Land, Power, and the Sacred

Home , Aki Province, Awa Province (Chiba), Awa Province (Tokushima), Bizen Province, Bungo Province, Chikuzen Province

The Estate System in Medieval Japan

edited by Janet R. Goodwin and Joan R. Piggott

23 22 21 20 19 18 6 5 4 3 2 1

Library of Congress Cataloging-in-Publication Data Names: Goodwin, Janet R., editor. | Piggott, Joan R., editor. Title: Land, power, and the sacred : the estate system in medieval Japan / Janet R. Goodwin and Joan R. Piggott, [editors]. Description: Honolulu : University of Hawai‘i Press, [2018] | Includes bibliographical references and index. Identifiers: LCCN 2017052550 | ISBN 9780824875466 (cloth alk. paper) Subjects: LCSH: Manors—Japan—History—To 1500. | Land tenure—Japan—History—To 1500. | Japan—History—To 1600. Classification: LCC HD914 .L36 2018 | DDC 333.3/234—dc23 LC record available at https://lccn.loc.gov/2017052550

University of Hawai‘i Press books are printed on acid-free paper and meet the guidelines for permanence and durability of the Council on Library Resources.

Cover art: Map of Hineno Village on Hine Estate, dated 1316. Courtesy of the Imperial Household Agency (Kunaichō). To the memory of our professors Delmer M. Brown, University of California at Berkeley Jeffrey P. Mass, Stanford University

Contents

List of Figures, Plates, and Tables xi Acknowledgments xv Abbreviations xvii Periods of Premodern Japanese History xix Premodern Provinces and Modern Prefectures xxi

Introduction Janet R. Goodwin xxv

Part I The Big Picture 1 Estates: Their History and Historiography Joan R. Piggott 3 2 Medieval Japan’s Commercial Economy and the Estate System Sakurai Eiji (translated by Ethan Segal) 37

Part II How Do We Know about Estates? 3 Ōbe Estate in the Archaeological Record Nishida Takeshi (translated by Michelle Damian) 61 4 Tōdaiji’s Estates in Its Documentary Record: Perspectives on Ōbe Estate Endō Motoo (translated by Janet R. Goodwin) 81 5 Hine Estate in Izumi Province: Archaeology, Landscape Reconstruction, and Village Structures Hirota Kōji (translated by Janet R. Goodwin) 105

vii viii Contents

Part III. Making the Land Productive 6 Agricultural Expansion and Irrigation in the Early Medieval Age Kimura Shigemitsu (translated by Kristina Buhrman) 143 7 Loggers and Cultivators of Nabari: Tōdaiji’s Kuroda Estate in Heian Times Joan R. Piggott 164

Part IV: Secular and Sacred 8 Hijiri and Temple Monks: Contrasting Styles of Estate Management Nagamura Makoto (translated by Janet R. Goodwin) 197 9 Beyond the Secular: Villages, Estates, and the Ideology behind Chōgen’s Land Reclamation Projects Ōyama Kyōhei (translated by Janet R. Goodwin) 211 10 Claiming the Land: Chōgen and the Development of Ōbe Estate Janet R. Goodwin 231 11 The Jōdoji Amida Triad: How Its Iconography Advanced Chōgen’s Mission Yoshiko Kainuma 253

Part V. Power, Space, and Trade 12 Nyoin Power, Estates, and the Taira Influence: Trading Networks within and beyond the Archipelago Sachiko Kawai 281 13 Networks of Wealth and Influence: Spatial Power and Estate Strategy of the Saionji Family in Early Medieval Japan Rieko Kamei-Dyche 319 14 As Estates Faded: Late Medieval Maritime Shipping in the Seto Inland Sea Michelle Damian 351 Contents ix

Part VI. Power: Challenges and Conflicts 15 Bad Neighbors and Monastic Influence: Border Disputes in Medieval Kii Philip Garrett 377 16 The Akutō on Ōbe Estate: Lawsuits, Evidence, and Participation in the Late Kamakura Legal System Dan Sherer 403 17 Warriors and Estates in Muromachi-Period Harima Noda Taizō (translated by David Eason) 427

Part VII. Getting the Word Out 18 Teaching Japanese Estates: Old Challenges and New Opportunities Ethan Segal 441

Some Afterthoughts Janet R. Goodwin 457

Glossary 461 Bibliography 473 Contributors 511 Index 517

Color plates follow pages 140 and 374.

List of Figures, Plates, and Tables

Figures Figure I.1 Map of Japan’s medieval provinces Figure 3.1 Site of Ōzuka tomb, Ōbe Estate Figure 3.2 Kiyotani archaeological site, Ōbe Estate, early medieval period Figure 3.3 Jōdoji subtemples Figure 3.4 Site of a Jōdoji building Figure 3.5 Remains of pit dwellings, Ōji Shiro-no-shita Figure 3.6 Yayoi period pottery, Ōji Shiro-no-shita Figure 3.7 Tumulus period artifacts, Ōzuka tomb Figure 4.1 Map of medieval Tōdaiji Figure 4.2 Tōdaiji organizational chart Figure 4.3 Classification and disposition of Tōdaiji documents over time Figure 5.1 Map of Ōgi Village Figure 5.2 Map of Tsuchimaru Village Figure 5.3 Map of Hineno Village Figure 7.1 Map of Kuroda Estate Figure 9.1 Map of villages drawing irrigation water from Sayama Pond Figure 11.1 Jōdodō section diagram, Jōdoji, Hyōgo Prefecture Figure 11.2 Jōdodō ground plan, Jōdoji, Hyōgo Prefecture Figure 11.3 Amida Triad, Jōdoji, side view Figure 11.4 Amida Triad by Kaikei, c. 1221 Figure 12.1 Royal family genealogy Figure 12.2 Jōsaimon-in’s estates (all) Figure 12.3 Map of Jōsaimon-in’s estates, Yamashiro Province Figure 12.4 Map of Jōsaimon-in’s estates, Ōmi Province

xi xii List of Figures, Plates, and Tables

Figure 12.5 Map, properties of Hachijōin and Kōkamon-in in the capital Figure 12.6 Map of Hachijō-in’s estates Figure 13.1 Map of the Saionji family estates Figure 13.2 Map of maritime trade routes Figure 13.3 Map of the River Yodo basin Figure 13.4 Map of waterways around the capital Figure 15.1 Map of northern Naga District, including Nate Estate Figure 15.2 Map of the upper Kinokawa Valley Figure 16.1 Map of Harima Province

Color plates Plate 1 Seated image of Chōgen, Jōdoji, Hyōgo Prefecture Plate 2 Terraces of Ōbe Estate Plate 3 Pre-Kamakura archaeological sites on Ōbe Estate Plate 4 Kamakura-period archaeological sites on Ōbe Estate Plate 5 Model of the abandoned Kōdoji Temple, Hyōgo Prefecture Plate 6 Edo-period illustration of Jōdoji buildings Plate 7 Nigori Pond, Ōbe Estate Plate 8 Map of the site of Hine Estate Plate 9 Map of Hineno Village on Hine Estate, dated 1316 Plate 10 Site of three inner villages, Hine Estate Plate 11 Terraced rice fields on the site of Hine Estate Plate 12 Location of Tsuchimaru Village, Hine Estate Plate 13 Location of Hineno Village, Hine Estate Plate 14 Yugawa Canal, Hineno Village, Hine Estate Plate 15 Plaque with account of Chōgen’s repairs of Sayama Pond, dated 1202 Plate 16 Amida Triad by Kaikei, c. 1195, Jōdoji, Hyōgo Prefecture Plate 17 Chōgen, ﬁrst half of thirteenth century Plate 18 Jōdodō, 1194, Jōdoji, Hyogo Prefecture Plate 19 Amida Triad, 1148 Plate 20 Head of Amida (before restoration) by Kaikei, c. 1202 Plate 21 Amida Triad, Ming period, China Plate 22 Amida Nyorai by Kaikei, 1201 Plate 23 Masks by Kaikei and others, c. 1201 Plate 24 Map of Kōkamon-in’s estates List of Figures, Plates, and Tables xiii

Plate 25 Map of all ports shipping salt to the Hyōgo checkpoint Plate 26 Map of major ports shipping salt to the Hyōgo checkpoint Plate 27 Map of all ports shipping rice to the Hyōgo checkpoint Plate 28 Map of major ports shipping rice to the Hyōgo checkpoint Plate 29 Map of Kii Province

Tables Table 4.1 Superior management rights to Ōbe Estate, twelfth to fifteenth centuries Table 4.2 Current locations of Tōnan’in and corporate temple documents Table 4.3 Selected documents related to Ōbe Estate, Muromachi Period Table 5.1 Chronology of Hine Estate in Izumi Province Table 5.2 Archaeological sites on Hine Estate Table 5.3 Archaeological remains of the Hine site according to type Table 5.4 Hine Estate communities: Kamakura period Table 5.5 Hine Estate communities: Muromachi, Sengoku periods Table 6.1 Extant decrees: “Measures for the Increased Production of Miscellaneous Grains” Table 6.2 Cultivation at Ōbe Estate in the Nanbokuchō period Table 6.3 Approximate acreage of “loss” and “flooded” fields at Ōbe Estate Table 8.1 Strata within Tōdaiji in the Kamakura period Table 11.1 Sculptures at Chōgen’s Bessho (Namu Amidabutsu Sazenshū) Table 13.1 Landholdings of the Saionji family Table 14.1 All goods shipped on Onomichi boats to Hyōgo, 1445 Table 16.1 Akutō in the second Ōbe Estate case, 1322

Acknowledgments

Even more than a monograph, a collective work such as this one is the product of many contributions and many interactions. As detailed in the introduction, the chapters in this volume are based on talks first presented at a conference on the estate system at the University of South- ern California in June 2012. We would like to thank each contributor to the volume not only for his or her chapter or translation, but also for the questions and comments on other contributions at the conference and for lively participation in discussion and analysis of the estate system. Participants at the conference shaped the volume as a whole, and not only through their own presentations and essays. We would especially like to thank Professor Endō Motoo of the University of Tokyo Historiographical Institute, who introduced us in the first place to Ōbe Estate, a major topic of the conference and this volume, and who of course contributed one of its chapters. His assistance, in fact, went much further than that: along with his colleague, Professor Takahashi Toshiko, he helped to organize the conference from the Japan side and assisted with editing the bibliography, and he was our “point man” who contacted the Imperial Household Agency and secured their permission to reproduce the map that appears on the cover of our book. We are very grateful for the efforts of Professors Endō and Takahashi; without them there would have been neither a conference nor a conference volume. Others who helped greatly with the conference and the volume are Kainuma Ken’ichi, who secured permissions and photographs for many of our illustrations; Kana Yoshida of the Center for Japanese Religion and Culture (CJRC) and Lori Rogers of the Department of History, both at the University of Southern California, for help with organizing the conference; Yoshiko Kainuma for invaluable translation assistance; Emily Warren, PhD candidate in Japanese history at USC, for

xv xvi Acknowledgments

proofreading the manuscript; and Philip Garrett, Jim Goodwin, and the late Arnie Olds, for yeoman work on the maps and illustrations. Ono City in Hyōgo Prefecture and the Ono City Education Committee (Kyōiku Iinkai) are to be thanked for maintaining museums and archaeological sites, and producing publications, that became the sources for many of the chapters in this volume. We wish to thank the two anonymous readers for the University of Hawai‘i Press, whose constructive criticism assisted us to shape this volume. Patricia Crosby and Stephanie Chun, Hawai‘i Press editors, were both encouraging and helpfully critical; we hope that Pat, who is now retired, will enjoy reading our work after evenings at the Met and that Stephanie will find our volume one of many bright spots in her editing career. Thanks also to managing editor Cheryl Loe for seeing this volume through the production process, and to copy editor Drew Bryan for his careful reading. We are grateful for financial help from various sources. The Cressant Foundation provided substantial funding for both the conference and the volume. The Northeast Asia Council of the Association for Asian Studies, the CJRC, and the USC History Department helped to support the conference with generous contributions of funding or personnel. We are also grateful to the following organizations for allowing us to reproduce their images or supplying us with photographs: the Impe- rial Household Agency, Jōdoji, Kōdaiin, the Nara National Museum, the National Research Institute for Cultural Properties, the Ono City Kyōiku Iinkai, the Osaka Prefectural Sayama Pond Museum, Saihōji, Sanzen’in, Shin Daibutsuji, Stanford University Press, and Tōdaiji. Finally, we would like to thank Arnie Olds (Joan Piggott’s husband) and Jim Goodwin (Janet Goodwin’s husband) not only for their concrete help with the volume, but for their support through many years of our research history. We regret that Arnie passed away before he could see the results of our hard work together. Abbreviations

DNK Dai Nihon komonjo DNS Dai Nihon shiryō HI Heian ibun KI Kamakura ibun KM Kōyasan monjo OS Ono-shi shi SHM Shōsōin hennen monjo

xvii

Periods of Premodern Japanese History

Jōmon 縄文 To ninth or Hunting and gathering culture sixth c. BCE leading to beginnings of (scholarly agriculture debate rages!) Yayoi 弥生 To 250 CE Rice agriculture; small chiefdoms Tumulus 古墳 250–552 Chiefdoms expand and (Kofun) consolidate power; seeds of a “realm” that came to occupy much of the archipelago Asuka- 飛鳥— 552–710 Early royal polities; expansion Hakuhō 白鳳 of Buddhism as an officially supported religion Nara 奈良 710–794 Throne consolidates power through law codes and control over the land Heian 平安 794–1185 First long-term capital established in present Kyoto. By the tenth century a courtier family, the Northern Fujiwara, comes to dominate the throne and government.

(Insei) (院政) (c. 1100–1185) Retired sovereigns lead the court, involving themselves in accumulation and development of estates. Warriors, especially the Ise Taira, dominate the court at the end of this period. Kamakura 鎌倉 1185–1333 After a civil war with the Ise Taira, victorious Minamoto forces establish a military government (shogunate) in Kamakura; at first the shogunate shares power with the throne and court, and then it becomes dominant.

xix xx Periods of Premodern Japanese History

Kenmu 建武新 1334–1336 Go-Daigo Tennō attempts to Restoration 政 restore the power of the throne

Nanbokuchō 南北朝 1336–1392 Civil war between supporters (Northern of two courts: the northern and Southern court in Kyoto supported by Courts) the new Ashikaga shogunate, and Go-Daigo’s court-in-exile at Yoshino

Muromachi/ 室町/ 1392–1568 Power is consolidated under Ashikaga 足利 the Ashikaga shogunate and provincial overlords (shugo)

(Sengoku) (戦国) (1467–1568) Civil wars. Warlords (daimyō) control territories and vie to expand them.

Azuchi- 安土桃 1568–1603 Civil wars come to an end under Momoyama 山 three successive unifying generals

Edo/ 江戸/ 1603–1867 The final unifier, Tokugawa Tokugawa 徳川 Ieyasu, establishes a shogunate in Edo (modern Tokyo). Japan enters its early modern age. Premodern Provinces and Modern Prefectures

5 Inner provinces (Kinai) and 7 circuits (Shichidō) 五畿七道 Province Japanese Prefecture Honshu and nearby small islands Yamashiro 山城 Kyoto Yamato 大和 Nara Kinai 畿内 Kawachi 河内 Osaka Izumi 和泉 Osaka Settsu 摂津 Osaka Iga 伊賀 Mie Ise 伊勢 Mie Shima 志摩 Mie Owari 尾張 Aichi Mikawa 三河 Aichi Tōtōmi 遠近 Shizuoka Suruga 駿河 Shizuoka Tōkaidō 甲斐東海道 Kai Yamanashi Izu 伊豆 Shizuoka Sagami 相模 Kanagawa Awa 安房 Chiba Kazusa 上総 Chiba Shimōsa 下総 Chiba Hitachi 常陸 Chiba Musashi 武蔵 Saitama, Tokyo, Kanagawa

xxi xxii Premodern Provinces and Modern Prefectures

5 Inner provinces (Kinai) and 7 circuits (Shichidō) 五畿七道 Province Japanese Prefecture Ōmi 近江 Shiga Mino 美濃 Gifu Hida 飛騨 Gifu Shinano 信濃 Nagano Tōsandō 東山道 Kōzuke 上野 Gunma Shimotsuke 下野 Tochigi Mutsu 陸奥 Fukushima, Miyagi, Iwate, Aomori Dewa 出羽 Yamagata, Akita Wakasa 若狭 Fukui Echizen 越前 Fukui Etchū 越中 Toyama Hokuriku 越後北陸 Echigo Niigata Kaga 加賀 Ishikawa Noto 能登 Ishikawa Sado 佐渡 Niigata Tanba 丹波 Kyoto Tango 丹後 Kyoto Tajima 但馬 Hyōgo

San’indō Inaba 因幡 Tottori 山陰道 Hōki 伯耆 Tottori Izumo 出雲 Shimane Iwami 石見 Shimane Oki 隠岐 Shimane Premodern Provinces and Modern Prefectures xxiii

5 Inner provinces (Kinai) and 7 circuits (Shichidō) 五畿七道 Province Japanese Prefecture Harima 播磨 Hyōgo Mimasaka 美作 Okayama Bizen 備前 Okayama

San’yōdō Bitchū 備中 Okayama 山陽道 Bingo 備後 Hiroshima Aki 安芸 Hiroshima Suō 周防 Yamaguchi Nagato 長門 Yamaguchi Kii 紀伊 Wakayama Awaji 淡路 Hyōgo Shikoku Nankaidō 阿波南海道 Awa Tokushima Sanuki 讃岐 Kagawa Iyo 伊予 Ehime Tosa 土佐 Kōchi Kyushu Chikuzen 筑前 Fukuoka Chikugo 筑後 Saga Buzen 豊前 Fukuoka Bungo 豊後 Ōita

Saikaidō Hizen 肥前 Nagasaki 西海道 Higo 肥後 Kumamoto Hyūga 日向 Miyazaki Satsuma 薩摩 Kagoshima Ōsumi 大隅 Kagoshima Iki 壱岐島 Iki Island Tsushima 対馬 Tsushima Island

Introduction

Janet R. Goodwin

Landed estates known as shōen played a central role in the society and economy of late classical and medieval Japan. Estates produced much of the material wealth that supported all levels of society, from the aristocratic court in the capital of Kyoto to the farmers who worked the land. At various times during the tenth through sixteenth centuries, estates were locations of de facto government, homes to communal structures, nodes along trade networks, sites of developing agricultural technology, and centers of religious practice and ritual. Although mostly farmland, they yielded many nonagricultural products: lumber, salt, fish, and silk, to name a few. As Amino Yoshihiko has shown, estates also provided livelihood for craftsmen, seafarers, peddlers, and performers as well as for cultivators.1 By the twelfth century at the very latest, we can talk about an estate “system” that permeated all institutions and social classes across much of the Japanese archipelago. (For a list of premodern provinces and their modern equivalents, see page xxi.) While many of the first estates arose as simple holdings of a single family or institution, they eventually took on a complex system of ownership.2 By the later Heian age, no single entity had absolute ownership rights to an estate; rather, a hierarchy of individuals and institutions held rights (shiki ) to income from the land or specified portions of it. These rights also involved various responsibilities. Moreover, through the system that granted income rights to various parties, estates created vertical ties between those on different social levels, a point made eloquently some years ago by Cornelius Kiley.3 At the very top level, a high-ranking courtier or member of the royal family, well-placed to protect the estate from confiscation or increased taxation, might serve as supreme proprietor (honke, also translated as guarantor or patron). Not all estates had supreme proprietors, however, and superior tenure was often held

xxv Fig. I.1. Japan’s medieval provinces. From Jeffrey P. Mass, ed., The Origins of Japan’s Medieval World, Stanford University Press, 1997. Originally drawn by Arnie Olds. Used with permission. Introduction xxvii by a court noble or a powerful religious institution—a proprietor or ryōke —that controlled the estate as an absentee landlord. On- and off- site managers, known by a number of different titles depending on their duties and the estate in question, also had rights to income; these rights, and the duties that came with them, often became heritable and their holders could not be easily dispossessed. At the next lower level were headmen called myōshu, who supervised cultivation and collected rents from designated fields—myōden or named fields—and delivered them to estate managers. Such headmen were the most prominent among the hyakushō, prosperous local residents analogous to medieval English yeomen, who increasingly claimed and defended their own rights to occupy and exploit the land.4 In the later twelfth century, leading estate residents of the managerial and yeoman class were sometimes warrior retainers of powerful military commanders.5 And later, in Kamakura times, the most prominent of these warriors were military stewards (jitō ), who oversaw individual estates on behalf of the warrior government (shogunate) at Kamakura. The stewards were initially Minamoto Yoritomo’s retainers placed on estates confiscated from the Taira after the Genpei War (1181–1185), but they eventually were assigned to more estates and made responsible for policing and ensuring that rents were delivered to proprietors. Like other holders of rights to income, stewards took a portion of the estate’s receipts. Estates were not always contiguous territories, and some fields within them, as part of the provincial domain, might be subject to taxation by the provincial government. In many cases as time went on, however, borders were set to enclose their territory and provincial officials were forbidden to enter the estate to conduct cadastral surveys, pursue criminals, or levy taxes. Often no taxes at all were paid to the province. This situation, increasingly common from the twelfth century on, has led some observers to refer to estates as private holdings, but the boundary between the public and private spheres was fuzzy, and the individuals, families, and institutions in the proprietary class were in essence the government of the realm. Major landholders, in fact, belonged to a category that the historian Kuroda Toshio has dubbed kenmon —“gates of power”—the elite triumvirate that included the throne, the retired sovereign, and leading nobles in Kyoto; important Buddhist temples as well as shrines where mostly indigenous Japanese deities called kami were propitiated; and from the end of the twelfth century, the military xxviii Introduction

establishment.6 Over time, the dominant political power shifted from the throne itself to the court, then to the retired sovereign, and finally to the military. By around 1200, however, all three elements of the “gates of power” derived much of their economic support from the estate system and exercised governmental powers on their holdings.7 (For an overview of premodern historical developments, see page xix.) Estates existed in tandem with lands taxed by the province, and neither management structures nor agricultural practices differed drastically from one to the other. In fact, Nagahara Keiji refers to the estate and provincial land (shōen-kokugaryō) system, emphasizing the overlap between the two landholding categories. While revenues from provincial lands went to the government rather than to royals, nobles, or religious institutions, local officials played roles similar to those of estate managers, forwarding taxes and keeping a percentage of the take for themselves. Eventually, like estate managers, many such local officials secured hereditary rights to their positions and income.8 And as Ishii Susumu has pointed out in a study of Tashibu Estate in Bungo Province, landholdings called villages (gō) or bays or ports (ura) often operated in the same way as those called estates (shō).9 While provincial lands were often incorporated into estates in the later twelfth century, provincial and local governments and officials continued to reap benefits from publicly taxed holdings, and in some cases dues were shared: those calculated in rice were paid to the government, and those in labor (cor- vée) to the estate proprietor, or vice versa.10

How and Why Do We Study Estates? Estates have long been a fertile field for Japanese historians. As Joan Piggott demonstrates in chapter 1 of this volume, scholars have offered a number of different interpretations of the inception and evolution of the estate system and its relationship to the economy and society. How do we know what we do about estates? More broadly, how do we go about studying a distant past that supplies us with only fragments of evidence? Chapters by Nishida Takeshi, Hirota Kōji, and Endō Motoo help us to do just that. Archaeological investigations such as those described by Nishida and Hirota provide one view, based on remains of paddy fields, irrigation systems, dwellings, and official structures. One way modern historians learn about an estate’s past is to “walk the land” of the estate, a pro- Introduction xxix cess discussed in Joan Piggott’s chapter 1. Several years ago, guided by Nishida, Endō, and Noda Taizō, she and I did just that: we explored the land that was once Ōbe Estate. We tried to envision what that land must have looked like to Chōgen (1121–1206), the monk who developed the estate on behalf of its proprietor, Tōdaiji, and how it would have evolved through land reclamation and irrigation projects over the centuries. We viewed the remains of villages, irrigation canals, and an old castle, and we visited the still-extant temple Jōdoji on a bluff overlooking the estate. Joan Piggott accompanied Hirota Kōji on a similar journey across the landscape of Hine Estate, viewing the sites of old temples, graveyards, and irrigation canals, and both of us have visited the Izumisano Museum, which contains maps, artifacts, miniature reproductions of Hine Estate sites, and even an animation demonstrating the way proprietors, on-site managers, and cultivators quarreled over the estate’s harvest. Much of what we saw is detailed in the chapters by Nishida and Hirota, and we hope readers will virtually “walk the land” along with us. The written record, too, is extremely important and often requires considerable detective work for proper evaluation. Sometimes we are fortunate to have diaries of those involved in estate management, such as the sixteenth-century proprietor of Hine Estate introduced by Hirota. More commonly, however, we rely on documents that relate everyday matters such as rents assessed, profits distributed, and quarrels launched and settled. Sometimes a historian needs to examine documents as if they were archaeological artifacts, finding clues to their significance even in where and how they were stored, as does Endō in his chapter. Exploring documents in detail this way helps us to understand the way in which estates were developed and managed. Research on estates, moreover, reveals concerns of and interactions among human beings that might otherwise be occluded from history. Many of these concerns and interactions are featured in this volume. Land surveys reporting fields lost to drought, as discussed by Kimura Shigemitsu, for example, give some sense of the suffering of ordinary farmers from disastrous weather. Conflicts within estates and between estate residents and outsiders were often presented to higher authorities for judgment, and as detailed here by Hirota, Dan Sherer, and Noda, we learn just how the disputants related to one another. Ōyama Kyōhei examines a plaque, discovered in archaeological investigations of an irrigation pond, that shows how common people, including those considered outcasts, crossed village and estate boundaries to collaborate in xxx Introduction repairing the pond (see plate 15). Close study of estates shows us how proprietors of the highest class developed tactics for increasing both power and income; see the chapters by Sachiko Kawai and Rieko Kamei- Dyche. But we are also given a window into the lives of the other 99 percent in medieval Japan. In part because they are messy, estates challenge historians. Case studies of estates abound, proving that not all estates looked alike. Ōbe Estate in Harima Province, the focus of several chapters in this volume, developed along a different trajectory than Hine Estate in Izumi (studied by Hirota), the Kii Province holdings of the temple complex on Mount Kōya (examined by Philip Garrett), or Kuroda Estate in Iga Prov- ince (researched by Piggott). Estates were formed in different ways and at different times, as Piggott shows us in chapter 1. While the same titles were often used for officials on various estates, moreover, we should not be fooled into thinking that everyone who had the same title also had the same duties or even lived at equivalent places. For instance, the custodians (azukaridokoro) who represented the interests of estate proprietors included the capital-based female attendants of royal ladies (see the chapter by Kawai) as well as persons who lived on the estate and directly managed the land (see the chapter by Goodwin). Important as it is to look at estates one by one, it is also important to contextualize estates, as have many Japanese scholars, such as Amino Yoshihiko, who has looked at estates within a framework that included sea transport and nonagrarian people.11 In our volume, Ōyama’s chapter helps us to view estates, with their vertical ties between central elites and the local population, as only one structure in a complex system that included lateral ties among villages as well. Since developments in agriculture, commerce, and transport contributed to the evolution of estates, we need to look at the estate “system” not as a closed structure, but as a component of a broader sociopolitical order. Estate structures and practices also cast a long historical shadow: for example, travel and shipping routes that carried goods from estates to proprietors, and proprietors’ representatives to estates, continued as trade routes even as estates crumbled. Michelle Damian’s chapter illustrates this quite suc- cinctly. Moreover, scholars have cited the vertical estate structure as a template for the early development of merchant and artisan guilds (za), which also depended upon patron-client ties.12 This volume provides views of a number of different estates, as well as a focused look at one example, Ōbe Estate, which several of the con- Introduction xxxi tributors to this volume have studied closely (see figure 16.1). What insights into the estate system as a whole might a study of Ōbe Estate provide? First of all, the estate was instituted by its would-be proprietor, Tōdaiji, in negotiation with court authorities, not through a process of commendation from below, which is more familiar to English readers. Thus the Ōbe case supports recent scholarship that points to central proprietors rather than local landholders as those who initiated estate development in the late twelfth and thirteenth centuries. Second, shortly after Ōbe Estate was established, Tōdaiji and the Harima provincial government quarreled over its tax-free status, a common conflict throughout the estate system. Third, Ōbe provides a prime example of complexities in ownership and control that characterize most if not all medieval estates. Fourth, the early development of Ōbe Estate depended ultimately on religious authority, a common thread in an institutional arrangement in which temples and shrines managed landholdings as major players in realm-wide economics and politics (see the chapters by Ōyama, Nagamura Makoto, and Yoshiko Kainuma, as well as my own; for comparison, see Piggott’s chapter on Kuroda Estate). Finally, like many estates, Ōbe Estate saw endemic conflict, both within its borders and with outsiders (see the chapters by Sherer and Noda, and for comparison see Garrett’s chapter on Mount Kōya holdings). Thus while there are many differences among estates in their genesis, management, and ultimate control, a focused look at a single estate can provide insights into the way the system operated as a whole.

How Our Volume Came to Be This volume, in fact, owes its genesis to documents from Ōbe Estate. Seeds were planted in the summer of 2005, when Professor Endō Motoo of the University of Tokyo Historiographical Institute, then teaching the Kanbun Workshop at the University of Southern California, introduced some examples from a volume of all extant available documents related to Ōbe Estate, from its founding in 1147 to its demise in the late sixteenth century.13 All the documents were conveniently contained in a single volume: how could we resist? The USC Library obtained Ono-shi shi (The history of Ono City), the series that contained this volume, and a group of us, including faculty, independent scholars, and graduate students, began to interpret its documents in English. xxxii Introduction

As time went by we found that we needed to study not only Ōbe Estate but also the entire estate system. In addition to launching a graduate seminar on the history of estates at USC to do just that, we organized an international conference to which we invited specialists in medieval history from Japan, the United States, and the United Kingdom. In three days in June 2012 at USC, some twenty of us explored various aspects of estates: developments in agriculture and irrigation systems, the role of religious institutions and individuals, aristocratic and temple proprietors, conflict on and among estates, and the role of estates in a developing commercial economy. We asked how we know what we do about estates, explored ways we might find out even more, and discussed the teaching of estates to undergraduates to give them more insights into the lives of people from all walks of life, from both the capital and the countryside. Most of the presentations at that conference have been developed into chapters, and you can read them in this volume. For the first time, a rich variety of perspectives on the field of estate studies has been made available outside of the Japanese language. As supplementary materials, translated documents and commentary can be accessed on Piggott’s website Shōen Documents and Materials, as can her database of published Tōdaiji archives.14 An important feature of both conference and volume is the collaboration between scholars from Japan and those from the West. Fellow editor Joan Piggott and I were fortunate to have dissertation advisors who encouraged such collaboration. Delmer M. Brown, my advisor at the University of California, Berkeley, is known for working with Ishida Ichirō on a study and translation of the thirteenth-century historical text Gukanshō,15 and Jeffrey P. Mass, Piggott’s advisor at Stanford, worked on several projects with Takeuchi Rizō, Seno Seiichirō, and Ōyama Kyōhei (one of our authors). Our advisors, unfortunately, have both passed away, but we hope that they would have been pleased with our work on this project.

