly half a century to supplant bows because they were only incrementally more effective in range and penetration. Gun wounds were often inflicted in close proximity to enemy forces, as were arrows. Several 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 limitations of sixteenth-century aim and firepower. Negoroji appears to have first recognized the power of a squad of gunners. Their marksmen gained fame for their prowess during the 1570s and mid1580s. 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. The adoption of guns, be they the primitive hiya or, for that matter, Portuguese teppō, seemed to cause no discernible change in tactics. Guns did not stop pikemen from fighting in close quarters even after 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 figure is consistent with fourteenth- and fifteenth-century military patterns. The 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 political “unification” of sixteenth-century Japan.

148 Thomas Conlan

Teppō were able to achieve parity with arrows by 1580, but they could only become an effective weapon when gunners were organized into cohesive groups, a process that only began during the 1570s. The influence of guns on changes in tactics is seen in the first rosters recording military units and the par ticu lar weapon used by each, which first appeared in 1575. Firearms influenced the outcome of battle when marksmen were organized much as pikemen had been during the course of the Ōnin War.

The O¯nin Legacy, 1477–1600 Even though the Ōnin War witnessed the use of bows, catapults, and firearms, 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 organization proved to be decisive and caused battles to shift from mobile encounters of horsemen to entrenched formations of infantry. Commanders focused their energies on disrupting enemy supply lines and, when this proved ineffectual, attempted to destroy their opponents’ regional support by fomenting the rebellion of deputy shugo. Organization became crucially important in maintaining and sustaining political and military power, while the ability to effectively use weapons, be they “old” pikes or “new” guns, merely reflected this more fundamental process. In the confusion of the Ōnin War, when all authority became contested, deputies were encouraged to rebel against their shugo. Delegated powers proved ripe for usurpation, as deputy shugo were best placed to organize, supply, and command regional military forces after 1467. These 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 defected 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 effective control over Echizen a mere three years later. Similar betrayals became endemic over the ensuing century. The Oda of Owari province, another deputy shugo family of the Shiba, would also overthrow their shugo and come to control most of central Japan by 1582 under the stewardship of Oda Nobunaga. The powers inherent in the office 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

Instruments of Change 149

capital and entrust de facto command to their deputies. That an individual might be simultaneously appointed as the shugo of several provinces mitigated against direct political supervision of any single province. After 1471, deputy shugo increasingly displaced their shugo in violent fashion, thereby giving 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 proficient in organizing their troops were best able to amass political authority. Although often overlooked in the process of sixteenthcentury “unification,” 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 effective use of units of pikemen, for the Ashikaga Kiseiki recounts how his army of 900 pikemen clashed with a similar force of Hosokawa troops, inflicting hundreds of casualties before ensuring Hosokawa Harumoto’s defeat. After 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 legitimacy and preferred instead to base his authority on military prowess. Miyoshi Nagayoshi’s victory, and his subsequent actions, reveal a new attitude 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 political endeavors by 1547, the year of Nagayoshi’s victory, and a mere four years after Portuguese firearms were initially introduced to the Japanese archipelago. In short, most of the innovations thought to have arisen from the introduction of Portuguese firearms were fully established before these weapons ever arrived.

Conclusion Ambiguities inherent in the word “technology” allow two interpretations of its role in Japanese history to be true, namely, that technology was instrumental in determining Japan’s fifteenth- and sixteenth-century transformations and was a reflection of these processes. When understood as a technique or method of organizing armies, technology—here characterized as organizational technology—proved decisive. When conceived as the innovation or adoption of new weapons, its most common definition, technology functioned as a barometer of historical and, for that matter, organizational change.

150 Thomas Conlan

Japanese warfare provides an ideal arena for assessing the importance of new weapons and technologies because during the years 1333–1600 pikes replaced swords as the preferred weapon for hand-to-hand combat, while guns also gradually displaced bows. Of these transformations, the hitherto overlooked shift from swords to pikes proved dramatic and significant, while the dissemination of guns was gradual and less consequential. These changing preferences were not based on the innovation of new weapons. In fact, the lag between a weapon’s initial appearance and its widespread adoption undermines materialist explanations of tactical change. References to pikes first appear in documents dating from 1333, but they were not widely used until 1467. Similarly, guns first arrived in Japan by 1466, but they only marginally influenced tactics for the ensuing century. Most of the transformations attributed to guns, or for that matter, pikes, stemmed from fourteenthand fifteenth-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. The influence of guns was limited to giving men from the most prosperous regions—the capital and western Japan—a slight military advantage 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 detriment in 1585. The widespread adoption of pikes signified a more important development than the gradual dissemination of guns. Troops wielded pikes only after regional authorities (shugo and their deputies) possessed the ability to sustain large armies in the field. In other words, the practical ability to transform significant numbers of foot soldiers into a cohesive fighting block proved to be vitally important. Only when troops could be mobilized, fed, and trained over time could they fight in formation, which proved indispensable for both pikes and, ultimately, firearms to be used effectively. Improvement in the ability to extract surplus revenue, supplies, and manpower enabled shugo to strengthen their political control over Japan’s provinces during the course of the fourteenth through sixteenth centuries. Instead of relying on autonomous warrior houses to forge fissile armies, as had been