Structure and Themes of This Volume The chapters in this volume focus largely on estates themselves and the people and institutions connected with them, but all the contributors have kept in mind the larger picture of Japan’s historical development. What we hope to present in this volume is a view of that development through the lens of individual estates and the people involved in them, Introduction xxxiii within the context of the broader estate system and Japan’s medieval economy. The volume begins with two contextualizing chapters: Piggott’s overview of the history of estates, based on the work of a number of distinguished scholars, primarily in Japanese; and Sakurai Eiji’s examination of the way the economy developed, from Kamakura times until the first century of the early modern age. Sakurai shows how commercial- ization fundamentally changed estates, which became sites of markets and monetary exchange. After setting the overall scene, we then turn to methodological issues: “How Do We Know about Estates?” The chapters by Nishida Takeshi, Endō Motoo, and Hirota Kōji not only indicate the bases for our current understanding of estates, but also suggest how future research might progress to expand our knowledge further. As the title indicates, this volume explores three closely interrelated themes: land, especially how it was made fruitful; power, as derived from the land and shaped by conflicts over its people and produce; and the sacred, aspects of which were intimately connected to the development and control of land and to the exercise of power. Just as these themes were interwoven in actual medieval practice, so they are in the chapters of our volume. As one example, Hirota’s chapter, besides informing us on methodologies, shows how one estate embodied these themes as he explores the development of irrigation facilities on Hine Estate, conflicts over control among proprietors and local warriors, and the role of temples and shrines in estate management. One important issue addressed in this volume is how land was developed into dry fields and paddies that provided livelihood for residents as well as revenues for absentee proprietors. In a section titled “Mak- ing the Land Productive,” chapters by Kimura Shigemitsu and Joan Piggott discuss, from two totally different perspectives, how arable was developed and managed and how estates were formed. Kimura uses examples from the eighth through the fourteenth centuries, from both estates and provincial land, to examine agricultural and irrigation technology; he points out the importance of dry fields, often overlooked in studying landholdings that calculated their tax or rent payments in terms of wet paddy rice production. Kimura, in fact, was a pioneer in revising the “paddy-centric” view of the medieval society and economy, a revision that has exploded not only in Japanese-language studies but in those in English as well. In her chapter, Piggott examines the establishment of Kuroda Estate in the Heian period and the often difficult xxxiv Introduction relationships of its proprietor, its residents, and outside entities such as provincial officials. The chapter clarifies the arduous and contested process of converting provincial lands into estates. Both chapters make it clear that cooperative relationships among the various social levels engaged in the estates were absolutely necessary, even though conflict, exploitation, and resistance certainly did take place. Next, in “Secular and Sacred,” we explore religious factors as they intersect with themes of land and power in the establishment and management of estates. Nagamura Makoto examines the different ways regular temple monks and unofficial religious practitioners called hijiri engaged in estate management. One crucial point here, of course, is the involvement of religious figures in such seemingly “secular” activities. This point is also explored by Ōyama Kyōhei, who demonstrates that religious figures such as Chōgen, who established and managed Ōbe Estate while rebuilding Tōdaiji’s Great Buddha and its hall in the late twelfth century, did not view “the secular” and “the sacred” as totally different spheres. The two chapters that follow investigate Chōgen’s activities on Ōbe Estate: my chapter traces the connections among religious proselytization, land reclamation, and community formation, while Yoshiko Kainuma’s focuses on the temple Chōgen established on the estate, asking how its architecture and images contributed to estate management and control. Not only monks and hijiri had religious motivations for establishing and managing estates. Aristocratic laymen and women often sought estate revenues to support “vow temples” they had founded to pray for the souls of deceased ancestors. Sachiko Kawai cites examples of such in her examination of royal ladies called nyoin as estates’ supreme proprietors. Kawai’s chapter is thus linked to the previous section, but it and the chapter by Rieko Kamei-Dyche primarily concern power as manifested by strategies of estate accumulation. Both these chapters, in the section titled “Power, Space, and Trade,” demonstrate the importance of estate location and the development of estate networks, which were often engaged in trade and transport. The power obtained by such proprietors was derived not only from agricultural produce from the land, but also from estates’ participation in new forms of income generation and control over dispersed population. Michelle Damian’s chapter also focuses on trade, in this case lateral exchange between estates and between ports in the twilight age of the estate system. This chapter resonates with those by Kawai and Introduction xxxv

Kamei-Dyche in demonstrating the role of the sea as a connective force and also with Ōyama’s demonstration of the lateral ties among villages that had begun to develop centuries before. Along with the chapter by Sakurai, the chapters by Damian and Ōyama point to horizontal forces that competed with, and eventually helped dissolve, the old vertical hierarchies that had formed the basis for both the estate system and the larger society. Neither lateral nor vertical ties prevented conflict. One might even argue that estates might be formed through the process of conflict and resolution, as we see in the battles between the proprietary temple Tōdaiji and the Iga provincial government from which Kuroda Estate emerged (see Piggott, chapter 7). Conflicts played out in courts of law and on the battlefield, and they involved such issues as border disputes, water rights, and the involvement of warrior authorities in estate operations, even on the holdings of religious institutions. Chapters by Philip Garrett, Dan Sherer, and Noda Taizō introduce some of these conflicts in a section titled “Power: Challenges and Conflicts,” which explores how disputes were settled (or not) and the way they were entwined with the exercise of power. These three chapters, along with the overview of Hine Estate history in the chapter by Hirota Kōji, emphasize the momentous change in medieval Japanese history as warriors displaced civilian aristocrats as top elites and chief local holders of power. Beginning with the “dual polity” of the Kamakura period, in which the warrior shogunate shared power with the court, the process reached a watershed in the mid-fourteenth century, when the sovereign Go-Daigo’s attempt to restore the throne to power resulted instead in the firmer establishment of warrior dominance in the form of the Muromachi shogunate. While Garrett and Sherer demonstrate the ubiquitous armed conflict on estates well before that, Noda’s chapter focuses on the Muromachi aftermath. Noda exhibits in detail the course and results of warrior interference in estates held by others, particularly the religious establishment. This was one way, of course, in which warriors ended up as the dominant figures in the elite triumvirate and it is part of the process by which Japan broke into “warring states” in the late medieval age. Finally, in this volume’s last chapter—in a section titled “Getting the Word Out”—Ethan Segal asks how best to teach the estate system to undergraduates in American universities, and he critically examines the textbooks and other materials available for the task. The chapter xxxvi Introduction explores both the value and the problems of teaching about estates, often regarded as difficult even by Japan specialists. At the conference, Segal’s presentation on this subject generated a great deal of interest among the Japanese scholars attending, who saw the issue as a crucial concern in their own teaching of this important aspect of their own country’s history. Our intent is that this book prove helpful to professors who endeavor to teach estates—the history of the 99 percent—to their students, as well as to students themselves; to scholars conducting research on medieval Japanese history; and to those in other fields—premodern China and Korea, or medieval Europe, for example—seeking comparative materials for the phenomena they study. The volume might also be used as a handbook for those seeking information on various aspects of medieval society. With these goals in mind, we introduce a series of chapters dealing with the significant issues of land, power, and the sacred in medieval Japan.

A Note on Measurements In premodern Japan, the standard measure for land area was the chō, equivalent to about 2.94 acres until Toyotomi Hideyoshi’s cadastral survey in 1582. (In medieval times, however, the area of the chō varied according to geographical location.) A tan was one-tenth of a chō and a bu was 3.95 square yards. The common equivalent given for the koku, a measure of capacity, is 180 liters, but the measure seems not to have been standardized throughout the archipelago in medieval times, and the actual equivalent may be less.16 A to is one-tenth of a koku and a shō is one-one-hundredth of a koku. As for linear measurement, the shaku is almost equal to a foot.

A Note on Style and Translations The editors have decided to accept minor variations in style, including the way some personal and estate names are written and in the translation of some Japanese terms. It should be noted that some terms changed in meaning over time and others were broadly and imprecisely applied. For clarification, see the glossary. In the text, we have translated titles of Japanese works unless, as in some cases, the English ren- derings do not make much sense.

All maps and diagrams are by the authors unless otherwise indicated. Introduction xxxvii

Notes

1. In English, see Amino 2012, 65–78. 2. Some early estates, called shoki shōen by modern scholars, were established in the eighth century, but structurally, they were quite different from later estates. For details see Piggott, chapter 1. 3. Kiley 1974. 4. The terms myōshu and hyakushō resist easy translation and are variously rendered. I have chosen Jeffrey P. Mass’ translation of myōshu (1976, 204) and Philip Garrett’s translation of hyakushō (see his chapter in this volume). In some cases “cultivator” may be an appropriate term for hyakushō, but for an argument against seeing hyakushō simply as cultivators, see Conlan 2003, 108–109, 112–114. 5. Kamakura 2008, 53–55; Conlan 2003, 116–121. 6. Kuroda 1976. For a discussion of Kuroda’s ideas in English, see Adolph- son 2000, 10–20. 7. Joan Piggott’s historical overview in this volume (chapter 1) covers many of these topics in more detail. 8. Nagahara 1975. 9. Ishii 1995a, 14. 10. Nagahara 1975, 295. 11. For example, Amino 2012, 65–78. 12. Endō 1933, 3–4; Tonomura 1992, 105; Gay 2001, 56–57. 13. OS 4. 14. Both forthcoming. 15. Brown and Ishida 1979. 16. See Farris 2006, 19–20, and the chapter by Michelle Damian in this volume. pa rt i

The Big Picture

c h a p t e r 1

Estates Their History and Historiography

Joan R. Piggott

Property units called shōen—“estates” in English—loom large in the history of classical and medieval Japan. In medieval times in particular the estate was a major landholding institution, a structure that linked capital and countryside together politically and economically, and a stage of everyday life for large numbers of the population. Since I began studying estates in the mid-1970s, I have been strongly attracted to the case-study approach, inspired by Ishimoda Shō (1912– 1986), whose work The Formation of the Medieval World (Chūseiteki sekai no keisei, 1946) on Kuroda Estate I read early on. I really like exploring the local world of an estate through intensive engagement with its written and archaeological records.1 Nevertheless, I have long been aware that a broader view is needed as well, to contextualize and integrate individual case studies. This volume, which might be used as an introduction to estate history for general readers as well as a compendium for specialists, will demonstrate both approaches. In this chapter I present a syncretic history of estates, based on research since the 1970s in both Japanese and English.2 In a later chapter I provide a case study of the formation of a particularly well-known estate, Kuroda in Iga Prov- ince, during the Heian period (794–1185). Together the two chapters give English readers a sense of how the estate evolved over time as a key structure characterizing the transition to and the development of Japan’s medieval age. This chapter has been inspired primarily by three important studies: (1) Nagahara Keiji’s The Estate (Shōen, 1978, 1998), (2) An Introduction to Shōen Studies (Shōen nyūmon, Kōza Nihon shōenshi vol. 1, 1989), edited by Amino Yoshihiko, Ishii Susumu, Inagaki Yasuhiko, and Nagahara

3 4 The Big Picture

Keiji, and (3) A Handbook for Research in Shōen History (Shōenshi kenkyū handobukku, 2013), edited by members of the Shōen History Research Group (Shōenshi Kenkyūkai). These volumes by leading researchers in the fields of estate studies and medieval Japanese history provide a variety of perspectives on the chronology, organization, and significance of estates through classical and medieval times. Such information helps us understand not only how estates developed and functioned, but also how the estate as a fundamental social structure affected other aspects of Japan’s premodern history. Moreover, given that publication of these works spans the last half-century, reading and thinking about them in tandem results in a good picture of the contemporary state of the field.

Nagahara Keiji: Fundamentals of Estate History In his very accessible three-hundred-page book titled Shōen, Nagahara Keiji provides a rich account of estate history over eight hundred years, with a useful glossary and chronology. He begins with a definition and basic terminology: the estate was a multilayered landed proprietorship wherein each layer constituted an officership (shiki) with responsibilities, prerogatives, and perquisites. By the late twelfth century, as the structure and the system it created reached maturity, urban absentee landlords—none other than the highest-ranking court nobility and great official religious institutions—took the role of supreme proprietors (honjo, honke) and proprietors (ryōshu, ryōke) at the apex of the landholding hierarchy. Further down in the chain of command, lesser nobles acted as liaison personnel, and local elites served as agents, managers, and rent collectors on estates. In Nagahara’s view, during medieval times the estate was the main social structure because its shiki hierarchy served as the system linking capital and provinces in what he called “the shiki - based polity.”3 In addition to analyzing the political and social functions of the estate, Nagahara urges readers to think of the estate as a stage where life took place. As one way of studying estates, Nagahara walked the land to learn about local topography, geography, and habitus. He often led other researchers out to estate sites—I walked what was once Ōta Estate in Bingo with him in the early 1980s—and he worked closely with local historians and archaeologists to gain clues as to how estate residents once lived.4 Such walks are an important way of making the Estates 5 written record come alive while enhancing our ability to read and interpret that record.5 Nagahara’s work underscores the complexity of estates. They were situated in mountain meadows and valleys, along riverbanks, and near coastlines. In addition to agriculture, estate dwellers were involved in fishing, salt making, lumbering, pasturing, pottery making, dyeing, tex- tile production, and transport by land and sea, among other occupations.6 There were regional differences as well: for instance, case studies have shown that local elites exercised less power near the capital where absentee proprietors were based, while in the thirteenth century, in the Kantō region of eastern Japan, the warrior-led Kamakura shogunate subjugated local elites to its power, weakening the independence of estate structures there. As a general chronology for his eight-centuries-long estate history, Nagahara narrates the story in three parts: (1) a developmental stage of “early estates” (shoki shō) from the eighth to the eleventh centuries, (2) a stage of mature estates characterized by the shiki system during the twelfth and thirteenth centuries, and (3) a stage of decline from the fourteenth to the sixteenth centuries.

Early Estates The earliest estates were instituted by the royal government in the eighth century to encourage the opening of new fields (konden), rent from which would also support official religious organizations like the royal temple Tōdaiji (Great Eastern Temple) in Nara. Tōdaiji itself was autho- rized to open more than 3,400 chō (one chō = 2.94 acres) in nineteen provinces. A good example of such a property that has been extensively studied is Kuwabara Estate in Echizen Province on the Japan Sea coast, where local notables including the district chieftain (gunji) donated land to the temple and then worked with provincial and temple officials to open it and keep it cultivated. Because there were no residents at the new property, the district chief recruited neighboring cultivators of officially distributed fields (handen) to rent and farm the new fields, on which they paid 20 percent of the harvest as rent to the temple. They were otherwise exempt from provincial taxes.7 Given the lack of long- term residential cultivators, however, early estates like Kuwabara that supported official temples proved difficult to keep in cultivation, making their operation short-lived.8 6 The Big Picture

After these eighth-century estates came a second wave of estate formation in the later tenth and early eleventh centuries, associated particularly with tax-exempt fields (menden). The process began when provincial governors recruited land openers of middling status, often called field contractors (tato) or private landholders (shiryōshu) in the records, to open fields and pay taxes on them. The governors’ objective was to increase the productivity of their bailiwicks. In the later eleventh century, however, which saw provincial governors becoming more aggressive tax collectors, these contractors and local elites associated with them often allied themselves with high-ranking courtiers or official religious institutions by “commending” their fields to the powerful patron, since that was the best way to protect against increased taxation or confiscation by provincial authorities.9 Thereafter openers stayed on the land, managed their commended fields, collected rent, and sent part of it to the patron, who was the equivalent of an absentee proprietor. Under these new circumstances, the proprietor was expected to provide surety in cases of trouble with provincial authorities or collectors of special levies, or in case of quarrels with neighboring properties. A well-known example of such a commended estate is Kanokogi Estate in Kyushu’s Higo Province, where the commending process by a land opener to a high-ranking courtier reportedly began in 1029.10 Nagahara also gives the example of the great regional temple Kanzeonji of Chikuzen Province, which received many commendations in the early eleventh century. It then transformed the commended fields into temple estates by getting borders established and gaining immune (funyū) status for the holdings, meaning that provincial officials were forbidden to enter to survey or collect taxes.11 Later in this volume we will see the specific case of Tōdaiji’s Kuroda Estate, the eleventh-century history of which resembles that of Kan- zeonji’s properties. Yet another vector pushing estate development in the eleventh century was the breakdown of the benefice support system for official religious institutions such as Tōdaiji in Nara. Originally, provincial officials were mandated to forward revenues from designated benefice residence units (fuko) to official temples and shrines, and when that duty was neglected, religious institutions were unable to support their personnel, rituals, and physical plant. These institutions then demanded that their benefice households be exchanged for tax-exempt fields, which might then be developed into estates. Good examples are Iga’s Kuroda Estate, Estates 7 as we shall see later in this volume, and Kohigashi Estate in Yamato Prov- ince, formed at Tōdaiji’s request in the late eleventh century.12 According to Nagahara’s analysis, the later eleventh century was a particularly dynamic time for estate development. Even as the court was issuing regulations that allowed provincial authorities to curb the expansion of estates—they were known as shōen seiri rei—cultivators were increasing their efforts to open land. When provincial surveys then revealed that newly opened fields had pushed beyond recognized estate boundaries into the official domain (kōryō), provincial authorities would confiscate the fields, following estate-regulating laws like those of 1045 and 1069. Cultivators then appealed for negotiation and protection from their estate proprietor, who sought special court orders (kanshōfu) to certify their properties as tax-exempt (fuyu).13 Kyoto’s Tōji, for instance, won this sort of extraterritoriality for its Ōyama Estate in Tanba Province in 1042. The history of that estate demonstrates, however, that such properties still experienced ups and downs, as proprietors and provincial governors quarreled and negotiated over the extent of tax exemption and other prerogatives.14 In the midst there were violent confrontations between cultivators, sometimes aided by their proprietor’s agents, and provincial authorities, aided by local officials such as the district chieftain (gunji ) and township heads (gōshi ). And in the Kyoto capital, there were judicial hearings where authorities—the court, the regent, and, in the twelfth century, the retired monarch’s office (in-no-chō )—sought, often with substantial difficulty, to resolve conflicts of interest between the court’s needs as opposed to claims by proprietors, provincial authorities, and cultivators. Proprietors spent much energy organizing their records for use as proof documents—some are still extant today, as we will see later in the case of Kuroda Estate—to prepare for these judicial contests.15

The Estate System Nagahara considers the mid-twelfth century, especially the era of the retired monarch Toba (Toba In) from the 1120s to the 1150s, as an important epoch for the expansion of estates and the development of the new estate-based system that knitted the realm together in new ways. Specifically, Nagahara estimates that 60 percent of cultivated land came to be operated as estates, leaving only 40 percent of cultivated fields in the official domain where regular taxes were collected. Nagahara also 8 The Big Picture points out that by that time, even in the east of Honshu where estates and the shiki hierarchy developed late, officials associated with the provincial office were opening land and gaining hereditary control of the fields of townships in Hitachi Province. Furthermore, Kamakura-period (1185–1333) documents confirm that in both Hitachi and neighboring Shinano, 50 percent of all fields came to be located within estates. Two powerful vectors pushing this advance were: (1) land openers of various strata, from local elites to middling nobles that included fifth- rank governors, were commending land to high-ranking patrons in the capital who could use their influence to gain special privileges for their property, including exemptions and heritability, and (2) in the official domain, provincial governors were encouraging local officials to open land that was then designated hereditary township (gō, ho, or betsumyō) holdings, as a reward for service by local notables to the province. An early example is the case of the Mita township head in Aki Province’s Takada District. He received his appointment from the Aki governor in 1030, after opening fields for which he was given the heritable prerogative of collecting rent from cultivators. At the same time, he inherited the position of Takada district chieftain, although his hereditary appointment there still had to be confirmed by the provincial governor.16 For Nagahara, a notable point here is that local officials in the official domain were receiving appointments as officers (i.e., as shiki holders) just as were those in the estate hierarchy, meaning that such hierarchies were shaping social relations both within estates and in the surrounding official domain. The result was what historians call “the estate and official land system” (shōen-kōryōsei ).17 Subsequently over the course of the twelfth century, many of the new townships were amalgamated into large territorial estates, that is, estates with designated borders and markers that came to cover an entire district or more, under the protection of “supreme proprietors” such as premier royal ladies (nyoin) or the royal vow temples (goganji ) of monarchs.18 Notable too is that those who arranged for the institution of these new properties were not commenders. Rather they were leading capital elites and provincial governors eager to “institute estates” (risshō ) in order to support royal projects such as a vow temple or ritual. An example of such is Ōta Estate in Bingo, whose opening was arranged by the courtier Taira Shigehira with Go-Shirakawa as supreme proprietor and Shigehira himself as custodian (azukaridokoro).19 Another instance is Kanokogi Estate in Higo, mentioned earlier, which began as commended fields and then was Estates 9 expanded as an instituted estate when a queen-consort and then a royal temple took over as the property’s supreme proprietor. Why estates developed and whether the estate structure was subversive of or integral to court government have been debated extensively. Economically speaking, Nagahara argues that throughout the Heian period the main purpose of estates was to provide urban proprietors with goods and services; they were part of the official economy. Hotate Michihisa agrees: Given that the court government was made up of urban absentee landlords, estates were critical for their provisioning, and therefore they were neither extralegal nor illegal.20 This counters earlier arguments. For instance, Asakawa Kan’ichi, who introduced estates to English readers, saw estates as beyond the power of the state judicially and fiscally.21 Takeuchi Rizō (1942), whose work set the foundation for studying the estates of temples like Tōdaiji, took that stance as well, as did Kuroda Toshio (1975) and Toda Yoshimi (1991). And for both Kuroda and Toda, cultivators of estates were more independent than farmers who worked fields in the official domain. They were more serf-like than slave-like, which led both scholars to characterize the authority of the estate hierarchy as “feudal.” On the other hand, Nagahara sees the shiki hierarchy as patrimonial and bureaucratic, like the court’s authority.22 These debates have influenced periodization and thinking about the changing character of political and social organization through time. While many high-ranking courtiers relied on estates for their livelihoods, there were nevertheless disagreements at court and provincial headquarters about their cost to the realm. In particular, administrators at court frequently found themselves vexed. For instance, when a special tax was levied for some important project—for a ceremony, or to rebuild the Ise Shrine every twenty years, or for reconstruction of palace walls—where and how were such levies to be enforced? Which estates, belonging to which proprietors, should be exempted? Courtier diaries, including the Chūyūki journal of Fujiwara Munetada (1062–1141), whose responsibilities included oversight of such projects, record fragments of this debate, especially as estates expanded rapidly during the later eleventh and twelfth centuries.23 Meanwhile, it is clear that depending on estates influenced organization and operation within noble households and religious institutions. Absentee urban proprietors had to develop the staff and methods to survey the land; assign, collect, transport, store, and utilize rents in 10 The Big Picture various forms; and provide police and judicial services when there was trouble.24 Above all, they needed to assure ongoing production and payments. At the local level, fields were organized into “named units” (myō), each in the charge of an overseer (myōshu) who collected and paid dues (whether taxes or rent). Sources indicate that each such cultivator- manager might have charge of two or three chō (six to nine acres), and he was often the head of an extended family with numerous dependents (genin), since his duties required extensive manpower. These named units were therefore key components of an estate.25 Archaeology and the written record also indicate that estates had clusters of residents, with each settlement headed by resident cultivators ( jūnin), whose appearance in later eleventh-century documents indicates their increasing prominence. At the end of the twelfth century, the Genpei civil war (1180–1185) erupted. Nagahara characterizes it—as have many historians writing in English—as a movement of provincial land-opening elites, especially regional notables who were residential provincial officials (zaichōkanjin), against the dominance of the Kyoto court.26 After the victory of forces supporting Minamoto Yoritomo (1147–1199) in 1185, those local elites that had allied themselves with Yoritomo sought appointments from the new Kamakura warrior government to military stewardships (jitō shiki) on estates or units of the official domain where defeated Ise Taira loyal- ists had held posts. In other words, it was through an expansion of the shiki system both on estates and in the official domain that Yoritomo’s shogunate empowered his shogunal influence in post-Genpei society. The victory of Kamakura’s forces over those supporting the retired monarch Go-Toba (1180–1239; r. 1183–1198) in the short Jōkyū War of 1221 resulted in expanded authority and power for Kamakura, and new military stewards reportedly were appointed to as many as three thousand estates associated with Go-Toba’s losing side. Nagahara points out that the Hōjō shogunal regents themselves gained control of three hundred estates, for which they built an extensive water transport network for the movement of goods, people, and cash, the latter then coming into wider use. And as monetization increased steadily after 1250, residents of estates became increasingly involved in the commercial market economy to exchange their dues in kind for coin to pay rent.27 Fur- thermore, on Tōji’s Tara Estate in Wakasa near the Japan Sea, residents bought or sold silk, food, oil, salt, jars, and iron goods at the nearby market, while fields within the estate supported craftsmen and purvey- Estates 11 ors active in the market. By mid-Kamakura times myō units at Tara also had grown larger: they comprised from ten to several tens of chō, and the best fields yielded thirteen to 28 of rice per chō, of which three to five to was paid as rent, at a rate of 23 to 38 percent. It was against this background that many estates in Kamakura times witnessed ongoing competition, frequently violent, between the proprietor, who usually sent an agent, the zasshō, to manage the estate; the military steward, who participated in rent collection and had police powers; and elite cultivator overseers (myōshu, hyakushō ), who acted as rent collectors and community leaders. There were arguments within and between groups over who was to conduct field surveys and when, who should be responsible for record keeping, the fairness of annual rents and exchange rates, and who would oversee water resources and exercise police powers. All parties hoped to maximize authority and income while securing rights over land and people according to their position in the shiki hierarchy. The dual system of administration for estates and official land was, Nagahara demonstrates, a system fraught with contradictions and quarrels within and between layers of the different hierarchies.29 Nagahara calls attention to a particularly important development for military stewards in 1239, when the shogunate issued a supplementary law (tsuikahō ) giving stewards clear authority to chase criminals from estates and confiscate their land.30 This strengthened the control of stewards over cultivator residents. And when military stewards could make rent collection difficult enough for proprietors’ agents, they could sometimes persuade an estate proprietor to let them manage part or all of an estate by a “steward’s contract” (jitō ukesho). Agreements to split the estate between proprietor and steward could be the next step; such arrangements were called “splitting the land” (shitaji chūbun).31 An estate that experienced both sequentially was Tōji’s Ōyama Estate, which was managed by its Nakazawa-family military steward after 1244, and then split between the steward and Tōji in 1295. Such splits brought military stewards to the fullest extent of independent proprietorship as “resident landlords” (zaichi ryōshu). On the other hand, Tōji’s Yugeshima Estate, a small salt-producing island in the Inland Sea, also experienced such a split, with the temple doing better than the steward.32 A big question for estate and medieval historians has always been the effect of steward appointments, usually viewed as threats to proprietors and cultivators. While the steward system did integrate the Kamakura shogunate into the existing shiki system, it was not done easily nor was it 12 The Big Picture always successful. In particular, there was a strong tendency right from the very beginning for armed military stewards to aggrandize their prerogatives and perquisites vis-à-vis land and cultivators. A particularly good example is seen at Kunitomi Estate in Wakasa in 1207, where cultivators sought redress by writing up a detailed list of grievances against the steward and sending it to their proprietor, who took up the complaints with Kamakura authorities, who subsequently denied many of the steward’s claims.33 And while some estates belonging to prominent religious institutions like Tōdaiji escaped the imposition of military stewardships, many did not, nor did the estates of courtiers.34 Ōta Estate in Bingo Province, for instance, which had been commended to Kōyasan by Go-Shirakawa after the Genpei War, saw aggressive steward activity over the next 130 years, as did Ōyama and Ategawa Estates. Proprietors including temples spent a good deal of energy trying to preserve their own prerogatives and perquisites against obstruction by stewards.35 Despite these issues, and even though over the course of the Kama kura period military stewards gained an ever larger portion of the proceeds from estates relative to that of proprietors, Nagahara nevertheless argues that steward appointments actually stabilized the estate system.36 In Nagahara’s view, it was the weakness of cultivator households, not aggression by stewards, that destabilized the Kamakura-period agricultural economy.37 He points out that many such households failed, with the result that independent cultivators became agricultural laborers (genin) or left their homes to become destitute wanderers or slaves. Only after 1330 does Nagahara find an increase in the agricultural surplus, improved conditions for some cultivators, and attendant development of new forms of stratification and subordination that helped to destabilize the estate system: “The surplus increased along with social specialization. Commerce broadened. This led to the appearance of lesser cultivator ranks in the hamlet, their subordination by local landlords, and the development of real territorial lordship (in the hands of daimyō warlords).”38 Another threat to the estate system was the Mongol threat in the later thirteenth century, Nagahara argues. Specifically, shogunal responsibility for defense of the realm after threats by Khubilai Khan in 1268 resulted in Kamakura’s provincial constables (shugo) exercising greater authority and acting more like provincial governors of old. A great number of supplementary laws were issued by Kamakura during this period of crisis. Under orders from both court and shogunate, con- Estates 13 stables compiled field surveys (ōtabumi) as tallies of estates and fields in their provinces.39 They also supported and oversaw shogunal housemen’s defensive efforts in Kyushu into the 1290s, and they fought gangs of brigands (akutō, “evil bands”) attacking estates, especially in western Japan. While each case of evil-band activity needs careful analysis and contextualization, as Dan Sherer argues later in this volume, Naga- hara sees the breakdown in law and order of this time as the result of impoverishment, desperation, and increasing lawlessness by local elites against both court and shogunate.40 A particularly famous case of akutō violence was that at Kuroda Estate, then a major Tōdaiji property in Iga. There in the 1260s and 1270s estate dwellers resisted orders from the temple and followed a local leader from the Ōe family that the temple had tried to fire as its head estate official (gesu). Trouble was ongoing, and by 1331 the province of Iga was under akutō control. Similarly Naga- hara interprets demands for debt-canceling laws (tokuseirei) by Kamak- ura housemen as the result of local economic tensions: monetization and the Mongols ruined the economic circumstances of many shogunal housemen and made them unwilling to honor old agreements concerning rent payments for estates, therefore endangering the estate system and the society based on it.41

Decline of the Estate System While the would-be reformer Go-Daigo Tennō (1288–1339, r. 1318– 1339) supported continuation of the estate system, he ultimately unset- tled it by confiscating military stewardships from warriors and awarding them to temples as rewards for their sacral support for his restoration.42 Then when Go-Daigo’s ambitions for restoring the monarch as actual ruler of the realm proved unsuccessful in 1336, the new shogunate established by the warrior leader Ashikaga Takauji (1305–1358) enrolled commanders who, in the midst of ongoing fighting between the rival northern and southern monarchs (1336–1392), were determined to strengthen their own control over land and cultivators in their domains, including those of estates. Promulgation of the half-tax (hanzei) policy by the Ashikaga shogunate in 1352 led to further decline of estates that were still paying rents to proprietors.43 The policy gave constables of certain provinces (and eventually of all provinces) the right to keep half the rents collected from every unit of land in their provinces, whether estate or official. Of course, the constables had long involved themselves 14 The Big Picture in estate affairs. Even in Kamakura times, proprietors had asked them for help in collecting rents and pacifying evil bands, giving the constables good opportunities to negotiate for contract management of estates (shugo-uke).44 Other threats to the shiki hierarchy were emerging as well. After a long history of evil-band activity that dated back to the 1290s, cultivators at Tōji’s Yano Estate (Harima Province) joined together as a collective in the 1330s to seek independence; they sought to make their own contract to pay the annual rent and thereby exclude the proprietor’s agent. Such agreements made cultivators more independent on their own fields and in their own communities. By then, too, estate residents included tradesmen, craftsmen, and monks, and even lesser cultivators (kobyakushō) were being welcomed into village assemblies to advance the cause of local unity.45 Even after achieving greater independence, Yano residents continued to complain: in the 1370s, the temple was sufficiently threatened to term their acts treasonous, amounting to “inferiors overthrowing superiors” (gekokujō). At that point Tōji had no choice but to enlist the provincial constable to enforce its claims, which generally proved a risky option.46 Nagahara attributes successful communal movements like that at Yano to the fact that households (ie) of leading cultivators had become more stable, as surplus from fields both wet and dry increased, along with socioeconomic specialization and commerce.47 Cultivators’ ability to sell their produce in the market also weakened the shiki hierarchy. At Yano the record shows that residents were paying rent with coin by 1377; the source of their cash was the periodic market at a nearby port, which was held three times monthly. And by the later fourteenth century, Yano residents were using bills of exchange, making them part of a more diverse realm-wide economic network.48 Another development that was empowering the overseers of named units as elite cultivators by the fifteenth century was their practice of adding a surcharge, “added rent” (kajishi), to what the overseers collected from cultivators. Moreover, they were also selling that perquisite, sometimes to absentee figures such as sake brewers and lenders in Kyoto or Nara.49 This meant that just as contract managership had made stewards and constables into landlords earlier, added rent now gave elite cultivators wealth and status as well. Meanwhile, at the top of the status hierarchy, as religious and aristocratic proprietors lost livelihoods from estates, they looked elsewhere Estates 15 to dues from guilds, marketplaces, or toll barriers. Taking Tōji as an example, Nagahara notes that by the later fourteenth century even that prominent official temple had lost the properties for which it once served as premier proprietor. He calculates that in the 1390s rent from just seventeen of eighty estates was still reaching Tōji, which managed to keep only a few estates on which it held actual management rights. Even then, however, the temple depended on contractors (daikan) to exercise its managerial rights.50 As a result, Nagahara calculates, rents received by the temple were about 5 percent of those received in Kamakura times.51 And while Tōji was still benefiting from rents from its Niimi Estate in Bitchū Province fifty years later, by then the property was completely surrounded by the domain of the provincial constable. Two decades later, when Tōji turned against the shogunal prime minister (kanrei) Hosokawa Katsumoto (1430–1473) during the Ōnin War (1467–1477), Katsumoto was in position to confiscate even Niimi Estate.52 In another case, in fifteenth-century Tanba the provincial constable began charging a special cash levy called tansen on all fields, whether in estates or the official domain. And at Ōyama Estate in 1464, protests against a manager’s “evil ways” led residents to flee the estate and then to conduct a protest at the temple in Kyoto. As at Yano in Harima, cultivators felt empowered to raise their voices. By the early sixteenth century the Hosokawa constable came to exert strong control over the entire province; in 1508 the scion of the steward’s family was killed in a fight between rival members of the Hosokawa band, to which he belonged.53 And in Bingo still further to the west, the provincial constable had taken over Kōyasan’s Ōta Estate by 1439. During the decade-long Ōnin War, warrior commanders often took over the fields and cultivators of former estates. For instance, provincial warlords (kokujin) like the Mōri and Kobayakawa of Aki Province and the Asakura in Echizen had accumulated wealth and power as contractors that collected and paid estate rents. Then during the later fifteenth and early sixteenth centuries they took over the provinces, including estates therein, as warlords. Owari saw that happen even earlier: in Kamakura times it had been given to Daigoji for its livelihood as a benefice province, but then was taken over by the provincial constable during the North-South conflict (1336–1392). Still another vector that weakened the estate system was recruitment of wealthy cultivators’ sons to serve in the forces of constables and warlords. The new roles and relations destroyed the old hierarchical bonds 16 The Big Picture and values of the court-centered shiki hierarchy. Commander-retainer relations, sometimes called “feudal,” spread deeply into the social relations of the cultivator class.54 Nagahara sees this as an ongoing process during the long decades of North-South warfare, and it reached a high point by the turn of the sixteenth century. Nagahara concludes his eight-hundred-year history of estates and the estate system by recounting the lonely journey of the courtier viceroy (kanpaku) Kujō Masamoto (1445–1516) out of the Kyoto capital to reside on his estate in Izumi Province in 1501. In the midst of the enduring chaos after Kyoto’s destruction during the Ōnin War, Masamoto’s desperate hope was to manage Hine Estate directly as resident proprietor. The journal that he kept, however, records his failure, not least because locals tended to support the provincial constable rather than Masamoto. The estate system, with its layered hierarchy of officerships based on the putative authority of the monarch’s court, had become untenable. Particulars of the Hine story are recounted elsewhere in this volume in the chapter by Professor Hirota Kōji, whose research institute, the Rekishikan Izumisano, overlooks the partially preserved site of the estate. Visitors there can view a model of the place based on the written and archaeological records. They can also see a plethora of artifacts and walk the property that Masamoto’s ancestors had held since Kamakura times. Museums such as this—there are two more on the site of Ōbe Estate in Hyōgo Prefecture—are a significant supplement to written analyses of the estate system.

Studies of Estates in the Late Twentieth Century Nagahara Keiji’s history provides us with a fundamental narrative for estate history. An Introduction to Estate Studies (Shōen nyūmon), published in 1989 just as Nagahara was preparing the second edition of his book, shines light on new issues, views, and approaches toward the end of the twentieth century. The volume is in fact the first of ten in an encyclopedic series titled Fundamentals of Estate History (Kōza Nihon shōenshi ), whose compilers included Nagahara and three eminent medievalist colleagues: Amino Yoshihiko, Ishii Susumu, and Inagaki Yasuhiko. While I focus the following remarks on volume 1 because it provides broad insights into new issues and methodologies, readers will want to know too that volumes 2 through 4 include essays detail- Estates 17 ing the development of estates from formation through maturity to decline. Volumes 5 through 10 present case studies region by region: one volume is dedicated to estates in eastern Japan, three to those in mid-Honshu, and one volume each to those in western Honshu, Shikoku, and Kyushu. There is a helpful cumulative index in volume 10. This publication, completed between 1989 and 2005, is good evidence that estate studies had reached a new stage at the close of the twentieth century. In his introduction to the first volume, Amino Yoshihiko urges readers to rethink the common idea that estates were private landholdings dedicated primarily to agriculture.55 While he agrees with Nagahara that the estate was the fundamental structure of the medieval landholding and social production systems, he argues that from the beginning of the medieval age in the twelfth century estates supported many social groups, including nonagriculturalists, and that estates constituted the core of social relations within which goods and people circulated. Amino urges too that historians give more attention to market exchange and the various networks of provisioners, craftsmen, entertainers, and other dependents that served proprietors, as well as to monetization, all of which Amino thinks were loosening the autarky of the estate system even by late Heian times.56 He also urges close study of artifacts such as ceramics and metalwork, and of roads, ports, ferry points, checkpoints, and markets because they provide clues to the movement of goods and people.57 These issues and approaches are emphasized throughout the ten-volume series. In his essay “Documentary Sources for Estates (Shōen monjo),” Uejima Tamotsu discusses the rich archive of temple sources known as Tōji hyakugō monjo, which contains twenty-four thousand records, many of them concerning the temple’s estates and how they supported the daily lives and ritual activities of Tōji’s monks.58 Uejima describes how during the thirteenth and fourteenth centuries Tōji received the donation of as many as eighty estates from the monarchs Go-Uda (1267– 1324, r. 1274–1287) and Go-Daigo, as well as from premier royal ladies, shoguns, and commoners.59 He reminds us, as does Endō Motoo in his chapter concerning the Tōdaiji archive later in this volume, that we need to be conscious of how our documentary sources came to be collected and preserved (or not), and how they can be studied not as single records but as parts of larger collections that have important histories and characteristics of their own. 18 The Big Picture

Walking estates—exploring the terrain with one’s own eyes—and archaeology, with an attendant concern for material culture and artifacts, are key imperatives for researchers in the group that compiled Fundamentals. In this vein Hattori Hideo urges researchers to leave the library and walk the land where estates were, to understand geography and topography.60 Place names, local legends, traces of roads and old irrigation works, and in situ inscriptions need to be found and documented to understand how life was lived on estates. Similarly Kuroda Hideo invites readers to study old estate maps to see how residents and officials perceived the landscape: “Look at the map first, then the documents,” he urges.61 The ceramic specialist Yabe Yoshiaki argues too that researchers need to give careful attention to the wares discovered on estates, since they provide clues to the trade and cultural networks in which residents participated.62 In her chapter later in this volume, Sachiko Kawai demonstrates this point. And Suitō Toshio writes about how inscriptions on metal and stone Buddhist objects contribute to our understanding of religious beliefs and practices in the daily lives of estate residents. These contributions in volume 1 of Fundamentals make it clear that reading only the written record is no longer adequate for studying estate history. In her concluding essay of this panorama of new approaches and issues in estate studies, Tabata Yasuko pieces together the process of estate decline, as warriors and cultivators themselves gained rights of proprietorship and thus escaped the dominance of absentee proprietors during the fourteenth and fifteenth centuries. Since this is an area for which little has been written in English, the story Tabata tells and the issues she illuminates for future study are particularly notable. In her introduction Tabata reminds readers that in addition to Naga- hara’s model of a shiki - based polity, other historians have described medieval society in different terms. Kuroda Toshio, for instance, envisioned a medieval polity with a “tripartite center”—the royal court, the Kamakura warrior government, and official religious institutions— called kenmon (literally “gates of power”), all of which were sustained by estates but through varying modes of relations with land and cultivators.63 Tabata herself privileges Kuroda’s model. She adds too how Ishii Susumu did much to increase our understanding of the lower ranks of the shiki system, especially the relationship between village (mura) residents and estates and the roles of warriors on estates during Kamakura times.64 While Nagahara, Kuroda, and Ishii would all agree that estates Estates 19 were fundamental to Japan’s medieval society, Tabata notes how their research trajectories varied considerably and have contributed different perspectives to the story of the estate system and its decline.65 Tabata then turns to the issue of how scholars have articulated the proprietary rights to which warriors aspired, and how warriors advanced their hold on those rights. Synthesizing the considerable oeuvres of Kuroda, Ishii, Ōyama Kyōhei (one of whose essays also appears in our volume), and Kawane Yasuhira, she finds two such rights emphasized: (1) authority to survey the land (kendanken) to enable its management, and (2) authority to manage cultivation (kannōken), which included assigning cultivating rights, assuring water supply, and keeping the peace. So how did warriors become local landlords capable of exercising these rights in Kamakura times and beyond? Tabata tells the story as follows. Key advances were made by local elites cum land openers who served the provincial headquarters (kokuga) in the eleventh century, who then became Kamakura houseman (gokenin) in the thirteenth century. As military stewards, they served as rent collectors and sheriffs, thereby expanding their control over fields and cultivators. Specifically they engaged in land opening, confiscation of land from “criminals,” and taking over fields by force (ōryō). Despite these successes, however, Tabata agrees with Nagahara that what most weakened the estate system was not the violence of military stewards. Rather, what most attenuated the command chains of both the shiki system and Kamakura’s warrior government was the dynamic expansion of commerce, transport, and nonagricultural production.66 She also cites Gomi Fumihiko, who has argued persuasively that there was a yawning gap between the worldviews of different sectors of early medieval society that added to the disorder: while the estate system hierarchy was essentially more traditional and patrimonial in its world view, the commander-subordinate relations (shujūsei) of the shogunate emphasized the master-man relationship often termed “feudal.” Tabata then turns to the cultivators, asking how did those who had been but rent collectors and parcel managers, the myōshu and hyakushō of Heian times, gain the ability to charge added rent, and even to sell that prerogative, by the later fourteenth century?67 She discusses research by Kuroda Toshio and others elucidating a critical advantage held by cultivators: the family house and garden, known as the zaike (home place), generally remained free from the control of military 20 The Big Picture stewards. In addition, cultivator solidarities, formed in the fourteenth century to negotiate contracts with proprietors, allowed some cultivators to collect and pay their own rents (hyakushō uke), significantly obvi- ating proprietors’ rights. Given this, where proprietors did not maintain management rights, they generally lost out on rent payments by the fifteenth century, just as Nagahara argued.

The Handbook for Shōen Studies, Current Research Turning now to the most recent compendium of shōen studies, the Shōen History Research Group began meeting in Tokyo in 2006 to investigate estate studies in the broad context of medieval history. Work on their Handbook for Research on Shōen History (Shōenshi kenkyū handobukku 2013) began in 2010, with the objective of elucidating the results of research published since 1990. Its contributors build on Nagahara’s history but also introduce a new chronology and thinking concerning the roles of central (rather than local) players in estate establishment during the twelfth century. In addition, the volume provides abundant new insights on estate development under the regents and during the fourteenth and fifteenth centuries as well. A short glossary of terms as well as an annotated bibliography for important works on estate history are included.68 In terms of chronology, the Shōen History Research Group calls for more historicization than afforded by Nagahara’s three-stage model. The group traces estate development through five stages: (1) the era of early estates in the eighth and ninth centuries, (2) aristocratic estates during the Northern Fujiwara Regency of the tenth and eleventh centuries, when proprietors in the capital district negotiated with or battled provincial governors to exempt estate lands and cultivators from taxes, (3) a particularly dynamic age of estate formation during the later eleventh and twelfth centuries when retired monarchs led the court and became supreme proprietors of massive numbers of estates, (4) an age of growing tensions and challenges as estates became more complex and harder to administer during Kamakura times, from the thirteenth to the early fourteenth century, and (5) an age of decline due to warfare between the northern and southern courts, which led to the rise of daimyō warlords from the late fourteenth to the fifteenth centuries. For each stage, researchers for the Handbook discuss exemplary estates, many of them different from those Nagahara wrote about, that show Estates 21 how estates formed and functioned and the difficulties faced by proprietors, managers, and cultivators over time and space. For the first stage, Togawa Tomoro discusses “estates of the classical age” (kodai shōen), problematizing the more familiar term used by Nagahara, “early shōen (shoki shō ),” because he sees little continuity between classical estates, which were essentially official operations, and later estates. Togawa also starts the story earlier than does Nagahara. He traces the formation of early estates back to royal landholdings (miyake) and to designated groups of cultivators (be, nashiro, koshiro) in many provinces in the late sixth and early seventh centuries. Later, when Shōmu Tennō built the Tōdaiji Daibutsu (Great Buddha) in the mid-eighth century, that temple’s agents worked closely with provincial governors and district chiefs, especially in the Hokuriku provinces facing the Sea of Japan, to open new rice fields that were to be rented out to neighboring cultivators attracted by the partial tax-exempt status of the new fields. Fortunately, there are eighth-century maps illustrating the topography and facilities of some of these temple properties, and in recent years considerable research has been done on them. In the ninth century, too, royal grant fields (chokushiden) represented a similar effort to provide livelihoods for retired monarchs and princes.69 Togawa notes too that archaeological work since the 1980s has provided new insights into material culture, literacy, and management on some of these early properties. At Chimori Estate in Echizen Province and Yokoe Estate in Kaga Province, for instance, inscribed pottery finds and wooden documents (mokkan) have identified the venue of estate offices. Chimori Estate was a property of Tōdaiji opened from 749 onward. It grew to great size, three hundred chō (882 acres), but as early as 766 temple administrators were complaining of their inability to manage the distant property.70 Another of Tōdaiji’s early holdings, Yokoe Estate, was made up of royal grant fields donated by one of the daughters of the monarch Kanmu Tennō.71 By the mid-ninth century, these classical estates had mostly failed. While Nagahara had blamed the lack of resident cultivators, Togawa emphasizes the decline of district chiefs as powerful local notables, since they had been the temple’s most important partners in land opening and subsequent cultivation of the new fields.72 Moreover, Yoshimura Takehiko has argued persuasively that a serious problem on these estates was the lack of a shrine community, which would have aided continuity of cultivation. Indeed, the evidence shows that estates with shrines did better.73 22 The Big Picture

For the second stage, that of aristocratic estates in the tenth and eleventh centuries, Kamakura Saho and Ono Takashi shine light on estate development during the Regency era, especially in the eleventh century. We have long needed this synthesis based on new case studies well contextualized by a broad reading of the historiography of the era. Despite the famous complaint by Fujiwara Sanesuke (957–1046) that in the time of Fujiwara Michinaga (996–1027) as court leader “the whole realm belongs to the Regents’ Line,” Kamakura and Ono agree that holdings of the Regents’ Line (Sekkanke) increased dynamically after Michinaga’s day, during the eras of his son and grandson, Yori michi (992–1074) and Tadamichi (1097–1164), who received many commendations of new fields from local openers. These estates, however, were never formally managed by the Regents’ Line; rather, relations were informal, as Yorimichi once famously remarked when asked about them.74 In addition to the new fields being commended to regents and other nobles, provincial governors were working diligently to get more fields opened, just as Nagahara maintained. Kamakura and Ono point out that at the beginning of the eleventh century the climate was warming, with lots of rain and damaged crops as a result. In this environment, the pressure to open new fields or reopen fields that had gone out of cultivation made governors beneficent toward openers. Trouble developed later in the century, however, as the climate improved, the number of exempt fields continued to rise, and the budgetary demands of the court increased as well. Then there was more aggressive surveying and record keeping by provincial authorities, together with determined collection of kanmotsu, a new comprehensive tax assessed on fields, as well as zōyaku, a tax on individuals.75 Given the new dangers, land-opening cultivators sought alliances with elite patrons in the capital who had the influence to gain and keep fields tax-exempt and their cultivators tax- exempt as dependents (yoriudo).76 Later in the volume we will see how Tōdaiji’s Kuroda Estate expanded under such conditions. As mentioned earlier, in 1045 and 1069 additional regulations were promulgated to slow the formation of new estates and the expansion of old ones. The Shōen Records Examination Office (Kiroku Shōen Kenkeijo) was established to check the enabling documents of all estates, and properties without satisfactory papers were confiscated, even those of prominent religious institutions.77 Special levies (rinji zōyaku) to rebuild Ise Shrine and reconstruct the royal palace—which Estates 23 burned down frequently after 960—also became more common. To facilitate tax collection, provincial governors pressed for equalized tax collection from all fields in a province (ikkoku heikinyaku), whether in an estate or in the official domain. In rare cases, a particularly influential proprietor like Kyoto’s Tōji might request and receive a court order of immunity that barred officials from entering its fields—as at Ōyama Estate in 1042—to prevent quarrels and violence over tax collection.78 Nevertheless, in the case of special levies, immunity did not automatically prevent provincial officials from surveying and collecting taxes on newly opened fields beyond those originally exempted. There were endless disputes over which fields were new or old. And for their part, court authorities often found it difficult to decide who should pay or not pay. Such pressures resulted in violence and unsta- ble incomes from estates, even those of the most powerful, including the Regents’ Line and Tōdaiji, as the study of Kuroda Estate shows later in this volume.79 Kamakura and Ono also spotlight a different sort of estate that was taking shape, one that was planned by high rankers at court with the cooperation of middle rankers, usually provincial governors. These properties were the “instituted estates,” mentioned but discussed in less detail by Nagahara, in contrast with earlier “commended” estates for which the initiative had come from below.80 An example of such was Suda Estate in Kii Province, created by Fujiwara Kaneie of the Regents’ Line in 987 with the cooperation of the Kii provincial governor. Its purpose was to support a new vow temple for Ichijō Tennō (980–1011, r. 986–1011), the Zanmaiin, within the precincts of Iwashimizu Shrine. Immunity for Suda Estate came gradually, however: the estate was declared exempt from kanmotsu and special levies and immune from official entry only in 1028. A similar process resulted in instituted properties to support the Byōdōin retirement temple of Fujiwara Yorimichi in Uji at mid- century.81 Nine of Yorimichi’s own properties were donated to support the Byōdōin in 1052; each was to have boundary markers put in place and to be partially tax-exempt. The late decades of the eleventh century also saw increasing local organization of management on estates: records of expanded local staffs (shōkan) that were both stratified and more functionally specialized first appear in the 1060s and 1070s. They served not only as a proprietor’s local agents but also as community leaders. And under them the cultivators, often called hyakushō or jūnin, were gaining a status of their own as permanent cultivators.82 24 The Big Picture

Morita Hayato takes the story of instituted estates into the twelfth century, when retired monarchs led the court. He agrees with Naga- hara that by the era of the retired monarch Shirakawa (1053–1129, r. 1072–1086), estates were the fiscal base for capital elites and many provincial residents. An example of a new estate at the time is Ushigahara Estate in Echizen, put together in 1086 by the faction of allies, friends, and clients from various strata to support the Enkōin, a royal vow temple for one of Shirakawa’s queens.83 Morita also describes the network around Taikenmon’in (1101–1145), a consort of Toba Tennō, that organized Shidoro Estate in Tōtōmi Province to support yet another royal vow temple.84 Besides showing the top-down and official nature of estate development, such examples also show how the shiki hierarchy was becoming more complex as it integrated additional members to manage and cultivate more land. Furthermore, many of the newly created estates were expansive, “territorial” in that they included multiple townships or even districts that were once under provincial administration. These represented quite a new type of estate compared to the old “exempted-field” estates of the past, and they were venues of considerable land-opening efforts (dry fields and paddy) by locals of various strata, including district notables and prosperous cultivators.85 Morita explains too how such developments led to debates on estate policy and increasing political and economic tensions at court, as senior nobles and subordinates charged with planning court rites or construction projects had to negotiate with estate proprietors, who were their own seniors in rank and prestige and whose agents therefore wanted exemptions from the special levies needed to fund court projects.86 The era of the retired Toba (1123–1156) was a time of special tension, as Toba In became a more active proponent of estate formation to support his increasing royal costs and as the Regents’ House also routinized its management of estates as the primary source of income.87 In this analysis Morita disagrees with Nagahara on the character of the Genpei civil war. While Nagahara and many others have seen the war primarily as a clash between court authorities and countryside elites, Morita calls for better analysis of the complexities of court society at the time. The war, he argues, was actually a clash between provincial land-opening elites and the faction led by the Ise Taira family that led the court during the 1170s. At that time the faction of Kiyomori (1118–1181) was aggressively placing its allies and dependents in provincial posts and shiki hierarchies across the archipelago. In so doing, Estates 25 they made implacable enemies of those they displaced, with the Genpei War as the result.88 Turning to new research on Kamakura-period estates, Satō Yūki and Itō Rumi discuss how the existence of two centers of authority—the Kyoto court and the Kamakura shogunate—complicated shiki hierarchies as Kamakura housemen were inserted and ordered to serve two lords: the proprietor and the shogunate. Satō and Itō concur with Naga- hara and others that Kamakura’s military stewards were a stabilizing force. Nevertheless, endless quarrels at every level of the shiki hierarchy, the dual roles of military stewards, the difficulties faced by judicial authorities in the capital and in Kamakura in settling disputes, and fac- tional enmity in both centers ate away at the bonds of the shiki hierarchy that held the polity together.89 Shimizu Ryō then traces the decline of the estate system from 1336 onward, through the Northern and Southern Courts era and beyond. Notably, he disagrees with the common wisdom that the Ōnin War was the key event; the demise came earlier, he argues. His narrative is this: The Ashikaga shogunate of the later 1330s attempted to divide the realm into the lands of great religious institutions, the holdings of which were to be sacrosanct from warrior intrusion, and warrior domains. But the shogunate’s policy of half-tax in the 1350s resulted in a deadly wound to the estate system, since the half-tax to be paid to provincial constables actually amounted to half-possession of the land itself.90 Moreover, even as temples like Tōji argued for unified territorial control (ichienryō) of their remaining estates, cultivators were building communal solidarity against outsiders’—courtier or warrior—involvement in their affairs. Meanwhile the contractor-representatives (daikan) that temple proprietors had turned to for rent collection, credit, cash exchange, transport, and warehousing created increasing distance between estate cultivators and proprietors. In Shimizu’s opinion, a key moment was the assassination of Shogun Ashikaga Yoshinori (1394–1441) in 1441, since it pre- cipitated major unrest in many parts of the country and left proprietors without a shogunal guarantor for any remaining estates. From my perspective a hugely important contribution of the Hand- book, in addition to its update on new approaches and debates on estate history, is Kimura Shigemitsu’s essay on “The Structure of Medieval Estates.” In recent years Kimura has published extensively on medieval agricultural technology, including techniques of land opening and the importance of dry fields for early medieval farming.91 In this essay 26 The Big Picture

Kimura urges us to think about the landscapes of estates, the different sorts of fields that cultivators worked (wet and dry), and such topographical elements as irrigation ponds, unopened land, rivers, and bays.92 Like Kuroda Hideo in the Fundamentals series, Kimura urges close study of medieval estate maps, which methodology he demonstrates by an analysis of the medieval map of Jingoji’s Kaseda Estate in Kii Province. Besides such maps, he argues that everyday issues of estate life such as forms of rent payments are important for study; various goods were frequently substituted for rice in a process known as dainō (“payment by substitution”), and the calendar for such payments was a factor that affected cultivators’ lives greatly.93 Through a case study of Ikeda Estate in the province of Tōtōmi, where dry fields were particularly important, Kimura discusses the proprietor’s key right called kannō—encouraging cultivation—which at Ikeda faced hindrance by a military steward who aimed for maximal control of land and cultivators. Kimura also urges those interested in estate history to study visual resources such as the magnificent illustrated scroll narrating the extensive travels of the holy man Ippen (1239–1289), the Ippen Shōnin den’e. The scroll shows Ippen, who crisscrossed Japan numerous times, visiting marketplaces in both eastern (Shinano Province) and western (Bizen Province) Honshu, providing clues to the emergence of markets and lodging places on Kamakura-period estates. It was at such places, argues Kimura, that estate dwellers and outsiders met with increasing regular- ity and where the insularity of the estate community was penetrated and challenged most deeply. Kimura’s insights into agricultural technology and how we can use maps and other visual evidence to study cultivators’ lives in the communities of estates themselves provide us with an array of engaging new issues and approaches for future study.94

When I began to study the history of estates in the early 1970s, it was extremely difficult to situate a case study for a given estate within the broader political, social, and economic context of Japan’s premodern history. In that regard, Nagahara Keiji and the members of both the Fundamentals and the Handbook teams have transformed the field of estate history. Nagahara’s Shōen gives us a clear sense of how the estate took form and changed over eight centuries. The story traces the emergence of the shiki hierarchy and how it developed into the predominant network that linked centers and peripheries in the twelfth and thirteenth centuries. His account of the subsequent decline of estates Estates 27 during the fourteenth and fifteenth centuries identifies vectors that attenuated the dual-centered polity made up of court and shogunate, giving warriors and cultivators tools that weakened and eventually ended the estate system.95 The introductory volume of the encyclopedic Fundamentals series discusses the state of the field and new approaches to estate studies at the end of the twentieth century. By that point, the study of artifacts and material culture as well as “walking the land of the estate” had become widespread. Part of my own training in estate studies included walking a number of estates in various regions of Japan. Preservation of estates has also become an important objective. Two estates now being partially preserved are Hine Estate in Izumi Province, on which see the chapter by Hirota Kōji later in this volume; and Tashibu Estate in Bungo Prov- ince, Kyushu. Donors to preservation of the latter site are even called “proprietors,” and they receive annual “dues” of rice from Tashibu fields. Reports by those “walking the land” as well as archaeological reports are adding tremendously to our ability to imagine lives once lived on estates.96 The Shōen History Research Group’s call for clearer historicity in how we study the development of estates, additional study of networks that initiated estates from above, and more detailed analysis of developments in the previously understudied twelfth and fifteenth centuries identify important directions for estate studies. At the same time, new research continues to be published. For example, Kamakura Saho has recently published her own overview of the current historiography.97 In it she reiterates her view that estates did not represent a clash between official and nonofficial interests. Rather the estate was itself an official structure, just as Nagahara and others have argued. Kamakura emphasizes too that research concerning central initiatives resulting in the official institution of estates by actors at the center needs more consideration. Furthermore, by means of estate regulations, enforcement of which became more aggressive after 1045, policy makers at court tried to create a new tax system wherein both estates and official fields were taxed for special needs of the court’s budget. This new policy increased tensions at all levels. And as provincial authorities tightened their purview, estate land only increased in response. The contradiction was real, dynamic, and violent. Kamakura’s work, which explores the complexities of estates developed through a combination of commendation and institution along with the expanding managerial role of warriors, is an 28 The Big Picture example of the new research that demonstrates the continued appeal of estate studies to new generations of scholars.98 I want to conclude by emphasizing three arguments that are likely to be new to English readers. First, the idea that estates were private holdings, and thus that the expansion of the estate system in late Heian times should be considered a process of “privatization,” is not supported by contemporary research. Even in classical times estates were a means used by the court government to get fields opened and to support official personnel and enterprises. While researchers have pointed out differences between earlier and later estates, the main purposes for estate formation remained continuous: they were always a facet of the official landholding and production systems, often closely associated with supporting official religious institutions. That is why when Yasuda Motohisa prepared a chart of estates, he found that of 151 well-known properties, 105 of them (70 percent) had temples and shrines as their proprietor.99 Second, researchers agree that appointments of military stewards in Kamakura times stabilized estates rather than causing their decline. Such appointments added, however, to the complexity and tensions within and between shiki hierarchies. And third, in the twelfth century, a particularly dynamic era for estate formation, the initiative for estates did not come primarily from local commenders. Contemporary researchers emphasize the formation of early medieval estates by networks of upper and middling aristocrats who used their good offices to institute new estates, mainly to support religious projects of the court and monarch.