Instruments of Change 151

typical through 1350, shugo created a network of warriors who could be readily mobilized, thereby enabling them to overshadow provincial rivals, and project power into neighboring provinces. This improved organization is reflected by the increasingly well-defined structure of Japanese provincial armies. As the fifteenth century progressed, organizational techniques became increasingly sophisticated as shugo relied on their economic and political 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 formations, resembling the phalanx, where a massed body of men wielded pikes in unison and thus could withstand cavalry charges on open ground. Thereupon, the need to conscript as many men as possible and forge them into a unified army superseded the need to entice fickle warriors into one’s camp. Those most skilled in governance ultimately proved most skilled at organizing, training, and maintaining provincial armies. Deputy shugo managed to train and organize troops regardless of the presence of firearms. During the latter half of the sixteenth century, a scion of a deputy shugo family named Oda Nobunaga managed to assert control (“unification”) over the capital and central provinces, but he could have accomplished this task even if he had possessed no guns. His military powers, and those of his compatriots, were predicated on the improved organizational abilities and logistical prowess of fifteenth- and sixteenth-century armies, which in turn hinged on increased political control over economic surpluses. Improvements in organizational technology determined the nature and pace of mechanical, tactical, and ultimately political change.

Notes 1. 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, Shibunkaku, 2001), pp. 1–35, 154–160. 2. Asao Naohiro, “The Sixteenth Century Unification,” in John Hall, ed., The 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 Varley in his “Warfare in Japan, 1467–1600,” in Jeremy Black, ed., War in the Early Modern World, 1450–1815 (London: UCL Press, 1999), pp. 67–73, and Geoff rey Parker in The 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 The Complete Oxford English Dictionary (New Edition). 6. See Hans Delbrück, History of the Art of War, particularly vol. 3, Medieval Warfare, pp. 649–656, and vol. 4, The Dawn of Modern Warfare, pp. 23–57 (Walter Renfroe, Jr., trans., Lincoln: University of Nebraska Press, 1990). 7. One of the first 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 revolution” stemming from improvements in fi repower, see Parker, The Military Revolution, pp. 24ff. 8. A concern for military organization pervades the writings of Clausewitz, who ignores the role of new weapons on battle. See Michael Howard and Peter Paret, trans., On War (Princeton, NJ: Princeton University Press, 1976). 9. For the Mongol invasions and the creation of petitions for reward, see Thomas 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 housemen,” were warriors who submitted petitions for reward demanding compensation for their military ser vice. Conlan, State of War: The Violent Order of Fourteenth Century Japan (Ann Arbor, MI: Center for Japanese Studies, 2003), pp. 107–140. 12. Conlan alludes to this process in State of War, pp. 104–106, 162–164, 222–229. 13. Dai Nihon Komonjo Iewake, Series 9, Kikkawa ke monjo (Tokyo: Tōkyō teikoku 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. The figure 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. The 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. These 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.

Instruments of Change 153

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 figures, 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 inflicted 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. The 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 thereafter. 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 (hereafter HKSC). For more on the hanzei, see Conlan, State of War, pp. 95–98, 225–229. 30. Cohesive, regionally based units were forged after 1350, some twenty years after 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, Ogasawara 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 Hosokawa Tōtōmi no kami kakikudashi an, p. 288. 32. The 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 difficulty 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 provisions 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 hundreds 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 daikanshiki, pp. 591–592. 40. Ibid., doc. 803, Daikanshiki buninjō an, and doc. 804, Nakazawa Motoki daikanshiki ukebumi, p. 613. 41. For the pioneering study on the origins of the Ōnin War, see Paul Varley, The Ōnin War (New York: Columbia University Press, 1967). For a more recent analysis, see Elizabeth Berry, The Culture of Civil War in Kyoto (Berkeley: University of California 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.

Instruments of Change 155

44. The Kuge, Nakazawa, Ogino, Honjo, Adachi, Yaku, and Ashida can be documented 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 Hokinoichi, 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 (hereafter 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. The document is located most conveniently 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 fighting 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 descriptions 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 fighting 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 edifice 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. These passages are difficult 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. That 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 final 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 evening edition of the Asahi shinbun, p. 12, for the discovery of three of these ceramic projectiles in the wreckage of the Mongol fleet. See also Conlan, In Little Need of Divine Intervention, pp. 12, 73. 80. The 1972 cata logue, Nihonshi ni mieru teppō ten contains pictures of these early weapons on p. 2. For a reference to firearms (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 fired. 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

Instruments of Change 157

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 photograph 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 softball. 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. The 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) (hereafter 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. The 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 finds 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 organized according to units of pikes, guns, and horsemen, 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. The 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 (Shōzuijōkanseki 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. The 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. This 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 jidai no kizoku (Kōdansha gakujutsu bunko, 2002), pp. 188–210.