Notes

1. For an English translation and an introduction to Ishimoda’s work, see Ishimoda 2006, 326–361. 2. In addition, English translations of estate documents can be found in Piggott, forthcoming, Shōen Documents and Materials in English Translation, https://usc.ppjs.org/sources-Estateresearchproject. 3. Fortunately, several essays by Nagahara concerning estates have been translated into English. Nagahara 1990, 1977, and 1975. 4. In English, see Nagahara 1990, 301–343. 5. That record is huge. Kimura Shigemitsu (2000, 189) thinks there are about five thousand estate-related documents extant from the Heian age (794–1185), and about thirty-five thousand from the Kamakura age (1185– Estates 29

1333). Those from yet later periods, which have not been compiled in a single series of documents, are more difficult to count. As for the number of estates known to have existed, there is an extremely helpful database compiled in 1997 by the National Museum of History and Ethnology (Kokuritsu Rekishi Minzoku Hakubutsukan, 1998): www.rekihaku.ac.jp/up-cgi/login.pl?p=param /soue/db. 6. See Nagahara 2004a for his extensive examination of such occupations. 7. Nagahara 1998, 15–17. 8. Nagahara 1998, 9–21. For another overview of studies of early estates see Katō 1995. In English, on early estates including Kuwabara, see Torao 1993, 448–452; Sato 1974, 91–108; Asakawa 1965a; Piggott 1987, 207–238. 9. In 988, for instance, Owari cultivators petitioned the court against various sorts of lawless behavior by their governor. For a discussion of the famous Owari Petition, see von Verschuer 2007a, 305–328. Bruce Batten (1993, 103– 134) and Cornelius Kiley (1999, 236–340) have also written about the changing nature of provincial governorship. 10. For a translation of the relevant document, see Piggott, forthcoming. It should be noted that of late the full historicity of the process traced by this document is being questioned. 11. Nagahara 1998, 49–50. 12. For a good discussion of benefice resident units, and their role in estate formation, see Wakita 1969 (1977), 24–46. 13. HI 2:752–753, #s 601–602, translated into English in Piggott, forthcoming. 14. A particularly helpful work is Ōyama 1996. In English see Sato 1979 and Piggott, forthcoming. 15. The essay by Uejima (1989) on Tōji’s archives, discussed below, includes such, and there are many such documents in other temple collections as well. On judicial practices of the time, see Kiley 1982. And a particularly informative essay on judicial developments related to estates is Ihara 1993. For an English translation, see Piggott, forthcoming. 16. For translations of this and other similar documents, and for a brief history of shiki, see Piggott, forthcoming. For an English essay on local lordship, see Arnesen 1984. For more on another township, Hisatomi, see the chapter by Kimura Shigemitsu later in this volume. 17. In English see Nagahara 1975. Kōryō is sometimes called kokugaryō (provincial lands). 18. In English on the Chōkōdōin and Saishōkōin clusters, which were made up of instituted estates, see Ōyama 1990, esp. 96–98. On the similar Hachijōin cluster, see the discussion in Kawai 2007, 59–77. Also see her chapter in this vol- 30 The Big Picture ume. In English, Asakawa Kan’ichi (1965b, 37–68) analyzed the history of one such instituted estate, Ushigahara Estate in Echizen, which supported a royal vow temple at Daigoji in Shirakawa In’s time at the turn of the twelfth century. 19. Establishment of Ōta dates from 1166. For documents see Piggott, forthcoming. 20. Hotate, “Toshiken to kizoku hanchū,” hotatelog.cocolog-nifty.com, viewed November 15, 2015. In English, concerning the products estates provided to their proprietors, see Ōyama 1990, 96–101. Sasaki Ginya (1972, 58–65) thinks that estates continued to provide much of what aristocratic households needed through Kamakura times. 21. Asakawa 1965a, 1965b. More recently in English, Thomas Keirstead (1992, 10–14) has also described estates as “private holdings,” supporting what he terms “private centers” of power. 22. Max Weber saw “patrimonial domination” as an extension of patriarchal authority, dependent on the loyalty of subordinates who benefit from their service, and he saw it resembling bureaucracy in its division of labor and hierarchy. The patrimonial official can pass posts and lands to heirs. In contrast Weber defined “feudal domination” as a dependent relationship between a “lord” and his “man,” with the latter identifying his own fortunes with those of his lord. The subordinate’s claim to legitimacy flows from his lord. Rights and obligations in the feudal mode of authority are more formalized than under patrimonialism, but they depend on the lord’s beneficence, not law. See Weber 1978, esp. 212–301. 23. Toda 1979, chaps. 1–2; Laffin, Piggott, and Yoshida, forthcoming. 24. On the judicial operations of estate proprietors, a key result of the increasing extraterritoriality of estates, see Ihara 1993. In English see Kiley 1974 and 1982. As noted earlier, it is this extraterritoriality that has led some historians to think of such estates as “feudal,” that is, outside the authority of the court government. 25. In English see Keirstead 1992, 46–51. Developments in later times are discussed in Mauclaire 2009. 26. For instance see Mass 1999, 74–75; Mass 1990, 59–87. 27. In English see Segal 2011 and 2009 on economic developments. Also see the chapter by Sakurai Eiji in this volume. 28. While the usual equivalent given for the to is eighteen liters, it was probably less than that in the Kamakura period; capacity measurements were not standardized at the time. See Farris 2006, 19–20. 29. In contrast, Elizabeth Sato (1974, 107–108) praised the functionality of the shōen in medieval society: “The mature shōen proved to be an extremely successful means of securing a balance between the demands of a ruling class Estates 31 for income and the demands of the populace for a stable means of livelihood.” Based on my research on Kuroda, Ōi, Akanabe, Ōta, and Ōbe Estates, however, I see the estate system as prone to conflict, a conclusion shared by others in this volume (see chapters by Noda Taizō, Dan Sherer, and Philip Garrett, as well as my own essay on Kuroda Estate). 30. Nagahara 1998, 182. This is a “supplementary law” promulgated after the shogunal Goseibai shikimoku (Judicial formulary) of 1232. For an English translation of this “list of protocols for judgment,” see Hall 1906. For the Japa- nese text of the supplementary law, see Satō and Ikeuchi 1955, 115 (Kamakura tsuikahō #119). 31. See Mass 1974, 157–183. A place where there was significant trouble over such matters from 1220 through the thirteenth century was Kōyasan’s Ategawa Estate in Kii Province. Its steward was famous for cruelty to the cultivators. For its story in English see Frölich 2007, esp. 41–118. 32. Shapinsky 2014, 74–77. 33. Mass 1976, 113–115. For background in English, see Ōyama 1990, 110– 112. This same essay by Ōyama contains an excellent description of the role of stewards within estates. 34. For an analysis of Tōdaiji’s estates in the early Kamakura period, see Piggott 1982, 45–91. 35. On the long history of the Nakazawa steward at Ōyama, see “On the Ōyama Nakazawa Jitō,” in Piggott, forthcoming. 36. Nagahara 1998, 193. 37. Nagahara 1998, 193–194; Farris 2006, chaps. 2–3, and esp. 92. Prob- lematizing Nagahara’s date of 1330 is Farris, who sees 1280 as “the dawn of a new era” and a turning point from economic stasis to growth. 38. Nagahara 1998, 195–196. In English, on the subject of slaves in medieval Japan see Nelson 2004. 39. Nagahara 1998, 86. Also see Ōyama 1990, esp. 89–96. 40. Nagahara 1998, 230–250. The Akutō Kenkyūkai continues to publish new studies on the subject. In English, see Harrington 1982, 221–250; Oxen- boell 2006, 1–21; my introduction to Ishimoda in Piggott 2006, 327–329; and Dan Sherer’s chapter in this volume. 41. Nagahara 1998, 216. In English on debt cancellations see Brown 1949 and Segal 2011, esp. 108–147. 42. See Goble 1996, 271–272, for his view of Go-Daigo’s policy on estates. 43. On hanzei in English, see Grossberg 1981a and Winterstein 1974, 210–220. 44. On shugo contracts in English, see Kawai 1977. On shugo involvement in Ōbe Estate, see the chapter by Noda Taizō in this volume. 32 The Big Picture

45. See Tonomura 1992 for a study of a collective that developed on an Enryakuji estate in Ōmi Province. 46. On Yano Estate in English, see Keirstead 1992, 72–87. See too my review of Keirstead in Piggott 1993, 41–42. On cultivators’ activism at Ōyama Estate, see Keirstead 1990, 357–388. 47. Nagahara (1977) presents a case study in English concerning the formation of another such village community at Kamikuze Estate in Yamashiro Province. There he notes that there was an earlier instance of such development in 1240, at Kubo Estate in Yamato. 48. For a discussion of these developments in connection with the estate system, see the chapter by Sakurai Eiji in this volume, and Segal 2011. 49. On the myō and myōshu development in English, see Keirstead 1992. 50. See Nagahara 1998, 99, for a detailed discussion of this contract deputy (ukeou daikan) mode of estate management and tax collection. Also see the chapter by Endō Motoo in this volume. Zen monks were active as deputies, as were monks from Mount Hiei. In English see Nagahara’s essay on Kamikuze Estate (Nagahara 1977), where a monk was the contract deputy in the 1340s. The roots of the system went back to later Kamakura times, when appointees of proprietors called custodians (azukaridokoro) collected rents for many temples and aristocrats. 51. Nagahara 1998, 244–250. 52. Recent research on developments at Niimi Estate, often cited as the best example of a long-enduring estate, include the essays in Ebisawa 2014; Ebisawa et al. 2014; and Ebisawa and Takahashi 2014. 53. In English see Keirstead 1992, and a short essay by Joan Piggott on the last days of the Nakazawa military steward in Piggott, forthcoming. For more on Hosokawa activities on Harima estates, see the chapter by Noda Taizō in this volume. 54. Nagahara 1975; Nagahara 1998, 308. 55. In Japanese a useful introduction to Amino’s perspective on estates is his “Shōen ni ikiru hitobito” (Amino 1995a). In English, essays from two of Amino’s books, Nihon no rekishi wo yominaosu (1991) and Zoku Nihon no rekishi wo yominaosu (1996), have been translated by Alan S. Christy (Amino 2012). Chapter 3 concerns estates specifically. 56. For details of the argument see Amino 1991a, 196–205. Amino’s views on estates can be found in Amino 1978a, 1–60. In English see Amino 2012, esp. 65–78. On markets and monetization in English see Segal 2011. 57. In English, Segal 2011 provides some discussion. For a helpful chronology with accompanying text and illustrations on the development of commerce in the capital district, see Shūkan Asahi hyakka Nihon no rekishi 1986, Chūsei, vol. 2, no. 20. Estates 33

58. Tōji archives printed to date can be accessed in three collections: Kyōōgokokuji monjo 1960–1970, Dai Nihon komonjo: Tōji monjo 1925–, and Tōji hyakugō monjo (Archives of Tōji in one hundred boxes), 2004–. For an English essay on the character and provenance of the Tōji hyakugō monjo, see Piggott, forthcoming. This collection was recently recognized by UNESCO as an outstanding resource for world historical memory, on which see Uejima, Ōyama, and Kurogawa 2015. 59. Uejima begins his story with donations of estates in Yamato and Iyo Provinces by the royal lady Senyōmon’in (1181–1252) in 1239. On Senyōmon’in herself and her estates, in English see Kawai 2015 and Shapinsky 2014, as well as Kawai’s chapter in this volume. Amino himself has written extensively about Tōji’s estates, in Amino 1966 and Amino 1978a. More recent are Ōyama 1996 and Tatsuta 2003. In English on Tara Estate, a Tōji holding on the Japan Sea coast, see Yamamura 1981. 60. For the methodology of “walking the estate landscape,” see Hattori 1995 and 2004; Satō and Gomi 1996; and Ishii 2000. Examples of what one might observe by walking an estate are described in this volume by Nishida Takeshi and Hirota Kōji. 61. Tōkyō Daigaku Shiryō Hensanjo, the Historiographical Institute at Tokyo University, has been publishing these maps with commentary and research. See the seven-volume Nihon shōen ezu shūei (1988–2003) as well as the first volume of Nihon shōen ezu shūei shakumon (2007–). 62. In this regard, announcements of finds of estate sites and artifacts in every part of the archipelago appear in the monthly journal Gekkan bunkazai hakkutsu shutsudo jōhō. In addition, frequent reports in the section “Bunkazai repōto” in the journal Nihon rekishi provide useful abstracts of estate site reports. And critically important on the material culture of estate studies are Uno 2001 and Ono Masatoshi 2001. See too the chapter by Nishida Takeshi in this volume. 63. A good sense of Kuroda’s views can be found in his “Chūseishi josetsu” (1975), 2–34. In English, see Dobbins 1996, 217–232, as well as other essays in the same issue of the Japanese Journal of Religious Studies, which is dedicated to Kuroda’s research. See too Adolphson 2000 (10–20), especially his discussion of Kuroda’s “gates of power” model of medieval society. Concerning Ishii Susumu, several of his essays are referred to here and in my essay in this volume concerning Kuroda Estate. In English, see Ishii 1985, 1–14. 64. Satō Makoto and Gomi Fumihiko (1996) are carrying on Ishii’s work (and that of Ōyama too) by studying the relationship between villages and estates, focusing on archaeology and fieldwork as well as written records. A good overview of the field of village (sonraku) development is Tamura 1994, esp. the introduction. 34 The Big Picture

65. For helpful background on these models, see Inaba 2012, 54–58. Inaba also points to the work of Satō Shin’ichi, who emphasized the tendency for posts to become hereditary as a key characteristic and vector in the medieval polity. Kimura 2012 discusses main themes in the research of these and other medieval historians. For Ishii’s views specifically see Ishii 2000, 3–49. 66. In English see Kudō 1983, 1–27 (in Japanese, Kudō 1969). Kudō is less focused on the shiki chain than is Nagahara and seems to echo Kuroda in his definition of the estate as “a stratified structure composed of different offices and their stipends, corresponding to each of the medieval social strata such as nobles, warriors, commoners, all guaranteed by state power” (9). Similarly, a historian of commerce, Sasaki Ginya (1972, 167–169), also sees the mid-thirteenth century as a time of decline for the shiki hierarchy as resident landlords and cultivators grasped the initiative. 67. Tabata follows the historiographical debates on “named fields” all the way back to the 1930s, so for a detailed view of myō development, the essay is particularly enlightening. 68. Another useful glossary, in addition to that in Nagahara’s volume, is found in Abe 2011. Also useful is Seno Seiichirō, ed. Nihon shōenshi daijiten (2003). 69. On royal grant fields, see Piggott, forthcoming. 70. Togawa 2013, 22–26, esp. 22. 71. See the report in Gekkan bunkazai hakkutsu shutsudo jōhō 1991.1, 62. On estate archaeology, see Uno 2001. 72. An estate for which we have documentation actually explaining the failure is Takaniwa-no-shō in Inaba Province. See Togawa 2013, 15; HI 1:62 (#s 71–72). Failure of Tōdaiji’s early estates by 950 is documented in the Tōdaiji yōroku (Fundamental records of Tōdaiji), a compendium of temple documents compiled around 1106. 73. See Katō 1995, 326–327, for a discussion. 74. See Kamakura and Ono 2013, 38–43, and Kawabata 2000, 59–65. The latter calls for a new look at shiki, given differences in their character in the estates of different proprietors, such as the Regents’ House and the royal family. The argument is that there were different sorts of estates with different shiki with very different powers and functions at various levels. What is critical to know is whether the supreme proprietor actively managed the property. In Eng- lish, on estate policy and history during this era, see Hurst 1999, esp. 588–614. 75. See Kiley 1999, esp. 318–319. 76. These changes in provincial administration are associated with the emergence of the court-centered polity, on which, in English, see Batten 1993; Kiley 1999; and Sasaki 2006, 227–244, especially the chart on page 241. Estates 35

77. In English, see Sato 1979 and Keirstead 1992. 78. HI 2:752–753 (#s 601–602). For a translation see Piggott, forthcoming. 79. See, for instance, Sato 1974 and 1979 on the difficulties faced by Tōji at Ōyama Estate. 80. Kawabata Shin’s research (2000) on instituted estates is particularly informative. He notes that Kudō closely connected their expansion in the twelfth century with the proprietary province system (chigyōkokusei), whereby top royals and regents were given control of administration and tax collection in a province as a benefice (on which see Gomi 1984, 123–169). The historical importance of instituted estates has increased too as many documents from commended-type estates have been found to be late Heian forgeries. Satō 2001, 2–3. 81. On construction of the Byōdōin at Uji in English, see Yiengpruksawan 1995. And on the structure and images of Byōdōin, see the chapter by Yoshiko Kainuma in this volume. 82. Satō 2001, 5–10. 83. Asakawa Kan’ichi (1965b, 37–68) wrote an early case study of Ushiga- wara Estate in English. 84. Morita 2013, 67–68. Satō 2001 also provides useful background and context concerning instituted estates as does Hatano 2005. Morita uses the term “network,” but Cornelius Kiley (1974, 109–126) called it “a vertical faction.” G. Cameron Hurst (1976, 254–274) also provides perspectives on estates founded during the period when retired monarchs led the court. 85. As one example, Kimura points to Ikeda Estate in Tōtōmi, which was set up in 1171 with ten different named units (sato, ho, etc.), including ponds, dry fields, paddy, orchards, forest, hills, and mountains (Kimura 2000, 31; see HI 7:2774–2790, # 3569, dated 1171). Kimura and others see these as places for considerable land opening in late Heian times, but there is debate about the success of such efforts. 86. As I have pointed out, that tension was hardly new. It was already evident in the early 900s, in the early regent Tadahira’s time. See Piggott and Yoshida 2008, 23–81. 87. See too Kawabata 2000, 14–16, 23–47. Kawabata notes that even late in Shirakawa In’s time many new instituted estates were formed by royal order to replace benefice household payments that were no longer reaching beneficiaries, none other than shrines and temples that were official ritual centers. The process of exchange was called benpō, and it essentially created “official (kokkateki) estates” (69). Kawabata concludes that instituted estates were more important for the estate system than were commended estates. And they ultimately represented a serious fiscal and ideological contradiction for court gov- 36 The Big Picture ernment: top court elites relied on estates for their livelihood, but also felt compelled to promulgate estate regulations as in the past. 88. Another member of the Shōen History Research Group, Kamakura Saho, made this the theme of a recent presentation titled “Twelfth-Century Innovations in the Formative Process of the Shōen System” at the University of Southern California in November 2015. For an abstract, see dornsife.usc.edu/ shoen. 89. In the English historiography we do not yet have such an overall evaluation of the Kamakura age, but various essays in Mass 1982 and Yamamura 1990 point to this conclusion. 90. Another good look at the Ashikaga shogunate, shugo, and estates is Kishida 1999. In English see Miyagawa 1977. 91. See Kimura’s chapter in this volume. For his quite readable essays on farming, see Kimura 2000 and Kimura 2010, esp. 82–109. 92. Takahashi Toshiko (2012) has stressed regional and topographical differences for estates as well. For a recent case-study approach, see Ebisawa and Takahashi 2014b. Ebisawa is well known for his preservation efforts at Tashibu Estate in Takada City of old Bungo Province, on which see Ebisawa et al. 2014; Ebisawa 2011; Ishii 1995b; and Asahi Hyakka 1995, 2–17. 93. For more on payment by substitution, including the use of cash, see the chapter by Sakurai Eiji in this volume. 94. James Foard (1977) and Laura Kaufman (1980) have written in English about Ippen’s travels, his faith, and the scroll. 95. A thoughtful review of Nagahara’s Shōen can be found in Kurushima 2000. 96. A good discussion of Tashibu Estate is Ishii Susumu’s “Kunisaki no shōen no miryoku,” in Ishii 1995b, 3–18. 97. Kamakura 2013, 129–162. 98. Kamakura 2008, 47–55. 99. Yasuda 1977, 250–253. An important factor is that the records of the great religious institutions have been better preserved than those of other proprietors, including the royal house, the Regents’ House, and aristocratic families. I also note that the properties of such notable courtier houses as the Saionji are missing from Yasuda’s list, perhaps because the record is not robust. Nota- ble too is that in his province-by-province list, many more estates are included. c h a p t e r 2

Medieval Japan’s Commercial Economy and the Estate System

Sakurai Eiji Translated by Ethan Segal

Some readers might be surprised to discover that this chapter links two topics that are not usually associated with each other: economic history—the history of trade, markets, and coinage—and the history of medieval estates. Although we tend to think of economic history (and especially monetary history) as something distinct from the study of landed estates, in fact the fields did not develop in isolation from each other. To the contrary, medieval Japan’s economy grew in large part out of changes to the means and methods of collecting rents (nengu) under the estate system. Therefore I hope to demonstrate in the pages that follow why there is meaning in introducing to a non-Japanese audience recent advances in the field of commercial economic history (ryūtsū keizaishi), focusing in particular on its relationship to the estate system.1

Medieval Japan’s State Structure and the Evolution of Currency Recent research breakthroughs in commercial economic history have come from the subfield of the history of currency, so I will start by introducing some of those findings. Starting in the late seventh century, the imperial court produced thirteen different types of bronze coins over a period of approximately three hundred years.2 These coins were inspired by the types of coins made by Chinese imperial dynasties: they were similar in size, shape, design, and metal content. But when the Japanese government

37 38 The Big Picture stopped producing new bronze coins in the mid-tenth century, its existing coins quickly lost the trust of those who had been using them. As a result, the circulation of bronze coins ceased in the early eleventh century, and for the next 150 years or so, Japanese did not use metal currency. Instead, late Heian society was an age of commodity money, relying on rice, silk, and linen to facilitate exchange. The timing of this development may strike some readers as unusual, for in the mid- tenth century, neighboring countries such as Goguryeo (Korea) and the Dinh dynasty (Vietnam) were just beginning to cast bronze cash, while at the same time Japan was moving in the opposite direction by abandoning its cash economy. Behind the court’s decision to forsake bronze cash was the practical problem that the supply of copper was dwindling. An even more important factor, however, was the transformation in the nature of state finances, which minimized the need to cast bronze coins. Generalizing from the examples of China and Japan, the motivation for producing bronze cash was not to provide a common medium of exchange for ordinary people but rather to create a means of payment, that is, a way for the government to pay soldiers and workers. It is a characteristic of autocratic states to spend huge sums on public works and the military and thereby increase the demands on the state treasury by soldiers and workers who need to be paid. Most of the Chinese dynasties behaved in this way, as did ancient Japan, which, especially in the seventh century, was concerned about military threats from the Korean peninsula. Japa- nese leaders quickly strengthened and centralized state power at that time by modeling their government on that of China. They studied and implemented Chinese notions of how to build cities and design legal and governmental structures, as well as other aspects of an advanced society. It is well known that there were deep connections between ancient Japan’s casting of bronze cash and the construction of its capital cities. It is considered established fact that the Wadō kaichin coin, considered for many years to have been Japan’s oldest domestically produced coin, was made in order to cover the labor costs of constructing Heijō-kyō (Nara). And some researchers believe that the fuhonsen coin, which has recently displaced the Wadō kaichin as Japan’s oldest coin, was produced to pay for the construction of Fujiwara-kyō, which was the capital prior to Nara.3 With the movement of Japan from the classical into the medieval age, however, there was a transformation in the structure of the Japan’s Economy and the Estate System 39 economy. By the eleventh century—the late classical age—the court no longer found it necessary to spend huge sums like an autocratic state. Those types of states, like imperial China and classical Japan, took advantage of the fear of foreign invasion to successfully build up state structures, but the way they did so was inefficient due to their fluctu- ating demands for large quantities of men and materials.4 In Japan’s case, the seventh-century court used concerns that Tang China might invade Japan to justify strengthening central government institutions. But Japan’s imperial court began to move away from such structures as the Tang dynasty declined and eventually collapsed in 907 and the fears of war were relaxed. Although rulers had often lived in royal palaces in earlier times, during Japan’s medieval age there was no regular practice of the sovereign living in a special palace. Instead, whether emperor or shogun, medieval rulers typically lived in residences that were at best the equal of those of the upper aristocrats. In fact, the homes of the medieval emperors were known as “sato dairi ” and were situated among the homes of the nobles across the city of Kyoto.5 In addition, in medieval Japan there was no equivalent to the chronic pressures placed on state finances that the Chinese dynasties had to deal with due to the threat of northern nomadic invaders. Even on occasions when domestic battles were fought in Japan, the government did not have to provide salaries to its fighting men because warriors were expected to provide for themselves. As explained above, the primary reason premodern societies created bronze cash was to have a means of payment for soldiers and workers, so the (otherwise puzzling) fact that medieval Japan never issued its own coinage makes perfect sense when understood in light of these circumstances. From Heian times on, in fact, Japan was for the most part spared from the stringent budget pressures created by military expenditures and spending on public works. Due to these and other changes in the structure and thinking of government in the middle of the Heian period, the nearly three hundred years of Japan making its own bronze coins came to an end, and the circulation of cash stopped soon thereafter. Rice, silk, and linen took on the functions of currency for the next 150 years. Strange as it might seem, the early eleventh century, best known as the “golden age” of aristocratic culture that produced The Tale of Genji and other great literary works, actually was a period during which metallic currency was not used. 40 The Big Picture

In the middle of the twelfth century, however, people in the Japanese islands suddenly began using bronze cash again. Unlike the case in the seventh through tenth centuries, during which the economy relied on domestically produced currency, the newly emerging economy at the end of the Heian age was based on imported Chinese coins. Although the volumes of coinage in circulation would fluctuate over time and would sometimes be supplemented by domestically produced “nongov- ernment issued” (i.e., counterfeit) cash, this reliance on Chinese coins would continue until the successful production of Kan’ei tsūhō coins by the Edo bakufu in the middle of the seventeenth century. Most of the Chinese coins that entered Japan in the twelfth century were of Northern Song manufacture. In order to understand the spread of Northern Song coins at this time, we must recognize that the economic policies of the Northern Song resulted in larger quantities of bronze coins than from other historical Chinese dynasties.6 The North- ern Song dynasty, however, collapsed after being attacked by the Jurch- ens in the 1120s. When its coins started to appear in Japan, the remnants of the previous dynasty had formed a new dynasty known as the South- ern Song (1127–1279), which did not place as much emphasis as its predecessors on producing metal currency. The twelfth-century Japa- nese were fortunate in that many coins from the previous dynasty were available for export to Japan. But since Japan at that time was still operating as a commodity economy, there was no guarantee that imported coins would immediately be accepted as currency. If merchants or other individuals brought Chinese coins to Japan expressly for the purpose of circulating them as money, then they were taking quite a risk. Here again, recent research offers new insights on the question of how bronze cash was used in late Heian and early Kamakura-period Japan. A research group at Beppu University led by Iinuma Kenji measured the lead isotope ratios of sutra containers from all over Japan. Most of the containers were made of bronze, an alloy that includes copper as a primary ingredient. The researchers found that those containers exclusively used Japanese copper until the middle of the twelfth century, but after that point almost all of them included significant amounts of copper from southeastern China.7 Iinuma’s results strongly suggest that Northern Song coins were being melted down and used in the fashioning of bells, mirrors, and sutra cases. Even though the Japa- nese had a history of using coins as money in earlier times, that was part of the distant past from the perspective of twelfth-century people who Japan’s Economy and the Estate System 41 had no personal memory of or experience with metal cash. To export coins to a place like Japan that did not have a money economy would have entailed tremendous risk unless the coins were originally imported for use as something other than money. Very likely, the Japanese came to value them initially for their copper content. In this way, the monetary history of premodern Japan followed a complicated progression from an age of domestically produced coinage (seventh to tenth century), to an age of commodity money (eleventh to early twelfth century), to an age of Chinese coins (late twelfth to mid- seventeenth century), followed by a return to domestically made money (but of three metals: gold, silver, and bronze, from the mid-seventeenth to the mid-nineteenth century). In reviewing these stages of monetary evolution, one noteworthy observation is that the age of Chinese coins corresponds almost exactly to the years of Japan’s medieval age.