Postscript John A. Ferejohn and Frances McCall Rosenbluth

WAR IS CREDITED WITH THE CREATION of the modern nation-state as we know it. Po-

litical 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 organized feudal entities or isolated merchant city-states into large and centralized territorial states. In meeting external threats, governments had to find better ways to squeeze resources out of their populations and in the process developed greater internal capacity to govern, administer, and control their own territories. The macro-sociological accounts may be right that the nation-state was a side effect of the fiscal needs of early modern governments waging wars. In Japan, as in Europe, 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 political organization. The explanation offered 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 effort of each fighting man is valued, territorial scale can be small without losing strength. Scale economies in defense explain why territorial consolidation originated in flat, fertile regions where farmers were willing to trade taxes for protection out of dire necessity. Eventually, national boundaries radiated outward until more remote areas were swallowed up 159

160 Ferejohn and Rosenbluth

under centralized control but not until communities living in prosperous lowlands supported enormous and opportunistic armies. In Japan, as in early modern Europe, farmers living in fertile plains accepted—often under dire circumstances—strong, centralized government as the solution to their physical vulnerability. Scale economies account for considerable variation across cases in the effects of war on territorial consolidation, but an explanation centering on the relative costs of expansion is as incomplete as the macro-sociological conventional wisdom in accounting for the onset of wars in the first place. After all, Japan enjoyed freedom from foreign invasion, and a decentralized system of overlapping jurisdictions lasted for centuries after 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, engulfing Japan in internal wars of conquest? One explanation suggested in the literature on the military revolution in Europe is that firearms changed the balance of power in favor of aggression over defense by making castles more vulnerable to attack. Thomas Conlan, in Chapter 7, warns against the technical change argument, pointing out that guns were not always decisive in battle. The timing of firearms and territorial consolidation, moreover, is not quite right. Conlan argues instead that military success rested on superiority in administering territories, raising revenues, and organizing large armies. This still leaves open the question, of course, of why some domain lords became more successful than others in fiscal, military, and logistical administration. Ike’s and Inaba’s analyses of the relationship among warlords, estate owners, and farmers, in Chapters 3 and 4, supply some missing pieces. For centuries, Japanese farmers were enmeshed in clientelistic relationships with local estate owners for their livelihood. Gradually, farmers became more economically self-sufficient, particularly in the Tokai region along the thoroughfare between Kyoto and Kamakura where proximity to resources and markets gave farmers alternatives to clientelistic dependency. Once more independent, farmers were free to shift alliances for self-protection and advancement in a way that shook the entire ladder of dependency, 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 confiscation. Insecurity and instability intensified, which increased the farmers’

Postscript 161

willingness to pay taxes to whichever coalition of leaders was perceived as being best able to secure life and property. Once political authority and military power lined up territorially, relatively small perturbations in force sizes unraveled a delicate balance of power among warlords, unleashing the widespread 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 firearms were an instrument of conquest, the reasons for upheaval and the ensuing acceptance of a repressive peace lay deeper in changing societal structure. Ironic though it is that growing agricultural self-sufficiency should have triggered the chain of events leading to more repressive governance, the episode showed that economic autonomy is not as empowering as mobility or defensibility. The specificity of agricultural assets in land meant that farmers—even those with multiple suppliers and markets and who could therefore sell their products at market prices—still had to contract with someone for physical security. The process of social turmoil and ensuing territorial consolidation in medieval Japan displays some striking similarities and contrasts with what happened in Europe. Had Charles Tilly and Sidney Tarrow (2003) turned their attention to Japan, they would have noted that in sixteenth- and seventeenthcentury Japan, as in Prussia where there were relatively few owners of mobile assets, owners and tillers of land were the main players in the struggle for political 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 decades had come from farming villages desperate to make deals with one overlord or another. By contrast, in the Italian city-states or England, or even France where monarchs sold influence to increasingly wealthy merchants, the groups bargaining with the monarch were more mobile and less vulnerable than farmers, tempering the nature of centralized rule. In medieval Italian city-states as well as in England following its Glorious Revolution of 1688 and the Netherlands after 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 uninterrupted history would have played out, but in England 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

162 Ferejohn and Rosenbluth

which it marches all the way to full, nationwide absolutism, as in seventeenthand eighteenth-century Prussia or France, and when instead does it stop short at federal bargains in which provinces retain some autonomy as in seventeenth- and eighteenth-century Japan or even Italy and the United Kingdom (if we consider Scotland, Ireland, and Wales)? The 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 measure of local economy can be a low-cost way to govern. Aristocratic rule is likely to be a weak form of political organization for the purposes of war-readiness against external threats. In the first place, as Machiavelli noted of his native Florence, aristocracies (of the merchant variety 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 local monopolists, they have the power to do so. The absolutist monarchies that arose in early modern Europe can be interpreted as a broader, rather than narrower, coalition than the feudal governments they replaced in response to perceived defensive need. Unlike the earlier 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 Prussia a strong monarchy balanced 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 Prussians and Russians in the eighth century for fear that a victorious peasant army might turn on the Polish nobility. Poland survived for many years as a narrowly based oligarchy, but was caught unprepared by the large armies from Prussia and Russia that dismembered the proud Polish republic in 1797. The Polish anomaly aside, the contrast between European absolutism and Tokugawa’s loose federation of domains seems to suggest that tighter centralization 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 Prussia faced that would have pressed territorial domains into common cause. Without real fear of foreign invasion, there was no pressure from the center to elicit extensive mobilization, nor did the domains have any interest in providing it. The domains paid the Tokugawa nominal allegiance 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. Insufficiently paid as war-less warriors, many in the military class became bureaucrats. As in France where the monarchy lured aristocrats into becoming 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 domainal treasuries. 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 finally faced a foreign threat in the form of the “Black Ships” of the U.S. Navy steaming into Edo Bay beginning in 1854, samurai-bureaucrats in the enemy domains launched a revolution that toppled the Tokugawa in short order, given the lack of national preparedness. The 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.

Notes 1. Of course, not all states landed on efficient solutions to their need for funds. Beik (1985) argues that the French crown raised war funding by making office into secure property that could be sold, but in so doing, increased the group of nobility exempt from taxation. The French monarchy’s short-term solution to its financial problem created long-term shortfalls.

164 Ferejohn and Rosenbluth

2. Bean 1973. 3. On the Tokugawa regime, see, for example, Banno and Mitani 2006; Maruyama 1974; Totman 1967; Najita 1987. 4. The 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 affairs, they would settle for nominal Tokugawa overlordship. The 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. This 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 Renaissance Florence had in mind a different 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 decision in about 450 b.c. to give fighting men the right to vote, but Cleisthenes’s “democratization” of Athens in 508 b.c. in order to overthrow Spartan-backed aristocrats was an even more ancient precedent. In Japan, as in Switzerland, only the isolated mountain or monastic communities came close to this model. 6. England, which enjoyed some protection from the Channel, also avoided the political consequences of full-scale military mobilization, but if we are correct, developed a form of mixed government rather than territorial federalism because of the power of merchants to protect themselves from the arbitrary rule of monarchs.

References Banno, Junji, and Ichiro Mitani. 2006. Nihon no kindaishi jukko: Rekishi o tsukuro mono [Essays in Modern Japanese History: What History Is Made Of]. Tokyo: Iwanami shoten. Bean, Richard. 1973. “War and the Birth of the Nation State.” Journal of Economic History, 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. The Military Revolution and Political Change: The Origins of Democracy and Autocracy in Early Modern Europe. Princeton, NJ: Princeton University Press. Ertman, Thomas. 1997. Birth of the Leviathan: Building States and Regimes in Medieval and Early Modern Europe. New York: Cambridge University Press. Maruyama, Masao. 1974. Studies in the Intellectual History of Tokugawa Japan. Translated by Mikiso Hane. Tokyo: University of Tokyo Press.

Postscript 165

Najita, Tetsuo. 1987. Visions of Virtue in Tokugawa Japan: The Kaitokudo Merchant Academy of Osaka. Chicago: University of Chicago Press. Tilly, Charles. 1975. “Revolutions and Collective Violence.” In F. Greenstein and N. Polsby, eds., The Handbook of Political 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. The 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

Glossary

Akutō ᝇඨ, “evil bands”; groups of peasants and minor warriors involved in looting, raiding, and other local crimes during the late thirteenth and fourteenth centuries ashigaru ㊂㍅, literally, “light feet”; low-ranked infantry; drafted peasants bōsen 㜭ᡋ, defensive warfare chakutōjō ╌฽≟, reports of arrival chūka ୯⳱, literally, “center of civilization,” often used to refer to China chūnin-sei ୯ெโ, mediation by powerful individuals Daijōkan ኯᨳᏻ, Council of State daimyō ኬྞ, domain lords gekokujō ୖ๏୕, literally, “lower overcoming higher”; proverb that aptly characterizes the turbulent Warring States era which experienced the widespread phenomenon of subordinates rebelling against their lords genin ୖெ, shoju ᡜᚉ, direct laborers gō 㒋, village or district gokenin ࡇᐓெ, literally, “housemen” of the Kamakura bakufu who were obligated to perform guard duty but were otherwise elite members of provincial society. The term also refers to a low-ranking warrior in the Tokugawa era. Goseibai-shikimoku (Jōei ㇾỄ Formulary) ᚒᠺᩃᘟ┘, the law code of the Kamakura bakufu, promulgated in 1232 and appended frequently thereafter through the remainder of the Kamakura age gunchūjō ㌯ᚽ≟, 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 ㌯ᅆ, provincial regiments 169

170

Glossary

gunnai ichizoku 㑾හୌ᪐, regional alliance of samurai that in the sixteenth century would eventually give way to a local league of fighters, the Iga ikki, which had one main stated objective: restoring peace and order in the province gusuku ᇖ, castles han-kenryoku ⸤ᶊງ, daimyō authorities hanzei ༖ῥ, 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 ඹ㎨ฦ㞫, separating farmers and warriors into two classes. Toyotomi Hideyoshi’s Sword Hunt (1588) formalized a process of specialization that had already begun under military governors heishi ඹኃ, a conscript soldier enrolled in one of the provincial regiments under the ritsuryō codes hihō hisō 㣍◑ℾᵍ, literally, a “flying projectile fire spear”; that is, firearm hikan ⿍ᏻ, retainer hiryūsō ℾ❫ᵍ, Chinese sources refer to a primitive gun, known as a “fi re dragon spear,” being created in 1355 hiya ℾ▦, primitive firearms composed of three metal tubes welded together, known as fire arrows hō ◑, a catapult hōkōshū ዅප⾏, direct retainers of the shogun honryō ᮇ㡷, right to the land hyakushō Ⓤጞ, common people; that is, persons without court rank and subject to taxation Iga ikki ௿㈙ୌᥙ, local league of fighters in Iga with one main stated objective: restoring peace and order in the province Iga sōkoku ikki ௿㈙⥪ᅗୌᥙ, the league of all the commons of Iga. 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 Ikkachū ୌᐓ୯, “families”; name for the forty-six signatories who sought to ensure the regional peace, vowing not to fight over taxes on peasants and to try to prevent insurgent acts by subordinated people who might “cause trouble” ikki ᥙ, a pledged organization with the clear aim of maintaining order within the domain; squads ikkō ikki ୌྡྷୌᥙ, literally translated as leagues of members of the Honganji sect of True Pure Land Buddhism (Jōdo Shinshū) in the sixteenth century Imagawa kanamokuroku ௑ᕖ௫ྞ┘㘋, code of laws, enacted by the Imagawa, Sengoku daimyō of the province of Suruga in 1526