The Cash Commutation System and the Medieval Market Economy Chinese money spread throughout Japanese society in several stages. It really began to be used in the mid-twelfth century. From that time, coins started to absorb most of the medium-of-exchange function that had been previously performed by rice, silk, and linen. It was not until the thirteenth century, however, that cash began to function as a means of payment. In short, although cash quickly became popular as an instru- ment for buying and selling land and goods, material commodities continued to be used to pay estate rents and taxes. According to Matsu- nobu Yasutaka, coins gained acceptance as a means of payment in two stages: first, in the late 1220s, when they overtook silk and linen, and then approximately half a century later, in the 1270s, when they came to replace rice.8 There have been many important changes over the long history of estates, but among them, the shift to cash commutation—to paying estate rents in cash rather than goods, known in Japanese as daisennō— must rank among the most significant. Commutation became widespread in the second half of the thirteenth century, particularly in the 1270s.9 Until recently, this shift had been attributed solely to domestic factors such as increased economic activity and commodity exchange. Relatively recent research by Ōta Yukio, however, makes a convincing case that commutation was actually connected to the changing condi- 42 The Big Picture tions in China that accompanied the collapse of the Southern Song and the founding of the Yuan dynasty.10 In 1276, the Southern Song capital of Lin’an (Hangzhou) fell, and the Mongols came to control China. The following year, they issued a policy requiring the use of paper money and forbidding the use of bronze cash. As a result, bronze coins lost their domestic value in China and began to flow overseas in large numbers to Japan, Java, Vietnam, and other places, leading to the common use of Chinese coins throughout East Asia. There should be no doubt that the switch to commuting rents to cash payments owes much to the ready availability of exported Chinese cash, which resulted from the monetary policies of Yuan dynasty China. With regard to the process by which rent payments were converted into cash, scholars have come to realize a number of important insights. Among those, one of the most significant was Amino Yoshihiko’s conclusion that medieval rents usually required the payment of much more than just rice. Drawing on assessments that were often valued in terms of rice, earlier scholarship had made it seem that the grain was the primary (or even the only) item paid as rent. But depending on the location of the estate, the needs of the proprietors, and results of demands and negotiations over the years, rent obligations might also consist of salt, salmon, abalone, and other products from the ocean; lumber, cypress bark, and other goods from the woodlands; iron, gold, and other miner- als; silk, linen, and other textiles; straw mats or other craft goods; etc. In other words, estate rent payments included many products that were not the result of wet rice agriculture. Moreover, those products came to be associated with the certain specific regions in which they were produced.11 Rent payments in rice were quite common in western Japan, but in Mino, Owari, and points farther east, rice was less commonly used to meet estate obligations. Many estates in Mino, Owari, Ise, Hitachi, and Shimotsuke Provinces paid in silk, whereas those in Suruga, Izu, Kai, Sagami, Musashi, Kazusa, Shinano, Kōzuke, Mutsu, and Dewa tended to pay in other textiles. In the Hokuriku region, along with estate rent obligations in rice, silk floss was also widespread, and in western Japan, there were some provinces like Tango that had relatively high amounts of silk and thread as rent payments. In general, estates in western Japan provided rice whereas those in eastern Japan paid in textiles. There were, however, many exceptions, such as salt from the provinces bordering the Inland Sea, iron from the mountainous Chūgoku region, paper Japan’s Economy and the Estate System 43 from Tajima, straw mats from Hōki and Izumo, lumber from Suō, gold from Mutsu, oxen from Nagato, and horses from Mutsu, Dewa, and Shimotsuke. In truth, each province became known for multiple types of goods. According to Katsuyama Seiji, the earliest precursors of this regional specialization in the production of commercial goods began to form in the early eleventh century.12 As Amino went on to stress further, an important characteristic of the medieval estate rent system was the seeming contradiction that it taxed wet rice fields in goods that could not be produced by wet rice fields. Amino contends that the ancient chō and yō —tax obligations that were originally levied on people—evolved into a land tax, leading to the rent obligations levied on rice fields. For the most part, scholars agree with him, but they are divided on an important question: whether the producers of non-rice goods were the same individuals as the people who were required to pay rents. In other words, did rice farmers make non-rice commodities as secondary work or were there full-time, nonrice-producing specialists who only made other commodities? Accord- ing to advocates of the first view, communities that were obligated to pay estate rents would grow rice and also engage in non-wet-rice agriculture during the times when they were not busy in the fields. Aside from the portions of production that they paid as rents, almost all the rice they harvested was for their own consumption or profit. In contrast, advocates of the other view hold that specialist craftsmen (who lived on or off of estates) produced non-wet-rice agricultural goods, in short, that these craftsmen were distinct from rice farmers. Communities that were required to pay rents would use their rice to purchase commodities from those specialists and then use those commodities to pay proprietors as part of their obligations. Those who follow this view tend to assume that the main actors of trade were not the rent payers themselves but more likely the estate managers (shōkan) or their agents (daikan). Amino favored the view that rent payers were also craftsmen (i.e., the same people produced rice as well as other goods). He believed that many of those commodities, from the salt made along the Inland Sea to materials such as iron and silk, did not necessarily require specialists’ skills; he referred to them as “the production of average people.” But not all scholars concur on this point. For example, Nagahara Keiji contended that goods woven from silk were beyond the skill of the average farming family. In classical Japan, they were made at the provincial or district office (where, according to Nagahara, officials employed special- 44 The Big Picture ist weavers—who were distinct from ordinary peasants—expressly for that purpose). In medieval times, silk production was generally a large- scale enterprise undertaken by members of the military land steward or estate manager class; or in Mino, Owari, and a few other provinces, there were special powerful production areas where specialist families or villages worked. Either way, Nagahara asserted that specialized producers were involved in silk production.13 A counterexample is the salt made along the Inland Sea at estates such as Yugeshima, a Tōji holding in Iyo Province. Amino’s research has shown that on this estate, the producers were the same individuals as those who had to pay rents, confirming that they made salt as a secondary occupation to their farming.14 Recently, Ebisawa Tadashi has supported Amino’s theory with a study of the payments made in iron by Yoshino Village on Bitchū Province’s Niimi Estate, another Tōji holding. His study, which focused on the method of collecting iron sand known as “kan’na nagashi,” concludes, “It is clear from Yoshino Village that each myō (unit of rent payment) was meeting its rent obligation by paying in iron. Until the end of the thirteenth century, the head of the myō and the estate residents under his control produced the iron needed for rent payments by working in the agricultural off-season— that much is certain.”15 On the other hand, there is abundant evidence that silk producers were not always the same people as those who owned the rice fields and paid rents. For example, consider a Rokuhara judicial pronouncement from 1278 concerning Akanabe Estate, a Tōdaiji holding in Mino Prov- ince famous for its silk.16 The document includes testimony from an assistant land steward, who clearly states that “recently silk and silk floss have been very expensive, so it is difficult to purchase them with nengu rice.” As this comment indicates, the estate obtained the silk and silk floss that it paid as rent through trade. Some challenge this view, arguing that it only reflects the late thirteenth century as an age of declining silk production and that in earlier times the estate residents wove their own silk. But we also find a legal petition from 1279 which, quoting Records from the Distant Past (Ōko kiroku), states that “no matter how terrible the conditions (drought, flood, losses) might be, we must strive to ensure that the peasants continue to work with all of their strength to provide one hundred hiki of silk and one thousand ryō of cloth.”17 We are not able to assign a definitive date of composition to Records from the Distant Past, but it makes it seem as if expecting estate residents to Japan’s Economy and the Estate System 45 use the wealth they obtained from rice production to acquire silk or silk floss was not something new. Challenging these examples, however, Sasaki Gin’ya stresses the continuity with the codal chō-yō tax system, proposing that this type of purchase method was already in place by the late Heian or early Kamakura period.18 Although it may seem like throwing water on a heated debate, looking at the evidence in this way reveals that when it comes to the production of rent goods other than rice, there are examples of both: some producers were the same people who paid estate rents and who therefore produced commodities as secondary work, while others purchased commodities from distinct specialists who were not also engaged in rice production. Further complicating our picture, we can anticipate small differences case-by-case in the process of commuting rents into cash. In the former instance, in which the cultivators were the producers, we think they only paid a small portion of their production to the estate proprietor and sold the rest to merchants in the marketplace. If the goods were of insufficient quality to become market commodities, then they surely were not good enough for the estate proprietor either. The difference between before and after the institutionalization of cash commutation was only that, formerly, the goods were paid directly to the proprietor, but later they were sold and the money earned from their sale was in turn used to pay the proprietor. The estate residents’ economic activity did not actually change very much. But many questions about this process remain to be answered, such as how much effort the leading estate residents put into rice production as opposed to how much they foisted it off onto lower peasants so that they could spend more time on commodity production. On the other hand, in the latter instance—when cultivators or estate managers purchased commodity goods from separate specialists—the spread of commutation had little impact on those specialists. Whether before or after cash payments became commonplace, those producers remained specialized commoners who received rice or coins in exchange for their products. The difference was for the buyers, who began to pay in cash rather than silk or floss, and the people with whom they traded shifted from the handicraft producers to merchants. In either case, even before the advent of cash commutation, the exchange of goods (for rice or cash) was an essential element in local economies and can be seen as having smoothed the path to the commutation of 46 The Big Picture rents. In those regions where commodity exchange was not very active, such as regions in which rice production was ascendant, we might sus- pect that cash commutation developed more slowly. It seems logical to assume that those areas that were not as dependent on rice production should have been the quickest to commute their rent obligations. Ōyama Kyōhei was able to confirm this with groundbreaking research on Nishidai Village on Tanba Province’s Ōyama Estate. His analysis showed that cash commutation emerged not in the places that were primarily focused on rice production, but rather in the areas that did not have good growing conditions for rice and were forced to engage in other kinds of production.19 We must conclude that commutation fundamentally changed the economic structure of medieval Japan. First of all, it gave rise to the production of an enormous volume of commodities. Because most of the products that had in earlier times been sent to the capital as rent were now being sold locally, from this time forward most of the goods that were shipped to the capital were commercial commodities rather than rent payments. Facilitating these transactions were fixed-day markets that emerged in each locality and became the sites for goods to be converted into cash. As commutation began to spread throughout the Japanese islands from the late thirteenth century, the volume of goods circulating as traded commodities came to exceed the volume of goods circulating as rent. In short, Japan entered a market economy stage of society.20 Evidence to support this claim can be readily found in the Regis- ter of Incoming Ships at the Hyōgo North Checkpoint (Hyōgo Kitaseki irifune nōchō), a register that recorded all the port taxes paid by boats in 1445 (Bun’an 2) at the Hyōgo North Checkpoint, a Tōdaiji holding in Settsu Province.21 According to this document, in just that one year, approximately 100,600 koku of salt, 37,000 koku of lumber, 20,400 koku of rice, and many other goods of all kinds—barley,22 beans, and other grains; sardines, sea cucumbers, dried and salted sea bream, shrimp, abalone, kelp,23 and other kinds of seafood; iron, sesame, ramie, straw mats, indigo, vases, and earthenware mortar bowls—passed through the port. As the register reveals, huge volumes of variegated goods were transported from Shikoku and the western end of Honshu through Hyōgo North Checkpoint on their way to the Kinki region. Since most of the goods had checkpoint taxes (sekisen) levied on them, we know that they were not rent payments but rather commercial goods being transported Japan’s Economy and the Estate System 47 for sale.24 And we cannot ignore that among them were salt, rice, lumber, iron, woven mats, and other items that were once supplied as rent payments. The register shows in detail how goods that had formerly been collected as estate rents were now being sold for cash locally and then transported to the Kinki region, which had a large population of consumers. The spread of cash commutation was also a major factor behind the birth of a credit economy. Amidst the growing circulation of commodity goods and expansion of the economy, the demand for a lighter weight form of payment that was less expensive to transport than metal coins led to the appearance of “saifu.” These were bills-of-exchange similar to the promissory notes and drafts of today.25 Most of them were issued by toiya, who were transport specialists and trade brokers around the Kinai region and in the major port towns. Merchants traveling to the provinces would bring these bills with them to use in buying the goods produced in each rural locality. There can be no doubt that the huge volumes of goods being moved (as we saw recorded in the Register of Incoming Ships at the Hyōgo North Checkpoint) were due to the operations of these long-distance traveling merchants. The bills that were given to the estate manager or leading cultivators in exchange for their produce would in turn be forwarded by them to their proprietor in the capital as rent payment, and the proprietor would then bring the bill he received to the trade broker who had issued it to redeem it for cash. In this way, things advanced even further beyond commutation, and more than a few estates paid their rent obligations by means of these medieval bills-of-exchange.26 What especially draws our attention is that most saifu were of the fixed value of ten kanmon.27 If you were to count them, one saifu generally represented ten kanmon, two saifu equaled twenty kanmon, etc. This fixed value meant that they could be circulated from person to person as if they were a type of paper money. Ten kanmon at that time was the rough equivalent of ¥600,000 to ¥1,000,000 today ($6,000 to $10,000 if the exchange rate is ¥100 = $1), revealing that the late medieval Japa- nese economy had advanced to such a degree that it needed to produce this type of special bill-of-exchange to permit the easy transmission of such large sums. Incidentally, because most of these long-distance traveling merchants paid checkpoint taxes when they traveled around the countryside buying local produce, we know that they were different from the 48 The Big Picture specially privileged, tax-exempt merchants of earlier times. In the early medieval era, purveyors to the imperial court, temples, and shrines were bestowed with the title kugonin or jinin in exchange for providing the court or religious houses with certain tribute goods. These merchant- like purveyors received exemptions from the taxes levied on other travelers or were given a monopoly on business in a certain region. In early medieval times there were many of these special merchants operating actively, but the only merchants of that type listed in the register of the Hyōgo North Checkpoint were the Ōyamazaki jinin, who were oil merchants affiliated with Iwashimizu Hachimangū.28 This suggests that special merchants of that type were no longer the main actors of economic activity in late medieval times. It is also very likely that the spread of cash commutation increased the production of commercial crops. Although commutation meant a greater demand for cash to make estate rent payments, no one cared in the least how the payer obtained that cash. Therefore, some villages and estates that had previously been forced to grow rice no longer had the need to do so. Instead, they could raise or make the product which best suited soil and climate conditions and would most readily be sold (i.e., converted to cash); in short, commercial crops were the clear, logical choice, and at a minimum, the cash commutation system must have systematically supported such actions on the part of producers. Products that were until recently believed to be typical early modern commercial crops have been found in earlier records.29 For example, indigo from Awa Province is listed in an entry in the Register of Incoming Ships at the Hyōgo North Checkpoint, which notes that 442 koku of indigo cleared customs at the checkpoint. This example, which dates to the early fifteenth century, is of great significance since it demonstrates that large quantities of commercial indigo were being shipped to the Kinki region some two to three hundred years earlier than what we would expect from the accepted historical narrative.30 According to Letter-Writing Samples for Study at Home (Teikin ōrai), which dates to the fourteenth century, as well as “Various Fine Prod- ucts for Sale in Kyoto” (Kyōto shogeisai baibai daimotsu no koto), from the late fourteenth or early fifteenth century, new kinds of products— things that up to that point had been included neither in textbook-like materials such as ōrai mono nor in lists of estate rent products—began to appear.31 Surely it was the spread of cash commutation in the late thirteenth century that created the conditions for these special products Japan’s Economy and the Estate System 49 and production areas to thrive.32 Recently, Takahashi Kazuki has identified the mid-thirteenth century as the period when building materials began to be branded with the name of the locality in which they were produced or with the name of the port from which they were shipped. This matches closely to the timing of the spread of cash commutation.33 In truth, there are significant constraints on what we can learn from historical documents since they reveal very little about the actual conditions under which commercial crops were produced in the late medieval era. But among the few that at least offer a glimpse is an early sixteenth-century legal complaint filed with the Muromachi shogunate by Henjo Shin’in, the proprietor (ryōshu), against the manager of a rice field in Nishi Hachijō, Kyoto. According to the lawsuit, although the manager claimed that he needed a reduction in estate rent every year because of agricultural losses, he simultaneously enjoyed a flourishing business cultivating commercial crops such as indigo and melons.34 The implication is that his fields were fine but he preferred to grow commercial crops for the revenue that they produced. We also saw examples of indigo cultivation in the Register of Incoming Ships at the Hyōgo North Checkpoint, but the timing of the Nishi Hachijō case, which overlapped with the Kujō Indigo Guild, suggests there may have been a connection.35 In this way, fields that should have been used to produce rice were actually converted to the production of commercial crops, and not merely as secondary crops. Surely there were many more examples like this at the time. This particular example, on the outskirts of Kyoto so close to the estate proprietor’s residence, became a problem because it happened to receive an illegal rent exemption. Apparently the estate manager thought that if he continued to pay a minimal amount of rent while acting innocent, then the proprietor would not notice, or at least would not blame him for the illegal cultivation. On the one hand, one might think that the proprietor would immediately notice if cultivators or managers on their own decided to convert wet rice fields into commercial crop production, and that conflicts would arise when what was paid did not match what was due, but in fact we have no examples of such conflicts. This is most likely because over time, as the members of agricultural communities came to pay their obligations as entire villages, the records maintained by the estate proprietors increasingly became works of fiction, divorced from the reality of what was being grown in the fields. Katsumata Shizuo explains that with the spread of “whole village” payments, the villages started mak- 50 The Big Picture ing their own records, while at the same time the estate proprietors no longer had much need for detailed records on individual fields. As a result, gradually, their records became more fictional than a reflection of reality as they “cooked the books” by manipulating the numbers.36 This surely is the reason that commercial crops, even if they were in fact being grown, rarely appeared in the records of the estate proprietors. One additional reason why commercial crops do not appear in the written documents is that most were planted in dry fields. Kimura Shi- gemitsu has shown that dry fields were traditionally difficult for estate proprietors to control. Therefore, advances in dry field cultivation surely “encouraged the production of commercial crops and were a basic process that helped create a medieval division of circulated goods system.”37

The Demise of the Medieval Market Economy The cash commutation system continued for some three hundred years until, in the late sixteenth century, warlords known as sengoku daimyō revived payments in rice under the kokudaka (rice-based assessment) system.38 The most powerful explanation for the demise of cash commutation can be found in late sixteenth-century monetary trends. As Kuroda Akinobu has made clear, the supply of cash from China to Japan was interrupted due to a number of complex factors operating on a world- wide scale, and the result was the decline of the monetary economy in Japan.39 First, we must recognize that underlying the premise of his argument is the knowledge that “good money” (seisen, or official, government- issued coins) was becoming scarce in both China and Japan in the sixteenth century, and that even if people supplemented it with counterfeit coins, the volume in circulation was barely adequate to maintain a currency economy. In Japan, along with domestically produced counterfeit coins, there were many counterfeit coins that were made in China and imported along with “good money.” With those conditions as background, according to Kuroda, the smuggling rings that brought coins to Japan suffered two devastating blows in 1566–1567 when the Ming government attacked the wakō pirate stronghold of Zhangzhou in China and at the same time partially lifted the ban against ocean travel. Then, with the establishment of regular ship service between Manila and Acapulco, silver from the Japan’s Economy and the Estate System 51

South American Potosí mine began to flood the southeast China coast, which transformed that region from a coin-using area to a silver-using area and drove out the counterfeit coin producers. Their organization, which had relied on both domestic and foreign demand, lost its domestic base. As Kuroda sees it, the coin economy declined around the year 1570 in two stages: first the smugglers disappeared, then the counter- feiters disappeared. It is a view that we would do well to accept for the most part. With one of the two sources of coins (those from China, with the other source being from within Japan) drying up, coins became increasingly scarce in Japan after the 1570s, making it impossible to maintain a cash-based economy. It was a remarkable phenomenon, but it led to western Japan transforming itself from a region that used cash to a region that used rice. The rise in cash prices that has been observed in Nara during the Tenshō years (1573–1592) is attributable to the same factors.40 And so the curtain fell on three hundred years of daisennō. The payment of rents began to regress to a system once again based on rice. This was, of course, the basis for the switch from the kandaka system (calculating taxes in cash) to the kokudaka system (calculating taxes in rice). Due to these changes, the volume of goods circulating as estate rent payments came once again to exceed the volume of goods circulating as commodities. This was because among the goods that circulated in late medieval times, rice and other grains returned to their status as fulfilling rent payments. As a result, Osaka emerged as the exchange market for the entire country, and the regional fixed-day markets that had emerged in the countryside for converting local agricultural produce into cash became less important and were reduced in number. This led to the end of the medieval market economy society that had been created by the lifeline of cash commutation. What was the fate of the regional industries that emerged in conjunction with the market economy of late medieval Japan? The variety that was seen in early medieval rent payments disappeared following the return to paying obligations in goods. But this does not mean that those various industries declined immediately. The return to payments in rice coincided with two other phenomena: first, the unification of the country under Toyotomi Hideyoshi (1537–1598), and second, increased demand for military provisions (including food supplies) to support 52 The Big Picture

Hideyoshi’s invasions of Korea. Under these circumstances, even if the variety of industries survived, it would be hard to see that reflected in the items paid as rents. Although the wide range of goods paid as rents in early medieval times offers insights into contemporary industries— not only farming, but also weaving, mining, paper production, and salt production—the switch to paying taxes and rents in rice means that we cannot employ a similar approach to understand the industries of the late sixteenth and seventeenth centuries. There are few documents that reveal much useful information about the trends of industry and commercial production during the transition period from late medieval to early modern. One of the precious few is Kefukigusa, a guide to haikai poetry compiled by Matsue Shigey- ori (1602–1680). Its publication date is not entirely clear: although the introduction mentions the year 1638, various detailed evidence has led to the more convincing theory that it was published in 1645. Either way, it clearly dates to the first half of the seventeenth century. The fourth volume of Kefukigusa includes a list of the famous products of each province, and it is from that list that scholars today have come to a detailed understanding of the specialized goods of the early Tokugawa period. The number of special products listed in Kefukigusa is 1,807 items, which is actually thirty times as great as the list of 61 special products found in the fourteenth-century Letter-Writing Samples for Study at Home. The breakdown is 276 agricultural products (15.3 percent), 171 items made from wood (9.5 percent), 334 ocean products (18.5 percent), 115 mineral and animal products (6.4 percent), 167 items of clothing (9.2 percent), 33 types of fuel (1.8 percent), 269 kinds of household goods (15.9 percent), 55 different production tools (3.0 percent), 164 craft items (9.1 percent), 57 military implements (weapons or armor, 3.2 percent), 113 medicinal items (6.3 percent), and 53 other various items (2.9 percent).41 Of course, neither book was created for the purpose of recording statistics, so a straightforward comparison is difficult. On the one hand, it seems safe to assume that the various provincial industries did not immediately suffer a devastating blow due to the abolition of provincial cash commutation. To be clear, although Kefukigusa refers to provinces across the country, in fact 40 percent— approximately seven hundred items—were concentrated in the Kinai region, revealing that there was an unequal distribution of industry and Japan’s Economy and the Estate System 53 technology that favored the center. During this time, the possibility that industry declined in the rural regions is real, because the shrinking of local markets must have had a negative impact on the surrounding industries. That some items that appear in Letter-Writing Samples for Study at Home do not also appear in Kefukigusa seems to confirm this possibility, although there was no effort by the author of the later work to coor- dinate with or reflect on the list in the earlier one. On the other hand, it is unlikely that so many varied industries managed to grow and prosper from nothing in the (at most) half a century following the abolition of cash commutation. Certainly the Kefukigusa list includes mirin (a type of rice wine), soy sauce, playing cards, and other new products that had not existed in earlier times, but more than half emerged from traditional industries that date from the medieval era. For example, the list makes mention of melons and indigo from Kyoto’s Ninth Avenue as well as Yamazaki oil, all of which were products of the medieval age (and discussed earlier in this chapter). Presumably these products never disappeared from the marketplace. Among economic historians who study the sixteenth and early seventeenth centuries, the most significant area of investigation and debate centers on attempts to evaluate the impact of the transition from rural provincial markets to an integrated, “countrywide” economy based on rice. Scholars believe this change to have been more important than the increase in commercial crops or handicrafts. The emerging nationwide market at Osaka took over the function that local markets had previously borne for converting rice into cash, and so the number of long-distance traveling merchants (who came to the countryside in part to buy rice) sharply declined. As a result, the fate of other industries including the cash crops and handicraft goods of each region (i.e., the non-rice goods sold in those provincial markets) was determined by how much the remaining merchants would purchase. Unfortunately, clarifying those conditions, including the basic challenge of devising a specific approach to the subject, are all topics that remain for future research. Incidentally, the Edo bakufu attempted to address the late sixteenth-century dearth of coins by casting the Kan’ei tsūhō coin in 1636. It was the first time a Japanese government had produced metal currency since the middle of the tenth century, almost seven hundred years earlier. Also from the early seventeenth century, the bakufu began producing money in gold (the Keichō koban) and silver (the 54 The Big Picture

Keichō chōgin). By issuing coins in gold, silver, and bronze, it shifted the national currency policy to one based on three metals for the duration of the Edo period.

It has become clear in recent years that the economic system did not develop in a simple, steadily rising progression. Instead, it was more complicated: it included many steps forward and back, with new practices and patterns forming and collapsing repeatedly. According to various theories of stages of development, whether with respect to a market economy or a credit economy, we assume that one stage followed on the next naturally, but actual history often follows a different path. The premodern economy of Japan seems a poor fit for a “stages of development” model. As I have argued elsewhere, rather than attempt to force such a model on the evidence, instead it seems more appropriate to utilize a model based on a theory of civilization that includes repeated rises and falls.42 This chapter has emphasized the discontinuities between the medieval and early modern periods rather than the connections. That was not intentional, although I must admit that underlying my approach was an awareness of these issues. It reflects in part some of the dramatic changes that we observe in the history of the economy during the transition from medieval to early modern. Of course, research into the history of estates—which is intimately connected to the history of the commercial economy—is also an inseparable part of this transformation. One final point that I would like to emphasize is that, for many years, research on premodern Japanese economic history was conducted within a national framework, as if Japan were isolated from the rest of Asia. But recently, scholars are reappraising Japan and its relationship to the rest of the world, and their findings—as some of the work cited in this chapter suggests—are leading to new insights and the rewriting of history. These changes to our perceptions of the past have broader implications for our understanding of premodern Japanese history that make it no longer sufficient to view estates through a national lens. Instead, just as we now recognize the links between medieval Japan’s economy and the wider world, so too we must try to understand estates from a world history perspective. Japan’s Economy and the Estate System 55

Notes

The translator gratefully acknowledges Bruce Batten’s assistance in identifying the English equivalents of certain commercial products, and the constructive feedback of So Mizoguchi and Janet Goodwin, who read and commented on drafts of the translation. 1. A literal translation of “ryūtsū keizaishi ” would be “distribution economy” or “circulation economy,” but neither term is commonly used by economic historians publishing in English. Instead, we have adopted the term “commercial economy,” which calls to mind changes in the economy of medieval Europe described by Robert Lopez and others. Lopez 1971. 2. Translator’s note: Until recently, most scholars referred to twelve types of imperial coins issued over a period of 250 years, starting with the Wadō kaichin coin of 708. These coins are commonly referred to in Japanese as the “kōchō jūnisen.” Professor Sakurai refers to thirteen coins because he includes the fuhonsen coin, which probably dates to the 680s. He is not alone in this opinion, though scholars continue to debate the function of fuhonsen. On kōchō jūnisen and fuhonsen, see Segal 2011. 3. Matsumura 1998. 4. For a detailed discussion of this point in English, see Batten 1986. 5. Translator’s note: The term “sato dairi ” was first used in the eleventh century to refer to a temporary residence for the emperor during the time when he could not live in the regular palace (such as following a fire), but by the twelfth century it had become permanent practice for the emperor to reside outside of the original Heian Palace grounds. 6. Northern Song mints produced between one and two billion coins per year during the first half of the eleventh century and production achieved a peak of six billion coins under the policies of Wang An Shi in the year 1073. Gernet 1972. 7. Iinuma 2008. 8. Matsunobu 1989. 9. Sasaki 1972. 10. Ōta 1994. 11. Amino 1980a, 1998. 12. Katsuyama 1995. 13. Nagahara 2004a. 14. Amino 1980b, 1985. 15. Ebisawa 2013. Translator’s note: The myō was a unit of taxation and also a unit of land registered in the name of a cultivator family. It was a subunit or 56 The Big Picture parcel on many estates, that is, a given estate might contain several myō within its boundaries. Segal 2012. 16. KI 18:36–37 (# 13316; 1278/12/8). 17. KI 18:70–73 (# 13402; 1279/1). Translator’s note: Records from the Distant Past is a source that is no longer extant; we only know of it from this reference in Tōdaiji documents. As for hiki and ryō, they were units of measurement comparable to “bolts” in English. 18. Sasaki 1972. 19. Ōyama Kyōhei, “Chūsei sonraku ni okeru kangai to senka no ryūtsū.” In Ōyama 1978. (Originally published 1961, Hyōgo shigaku 27.) 20. Sakurai and Nakanishi 2002. 21. For more on this source, see the chapter by Michelle Damian in this volume. 22. Translator’s note: The same character can also be used to refer to wheat. 23. Seersucker kelp, Casteria costata. 24. This is because rents were supposed to be exempt from transport taxes such as sekisen. 25. For more on saifu in English, see Segal 2011. 26. Sakurai 1996. 27. Translator’s note: The kanmon was a large unit of currency, consisting of one thousand coins grouped together by running a string through the holes in their centers. 28. For an English-language study of the Ōyamazaki oil merchants, see Gay 2009. 29. The “four trees and three grasses”: tea, paper mulberry, lacquer, mulberry, linen, indigo, and safflower. Note that paper mulberry is a common name for Broussonetia kazinoki, whereas mulberry is a common name for Morus bombycis. 30. Imatani 1981. 31. Letter-Writing Samples for Study at Home dates to the Nanbokuchō period (1336–1392). It contains samples of correspondence that not only serve as examples of letter writing but also as a kind of dictionary of items used in daily life. A printed copy is available in Zoku gunsho ruijū 13.2, 1128–1143. An English translation from the USC Kanbun Workshop, 2017, is forthcoming: http://www.uscppjs.org/kw. “Various Fine Products for Sale in Kyoto” dates to the late fourteenth or early fifteenth century and includes a list of prices in Kyoto. Although it is not available in printed form, there is an almost identical printed version, entitled Shogeikata daimotsu zuke, in Zoku gunsho ruijū 33.1: 421–424. 32. Sakurai 2004, 2007. Japan’s Economy and the Estate System 57

33. Takahashi 2010. 34. Hōgetsu 1963. 35. Translator’s note: The Kujō Indigo Guild is mentioned in a Muro- machi shogunate document found in the Tōji hyakugō monjo, dated Eishō 7 (1510)/6/3. Taken together, these two documents suggest that there was active production of indigo in early sixteenth-century Kyoto around Hachijō (Eighth Avenue) and Kujō (Ninth Avenue). 36. Katsumata 1996. 37. Kimura 1982. 38. Translator’s note: Sengoku daimyō were warlords who assembled regional domains that included the territory of estates. In some cases, they abolished estates in their holdings altogether. On sengoku daimyō, see Hall 1961. On the transition from coin-based assessments (kandaka) to rice-based (kokudaka) assessments, see Yamamura 1988. 39. Kuroda 1994, 2003. 40. Mōri 1974. 41. Asao 1961; Wakita 1963; Mori 1977; Kitō 2002. 42. Sakurai 2013. pa rt i i

How Do We Know about Estates?

c h a p t e r 3

Ōbe Estate in the Archaeological Record

Nishida Takeshi Translated by Michelle Damian

Ōbe Estate, a possession of the Nara temple Tōdaiji, was established as an estate in 1147, the late Heian period.1 Its territory was not originally a Tōdaiji holding, but was established by exchanging the three Tōdaiji estates of Akō, Tarumi, and Ao for land administered by the province. Why was Tōdaiji interested in the Ōbe lands and why did the temple deliberately act to make them into an estate? When was the estate developed, and when did redevelopment begin after an apparent hiatus? To understand these two issues, I will examine the results of archaeological excavations carried out in conjunction with site distribution and development to clarify the state of Ōbe lands prior to their establishment as an estate. These excavations reveal that before Tōdaiji assumed control, the western portion of the territory contained a number of agricultural settlements with irrigation facilities, but the eastern portion was largely unused except for burial grounds and temples. Development of the eastern portion began only in 1192, when the court dispatched the monk Chōgen (1121–1206) to oversee the estate. (For sculptural portraits of Chōgen, see plates 1 and 17.) Broadly speaking, the establishment of estates involved both the opening of uncultivated land and the incorporation of existing villages, cultivated fields, religious facilities, and gravesites, and both processes are important to understanding the estate system as a whole. Estates did not completely reconfigure the landscape, but instead depended upon features that were already in place, and proprietors such as Tōdaiji must have benefited from an existing nucleus of cultivated land.2 Archaeological

61 62 How Do We Know about Estates? research can give us a good idea of what Tōdaiji representatives might have seen when they initially examined the land, and what steps they took to exploit their holding further. This chapter will first examine the condition of the territory that became Ōbe Estate before Chōgen’s arrival and then turn to his efforts to expand the estate and make it bountiful. From 1989 through 1995, the Board of Education of Ono City, Hyōgo Prefecture, took the lead on a project titled “Survey of the Detailed Dis- tribution of Ōbe Estate Archaeological Sites.” Upon receiving funding from the national government, an Ōbe Estate Survey Committee was organized. Headed by Ōyama Kyōhei and with the cooperation of the Japanese History Research section of the Faculty of Letters, Kyoto Uni- versity, the committee recorded the current state of all the lands that had once been part of Ōbe Estate. The objectives of the survey spanned a wide range, including the state of arable land, irrigation, place names, folk practices, documentation, ruins, and stone structures. Oral histories and physical surveys were conducted in each subdivision of the current city. The survey did not end at the borders of Ōbe Estate, but spread out into the area beyond it. The results of the survey were published in the “Report on the Survey of the Current State of Ōbe Estate, Harima Province,” volumes 1–7,3 and they are a valuable resource in the study of this important medieval landholding.

The Location of Ōbe Estate and the Surrounding Topography Ōbe Estate lay within the borders of present Ono City in Hyōgo Pre- fecture, part of the Kinki region in the central part of Japan’s Hon- shu island. Ono City itself is in the southeast part of Hyōgo Prefecture, in the northeast of the Harima region, an area called Kita-Harima or Hokuban. According to the Japanese national geospatial specifics, its coordinates are latitude 34° 57ʹ, longitude 134° 57ʹ. One of the largest rivers in Hyōgo Prefecture, the Kako River, runs from north to south through the center of the city, and as a result of its rich blessings a vibrant culture developed here in ancient times. The Kako River has carved out a vast river terrace on both banks, and the city of Ono was built on those river terraces. The land that was once Ōbe Estate occupies the area between the central part of present Ono City and its northern borders. The river terraces are situated on the left (eastern) bank of the Kako River. Broadly speaking, they form three levels: a lower terrace thirty meters above sea Ōbe Estate in the Archaeological Record 63 level, a fifty-meter middle terrace, and a seventy- to eighty-meter upper terrace. Looking eastward from the Kako River embankment, it is possible to see the shape of the land as it gradually rises level by level from the alluvial terrain, through the lower, middle, and upper terraces, and finally to a hill of about one hundred meters above sea level (see plate 2).

A Summary of the Sites on Each Terrace Level Lower Terrace

The lower terrace, the one closest to the Kako River, has supported settlements since prehistoric times. Ruins have been confirmed from settlements as far back as the Yayoi era nearly two thousand years ago, and the area includes traces of occupation from the Tumulus, Nara, Heian, and Kamakura periods. In fact, it has been continuously occupied until the present time. Among the local place names for current settlements on the lower terrace such as Ōji, Takada, and Shikiji, many correspond to those recorded in medieval land surveys and other documents in the archives of Jōdoji, a temple established on the grounds of the estate, and in those of the proprietor Tōdaiji (see plates 3 and 4). Along the border of the lower and middle terraces there are natural channels through which water flowed from the middle terrace. Vestiges of those channels can be confirmed in the lower terrace area. Although the original channels are no longer used today, their traces can be seen along the borders of modern hamlets. Today, holding ponds have been created in the valleys, and paddy fields were developed in the areas near the source of the streams, obscuring the original water channels, but their traces can be confirmed through excavation surveys.

Middle Terrace

The middle terrace stretches two kilometers from east to west and three kilometers from north to south, forming a wide, flat plain. Today it encompasses the villages of Kurokawa, Nakashima, and Kōdo, which are surrounded by paddy fields. The oldest ruins verified on this terrace are from the mid-Tumulus era, such as the mid-fifth-century Ōzuka mounded tomb (see figure 3.1) and the Shikiji Ōtsuka mounded tomb, large tombs used to bury powerful people of the time. By the end of the Tumulus era, a number of tombs had been constructed, mainly at the edges of the 64 How Do We Know about Estates? middle terrace. Representative sites include the Ōji grave complex in the vicinity of the Ōzuka mounded tomb (more than twenty-four graves), the Shikiji grave complex in the Shikiji Ōtsuka mounded tomb area (more than thirty graves), as well as the Takada grave complex (eighteen graves), the Minamiyama grave complex (nine graves), and the Kuboki grave complex (eighteen graves). As for Nara-period sites, the Kōdo abandoned temple ruins sit near the edge of the terrace. This site is the remnant of a temple built in the late seventh century that disappeared around the end of the Heian period. (For a model of this temple, see plate 5.) We have no evidence of settlements on the terrace prior to the establishment of Ōbe Estate. Except for the mounded tombs and temples built at the edges of the terrace, much of the area was unused. Kamakura and Muromachi period sites indicate that after the estate was established, the terrace was opened to cultivation, settlement, and production. At present, the post-Heian sites that currently have been confirmed on this terrace include the Kiyotani ruins, thought to have

Fig. 3.1. Site of Ōzuka tomb, Ōbe Estate. Photo by the author. Ōbe Estate in the Archaeological Record 65 served as an administrative facility from the early Kamakura period and used at least through the fourteenth century (see figure 3.2). Others include the Minamiyama ruins (early-mid-Kamakura period) that hold a kiln site for firing pottery, the Kurokawa Totta ruins that have confirmed waterways and the ridges of former paddies, the Kurokawa Umawatari ruins, and the Kurokawa Oka-no-jō ruins (all from the early- mid-Kamakura period through the Muromachi period). After the Kamakura period all the sites other than the Kiyotani site were production sites, and all postdate the establishment of Ōbe Estate.

Upper Terrace The upper terrace is narrower than the lower or middle terraces, and above it lies a hilly area. From those hills down to the upper terrace one can see several naturally formed channels, and a holding pond that was created at each channel. The construction dates for the ponds are unclear, but even today they are an important source of water for the paddy fields that reach down to the middle terrace.

Fig. 3.2. Kiyotani archaeological site, Ōbe Estate, early medieval period. Photo courtesy of Ono-shi Kyōiku Iinkai. Fig. 3.3. Jōdoji subtemples. Drawn by the author. Ōbe Estate in the Archaeological Record 67

Sites on this terrace include Jōdoji Temple, built at the beginning of the Kamakura era and a central point on Ōbe Estate, with its environs including the ruins of its monks’ cottages, and the Tōriike site, with ruins of a kiln used for firing pottery (see figures 3.3 and 3.4; plate 6.) In addition there is the Kitsunezuka grave complex (two tomb graves). Like most sites on the middle terrace, all sites are dated after the establishment of Ōbe Estate, mostly from the Kamakura era and later. One earlier site, the Takazukayama circular mounded tomb at the southeastern edge of Ōbe Estate territory, was probably used as a border marking when boundaries for the estate were set. The tomb, on the southern edge of the hills that border the upper terrace, is approximately twenty meters wide and more than two meters high; it is the sole grave extant at the height of 125 meters above sea level. As it has not yet been surveyed, its details are unclear, but its style indicates that it was probably constructed around the early part of the middle Tumulus period (early to middle fifth century). It is thought to be the “enormous tomb” cited as the marker of Ōbe Estate’s eastern border in the Council of State (Dajōkan) order establishing the estate.4

Fig. 3.4. Site of a Jōdoji building. Photo courtesy of Ono-shi Kyōiku Iinkai. 68 How Do We Know about Estates?

The Lands of Ōbe Estate

According to a copy of that order dated Kenkyū 3 (1192)/8/25, the boundaries of Ōbe Estate were: “on the east, an enormous tomb; on the west, the Kako River; on the south, Kawauchi Village; on the north, Nanjō.”5 Today, with the exception of the Kako River, no currently used place names evoke those boundaries. In practice, however, if we consider the boundaries of the Tōjō River to the north in the vicinity of Shigei-chō and the middle and lower terraces to the south, as well as the above-noted Takazukayama mounded tomb to the east, the territory of Ōbe Estate occupied an area of about five kilometers from east to west, and about four kilometers from north to south.6 The Ono City Board of Education survey of Ōbe-related archaeological sites examined not only the districts noted above that are considered part of Ōbe Estate, but also the surrounding areas.7 The resulting information on such features as irrigation structures and residence locations of shrine affiliates supplies important evidence for conclusions about Ōbe Estate territory.

Major Sites Within the Territory of Ōbe Estate The following sites date in part from the time before the estate was established.

Takada Jizō-no-moto Ruins

These ruins of a settlement lie in the southwest area of the modern Takada hamlet, at the edge of the lower terrace. An archaeological survey of the ruins was conducted between 1989 and 1991. Ruins of Yayoi- and Tumulus-era pit dwellings as well as pillared buildings8 from the Nara and Heian periods were discovered. This site also contains a number of large holes (dokō, “earthen graves”) that are thought to be graves. They are mainly from the Yayoi and Kamakura periods. In the Yayoi period, it seems that the grave area was demarcated by a trench, over which a simple bridge was laid. Within the area bounded by the trench, two groups of graves containing wooden coffins as well as burial pots with bones have been confirmed within an area called the “square, ditched grave” (hōkei shūkōbo). The Kamakura era sites were dug as Ōbe Estate in the Archaeological Record 69 large, elliptical holes and the remains placed inside. Sueki ware teacups and other grave goods have been recovered from these sites.

Ōji Shiro-no-shita Site This settlement site is situated in the southwest near the Oinokawa natural spring on the lower terrace. An excavation survey was undertaken in 1995 and 1996 in conjunction with the construction of a communal residence, and sixteen Yayoi-era pit dwellings, sixteen Tumulus-era pit dwellings, and two Kamakura-era pillared buildings were verified (see figures 3.5 and 3.6). Of particular note is that from the Yayoi through the Tumulus eras, the pit dwellings were rebuilt numerous times within a narrow confine. This was likely due to the need for continuous on-site management of the Oinokawa water source, important even today, that resulted in uninterrupted habitation during that time. Within the verified pit dwellings were remains of fallen roofing materials that had car- bonized due to fire. An iron blade that had been attached to a wooden shovel or hoe was also unearthed. It is the only example of its kind in

Fig. 3.5. Remains of pit dwellings, Ōji Shiro-no-shita. Photo courtesy of Ono-shi Kyōiku Iinkai. 70 How Do We Know about Estates?

Fig. 3.6. Yayoi period pottery, Ōji Shiro-no-shita. Photo courtesy of Ono-shi Kyōiku Iinkai.

Hyōgo Prefecture, and it is noteworthy that someone there owned such a valuable iron tool.

Ōji Tsuji-no-uchi Site This settlement site is on the lower terrace directly below the Ōzuka mounded tomb. Survey excavations were conducted in 1994, 1999, and from 2003 to 2006. These investigations revealed nine pit dwellings and three pillared buildings from the Yayoi period, thirty-seven pit dwellings and two pillared buildings from the Tumulus era, fifty- five pillared buildings from the Nara period, and seventeen pillared buildings from the Kamakura period. This is a large-scale settlement that continued from the Yayoi to the Kamakura periods, and its ruins can be found even within the borders of Ono City. We may surmise that this was a focal settlement on the lower terrace from ancient times. Ōbe Estate in the Archaeological Record 71

Ōzuka Mounded Tomb and Ōji Grave Complex This tomb was constructed in the mid-Tumulus era (mid-fifth century) and is situated on the edge of the middle terrace (see figure 3.1 above). It is a large circular tomb with a diameter of forty-five meters and a height of seven meters, surrounded by a ditch eight meters wide and one meter deep. In an excavation survey conducted in 1952, a pit-style stone chamber was discovered. Artifacts unearthed from the chamber include a mirror, two sets of armor, three swords, three daggers, a halberd, more than one hundred iron arrowheads, and other numerous iron objects (figure 3.7).

Fig. 3.7. Tumulus period artifacts, Ōzuka tomb. Mirror, armor, weapons. Photo courtesy of Ono-shi Kyōiku Iinkai. 72 How Do We Know about Estates?

These artifacts indicate that the person interred in the tomb was a powerful figure. At present, only this tomb and a smaller circular one remain at this site, but the nearby area called the Ōji grave complex includes one keyhole tumulus grave, eight graves within a circular tumulus of about the same size as the one at Ōzuka, and smaller circular tombs with fourteen graves. Approximately 750 meters north of this grave complex lies the Shi- kiji Ōtsuka mounded tomb, a large circular tomb with a diameter of fifty-four meters and a height of five meters, surrounded by a ditch that ranges in width from eight to fourteen meters. In a soil sample taken in 1924, a clay-covered coffin dating from the first half of the middle Tumulus period was discovered, and five decorative beads and seven mirrors were unearthed. According to the report on this finding, a bed of small stones four centimeters in depth covered a four-by-one-meter area. A foundation of clay sixty centimeters thick was laid on top of this bed, and a wooden coffin was placed on the foundation. It was surmised that the coffin was then covered with additional clay in the form of a roll of kamaboko (steamed fish paste). Part of the wooden coffin, coated with cinnabar, has been discovered.9

Minamiyama Tombs 1 and 2 This grave complex site, made up of eight circular tombs, is on the northwestern edge of the middle terrace. In 2013, a survey excavation was conducted on tombs 1 and 2. The first, surrounded by a trench, has a diameter of fifteen meters and a height of eighty centimeters. The second is eleven meters in diameter and sixty centimeters in height, with a three-meter-wide protrusion at the southern end that gives it the appearance of a scallop shell, leading to its name, the scallop-shaped tomb. It is also surrounded by a trench. In the burial chamber of both tombs, a hole was dug and a wooden coffin placed inside in a direct burial. One tomb, constructed in the early seventh century, contained multiple interments. In this survey excavation, a twenty-thousand-year- old Paleolithic-era knife-shaped stone tool was also discovered, suggesting that people had traversed this area from a very early age.

Kōdo Abandoned Temple Ruins These temple ruins sit on the edge of the middle terrace. The temple was established in the mid-seventh century, probably by powerful people of Ōbe Estate in the Archaeological Record 73 the time to appeal for prosperity and peace in this region. It was rebuilt a number of times but seems to have fallen into ruin in the late Heian period. According to the version of Jōdoji engi held by that temple itself, “The holy man Chōgen moved the main image from this dilapidated temple to the Yakushi Hall at Jōdoji to be its main image.”10 Excavation surveys were conducted at this site from 1973 to 1975 and again from 1993 to 1997. Although the temple structure no longer remains, its location is marked by a building foundation called a kidan made from packed earth. At this site, the main hall (kondō) that housed the main image was in the center, with two three-tiered pagodas situated slightly in front of it on the east and west. A lecture hall was situated behind the main hall, and these buildings were enclosed by a corridor called a kairō (a type of covered passageway). The setup of these buildings is the same as that of the well-known Yakushiji Temple in Nara. The covered passageway, however, extends to the north gate instead of to the lecture hall as at Yakushiji. There are also two foundations connected to the sides of the main hall that appear to be those of a bell tower and a sutra repository, endowing this site with its own unique characteristics. The ruins have been designated a National Historic Site, and it is currently maintained as an archaeological park. The park contains a 1/20 scale ceramic model of the temple that displays at a glance the layout of the entire site (see plate 5).

These sites postdate the founding of Ōbe Estate.

Kiyotani Ruins This site lies in the southeastern part of the middle terrace. In conjunction with the construction of National Highway 175 in 1987, the Hyōgo Prefectural Board of Education conducted an archaeological survey (see figure 3.2 above). Nara-period artifacts were unearthed, but at the time of discovery the earliest building remains dated to the late Heian period. A road stretches from south to north, and the residential area was a planned arrangement along the road. Artifacts discovered at the site also indicate that the road was constructed in the late twelfth or early thirteenth century, precisely the time that Chōgen arrived to develop Ōbe Estate. Within the residential area demarcated by a trench alongside the road are four buildings from the mid-thirteenth century and two buildings from the fourteenth century, again concurrent with 74 How Do We Know about Estates? the development of Ōbe Estate. No other settlement remains have been discovered that predate or postdate that time. The Kiyotani site is about one kilometer southwest of Jōdoji Tem- ple, and from the apparently planned layout of the road and the buildings, it seems that this area contained facilities that formed the locus of development for the middle terrace. When that development was partly completed, the hamlets that are today Nakashima, Kuro- kawa, Kiyotani, and Kōdo grew up in its place. As no Kamakura-period settlement remains other than the Kiyotani ruins have been found on the middle terrace, it seems likely that the modern hamlets have been built atop earlier ones. This site has also yielded rounded and flat eave-end roof tiles with the same crests as those used at Jōdoji. The aforementioned Kurokawa, Umawa- tari, and Kurokawa Totta sites contain vestiges of paddy and dry fields.

Jōdoji Monks’ Cottage Ruins At Jōdoji today, two monks’ cottages known as Hōjiin and Kangiin remain, but once there were twelve additional semiautonomous subtemples (figure 3.3).11 According to a drawing preserved in the Kangiin that is based on an Edo-period sketch, in addition to the two monks’ cottages, there were seven more buildings including Jizōin, Chimyōin, An’yōin, Hōshuin, Daishōin, Shin’ōin, and Chūshōin (plate 6). Further- more, several priests’ residences and interior buildings were drawn on the map. Another survey excavation was conducted between 1985 and 1987, when improvements were made in arable land in the area around Jōdoji. Upon excavating near the rice paddies to the northwest of the Hōjiin cottage, foundation stones for building pillars laid in a square were discovered at the place where the drawing shows the location of Jizōin. Since it was part of a building that used foundation stones, it is thought to have been one of the monks’ cottages. Because the excavation survey confirmed the remains of the temple buildings, the plans for creating arable land were altered. Instead of leveling the area and constructing low-lying paddies, additional earth was piled up on the site and paddy land was created on top of it. This allowed the site to be preserved while still providing space for additional arable land. Therefore major excavations were not conducted, and the subtemple remains, including those that have already been confirmed, are preserved under the earth. Ōbe Estate in the Archaeological Record 75

Land Development on the Middle Terrace Excavation surveys on the middle terrace show that except for the mounded tombs from the Tumulus period and temple remains from the Nara period constructed at the edges, most of the area remained undeveloped. That area, now called Haradano, is thought to correspond to the area known as Kanohara in Tōdaiji documents. Tōdaiji authorities likely felt that the wide, as-yet undeveloped lands of Kanohara had considerable potential for agriculture, and they exchanged some of the temple’s previously held estates to obtain the lands that became Ōbe Estate. Since except for the Kiyotani ruins no notable sites have been confirmed, however, it seems that after the establishment of the estate the development of those lands did not proceed smoothly. Only natural springs provided water, and it was not possible to secure enough to cultivate the entire terrace. Furthermore, in order to make this terrace into arable paddy fields, it was necessary to dam the channels that drew water into the valleys and to build large holding ponds to collect the water. At present, the South Pond on the southern side of the terrace and the North Pond to Jōdoji’s west are still important water sources. At the South Pond, the watershed area called “Gō” has long been useful for the three districts of Kiyotani, Kurokawa, and Nakajima, sending water toward the rice paddies in the southern part of the terrace. Technological prowess in civil engineering was necessary to meet the urgent demand for rapid, large-scale development of arable land on this terrace. Chōgen had this expertise; he had constructed ponds, bridges, and roads throughout the realm and had assembled technical experts for his work rebuilding Tōdaiji after it was burned to the ground in 1180 at the outset of the Genpei War.12 Upon arriving at Ōbe Estate, Chōgen erected boundary markers on the four sides of the estate and established Jōdoji as its focal point. It is thought that he also created the North and South Ponds, established irrigation systems within the middle terrace, and accelerated the development of arable land. There are many local traditions surrounding Chōgen’s activities. The “One-Night Folktale” holds that he opened the Kanohara lands to cultivation in a single night, and the tale known as “One-Strike Spring” claims that when he dug in the ground with his hoe, water bubbled up and became a valuable water source. Another legend relates that the 76 How Do We Know about Estates?

Muddy (Nigori) Pond (see plate 7) was formed by the mud dripping from his hoe when he washed it after opening the Kanohara lands in a single night. The water is said to clear only on the fourth day of the sixth month of the lunar calendar, the anniversary of Chōgen’s death. Finally, the Saddle-Hanging Pine commemorates the place where, to rest, Chōgen leaned on a pine tree to place a saddle on an ox that had labored doing land reclamation. Until recent years a pine tree and a small mound marked that spot. Due to a road-widening project, however, the tree was moved to the western side of the temple, and it has been replanted three or four times since. Excavation surveys on the middle terrace have confirmed the vestiges of waterways and ridges of rice paddies that suggest multiple reconstructions and reuse, some of them perhaps initiated by Chōgen. Finally, Chōgen was not only deeply involved with Jōdoji, but also with the construction of Nagaodera (the Unkōji ruins) and Raigōji outside the Ōbe Estate grounds. Nagaodera in particular is mentioned in Chōgen’s autobiography, The Good Deeds of Namuamidabutsu (Namuamidabutsu sazenshū).13 In a survey from 1994 to 1996, many Kamakura-era roof tiles were excavated from the site. Local residents may have diverted irrigation water to Ōbe Estate from the Manshōji River in the estate’s southern portion. Even today there is a waterway called the Nagao Canal that extends from a dam on the Manshōji River through the Tenjin and Kurokawa districts to a large pond in the Ōji area. Raigōji is in the present Ichiba-chō district of the southern part of Ono City. This district has a mooring point that is important for shipping by boat, and in the Edo period rice from the local area was loaded here. Chōgen, who was involved in regional land development, probably established this temple as a focal point for shipping.

Today, the sites that have been confirmed on the middle terrace are less settlement sites than administrative facilities, kilns for firing pottery, and the remains of paddy fields. The beginning of this buildup was the early Kamakura period, coinciding with the era of land development by Chōgen on Ōbe Estate. In other words, the present configuration of this terrace was shaped at the beginning of the Kamakura era. This evidence suggests that the establishment of the middle terrace settlements of Kurokawa, Kiyotani, Nakajima, and Kōdo ushered in a new era, after the earlier development of arable land on the terrace. Though development of both the lower and middle terraces Ōbe Estate in the Archaeological Record 77 occurred concurrently, from the Kamakura through Muromachi eras, the settlement patterns were different. On the lower terrace, spread- out settlements gradually converged into the present hamlets, while on the middle terrace the old settlements were in the same places as the hamlets today. One factor supporting the hypothesis of different settlement patterns is that residents of the lower and middle terraces are even now affiliated with two different shrines: lower terrace households with Kumano Shrine at the edge of the middle terrace, and middle terrace households with Hachiman Shrine, within the precincts of Jōdoji. In this chapter I have considered the results of excavation surveys and folklore concerning Ōbe Estate, particularly with regard to the period of development of the middle terrace. In addition, new ponds and dams called “terai ” (temple dams), constructed to develop arable land in alluvial areas and on the higher terraces, were developed on the lower terrace that had supported settlements prior to the establishment of Ōbe Estate. The construction dates for these irrigation facilities have not been determined, but tales recount Chōgen’s involvement. The temple dams especially restricted the main flow of the Kako River, diverting that water to its tributary the Tōjō River and letting the water flow from there. That type of construction requires a great deal of political and economic clout as well as knowledge of advanced civil engineering technology, which points to Chōgen or someone like him as the one who initiated such projects. The development and redevelopment of Ōbe Estate would not have progressed without the abilities of Chōgen. Because Chōgen, called “the most prepared,” was there, even today the lands within the former Ōbe Estate are known for their rich production of rice.

Notes

1. HI 7:2565 (# 3218). 2. For a discussion of land development that expanded upon existing arable land, see the chapter by Kimura Shigemitsu in this volume. 3. Ōbe no shō Chōsa Iinkai, ed., Ono-shi bunkazai chōsa hōkoku, vols. 9, 10, 14, 16, 17, 20, and 22. 4. OS 4:526–527 (# 217). 5. OS 4:526–527 (# 217). 6. On modern maps, former Ōbe Estate lands comprise the districts of Kako, Kita, Takada, Kano, Shikiji, Ōji, Hata, Kumo, Katayama, Shimo-ōbe, Den’en, Kōdo, Nakashima, Kurokawa, Kiyotani, Kitaoka, and Kyū-onochō 78 How Do We Know about Estates?

(Kami-shinmachi, Kami-honmachi, Honmachi, Nishi-honmachi, Honmachi- icchōme, and Higashi-honmachi). 7. Surveys extended through the districts of Kuboki, Sumiyoshi, Tenjin, Naka, Shigei, Shinmei, Ōshima, and Nagao. 8. Translator’s note: The term used in the Japanese here is hottate bashira tate- mono, indicating that holes were dug in the earth and pillars inserted into them to construct a building. Evidence on most archaeological sites would be postholes, indicating the locations of the original pillar foundations. Though I use the term “pillared buildings” throughout the rest of this chapter, it should be understood that unless foundation stones for pillars are specified on a site that the construction style was that of postholes dug for pillar insertion. 9. Kōbe Shinbunsha 1959. 10. There are two extant copies of Jōdoji engi, which differ in some details. The one held by the temple is a copy dated 1687. Another copy, in the archives of the Kōbe University Library, is dated 1614. For a view of the 1614 manuscript, see http://www.lib.kobe-u.ac.jp/kichosyo/jyodoji/ (accessed September 16, 2016). For a printed version of that copy, see OS 4:716–19, # 446. 11. This number does not include the extant buildings of Jōdoji on the upper terrace. 12. See the chapters by Ōyama Kyōhei, Nagamura Makoto, Yoshiko Kai- numa, and Janet Goodwin in this volume. 13. In his autobiography, Chōgen noted that he placed an Amida triad in Nagaodera. See OS 4:537–545 (# 227). In Ono City’s Nagao district, a large number of roof tiles dating to the Kamakura era were excavated at the Unkōji temple site. Though it has not been confirmed, it is thought that the Unkōji site is actually the remains of Nagaodera.

Archaeological Reports For primary sources and other secondary sources, see the combined bibliography at the end of this volume. The page numbers listed below indicate sections that concern Ōbe Estate.

Hyōgo-ken Maizō Bunkazai Chōsa Jimusho, ed. 1993. Hyogo-ken bunkazai chosa hokoku 121, Kiyotani iseki Minamiyama kofungun Tamatsu Tanaka iseki Minami Ōyama chiten (Report on the survey of cultural assets of Hyōgo Prefecture, no. 121: Kiyotani ruins, Minamiyama Tumulus complex, Tamatsu Tanaka ruins, and the Minami Ōyama site), 133–135. Hyōgo-ken: Hyōgo-ken Kyōiku Iinkai. Hyōgo-ken Maizō Bunkazai Chōsa Jimusho, ed. 1995. Hyogo-ken bunkazai chosa hokoku 146, Takada Koyama-no-shita iseki hakkutsu chosa hokokusho (Report on Ōbe Estate in the Archaeological Record 79

the excavation survey of Takada Koyama-no-shita ruins), 35–38. Hyōgo-ken: Hyōgo-ken Kyōiku Iinkai. Hyōgo-ken Ono-shi Kyōiku Iinkai, ed. and pub. 1999. Tsutaetai Ono kara no messeji 4 Shinpojiumu genzai ni ikiru chūsei no keikan (A message from Ono, vol. 4: Symposium: The medieval landscape that lives on today), 45–52. Hyōgo-ken Ono-shi Kyōiku Iinkai, ed. and pub. 2000. Tsutaetai Ono kara no messeji 5 Shinpojiumu gyōseiman ga kataru kogun jidai (Symposium: The Tumu- lus period revelations about local administrators), 19–22. Hyōgo-ken Ono-shi Kyōiku Iinkai and Kōdoji Haiji ato hakkutsu chōsadan, eds. 1977. Harima Kōdoji haiji ato hakkutsu chōsa hōkokusho (Report on the excavation survey of the ruins of the abandoned temple Kōdoji in Harima), 43–52. Hyōgo-ken. Ōbe no shō Chōsa Iinkai, ed. 1991. Ono-shi bunkazai chōsa hōkoku, vol. 9 (Report on the survey of cultural assets of Ono City, vol. 9), Harima no kuni Ōbe no shō genkyō chōsa hōkokusho I (Report on the current state of Ōbe Estate in Harima Province, vol. I), 3–6, 112–121. Hyōgo-ken Ono-shi Kyōiku Iinkai. Ōbe no shō Chōsa Iinkai. 1992. Ono-shi bunkazai chōsa hōkoku, vol. 10, Harima no kuni Ōbe no shō genkyō chōsa hōkokusho II, 3–6, 102–109. Ono-shi Kyōiku Iinkai. Ōbe no shō Chōsa Iinkai, ed. 1993. Ono-shi bunkazai chōsa hōkoku, vol. 14, Harima no kuni Ōbe no shō genkyō chōsa hōkokusho III, 3–5. Ono-shi Kyōiku Iinkai. Ōbe no shō Chōsa Iinkai, ed. 1994. Ono-shi bunkazai chōsa hōkoku, vol. 16, Harima no kuni Ōbe no shō genkyō chōsa hōkokusho IV, 3–6, 91–102. Ono-shi Kyōiku Iinkai. Ōbe no shō Chōsa Iinkai, ed. 1995. Ono-shi bunkazai chōsa hōkoku, vol. 17, Harima no kuni Ōbe no shō genkyō chōsa hōkokusho V, 3–9, 81–98. Ono-shi Kyōiku Iinkai. Ōbe no shō Chōsa Iinkai, ed. 1996. Ono-shi bunkazai chōsa hōkoku, vol. 20, Harima no kuni Ōbe no shō genkyō chōsa hōkokusho VI, 3–7, 135–150. Ono-shi Kyōiku Iinkai. Ōbe no shō Chōsa Iinkai, ed. 1998. Ono-shi bunkazai chōsa hōkoku, vol. 22, Harima no kuni Ōbe no shō genkyō chōsa hōkokusho VII, 3–7, 35–53. Ono-shi Kyōiku Iinkai. Ono-shi Kyōiku Iinkai Bunkazai-ka. 1992. Ono-shi bunkazai chōsa hōkokusho, vol. 12, Ono-shi iseki bunpuzu (Map of Ono City archaeological sites), maps 15–34. Ono-shi Kyōiku Iinkai. Ono-shi Kyōiku Iinkai Bunkazai-ka. 1995. Ono-shi bunkazai chōsa hōkokusho, vol. 19, Oji Tsuji-no-uchi iseki hakkutsu chosa hokokusho (Report on the excavation survey of Ōji Tsuji-no-uchi ruins), 28–30. Ono-shi Kyōiku Iinkai. 80 How Do We Know about Estates?