Glossary 171

Inoue-shū chūbatsu-jiken ஬୕⾏ヸఅ஥௲, clan from the pact for invading a neighbor’s territory iru ᑏࡾ, to shoot; used for both bows and firearms iteki ኻ≦, term for neighboring “barbarian” countries jinai ᑈහ, temple grounds jinai mach ᑈහ⏣, temple town jitō ᆀ㢄, estate stewards or managers appointed by the Kamakura shogunate; chiefly concerned with tax collection and local police duties jitō-shiki ᆀ㢄⫃, the rights, duties, and perquisites associated with the position of jitō jizamurai ᆀ౜, literally, “samurai on the land”; yeoman farmers who also functioned as warriors Jōdo shinshū Ὧᅰ┷ᏺ, literally, “True Pure Land” Buddhism; this was the sect that sponsored a large self-governing league in Kaga province kaimin, ᾇẰ, people of the sea kami ♼, Shintō divinities kanjō វ≟, document of praise, which could include a recommendation that a military leader be rewarded for his actions in battle karita-rōzeki ศ⏛⊃⡘, rice plants kashindan ᐓ⮟ᅆ, 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 ྙᡋዅ⾔, battle reports to administrators kassen chūmon ྙᡋἸᩝ, battle reports, written shortly after every skirmish, with each document mentioning the damages incurred by warriors so as to ensure compensation for their actions kebiishi ᳠㟸㐢౐, literally, “Investigators of Oddities”; officers of the Office of Imperial Police (kebiishi-chō) in the capital; also a provincial military/police post in the provinces established in the mid-ninth century kenka-ryōseibai-hō ႒ბ୦ᠺᩃἪ, rule under which both aggressors and defenders were regarded with equal severity and private wars were prohibited altogether kinjo-no-gi ㎾ᡜࡡൢ, mediation kōgi පൢ, shogun’s officials koguchi ⹙ཾ, literally, “tiger’s mouths”; referred to extremely narrow paths kokuga ᅗ⾞, the provincial government office kokujin or kokunin ᅗெ, literally, “people of the country” or local warriors; during the fourteenth and early fifteenth centuries, principally used to distinguish powerful provincial warriors whose activities were confined to a single province from those who also held posts as shugo; also a person who resides in the province

172

Glossary

kokujin-ikki ᅗெୌᥙ, local social pact of equal-basis unification of local lords based on regional relations kokushū ᅗ⾏, emerging warrior leaders known as Warring States daimyō kosen ᨶᡋ, offensive warfare kosenbōsen-hō ᨶᡋ㜭ᡋἪ, law that restricted private wars kugyō ප༽, high-ranking aristocrats menka ඞᐓ, peasants miuchi ᚒහ, powerful retainers miyako no musha 㒌ࡡṂ⩽, 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 general  career advancement; sometimes called the samurai in English-language sources mura no shiro ᮟࡡᇖ, fortified places on the heights, what Fujiki Hisashi calls “castles of the village” mura-okite ᮟ᤽, rules governing local behavior mura-uke-sei ᮟㄫโ, system in which peasants paid land taxes to the village council, which in turn paid the lord his due musha taishō Ṃ⩽ኬᑑ, literally, “warrior general”; high-ranking warriors nengu ᖳ㈁, land taxes ninja ᚰ⩽, literally, “stealthy ones”; or more commonly read as shinobi mono; lightly armed warriors who fought with homemade implements and used tactics of surprise, stealth, and intrigue ōryōshi ᢪ㡷౐, literally, “Envoy to Subdue the Territory”; one of the two main titles granted provincial warrior leaders under the Heian provincial military/police system Ritsuryō ᚂ௦, the legal codes that defined the structure and operations of the imperial state saikyo ⿚ス, legal judgment saikyo-jō ⿚ス≟, letter of sentence sashidashi ᣞฝ, tax documents that reveal a contractual relationship between a village community and a warrior lord. When a warrior lord defeated his opponent and conquered his opponent’s territory, he first had to collect the tax documents from villages in the territory that he newly gained seii taishōgun ᙹኻኬᑑ㌯, literally, “Field Marshal for Subduing the Barbarians”; originally a temporary title for the officers 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 ᦜ㛭ᐓ, Regents’ House; the sublineage of the northern branch of the Fujiwara 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)