Ono-shi Kyōiku Iinkai Bunkazai-ka. 1996. Ono-shi bunkazai chōsa hōkokusho, vol. 21, Kendō Sumiyoshi Suminaga sen kairyō jigyō ni tomonau maizō bunkazai hakkutsu chōsa hokokusho (Report on the excavation survey of buried cultural assets in conjunction with the project for improvements on the Sumiyoshi Suminaga prefectural highway), 33. Hyōgo-ken Ono-shi Kyōiku Iinkai. Ono-shi Kyōiku Iinkai Bunkazai-ka. 2001. Ono-shi bunkazai chōsa hōkokusho, vol. 24, Oji Tsuji-no-uchi iseki hakkutsu chosa hokokusho 2 (Report on the excavation survey of Ōji Tsuji-no-uchi Ruins 2), 15. Hyōgo-ken Ono-shi Kyōiku Iinkai. Ono-shi Kyōiku Iinkai Bunkazai-ka. 2005. Ono-shi bunkazai chōsa hōkokusho, vol. 26, Kuni shiseki Kōdo haiji ato hakkutsu chosa hokokusho (Report on the excavation survey of the National Historic Site of the abandoned temple Kōdoji), 98, 99. Hyōgo-ken Ono-shi Kyōiku Iinkai. Ono-shi Kyōiku Iinkai Bunkazai-ka. 2005. Ono-shi bunkazai chōsa hōkokusho, vol. 28, Unkōji (Nagaodera) iseki hakkutsu chosa hokokusho (Report on the excavation survey of Unkōji (Nagaodera) Temple), 71, 72. Hyōgo-ken Ono-shi Kyōiku Iinkai. Ono-shi Kyōiku Iinkai Bunkazai-ka. 2013. Ono-shi bunkazai chōsa hōkokusho, vol. 31, Oji Yamanoshita iseki hakkutsu chosa hokokusho 2 (Report on the excavation survey of Ōji Yamanoshita ruins 2), 40. Hyōgo-ken Ono-shi Kyōiku Iinkai. Ono-shi Kyōiku Iinkai Bunkazai-ka. 2014. Ono-shi bunkazai chōsa hōkokusho, vol. 5, Minamiyama iseki kakunin chōsa hōkokusho (Report on the survey of Ono City cultural assets, vol. 5, Report on the Verification Survey of Minamiyama Ruins), 44, 45. Hyōgo-ken Ono-shi Kyōiku Iinkai. Ono-shi Kyōiku Iinkai Bunkazai-ka. 2014. Ono-shi bunkazai chōsa hōkokusho, vol. 32, Minamiyama 1gō fun, 2gō fun hakkutsu chosa hokokusho 2 (Report on the excavation survey of Minamiyama tombs 1 and 2), 21, 22. Hyōgo-ken Ono-shi Kyōiku Iinkai. c h a p t e r 4

Tōdaiji’s Estates in Its Documentary Record Perspectives on Ōbe Estate

Endō Motoo Translated by Janet R. Goodwin

Ōbe Estate in Harima Province was established in the mid-twelfth century as a holding of Tōdaiji, a great Buddhist temple in Nara with ties to the throne and a history that stretches back to its eighth-century founding by the sovereign Shōmu Tennō (701–756, r. 724–749). Over several centuries, Tōdaiji preserved many records concerning Ōbe Estate, and these documents have become the most important materials for understanding the estate’s long and complex history. Of course there are many other pertinent sources, such as documents in other collections and the findings of archaeological and other on-site investigations, but the collection known as Tōdaiji monjo is arguably the most important single location of source materials, and we all depend ultimately on its documents. Such an impression is particularly strong for me, since I have examined Ōbe Estate only through my work compiling Tōdaiji documents. Nevertheless, I think that conducting research using this perspective has considerable significance.

Research on Documents: A Quick Overview Let us begin with an overview of how we as researchers examine documents like those concerning Ōbe Estate and what questions we ask about them, in addition of course to reading them for content.1 Suppose I have a medieval document sitting on my desk. To understand that document correctly, I need to examine it from four different perspectives:

81 82 How Do We Know about Estates?

Who possessed or transmitted it; whether or not the physical document was reused in any way; whether it is an original, a copy, or a draft; and the relationship between the document’s sender and its recipient. Let me explore these perspectives in a bit more detail. Which individuals or organizations possessed the document? In this regard documents fall into one of two categories. In the first case, the document has been in the same hands since the medieval period. Usually there are other related medieval documents in the same location. Typically, such documents were kept by aristocratic or military houses or by shrines and temples, and the documents are materials that relate the history of the organization that owns them. In the Edo and modern periods, such organizations often assembled these documents, generally by pasting a number of them together in scroll for- mat (as “bound documents”). This was because there was a danger of losing important documents if they remained separate. If there were too many related documents, however, it was impossible to paste them together; loose documents such as this are called “unbound documents.” The second mode of possession is that of “collected documents.” Such documents were accumulated, often in modern times, by an individual, family, or organization that did not originally possess them, a practice that differs from those in the medieval and early modern eras. We cannot say that such documents relate the history of the entity that collected them. Tōdaiji documents often fell out of the temple’s possession and were collected by others. Many of the examples cited here fall into this category: documents originally held by Tōdaiji but obtained by others in modern times. Although most of the medieval documents extant today were preserved just as they were, in some cases documents were discarded but the paper on which they were written was used for other purposes. Thus such documents were fortuitously preserved until today. In medieval times the paper of discarded documents was used in one of two ways. One, the paper was recycled and dissolved for other purposes, so of course the characters written on it were no longer visible. In other cases the paper was turned over and characters were written on the other side. For instance, totally unrelated letters were often written on the other side of document drafts or notes for Buddhist study. In this way, the paper of discarded documents was reused. Such documents are known as shihai monjo, or reverse-side documents. Documents preserved Tōdaiji’s Estates in Its Documentary Record 83 in this way are a lucky find for scholars and a boon to research on the medieval age. We also classify documents as originals, copies, or drafts. As one might expect, originals are found mostly in the archives of their addressees, while most drafts are in the archives of the senders. Since account records and records of oral negotiations were composed by a representative of the organization that came to store them, there are many originals in these categories. Copies, however, can be found in various places. In the class-based society of medieval Japan, status distinctions were extremely important, and features of a document reflect such differences between a document’s author and its recipient. Documents may be classified into four categories depending on the status relationship between sending and receiving parties. The first is from a superior to an inferior: These are orders of several types sent from a powerful individual, institution, or head of an organization. For example, the retired monarch’s office issued orders known as inzen or in-no-chō kudashibumi, while the Kamakura shogunate promulgated orders called Kamakura-dono kudashibumi, Kantō migyōsho, or Kantō gechijō. Rokuhara, the shogunal office in Kyoto, also issued gechijō and migyōsho. Officials of Buddhist temples such as administrators (bettō) and head monks (inju) of branch temples also sent orders called migyōsho to their subordinates. Another document type in this category is the letter of appointment (buninjō) to those holding posts as estate officials. The language in the introduction of each of these document types makes it easy to distinguish one type from another, but differences in their usage are more subtle. Roughly speaking, a rinji (an inzen is one type) was a formal proclamation of a command, often oral, issued by the throne. A kudashibumi, an edict of permanent duration, was often used for appointment decrees, and a gechijō, also a decree for the long term, was often used for judicial decisions. The migyōsho, on the other hand, was a document type used for communications with short-term effect.2 Second, there are two general types of documents used by subordinates in communicating with their superiors. Petitions are called mōshijō; depositions in lawsuits from plaintiffs (sojō) and defendants (chinjō) are, broadly speaking, of this type. Documents known as ukebumi, on the other hand, were used to acknowledge orders from a superior. 84 How Do We Know about Estates?

In the medieval period, individuals or groups often issued oaths known as kishōmon, addressed to the buddhas or deities. These were similar in form to ukebumi, and formally speaking they were presented to transcendent supernatural “authorities.” In these oaths, the signato- ries pledged to perform certain duties or obligations, or swore the truth of their claims, on pain of divine punishment. Third, there were also communications between parties of equal status, but since there are no such documents mentioned in this chapter, I will not discuss them further. Fourth, many medieval documents were records of some sort. In contrast to the first three types, which were sent from one party to another, records often remained in the hands of those who compiled them. Land registers supplied the fundamental information for assessments of rents from estates and other holdings. One type of register was the list of responsible rent payers (nayosechō), which recorded the name of the payer and the amounts due in rice, for each unit of paddy or dry field. Account ledgers recorded how estate rents were apportioned after they were collected; for example, they would show which parties within a proprietary organization received what percentage of the rents. These are important documents for understanding the way an organization operated. In the medieval period and later, an organization such as Tōdaiji, which possessed many documents, listed these records in catalogs known as document lists or document management ledgers. The document types and features discussed above help scholars to identify and understand medieval historical materials. Please keep these details in mind as I turn now to the history of Ōbe Estate as portrayed in the documentary record.

Tōdaiji’s Internal Divisions and Document Preservation A large organization such as Tōdaiji contained a number of internal units that compiled and stored historical materials (see figure 4.1). Figure 4.2 shows the internal structure of Tōdaiji in medieval times. Particularly important for this study are the temple as a whole or, in contemporary terms, the corporate temple; Tōnan’in, a separately administered cloister within Tōdaiji’s precincts; and the Ordination Platform (Kaidan’in), with the Oil Storehouse (Aburakura) under its authority. The Ordination Platform was a single structure within Tōdaiji that seems to have operated as an independent entity. Tōdaiji’s Estates in Its Documentary Record 85

Fig. 4.1. Medieval Tōdaiji. Drawn by the author.

These internal units functioned in various ways, and in the process they generated a number of written records, including drafts and duplicate copies. In addition, records were obtained from outside organizations and preserved in the various Tōdaiji collections. Documents preserved from medieval times have been used as historical sources, and from the Meiji period on they have been arranged in the order shown in figure 4.3. In other words, documents were prepared at, transmitted from, and preserved at many subunits of medieval Tōdaiji. Some of these documents were assembled in collections, while others were scattered in various locations. Most of them are now held by the Tōdaiji Library, but others are stored in locations outside the temple. Through such processes of preservation and assemblage, Ōbe Estate documents have become available for our research. Fig. 4.2. Tōdaiji organizational chart.

Fig. 4.3. Classification and disposition of Tōdaiji documents over time. Tōdaiji’s Estates in Its Documentary Record 87

As figure 4.3 shows, in some cases materials preserved in medieval times in the archives of separate organizations were later assembled as a single set of historical records. In other cases, materials were kept at temples completely outside of Tōdaiji. The situation is fairly complex. Some documents have lost some of their distinguishing features. Documents that were originally kept by different parties seem—mis- takenly—to belong together, while others that do belong together have been disassembled. This situation exacerbates the great sense of distance between the medieval age and our own times, a gap I think is important to overcome in our research. This is particularly so in the case of Ōbe Estate, the management of which was undertaken by several different internal Tōdaiji units over time and involved many historical changes. As I will explore in detail below, from the early Kamakura to the mid- Muromachi period the chief managerial authority over the estate moved back and forth among Tōnan’in, corporate Tōdaiji, and the Ordina- tion Platform–Oil Storehouse (see table 4.1). Relevant documents were housed in the archives of each managerial unit during its period of control. Nagamura Makoto concisely defined questions regarding such historical materials as questions regarding Tōdaiji’s internal organization, an argument the importance of which I wish to emphasize.3 Such an approach focuses on materials of each internal unit that took charge of affairs, enabling us to understand the history of Ōbe Estate’s management in considerable detail. More broadly speaking, the documentary record also helps us to evaluate the relationship between the corporate temple and Tōdaiji’s subunits. With this in mind, I wish to examine several issues chronologically.

Early to Mid-Kamakura Period: Chōgen Restores the Estate, and Tōnan’in Assumes Management Rights

Ōbe Estate traces its history back to the Heian period, but in the tur- moil of the Genpei War in the 1180s, it seems to have temporarily been abandoned by its proprietor, Tōdaiji. In 1192, after the war had ended, the restoration of the estate was entrusted to Chōgen (1121–1206), the monk in charge of Tōdaiji’s reconstruction after much of the temple was incinerated in the course of the conflict. (For sculptural portraits of Chōgen, see plates 1 and 17.) 88 How Do We Know about Estates?

Table 4.1 Superior management rights to Ōbe Estate, twelfth to fifteenth centuries

Date Authority Remarks 1147 Tōdaiji (corporate) Estate established 1192 Tōdaiji (corporate) Estate revived by Chōgen 1206 Tōnan’in Chōgen’s testament 1295 Tōdaiji (corporate) 1307 Tōnan’in 1321 Tōdaiji (corporate) 1327 Tōnan’in 1334 Tōdaiji (corporate) c.1364 Tōnan’in The exact year when management authority reverted to Tōnan’in is uncertain. Rents were divided into 2/3 and 1/3 portions; Tōdaiji possibly received 2/3 and Tōnan’in 1/3. 1392 Tōdaiji (corporate) & Tōdaiji: 2/3 of rents; Ordination Ordination Platform– Platform: 1/3 Oil Storehouse This arrangement seems to have continued through the fifteenth century.

Two documents from 1192, both recently discovered in the Yale University archives, deal with this matter. One, dated 8/15, is a copy of an order from the absentee office (rusu dokoro) of the Harima provincial government; the other is dated 8/23 and is a communication from the former Tsushima governor Nakahara, the proxy who headed the Harima governor’s office.4 These documents order that Ōbe Estate be confirmed as Tōdaiji’s holding and that markers be set to establish its borders. The two documents are pasted together, with Chōgen’s seal appearing on the juncture. There is no doubt that these documents, written in Chōgen’s hand, were important records that officially legiti- mized Ōbe Estate at the time. In 1197, Chōgen willed proprietor’s rights (ryōke shiki) to the estate to Tōnan’in,5 and while research is still incom- plete, I think it is likely that these documents were transmitted to that cloister. Tōdaiji’s Estates in Its Documentary Record 89

Another document of interest is an order from the office of the retired monarch Go-Toba, dated 1206/4/15,6 when Tōnan’in held authority over Ōbe Estate. Tōdaiji had sought dismissal of Chen He-qing, the casting master of the Great Buddha (Daibutsu) image, as proxy manager (daikan) of Ōbe Estate and several other estates held at this time by Tōnan’in. Go-Toba thereby agreed to Tōdaiji’s demands and issued this document, an order (kudashibumi ) from the retired sovereign’s office addressed to the Tōdaiji monastic cabinet as well as to appropriate provincial officials. There are two problems with this document. First, the plaintiff in the case was Tōdaiji rather than the proprietor Tōnan’in, and second, the document itself was archived by the royal cloister (monzeki 7) Zuishin’in in Kyoto rather than by its addressee Tōdaiji. There seem to have been two reasons that Tōdaiji issued its complaint. It was the supreme landholding authority over Ōbe Estate, even though proprietor’s rights were held by the cloister Tōnan’in. Thus the complaint to Go-Toba had to be presented in the name of Tōdaiji. There were special circumstances involved, moreover, since Chen He- qing—who had already supervised the casting of the Tōdaiji Great Bud- dha under Chōgen’s leadership—was an immigrant from Song China selected by Chōgen as proxy manager. Subtle opposition between Chōgen and Tōdaiji figured in this situation: by removing He-qing from his position, the Tōdaiji monks perhaps hoped to reduce Chōgen’s influence over these estates. The second problem, the archiving of this document by Zuishin’in, can be explained as follows. In medieval Japan, the addressee of a document and its actual recipient were often different parties, and the recipient was the party that was actually in charge of the holding in question, in this case the proprietor Tōnan’in. So the document was probably first received and then archived by Tōnan’in, and then it was later moved to Zuishin’in, perhaps in the Muromachi period, when the two cloisters were headed by the same abbot and Zuishin’in assumed management of Tōnan’in. At this time Tōnan’in documents, including this one, were probably moved to Zuishin’in. In summary, documents from this period attest to Chōgen’s restoration of Ōbe Estate, Tōnan’in’s exercise of proprietor’s rights, and some subsequent management difficulties. The 1206 document shows quite clearly that to understand fully a document’s 90 How Do We Know about Estates?

meaning and importance, we need to take into account features other than content, such as a document’s recipient and where it was eventually stored.

Late Kamakura to the Early Nanbokuchō Period: Transfer of Management Rights between Tōnan’in and the Corporate Temple From late Kamakura times to the early Nanbokuchō (Northern and Southern Courts) period, rights to control and manage the estate were passed back and forth between the corporate temple and Tōnan’in, beginning in 1292 when two disputes arose: one with Jōdoji, an autonomous temple (bessho) on the estate, and another with the estate military steward’s deputy (jitōdai). Table 4.2 classifies selected documents by date and the identity of the proprietor (ryōke or shōen ryōshu) at the time.

Table 4.2 Current locations of Tōnan’in and corporate temple documents Document originals presented to the corporate temple from the estate are marked by { } around their numbers. The documents preserved at medieval Tōnan’in are underlined. The symbol ° indicates document originals preserved at the corporate temple in medieval times.

Storage and Date Document Type Preservation

Period 1 of Tōnan’in’s management (to 1293)

1 1207/1 Copy of letter of Shinpukuji Hōshōin appointment to reeve’s document position, from the Tōnan’in chancellery 2 1233/2/29 Copy of an order from the Unbound document shogunal headquarters at Rokuhara 3 1283/12/10 Copy of a letter of Shinpukuji Hōshōin appointment to the document reeve’s post at Ōbe Estate, issued by the Tōnan’in chancellery Table 4.2 (continued)

Storage and Date Document Type Preservation

Period 1 of corporate Tōdaiji management (to sometime before 1307) °{4} 1295/1 Ōbe Estate cultivators’ Unbound document petition {5} 1295/2 Copy of petition from Tōdaiji Unbound document monastic council °{6} 1298/5/4 Acknowledgment of unpaid Unbound document rents from additional cultivated fields from Ō Hisakiyo, Ōbe Estate reeve 7 1300/12/30 Tōdaiji monks’ oath Mizuka Naoya archives

Period 2 of Tōnan’in management (to sometime before 1321) 8 1307/6 Copy of a letter of Document collection appointment to the reeve’s of Kanō Kōkichi position, from the Tōnan’in chancellery 9 1307/6 Copy of a letter of Shinpukuji Hōshōin appointment to the reeve’s document position, from the Tōnan’in chancellery 10 1320/5/23 Copy of an order from the Document collection Tōnan’in chancellery of Kanō Kōkichi

Period 2 of corporate Tōdaiji management (to 1326) °{11} 1322/9 Plaintiff’s deposition in lawsuit Unbound document filed by the nun Kakushō, Ōbe Estate reeve °{12} 1323/2 Petition from Yūya, daughter Unbound document of deceased Ōbe Estate reeve Kan’en 13 1324/7/16 Oath signed by members of the Private archives of Ōbe Estate administrative office Takeuchi Bunpei °{14} 1326/10 2nd petition from the Ōbe Unbound document cultivator Heinai Table 4.2 (continued)

Storage and Date Document Type Preservation

Period 3 of Tōnan’in management (to 1334) 15 1327/11/20 Copy of an order from the Tōdaiji documents, Tōnan’in director (bettō) in the University of Tokyo Faculty of Literature archives 16 1331/12/5 Copy of a letter of Shinpukuji Hōshōin appointment to the reeve’s document position by the Tōnan’in chancellery 17 1331/12/13 Copy of an order from the Unbound document Ōbe Estate custodian

Period 3 of corporate Tōdaiji management (with the reeve’s position possibly under Tōnan’in jurisdiction) (to 1346) °{18} 1335/7/5 Petition from the holder Unbound document of Yoshitomi name field on Ōbe Estate °{19} 1335/10/29 Ship captain Yakushi Unbound document Taifu’s document of reply concerning rent-payment rice from Ōbe Estate 20 Around 1337 Report on final accounts Unbound document of Ōbe Estate rents °{21} 1340/6/20 Petition from the monk Unbound document Gyōken, a centrally appointed administrator of Ōbe Estate 22 1340/7/4 Copy of request for battle Shinpukuji Hōshōin rewards from Kose Mijirō, document deputy of the nun Seia, reeve of Ōbe Estate °{23} 1346/6 Petition from Seia, reeve Unbound document of Ōbe Estate Tōdaiji’s Estates in Its Documentary Record 93

Before the corporate temple assumed management rights from Tōnan’in, the head monk of that cloister made an appointment to the reeve’s (kumon) position, according to an extant copy dated 1283/12/10.8 This document has been preserved on the reverse side of some religious texts in the archives of the Hōshōin cloister at Shinpukuji, in present Nagoya city. Shinpukuji was not a branch temple of either Tōdaiji or Tōnan’in, but in the late fourteenth century, its monks often discussed Buddhist doctrine with the head monk of Tōnan’in. Religious texts were inscribed on the backs of documents that were no longer needed and sent back and forth between the two temples. Documents fortuitously preserved in this way are, of course, reverse-side documents such as those discussed earlier in this chapter. In other words, medieval Tōnan’in documents were preserved at Shinpukuji Hōshōin, just as at Zuishin’in as mentioned above. Since these documents were preserved as the result of temporary contacts between Tōnan’in and Shinpukuji, there must have once been others that were discarded or destroyed. The few we have now are lucky finds. The troubled times of 1292 are detailed in a number of documents, including an acknowledgment of orders (ukebumi), dated 1292/6/13 and signed by the two reeves of the Kanohara (eastern) sector of Ōbe Estate.9 In this document, the reeves acknowledge a communiqué from the corporate temple in connection with a dispute involving Jōdoji. Tōdaiji instructs the reeves to follow orders sent earlier by the estate’s proprietor, Tōnan’in. On this occasion it seems that the local managers of the estate could not rely on Tōnan’in and had to summon support from the corporate temple. In another acknowledgment of orders that is dated 1292/intercalary 6/16, the reeves Ō Hisamitsu and Enkai, after reporting earlier that the offices on the estate of the military steward’s deputy and the custodian had been attacked and burned, swore to follow the instructions of Tōdaiji, not Tōnan’in.10 Meanwhile, the steward’s deputy filed suit with the shogunate, accusing Tōdaiji’s agent (zasshō) of illegal activities.11 As a result of these difficulties, Tōnan’in relinquished its own proprietor’s right. This situation is further reflected in a communiqué from the head monk of Tōnan’in, dated 12/29, probably in the same year.12 Then in 1293, corporate Tōdaiji filed a lawsuit with the Kamakura warrior government complaining of violations by the estate’s military steward.13 This indicates, I think, that management rights had been transferred to the corporate temple, allowing its authorities to file the suit. 94 How Do We Know about Estates?

One unforgettable affair in the history of Ōbe Estate in this period is the dispute within the reeve’s family presented to Tōdaiji for legal judgment. In this long-running feud, lasting from late Kamakura times to the Nanbokuchō period, a woman and her daughter-in-law—and subsequently their heirs—both claimed the reeve’s position, which had been passed down through the Ō family. The dispute began in 1320, when control over the estate changed hands from Tōnan’in to corporate Tōdaiji. Subsequently control shifted back and forth between the two parties, as did the location where legal proceedings took place. Document number 15 in table 4.2 verifies that Tōnan’in had agreed to hear the case in 1327.14 But it is interesting that the only extant originals of statements by the plaintiff and the opposite side date from the time when the corporate temple was in charge (see table 4.2 numbers 11, 12, and 23,15 which are included among the unbound documents in the Tōdaiji Library archives). It is clear that many unbound documents were stored by the corporate temple, and thus the originals, which had been presented to Tōdaiji, were preserved just as they were until modern times. Those marked in table 4.2 by a curly bracket around their numbers were originals presented to the corporate temple from the estate itself. For the most part, extant originals are those that were kept in the corporate temple archives in the medieval age. On the other hand, hardly any originals have been preserved from the periods of Tōnan’in management. In fact, few of the documents once held by Tōnan’in have been preserved until the present day. Those that remain are found either on the reverse sides of sacred texts that once belonged to Tōnan’in, or were obtained as copies by the corporate temple. In short, the first peak in the quantity of research materials on Ōbe Estate comes from the late Kamakura period, when the corporate temple assumed management rights to the estate. It is necessary to keep in mind the limitations and biases in research materials based on this situation: in comparison to records from corporate Tōdaiji, there remains only a small portion of those from medieval Tōnan’in.

End of the Nanbokuchō Period to the Mid-Muromachi Period: The Corporate Temple and the Ordination Platform From the end of the Nanbokuchō era in 1392 to mid-Muromachi times, the corporate temple and the Ordination Platform shared management authority over Ōbe Estate. Tōdaiji’s Estates in Its Documentary Record 95

Harima Province was beset by military conflict from the onset of civil war in 1336 between two rival royal courts, one—the northern court—in Kyoto and supported by the new Muromachi shogunate, and the other in the mountains south of the capital, where the besieged monarch Go- Daigo had taken refuge. In the midst of these disturbances, sometime around the year 1349 the shogunate gave the rights and duties of the reeve of Ōbe Estate as battle rewards to the Toyofuku family, thought to be local warriors. Unlike the reeve’s position up to that point, the new reeve was not subject to Tōdaiji authority. Tōdaiji attempted to recover its control with the help of the shogunate and the Harima military governors (shugo; of the Akamatsu family), but the effort was completely unsuccessful until 1379. Playing an important role in this process were the Ordination Platform and its subsidiary unit, the Oil Storehouse. It seems that their eventual success was the result of tenacious negotiations with the Akamatsu by authorities of the Ordination Platform. With this as a trigger, the Ordination Platform–Oil Storehouse began to participate in the management of Ōbe Estate. As far as we know from extant documents, this arrangement was regularized sometime in the 1390s. The policy at this time was to divide the proprietor’s share of the rents from Ōbe Estate between the corporate temple and the Ordination Plat- form–Oil Storehouse. Proceeds were distributed thus: the corporate temple received two-thirds of the proprietor’s share, and the Ordination Platform received one-third plus the entire reeve’s share, as well as shares from several tax-collecting units (myō).16 The extant documents indicate that the distribution of the proprietor’s allotment referred only to rent proceeds, not to a division of the land itself. For details and a list of documents related to management rights over the land during this period, see table 4.3. First, let us examine the situation of the corporate temple. The only relevant extant document (a bound document) is a daikan ukebumi, an acknowledgment of orders from a contractor (daikan) who had arranged to collect the two-thirds share of proprietor’s income belonging to the corporate temple.17 The document is dated 1419/5/14. There are very few materials such as land registers or account books that deal directly with local land management. According to the 1419 document, the contractor had agreed to forward a predetermined amount of rent every year to the corporate temple. In other words, management, in the form of rent collection, had been turned over completely to this individual. Thus the documents related to local estate management were probably accumulated by such contractors. Table 4.3 Selected documents related to Ōbe Estate, Muromachi Period

Ordination Platform–Oil Storehouse Corporate Tōdaiji 1404/12/23 Record of rent payments (1419?) Letter of contract from from Ōbe Estate named fields Tōdaiji officials Narukiyo and (myō) Tsunekiyo, Iō, and Tōshichi Tomokiyo regarding the Ōbe 〔1/12/74〕 Estate proprietor’s 2/3 share of the rents 〔1/12/54〕 1404/12/23 Record of payment of the 1/3 portion of the rent 1429/6 Copy of the record of a due the Ōbe Estate proprietor meeting of Tōdaiji monks 〔1/12/109・3/12/4〕〔1/12/71〕 1439/7/11 Ōbe Estate’s final 1431/2/19 Copy of a letter of accounts of rents 〔1/12/136〕 appointment to the post of contractor for the Ōbe 1447/10 Ōbe Estate proprietor’s Estate proprietor’s 2/3 share records of damage from weather, 〔1/12/119〕 etc. to crops 〔1/12/127 ・ 1/24/756–1〕 1439/11/6 Letter of acknowledgement from the 1448/11/9 Record of rent payments Tōdaiji official Narukiyo for his from Tsunekiyo myō on Ōbe Estate appointment as contractor for 〔1/12/126〕 the Ōbe Estate proprietor’s 2/3 〔〕 1449/10 Report on final accounts of share 1/12/51 rents in the Ōbe Estate reeve’s portion 〔3/12/446・1/12/138– 1・6〕 1450/10 Report on final accounts of rents in the Ōbe Estate proprietor’s portion 〔1/12/124〕

1457/10 Report on final accounts . of rents on Ōbe Estate overall 〔1/12/139〕 1457/11 Records of household tax from Ōbe Estate 〔1/12/134〕（Mid-Muromachi）6/3 Petition from Ōbe Estate cultivators, sent to the Oil Storehouse 〔1/25/647〕 Bracketed numbers refer to the document number at the Tōdaiji library. Tōdaiji’s Estates in Its Documentary Record 97

In contrast, the Ordination Platform–Oil Storehouse, which controlled one-third of proprietor’s rights, the reeve’s rights, and rights to income from Tsunekiyo, Iō, and Tōshichi myō, directly managed its holdings by appointing personnel of the Oil Storehouse as on-site managers (shōshu), a term dating from the Nanbokuchō period and defined as “monks dispatched to manage temple holdings” (Nihon kokugo daijiten). Tōdaiji probably adopted the term following the example of Zen temples.18 As indicated in table 4.3, there are many land registers and account books compiled by such on-site managers. The proprietor’s list of responsible rent payers and records of damaged fields, including those from the corporate temple’s two-thirds share, provide evidence for conditions of cultivation on the estate, a significant point. Also, the documents include communications from local cultivators addressed to the Oil Storehouse. Such documents were compiled and preserved by the Ordination Platform–Oil Storehouse to manage the estate directly. At present these documents are mixed in with those of the corporate temple and housed in the archives of the Tōdaiji Library, but I think it is logical to regard them as once having been in a separate collection belonging to the Ordination Platform–Oil Storehouse. Next I will examine in more detail the relationship between the corporate temple and the Oil Storehouse. I have concluded that in the early fifteenth century, the Ordination Platform–Oil Storehouse assumed responsibility for all rent collection on the estate, and then transferred to the corporate temple or its contractor the designated two-thirds portion of the proceeds. Evidence comes from two records of rent payments dated 1402. The first document, dated 1402/11, is a copy of a ledger recording Ōbe Estate rent payments for the entire proprietor’s share.19 This ledger was submitted by one Ryōchin, the reeve and on-site manager. The latter term indicates that Ryōchin was a monk of either the Ordination Platform or the Oil Storehouse. The fact that the reeve’s post, which was formerly filled by a local resident, had been assumed by someone from Tōdaiji indicates that the estate was under the combined direct management of the corporate temple and the Ordination Platform within Tōdaiji. The ledger notes the allotment of rents between the two-thirds proprietor’s portion: 124 koku 8 to 9 shō 4 gō 3 shaku (124.8943 koku) for the corporate temple, and the one-third portion: 64 koku 3 to 7 shō 1 shaku (64.3701 koku) for the Ordination Platform–Oil Storehouse.20 98 How Do We Know about Estates?

The second document, also dated 1402/11, is a ledger for the same time period but concerns the two-thirds proprietor’s portion only, in other words, the portion due the corporate temple.21 The document is an original version, not a copy like the first document. It was submitted by the proprietor’s deputy (kyūshu) Ryūshun, who affixed his seal. The deputy was an individual (in this case a Tōdaiji monk) to whom proprietary authority was delegated, and thus he had the authority to act and make decisions on behalf of the proprietor. Ledgers such as this document were submitted in the course of his duties.