Glossary 173

sengoku ᡋᅗ, warring states sengoku daimyō ᡋᅗኬྞ, powerful warlords of the Warring States period shakizu ᑏ␽, a wound from gunshot shiki ᘟ, originally a function or office with attached perquisites; later the right to designated income under the title of an office; often hereditary shisen ⚶ᡋ, private wars; that is, conflicts fought without the sanction of government authority sho-ku, small districts shōen Ⲧᅧ, a landed estate, normally held in proprietorship by a high-ranking member of the central nobility or by one of the great religious institutions; the dominant form of landholding and land administration from the eleventh through the fourteenth centuries, its full history runs from the eighth through the sixteenth shugo Ꮼ㆜, province-level “constables” appointed by the Kamakura shogunate; provincial military governors under the Muromachi shogunate shugo-dai Ꮼ㆜௥, shugo deputies sō Ⲧ, district; most Japa nese historians have compared them to the medieval communes of the West sōbuji ᝯ↋஥, peace order forbidding neighboring territories to fight over boundaries sōkan-shiki Ⲧᏻ⫃, estate owners sōryō ⥪㡷, 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 ᝯᮟ, village; organizational units below the sō, which arose to replace large estates as the dominant units of authority tachi uchi ኯภᡬ, literally, “clash of swords”; a general appellation for battle taikō-kenchi ኯ㛯᳠ᆀ, cadastral surveys, part of the policies of the Toyotomi Hideyoshi regime tanshū no onzei ୸ᕗ஄ᚒເ, a typical way to describe mid-fi fteenth-century military units tebiya ᡥℾ▦, “hand fire arrow” teppō 㕪◑, explosive shells; firearms teppōshū 㕪◑⾏, marksmen Tsuibushi ㏛ᤍ౐, literally, “Envoy to Pursue and Capture”; one of the two main titles granted provincial warrior leaders under the Heian provincial military/police system Tsuitōshi ㏛ゞ౐, literally, “Envoy to Pursue and Strike Down” ; a temporary deputation used by the court from the mid-Heian period to grant special military or police powers to warrior leaders for specific missions uji Ắ, political units of the pre-imperial era

174

Glossary

waka ࿰ḯ, Japa nese poems wakashū ⱕ⾏, villages’ military forces wakō ೔ෝ, seafaring pirates yakizu ▦␽, “arrow wounds” yamabushi ᒜఄ, “mountain monks”: practiced asceticism in the mountains under the influence of Shingon esoterism and Buddhism yari awase senshi 㚱ྙᡋṒ, confrontation with pikes yoriai ᐞྙ, village councils zuryō ུ㡷, career provincial officials of the mid- to late Heian period

Index

Absolutism, 14n16 Akutō. See Bandits (akutō) Alliances, 30, 116–117; warrior bands and, 36–39, 48n49, 49n51 Amako clan, 64 Amida, 93–95, 102–104 Archery. See Arrows, use of; Mounted archery Aristocratic rule, 162 Aristocrats, court, 26, 29, 34. See also Nobility, court Arrival, reports of (chakutōjō), 126 Arrows, use of, 126–127, 131, 131t, 132–135t, 147–148, 153n28, 157n93 Asai house, 64 Asakura house, 64 Asceticism, 119 Ashigaru (light infantry), 140, 144–145 Ashikaga shogunate, 11, 57, 61, 64 Ashikaga Takakage, 148 Ashikaga Takauji, 57, 67n8 Ashikaga Yoshiaki, 64, 69n32 Ashikaga Yoshimasa, 68n22 Ashikaga Yoshimi, 68n22 Ashikaga Yoshiteru, 149, 158n98 Autonomy, 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, 119 Bargaining leverage, farmers’, 5–6, 8–11 Battle reports (kassen chūmon), 126–130

Battles. See Pitched battles; Skirmishes. See also under names of specific battles Beik, William, 163n1 Bodin, Jean, 14n13 Buddhahood, 93–95 Buddhism: 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, 73 Cannon, 8, 16n34 Career provincial officials (zuryō), 29–30, 32, 34, 46n28 Castles, 8; of the village (mura no shiro), 115 Catapult (hō), 145 Cavalry: 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, 14n14 Chieftains, regional, 22–23, 25, 28 China, 5, 77; coins from, 69n35; Tang, 22–23, 25 Christianity, 92, 106 “Chronicle of Ōnin,” 143 Client-patron relationships, 30, 33, 48–49n49, 66, 160 Commoners’ struggles, 93, 106. See also Ikkō-ikki 175

176

Index

Community vitality. See Village communities Competence over loyalty, 53–70 Conscription: Meiji government and, 86; privatized, 35–36; ritsuryō military system and, 23–26 Consolidation. See Territorial consolidation Contracts: oaths and, 63; village communities and, 73–74; warrior alliances and, 38–39, 48n49. See also Oaths Court politics, 29–31, 46n28 Currencies, 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, 62 Decision-making, consensual, 83–84 Democracy, bargaining hypothesis of, 73–74 Discourses (Machiavelli), 13n3 Dispute resolution system, 80–81 Districts (sō), 116 Downing, Brian M., 159, 164n5 Economies of scale: autonomy and, 3; changes in, 7–9; consolidation and, 160–161 Electoral system, multilayered, 84–85 Elites, provincial: alliances of, 30; military service and, 32–34; wealth and power of, 26–29 England, 14nn14–15, 161, 164n6 Erizeni, 69n35 Ertman, Thomas, 159 Estate system, 15n21, 115, 160; Niimi, 141; Ōyama, 137–139, 141 Europe, 5, 8–9, 14nn14–15, 16n36, 161–163, 164nn5–6 Evaluation, merit system of, 83 Farmers’ bargaining leverage, 5–6, 8–11 Federal system, Tokugawa’s de facto, 3 Feudalism, 5 Firearms: 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 (flatlands) Foot soldiers: as leverage, 8, 10, 16n34. See also Infantry Foreign invasion, 1–2, 24, 79, 160, 163 Forty-Seven Samurai, The (fi lm), 67n13 France, 9, 14n14, 161–163 Fujiwara regents, 30, 33 Gekokujō (“the low overthrow the high”), 62, 69n29, 69n31, 115, 149 Gempei War, 40 Genji preeminence, 32–33 Gentry, provincial, wealth and power of, 26–29 Go Daigo, (emperor), 57, 67n8 Gokenin (“shogunal housemen”), 55–56, 152n11 Governance. 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), 116 Guns. See Firearms Hand-to-hand combat, 131t, 152n17 Hanzei edict, 137–139 Hashiba (Toyotomi) Hideyoshi. See Toyotomi Hideyoshi Hatakeyama Masanaga, 143–144 Headmen systems, 80–85 Heian period, 21, 28–29, 34–39 Heiji Incident, 39 Heinō-bunri. See Separation of warrior and peasant classes (heinō-bunri) Hino Tomiko, 68n22 Hiya (fire arrows), 146 Hobbes, Thomas, The Leviathan, 14n10 Hojo family, 40, 55–57, 62, 64, 74–75 Hojo Soun. See Ise Nagauji (Hojo Soun) Hokkaido, 15n18 Honganji sect, 69n33, 92–109, 106n4; ideology of, 93–95, 102–104, 107n32; justification for fighting by, 102–104; power of, 101; social order of, 95–98; uprisings and, 97–102, 107n26 Honshu, 3 Horeki, Reform of, 81 Horikoshi Kubo, 68n24

Index 177

Hosokawa house, 58, 61–62, 79, 139, 141, 149 Hosokawa Katsumoto, 61, 140–142 Hosokawa Masamoto, 61–62 Hosokawa Yorimoto, 138 Household 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, 114 Ikki, 67n9, 112, 116, 120, 153n25. See also Ikkō-ikki; Kokujin-ikki (territorial pacts) Ikkō-ikki, 64, 69n33, 119; Honganji sect resistance and, 92–106 Imagawa code of law, 60–61, 68n15, 68n20 Imagawa house, 60–64 Imagawa Yoshimoto, 63 Imperial 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; stratification and, 29–30 Infantry, 148; light (ashigaru), 140, 144–145; ritsuryō military system and, 24–26 Inoue house, 60–63 Ise Nagauji (Hojo Soun), 68n24 Isshiki Yoshitada, 141–142 Italian city-states, 161 Japan, settlement of, 5 Japanese corporate society, 87 Jizamurai, 112, 115–116, 120–121 Jōdo Shinshū (Pure Land) Buddhism, 2–3, 92–95, 103 Judicial 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), 15n30 Kamakura shogunate, 21, 40–43, 54–58 Kammu Heishi, 33, 47n37 Kasai house, 138–139 Kasai Motosuke, 138–139, 154n35

Kashindan (pacts), 117 Kato Kiyomasa, 77–79 Kawanakajima, Battle of, 63 Kennyo (Honganji patriarch), 100–103 Kikkawa house, 142–143 Kitabatake Akiie, 67n8 Kōga district, 110–111, 113, 116, 119–121, 123n1 Kōgagun chūsō, 116 Koguchi (narrow paths), 111 Kokujin-ikki (territorial pacts), 53, 57, 62, 67n9 Kokushu, 58–59 Korean invasion, 76–79 Kosenbosen-ho (law), 57 Koza-ha (associated with Communist Party), 14n17 Kumamoto, self-governance study and, 79–86 Kumamoto house, 81–82, 88n17 Kumamoto Prefectural Assembly, 84–86 Kuniwake, 64 Kusunoki Masashige, 67n8 Kyoto, 26, 31–32, 53 Labor, scarce, 16–17n31 Landholding, 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, The (Hobbes), 14n10 Lords 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; The Discourses, 13n3 Marriage, among aristocrats, 34 Marxism, 9, 14–15n17

178

Index

Meiji government/Restoration, 67n8, 71, 84–86, 163 Merchants, 65, 160, 161, 164n6 Mikawa province, Honganji sect in, 98–99 Military organization. See Technology, organizational Military protection: economies of scale and, 8–9; private market and, 32–39, 43; 16th century unification and, 4–12; village communities and, 72–74 Military revolution, 126, 147, 160 Military rule/systems: centralization and restructuring of, 22–26; decentralized, 5–6; first 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, 55 Minamoto Sanetomo, 40 Minamoto Yoriie, 40 Minamoto Yoritomo, 38–42, 55 Minamoto Yoshitomo, 39 Ming Dynasty, 77 Miuchi (retainers), 138, 154n33. See also Lords and retainers Miyako no musha (“warriors of the capital”), 32–33, 41 Miyoshi house, 64, 157–158n96 Miyoshi Nagayoshi, 114, 149 Miyoshi Yoshitsugu, 158n98 Mochihito, Prince, 39 Monasteries, 2, 7, 9, 92, 110, 116 Montesquieu, 14n15 Morgarten, Battle of, 1 Mori house, 60–64, 100 Mountain regions: defenses and, 2, 68n25, 118–120, 159; taxes and, 16n38 Mounted archery, 24–25 Mount Kōya temple, 91 Nagao Tamekage, 62, 68n26, 69n29 Nagashino, Battle of, 15n30 Naitō house, 139–140, 142 Naitō Motosada, 138–139, 154n35 Naitō Sadamasa, 142 Nakazawa house, 138–140, 153n32 Nara period, 21 Negoroji marksmen, 124–125, 147 Networks of warrior groups. See Alliances, warrior bands and

Nichiren sect, 95, 105 Niimi estate, 141 Ninja, 1, 11, 13n1; birth of tradition of, 110–123, 123n1 Nitta Yoshisada, 57, 67n8 Nobility, court: emigration to the countryside of, 30–32; military service and, 32–34; wealth and power of, 26–29 Oaths, 63 Oda Nobukatsu, 118 Oda 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, 63 Okinawa, 15n18 Ōnin War, 58, 61, 115; organizational innovation and, 140–145, 148–149 Organizational change, 125, 149. See also Technology, organizational Organizational technology. See Technology, organizational Ōtomo house, 64 Ōuchi Masahiro, 142, 155n54 Ōyama estate, 137, 139, 141 Oyamoto declarations and domain, 111–113 Pacts: kashindan, 117; territorial (kokujinikki), 53, 57, 62, 67n9 Patron-client relationships, 30, 33, 48–49n49, 66, 160 Peace order (sōbuji), 71–72, 75–76, 88n10 Peasants, 3–4, 8–12, 15–16n31, 160; class, separation of warriors from (heinōbunri), 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, 137 Petrarch, 13n4 Pikes, use of, 126, 131, 131t, 132–135t, 135–137, 147, 148–150; in Ōnin War, 140, 142–144, 156n70 Pirates, seafaring (wako), 2 Pitched battles, 135, 144–145 Plains (flatlands), vulnerability and, 3–4, 9 Poland, 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 protection Provincial elites. See Career provincial officials (zuryō); Elites, provincial; Governors, provincial; Lords and retainers Provisions. See Supplies and supply routes Prussia, 9, 162 Rebirth, belief in, 94, 102 Record book (mura-daka), 88n3 Reincarnations, 94 Religious opposition to consolidation, 91–109; justification for fighting 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 retainers Revenue. See Taxation Ritsuryō 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, 147 Rokkaku house, 64, 143 Romans, 10, 16nn41–42 Rono-ha (associated with socialists), 14n17 Rousseau, Jean Jacques, 13n3 Ryūkyū Kingdom, 146 Samurai, 6–7, 21–52, 32–43, 163; first 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 scale Security, 3, 9, 13, 14n10, 160–162. See also Military protection Seiwa Genji, 32–33

Sekigahara, battle of, 3 Self-governance study, 79–87 Sengoku era, 115; daimyō, 59–60, 76, 117–118 Separation of warrior and peasant classes (heinō-bunri), 71, 74–75; Honganji sect and, 104–105 Shibata Katsuie, 91 Shimazu house, 65–66 Shogunates, 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, 54 Shōkokuji, battle at, 143 Shugo (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, 135 Social 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, 94–95 Stirrups, cavalry and, 8, 16n34 Stone bullets, 146, 156–157n83 Stone monuments, 79, 81 Stratification, in Heian society, 29–30 Sue house, 63, 69n31 Supplies and supply routes, 140–141, 144–145, 148, 150, 156n73, 161 Suzuki Masaya, 131, 135 Swiss mountain warriors, local autonomy and, 1–2, 120, 159 Swords, use of, 131, 131t, 132–135t, 135–137, 153n28 Taika Reforms, 22 Taiko-kenchi. See Cadastral surveys (Taiko-kenchi) Taira, 32–34, 47n37, 54–55 Taira Kiyomori, 39 Taira Masakado, 36, 54 Taira Sadamori, 54 Takeda house, 62–64 Takeda Shingen, 15n30, 62, 64

180

Index

Tanba province, 138, 140 Tang China, 22–23, 25 Taxation, 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, 125–126 Technology, 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, 13n2 Temples, 2, 41–42, 69n33, 91–93, 95–103, 105; towns of ( jinai machi), 97, 100–101 Tenji (emperor), 22 Teppō (firearms), 145–148, 157n93 Territorial 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, 14n14 Tilly, Charles, 159 Tokugawa Ieyasu, 99–100, 105, 121, 163 Tokugawa regime, 3, 87, 121–122, 162–163, 164n4; Toyotomi Peace Order and, 75–76 Torigoe, temple of, 91 Toyotomi 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, 69n29 Uji (political units), 5 Unification, 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, 140 Village 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), 72 Village headmen systems, 80–85 Wakushu (village’s military forces), 72 Waldstetten, Treaty of, 120 War: consolidation and, 1–12, 159–163; as political leveler, 4. See also under names of specific wars Warring States period, 58–64, 67n12, 161 Warriors, 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, 1–2 Watchtowers, 144 Weapons, 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, 87–88n2 Yamabushi (mountain monks), 119 Yamamoto Tsunetomo, 15n23 Yamana house, 140–142 Yamana Sōzen, 61, 141–142 Yamato house, 22–23 Yasuoka Ryosuke, 84–85 Yoriai. See Village councils (yoriai) Zuryō. See Career provincial officials (zuryō)