The second document: Ledger of rent payments to the proprietor of Ōbe Estate

(Tokyo University Faculty of Literature archives)

Ledger of rent payments, calculated in koku of rice, for the year 1402, payable to the proprietor of Ōbe Estate 22 Total: 124 koku 8 to 9 shō 4 gō 3 shaku Minus 20 koku payable to Lord Uno23

The remaining 104 koku 8 to 9 shō 4 gō 3 shaku [is distributed as follows]: 25 koku: the deputy’s [Ryūshun’s] salary 1 koku: expenses for ceremonies when rents (in rice) were presented to Ōbe Estate’s storehouses24 5 to: expenses for ceremonies when rents (in rice) were presented to the storehouse at Shikichi 1 koku: operating expenses for estate headquarters 3 to 3 shō: expenses for two shrines on the estate25 7 koku 8 to: payment to the assessor who surveyed the land and assessed the productivity of the fields 3 koku 5 to: expenses for the visit of a proprietor’s representative26 to the estate 1 koku 2 to: [salary for] an armed guard [to accompany the transportation of rents to the proprietor] Tōdaiji’s Estates in Its Documentary Record 99

4 koku 8 to: expenses for traveling to the capital27 1 koku: travel expenses for the armed guard Subtotal of the above: 45 koku 1 to 3 shō 28

The remaining 59 koku 7 to 6 shō 4 gō 3 shaku were converted into cash, totaling 71 kan 147 mon. [The cash was allocated as follows:] 450 mon: expenses for a banquet upon presentation of rents to the storehouses 3 kan 700 mon: travel expenses for a representative of the military governor 7 kan 150 mon: expenses for the construction and repair of estate office buildings 3 kan 337 mon: salary for drawing up maps 1 kanmon: 29 banquet expenses for yeomanry (hyakushō) who come to pay their respects 7 kanmon 123 mon: honorarium for the military governor’s representative upon his visit to the estate Subtotal of the above: 22 kan 763 mon

The remaining 48 kan 381 mon [were allocated as follows]: 12 kanmon 639 mon: portion of rents not yet sent to Tōdaiji Remaining portion: 35 kan 742 mon 21 kanmon: Expenses for lectures on the Kongōmyō Saishōō Sutra (with one portion designated for two extra ordination ceremonies): 1 kanmon: salary for this year’s master in charge, from Tōdaiji 1 kanmon: salary for functionaries (jinin) of Tōdaiji Hachiman Shrine 500 mon: salary for rent collector (shōgō) in charge for the year 100 mon: salary for other rent collectors 100 How Do We Know about Estates?

3 kan 933 mon: salaries for seventy-six monks who lectured on the Lotus Sutra at Tōdaiji Hachiman Shrine 8 kanmon: Expenses for ordination ceremonies at Tōdaiji Kaidan’in. Remainder in cash: 209 mon Not yet allocated.30 1402/11 Deputy, Ryūshun (Seal)

The amount at the beginning of this document, about 124.9 koku, was allotted for various purposes. This corresponds to the amount the first document notes was due the corporate temple as holder of its two- thirds of the proprietor’s rights and suggests that the second document is the original of a report from the overseer to the corporate temple, who deducted all expenses incurred in connection with his management duties, while the first document may be seen as a copy of a report submitted at the time as supplementary data. The documents of 1402 indicate that rent collection followed this process: First, the Oil Storehouse’s on-site manager collected the entire proprietor’s share of the rents. Then he divided them into the two-thirds and one-third portions, recording this in a ledger (the first document), and transferred two-thirds of the amount to the overseer in charge of the portion due the corporate temple. The overseer then allotted the portion for various purposes, compiled the second document (translated above), and presented it to corporate Tōdaiji. In other words, the Oil Storehouse directly managed the estate and took responsibility for collecting the corporate temple’s two-thirds portion of the rents as well as for its own. It seems that the corporate temple’s overseer relied for the most part on management by the Ordination Platform–Oil Storehouse. In the course of time, the entire two-thirds portion of rents due the corporate temple was collected through a contracting system (daikan ukeoi ). On-site management of the estate was no longer carried out by someone unaffiliated with Tōdaiji, such as the local reeves who had previously performed such duties. Rather, direct management was undertaken by Tōdaiji internal units, the Ordination Platform and Oil Storehouse. We can surmise that the corporate temple, secure in its expectation of two- Tōdaiji’s Estates in Its Documentary Record 101 thirds of the rent proceeds, could then convert the system to one in which a proxy (daikan) of the overseer was paid a fixed sum. By this time, rents were paid in cash, as we see from the second document cited above.31 This gives rise to a new question. Why did the corporate temple completely entrust the management of Ōbe Estate to the Ordination Platform–Oil Storehouse in the early fifteenth century? This problem is a topic for further research, along with an examination of the actual activities of the overseer. Whether or not other temples followed similar practices is another matter that warrants future investigation. To conclude this section, there were two peaks in the production of documents related to Ōbe Estate. One was in the late Kamakura period and the other was in the mid-Muromachi period (1430–1460). Late Kamakura documents, when the corporate temple was in charge of the estate, mainly concerned complaints by various parties that involved the temple. There are, however, no extant ledgers to provide details of on-site management at that time. We have few documents from the corporate temple other than contractors’ agreements dating from the mid-Muromachi period, when it shared rights to income with the Oil Storehouse. On the other hand, the Oil Storehouse, which directly oversaw the estate, retained many documents related to its management duties. Thus the documents assembled in Tōdaiji monjo are not homoge- neous but include many different types of records. This is a result of the varied practices of document preservation by Tōdaiji’s various internal units.

Remaining Issues: Investigating Circumstances in the Period of the Northern and Southern Courts This chapter has examined historical materials from various units of Tōdaiji that served as proprietors of Ōbe Estate. I have aimed at presenting a history of the estate through a concrete examination of its management. Up to this point I have focused on two periods, the late Kamakura and the Muromachi. I would like to finish by using two sets of historical sources to examine the period in between, that of the Northern and Southern Courts. These documents shed further light on the controlling authority over Ōbe Estate and the relationship between the two entities that exercised that authority, the corporate temple and Tōnan’in. One set is several documents, dated 1339–1340 and signed by Kose Mijirō, proxy of the estate reeve, the nun Seia. They are copies of Mijirō’s 102 How Do We Know about Estates? requests for battle rewards and of his announcements of arrival for military duty.32 These documents indicate that the reeve’s family had joined forces with the Akamatsu, Harima Province military governors and supporters of the northern court and that the family’s representative had served with distinction in battle. The years 1339–1340 fell within the period of corporate Tōdaiji control of the estate. But it is significant that the documents are inscribed on the reverse side of sacred texts kept at Shinpukuji and are thought to be documents that once belonged to medieval Tōnan’in. These documents were probably stored by Tōnan’in or the corporate temple along with records of appointments to the reeve’s position in the Kamakura period. In other words they seem to have been used to support the legitimacy of the appointment to the reeve’s post. One possibility is that later, when Tōnan’in reassumed control, the reeve’s family presented these documents to Tōnan’in to demonstrate their connections to the Akamatsu military governors. The likelihood that Tōnan’in had reassumed management at this time is reinforced by another document, a rent payments ledger from the estate dated 1364/12.33 There are two important points regarding this document: (1) it notes the two-thirds–one-third distribution of rents, and is signed by two monks, Jōken and “the honorable proxy Genshū”; and (2) it makes no mention of a portion due the reeve, indicating that the ledger concerns the proprietor’s portion only. The first point indicates the division of proprietor’s rights and interests into two-thirds and one- third portions. Later in the Muromachi period, as we know, the corporate temple received two-thirds of the proprietor’s take, while the Oil Store- house received one-third. In this document, Jōken was responsible for the two-thirds portion and Genshū was responsible for the one-third portion. Jōken can be identified as the deputy (kyūshu) for the corporate temple. But what about Genshū? He is referred to in respectful language as “the honorable proxy,” which would not have been necessary for an affiliate of the Oil Storehouse. Rather, he was probably a representative of Tōnan’in. Since Tōnan’in’s head monk at the time was the tonsured prince Shōchin, whose rank was very high, it was necessary to be especially polite in addressing representatives of his cloister. These factors lead to the conclusion that it was Tōnan’in that received the one-third proprietor’s portion of rents at this time. In the same period, moreover, the corporate temple and Tōnan’in seem to have also split proceeds from another holding, the Hyōgo northern toll barrier, in this Tōdaiji’s Estates in Its Documentary Record 103 same fashion, perhaps providing a precedent for Ōbe Estate at this time as well as in the Muromachi period. Inaba Nobumichi argues that the corporate temple denied Tōnan’in’s rights to Ōbe Estate and attempted to bring the estate under its own control;34 but on the contrary, the corporate temple seems to have treated Tōnan’in with respect. In analyzing historical materials, I think it is necessary to recognize the mutually dependent relationships between the corporate temple, on the one hand, and its internal divisions such as the cloisters and the Ordination Platform–Oil Storehouse. This will provide us with a thorough understanding of the broad range of Tōdaiji’s activities and the complexity of its internal organization. To conclude briefly, this study has explored documents related to the management of Ōbe Estate by corporate Tōdaiji and several of its internal divisions. I have emphasized research methodologies that consider the locations where documents were stored as well as their content, form, senders, and recipients. All these features can help us to piece together a historical narrative, in this case, the way the Ōbe Estate proprietorship changed hands over the course of the medieval age.

Notes

1. For an elaboration of some of the topics raised here, see Satō 2003. Researchers in Japan begin with this basic textbook on the reading and inter- pretation of documents. 2. Following definitions in the glossary of Mass 1976, 202, 196, and 203. 3. Nagamura 2000, 23. 4. Appointees to provincial governorships often remained in the capital, delegating their duties to a personal representative (mokudai) and staff at the provincial headquarters. 5. See Chōgen’s testament: OS 4:532–535 (# 224). 6. OS 4:543–545 (# 228). 7. A monzeki is a cloister headed by a member of the royal family or a high aristocrat. 8. Judging from later documents, the reeve was the chief local managerial authority on the estate during this period. 9. OS 4:565 (# 258). This is an unbound document kept in the Tōdaiji Library archives. 10. Inscribed on the reverse side of a sacred text and kept in the Shinpukuji archives. OS 4:566 (# 259). 104 How Do We Know about Estates?

11. OS 4:567 (# 262). 12. OS 4:566–567 (# 261). 13. DNK M182 (Tōdaiji monjo); OS 4:569–571 (# 267). 14. DNK B89 (3) (Tōdaiji monjo); OS 4:660–661 (# 357). 15. Respectively: DNK M1108‑1 (Tōdaiji monjo), OS 4:630 (# 333); DNK M1135 (Tōdaiji monjo), OS 4:637–638 (# 340); and DNK Ｍ1101 (Tōdaiji monjo), OS 4:696–697 (# 424). 16. A large portion of estate land was divided into these units, on which rents were assessed. These were presumably registered in the name of the chief cultivator’s family. 17. DNK (Tōdaiji monjo), S380. 18. When used by Zen temples, the term was pronounced shōsu (see the entry in Nihon kokugo daijiten). For a discussion of shōsu in English, see Collcutt 1981, 241, 277–279. 19. Stored in the archives of the Faculty of Literature at the University of Tokyo. OS 4:733–734 (# 466). 20. Capacity measurements are based on a metric system. 21. Also stored in the archives of the Faculty of Literature at the University of Tokyo. OS 4:732 (# 465). 22. Corporate Tōdaiji. 23. Uno was a powerful local warrior. 24. These were most likely storehouses operated on behalf of the proprietor. 25. One of these was the guardian shrine Ōji. The other has not been identified. 26. Perhaps a fudezukai (scribe) or the deputy. 27. It’s not clear who incurred the expenses. 28. This amount was paid in kind (rice). 29. A kanmon is the same as a kan. 30. Perhaps this remained in Ryūshun’s hands. 31. OS 4:732 (# 465). 32. They are stored in the Shinpukuji archives. OS 4:682 (# 396); 686 (#s 405, 406) (announcements of arrival); 683 (# 398) and 686–687 (# 407) (requests for rewards). 33. In the document collection of Kanō Kōkichi. OS 4:711–714 (# 443). 34. Inaba 1997. c h a p t e r 5

Hine Estate in Izumi Province Archaeology, Landscape Reconstruction, and Village Structures

Hirota Kōji Translated by Janet R. Goodwin

Hine Estate in Izumi Province (present Ōsaka Prefecture) was a holding of the Kujō house, one of the five branches of the regents’ (sesshō) family, the highest-ranking nobles of the court. There are many archaeological sites and landscape features that date from the time when the estate was in operation. Fortunately, because aristocratic control was maintained during the long period from the Kamakura to the Warring States ages—roughly the thirteenth through the sixteenth centuries— we also have many historical materials from that time. These include a comparatively large number of records that document estate facilities and archaeological remains.1 In other words, Hine Estate is a prime candidate for reconstructing local society on estates overall. (For a map showing a portion of Hine Estate and some of its features, see plate 8.) Estates such as Hine should not be regarded only as private landholdings; they were also basic units of local medieval society, made up of communities and communal land. They also served as administrative units of the court and shogunate.2 Previous research on the estate system has focused on the management system as well as on violent intrusions across borders, seeing estates as objects of control and plunder. The conditions of villages within estate borders have also been examined. Courtiers and powerful warriors produced considerable documentary evidence concerning control over estates, but it has been hard to preserve historical materials that illuminate the facilities and physical features of each holding. Thus

105 106 How Do We Know about Estates? we still do not clearly understand medieval estates as sites of communal organizations and local society, sites that were characterized by distinct physical structures and landscapes. To examine the nature of local society on estates, we cannot rely on documentary evidence alone. Rather, we need to reconstruct estates’ topography and spatial features and the structure of their local societies, based on archaeological findings and folklore traditions as well as extant features of the landscape that date from the time when estates existed. There has already been some research on the structure of paddy fields from that time, but it is necessary to consider more broadly other features, including the organizations and physical facilities that provided the basis for local society on the estates. This chapter will begin with a general outline of the history of Hine Estate3 and will then reconstruct the Hine Estate’s landscape based on archaeological findings. Since local society on estates was made up of village communities, the chapter will also discuss the structure of villages on Hine Estate, focusing on the Muromachi and Warring States periods (1392–1573), from which archaeological findings and documents are the most abundant.

Hine Estate in Izumi Province, a Holding of the Kujō House Hine Estate lay in the southern portion of Izumi Province, in the Kinai region close to the capital of Kyoto. In 1234, Kujō Michiie (1193–1252), scion of the powerful family that monopolized the posts of regent and chancellor (kanpaku) to the throne, persuaded the court to convert a broad stretch of land, including the uncultivated areas of Hineno and Tsuruhara, into an estate under the proprietorship of the Kujō house. Thus Hine Estate was formed and established as a Kujō holding. The common wisdom up to now is that medieval estates were generally established when local proprietary lords of the warrior class commended land they had opened to aristocrats in the capital. In the case of Hine Estate, however, the Kujō influenced the court to grant them a holding formed from Izumi Province public lands, in other words, land under the control of the provincial government. There were quite a few estates held by powerful elites that were created in just this way.4 So the Kujō house developed the uncultivated areas of Hineno and Tsuruhara, making them into paddy to form Hine Estate. The land that became the estate, however, was a broad expanse that included not only Hine Estate in Izumi Province 107 wilderness but also villages and cultivated areas developed previously by the provincial government and the great religious institution on Mount Kōya in Kii Province. While numerous estates were established by nominally opening wilderness to cultivation, there were many cases like that of Hine Estate in which some land had already been developed. There were more than seventy-nine chō of paddy under cultivation on Hine Estate in 1234. In the Kamakura period (1185–1333) and the Nanbokuchō period (1334–1392), the estate was composed of the following four villages: Tsuruhara (thirty-nine-plus chō), Ihara (thirteen-plus chō), Hineno (fourteen-plus chō), and Iriyamada (thirteen-plus chō). These four villages were large ones containing a number of hamlets. The Kujō house, proprietors of several dozen estates, did not personally manage each holding. Their retainers, from a Minamoto lineage later called the Shinanokōji, supervised all the Hine Estate management personnel for each of the four villages: the custodian (azukaridokoro), the military steward (jitō), the jitōdai (military steward’s proxy), the local manager (gesu), the reeve (kumon), the surveyor (tadokoro), and the centrally appointed administrator (zasshō). Thus there was a two-layered control structure over the estate and the villages. Over time, the role of the Nakahara family (later called the Hineno), who combined the posts of custodian and centrally appointed administrator, was strengthened, and in the late Kamakura period, new estate officials known as satanin appeared in each village. Much land—more than 163 chō —in Hineno and Ihara Villages remained uncultivated. Around 1310, the Kujō ordered irrigation canals and reservoirs dug, turning wilderness into paddy and encouraging cultivators to settle on the estate. In medieval times monks and temples, leading groups of artisans, often assumed land reclamation contracts, and so in that year the monk Jissen (otherwise known as Jitsugyō Shōnin) was instructed to open the Hineno and Ihara wilderness to cultivation. But neither the Minamoto nor the Nakahara cooperated with this effort, and in 1316 the Kujō designated Kumedadera, a temple in Izumi Province, to renew efforts to reclaim the land. Two maps from the late Kamakura period illustrate the two land reclamation efforts.5 (For the map dated 1316, see plate 9 and the cover and dust jacket of this volume.) During the Kamakura period there were many people on Hine Estate who served other elites and estate proprietors. For example, there were functionaries (jinin) who served Kasuga Shrine and its affiliate, the powerful Nara temple Kōfukuji, as well as purveyors (kugonin) who supplied 108 How Do We Know about Estates? the court with fish from Abiki tribute lands along the seacoast. Many of these groups were involved in the production and circulation of goods, and they were sometimes joined by Hine Estate cultivators in providing services to the Kujō house and other elites, including temples and shrines. Along with collecting rents from cultivators on the estate, the Kujō also controlled service groups called the ōban toneri —some from Hine Estate—who performed various functions for the regent’s house. In the mid-fourteenth century the archipelago’s political landscape was transformed in significant ways, and these changes of course influenced Hine Estate. In 1336 the sovereign Go-Daigo Tennō broke with his erstwhile allies, the Ashikaga, and was forced to flee to the Yoshino mountains in Yamato Province, where he set up the so-called southern court. The Ashikaga—who established a new military government known as the Muromachi shogunate—supported the rival northern court in Kyoto, and until almost the end of the century, civil war broke out periodically between warrior supporters of the two courts. Hine Estate was one site of such battles. At the beginning of these troubled times, the Nakahara relinquished their post as the proprietor’s representatives (zasshō).6 Conflict soon arose between this family and the Kujō. As warriors in the service of the Muromachi shogunate, the Nakahara stood guard over Tsuchimaru and Ameyama forts in Iriyamada Village on Hine Estate, and battling with forces of the southern court, they also repeatedly invaded Hineno Vil- lage. Thereafter, the Kujō met the Nakahara—now called the Hineno— in armed conflict over the estate, which was frequently occupied by armies of the two warring sides. In the 1370s, shogunal forces and those of a southern court ally, the warrior Hashimoto Masataka, fought to control the two forts, and many shrines and temples on the estate were burned to the ground. Even after the end of the civil conflict in 1392, Hine Estate was invaded by a military governor (shugo) and his retainers of the Hineno family (the former Nakahara), and portions of the estate were seized by military overlords. Tsuruhara Village on the estate became the possession of the Satake, shogunal retainers, and was renamed Tsuruhara Estate. Ihara Village was seized by the two Izumi Province military governors from the Hosokawa family, and the village’s name was changed to Sano Estate.7 Kawaraya Estate, which lay between Tsuruhara and Sano Estates, became the holding of the Hino family, powerful aristocrats related to the Muromachi shogun. The Muromachi period is often said Hine Estate in Izumi Province 109 to be a period of decline for the estate system, but it was also a time when new estates emerged. After the Tsuruhara and Ihara Villages were seized, Hine Estate under the Kujō was reduced to the two villages of Hineno and Iriyamada. But then in 1410 Hosokawa Yorinaga, the military governor of Izumi Province, donated revenues from half of Iriyamada to his vow temple in Kyoto, Eigen’an at Kenninji, and planned to assume control of the last two villages that remained from the Kujō holding. To maintain title to what was left of Hine Estate, the Kujō tried to make a deal with the two Izumi military governors, appointing them as proxies to control Hineno and Iriyamada Villages. But the governors continued their violations, and in 1430 the Kujō filed suit with the shogunate, which verified Kujō rights over the two villages. The shogunate, the most powerful military authority, thus guaranteed the rights of a civilian noble house and put a stop to warriors’ invasion of a noble family’s holding. Afterwards, the Kujō turned over management of Hineno and Iriya- mada to proxies—monks and warriors—but reassumed direct control after 1457, appointing their retainers, the Karahashi and Tominokōji, as administrators. Even while the estate suffered violations by the military governors and the Hineno, management rights at that time were concentrated in the hands of the Kujō and their retainers and administrative staff. In the Warring States period, the authority of the Muroma- chi shogunate declined, and once more the Izumi military governors plotted to take over the estate. To stop their violations, the estate’s proprietor, the former regent Kujō Masamoto, delegated authority over Hineno and Iriyamada Villages to monks of Negoroji in Kii Province, a large temple that rivaled the military governors in power. Beginning in the 1480s, forces from Negoroji advanced into Izumi Province and repeatedly battled the governors’ armies. The structure of Hineno and Iriyamada Villages also changed during the Muromachi and Warring States periods. To summarize briefly here, beginning in the Kamakura period a number of new paddy fields were formed through land reclamation projects in Hineno Village, which was split into eastern and western sectors. Hamlets that still exist were established in both sectors. Four smaller villages were developed within Iriyamada: Funabuchi, Shōbu, Ōgi, and Tsuchimaru, and within each a number of hamlets emerged. These four smaller villages and their hamlets have also continued as communities until today. The villages of Hineno and Iriyamada thus had a layered structure with smaller communities 110 How Do We Know about Estates? contained within the larger ones; these communities were managed by local elites (bantō) and elders (korō). Village structures originating from the Muromachi and Sengoku periods have persisted until modern times. In 1501, Kujō Masamoto, to re-establish control over Hineno and Iriya- mada, settled in Iriyamada along with a number of retainers. For the next four years he supervised his holding directly, a practice followed by many other aristocrats during the Warring States period. During Masamoto’s residence on Hine Estate, he wrote the diary Masamoto-kō tabi hikitsuke (Records of Lord Masamoto’s travels), a record of his direct management of the holding; the diary, a crucial historical source for research on estates and villages in the Sengoku period, is extant.8 In this work, Masamoto diligently recorded conditions of battle, pillage, calamity, drought, and famine, as well as the daily lives of residents of Hineno and Iriyamada Villages. Meanwhile, the two military governors of Izumi Province plotted to attack the villages, abducting local cultivators and continuing to battle Negoroji forces in the vicinity. Although village headmen and cultivators quarreled with Masamoto over paying him tribute during his years of residence, they cooperated with him in resisting the military governors and protecting their villages. The invaders, including the Hineno, were forced to retreat. When Negoroji forces defeated the governors in 1504 and took over Izumi Province, Masamoto acceded to the temple’s demands to make its monks deputy contractors of Hineno and Iriya- mada Villages, and then he left the estate for Kyoto. During this time the cultivators of Hineno and Iriyamada suffered the natural calamities of drought, bad harvests, and famine and the human-made calamities of war and pillage. Masamoto’s demands for yearly rent payments were also a heavy burden. Yet the cultivators continued to lead their daily lives, growing rice, barley, tea, and sesame and harvesting mountain peaches, matsutake mushrooms, persimmons, and bracken from mountain areas. In addition they governed their villages themselves and voluntarily maintained the irrigation facilities that supported agricultural production. They maintained religious facilities and activities: temples where public assemblies were held, shrines, and their festivals. Cultivators also petitioned to protest Masamoto’s orders and took up arms to protect their villages, joining with neighboring villages to defend their livelihood and possessions. When Masamoto’s direct management came to an end in 1504, monks from Negoroji once again contracted to collect rents from Hineno and Iriyamada for the Kujō. Hine Estate was close to Kyoto, and the Kujō con- Hine Estate in Izumi Province 111 tinued to receive rents until 1533, when the curtain finally fell on three hundred years of the history of the estate as a Kujō holding. Then, until Toyotomi Hideyoshi subdued Izumi Province in 1585, Hineno and Iri- yamada Villages were controlled by Negoroji (see table 5.1 and plate 8).

Table 5.1 Chronology of Hine Estate in Izumi Province

1205 The Kōya monk Banna begins to open uncultivated land in Hineno Village but fails because of obstruction from Nagataki Estate and Zenkōji (hereditary holdings of the regents’ house). 1223 Mount Kōya again tries to open Hineno land but fails. 1233 Having inherited the regents’ house rights to Nagataki Estate, Kujō Michiie appoints members of the Nakahara (later called the Hineno) family as its local manager (gesu) and reeve (kumon). 1234 Hine Estate is established as a holding of the Kujō family, and land surveys are carried out. The Kamakura shogunate orders the military steward (jitō) and proxy for the provincial military governor (shugodai) not to obstruct Kujō control. 1245 The harbors of Sano (a fishing village) and Tsuruhara in Ihara Village, Hine Estate, become part of Abiki tribute lands (mikuriya). 1246 The Minamoto family, household officials for the Kujō, assumes control of Hine Estate and Zenkōji, and later, of Muhenkōin in Hineno Village. 1266 Nakahara Moritsune, local manager and reeve of Nagataki Estate, is appointed custodian (azukaridokoro) of Ihara Village, and afterwards to the same position in Iriyamada Village. 1277 A quarrel breaks out between the Nakahara, managers (zasshō) of Hine Estate, and Kasuga Shrine and its affiliates (jinin) on the estate. 1292 The estate official (satanin) Kanetsuna seizes the annual rents (nengu) and becomes involved in a legal dispute with the manager of Tsuruhara Village. 1301 The residents of Ihara Village refuse to repay a loan from the monks of Kōfukuji, calling themselves “cohorts of the powerful.” 1309 The Kujō order the Nakahara, estate managers, to undertake a complete land survey of the estate. Table 5.1 (continued) 1310 The Kujō entrust the opening of land in Hineno and Ihara Villages to the monk Jissen, and issue registers of uncultivated lands and maps of the two villages. 1312 The Kujō have the local managers and reeves in Hineno, Ihara, and Iriyamada Villages replaced. 1315 Kujō household official Minamoto no Kanesada and manager (zasshō) Nakahara Moriharu are accused of forcibly interfering with land reclamation efforts. 1316 Kumedadera contracts to open land in Hineno, and is given rights to the Nagataki Estate local manager’s and reeve’s positions. A map of Hineno Village is made. 1330 Nakahara Moriharu donates the Nagataki Estate manager’s and reeve’s rights to Amano Nyūsha Shrine in Kii Province. 1332 The Kamakura shogunate orders Hineno (Nakahara) Moriharu to join the attack on Kusunoki Masashige and Prince Moriyoshi. 1336 Hineno Moriharu joins the forces of the Muromachi shogunate, moving through Izumi and Kii. The shogunate and southern court forces clash at Ōgi Fort in Iriyamada Village. 1340 Functionaries (yoriudo) of the Music Office at the court are present in Sano, Nakanoshō, and Tsuruhara Villages on Hine Estate. 1342 The Kujō acknowledge Hineno Moriharu as local manager and reeve of Nagataki Estate. 1343 Images of the twelve divine generals are made at Negoroji in Kii Province for Hokkeji, the Buddhist temple (jingūji) at Ōizeki Shrine in Hineno Village. 1347 Under shogunal orders, the Hineno guard Tsuchimaru and Ameyama Forts, and then are given control of Hineno Village. 1351 Armies of the Ashikaga and the southern court clash repeatedly at Tsuchimaru and Ameyama Forts, and battles continue at Sano Fort in Ihara Village. 1355 The Negoroji monk Kakugō seeks contributions for the reconstruction of Ōizeki Shrine in Hineno Village, destroyed by battle fires. 1379 Shogunal armies force the surrender of Hashimoto Masataka of Tsuchimaru and Ameyama Forts. The next year he is killed in battle by the shogun’s army. Table 5.1 (continued) 1396 Kujō Tsunenori gives Hine Estate to his houseman Kanemune, but the estate has been invaded by warriors (bushi) and Kanemune cannot take total control. 1408 The Satake, housemen of the Muromachi shogunate, are given proprietor’s (ryōke) and military steward’s rights to Tsuruhara Estate (once a part of Hine Estate). 1410 Hosokawa Yorinaga, one of the two military governors (shugo) of Izumi, donates revenues from Iriyamada Village to his family temple, Eigen’an at Kenninji. This initiates conflict between the military governors and the Kujō. 1413 The Satake military stewards (jitō) of Tsuruhara Estate, along with Hineno Kunimori, invade five villages on Hine Estate, and the Kujō complain to the shogunate. 1417 From this year on, reports of rents and special levies (tansen) on Iriyamada Village are issued, and around the same time the number of fields in Hineno Village is recorded. 1419 Hine Estate is invaded by the Izumi military governors but the shogunate orders them to stop. The Kujō make great efforts to guarantee receipt of rents. 1428 Deputies of the two Izumi military governors contract to collect rents from Hine Estate, but the governors continue to invade the holding, and Kujō jurisdiction is reduced to two villages, Hineno and Iriyamada. 1430 The Kujō file suit with the shogunate about the military governors’ violations, and the latter give the income from Hine Estate to the Hineno. The Kujō appoint a Zen monk as their contracting deputy (daikan or shōshu). 1431 Hineno and Ihara Villages, and Danbaramitsuji Village, having constructed Lower Jūnidani Pond, agree to provide irrigation water from the pond. 1433 Accused of misbehavior, Kusaka Kunimune is replaced as Hine Estate deputy. The deputy position is contracted out to Zen monks. 1457 The Kujō decline to appoint a contracting deputy for Hine Estate and assume direct control. Nine local warriors (kokujin) of Izumi Province Hine District, including Hineno Hidemori, issue a pledge of alliance (ichimi dōshin). Table 5.1 (continued) 1467 Under one of the military governors of Izumi, the Hineno take up arms in the Ōnin War. During the war, Kujō Masamoto retreats to his family’s holding in Ninomiya, Owari Province. 1472 Kujō Masamoto gives overlords’ rights to Iriyamada Village to his houseman Karahashi Arikazu. The two military governors take half the rents of Hineno Village, invoking their rights to do so for military purposes under the “half-tax” (hanzei) system instituted by the shogunate. 1481 A register of the number of fields in Hineno Village is compiled, providing evidence that new hamlets were established in former wilderness areas of Hineno Village. 1485 Kujō vassal Tominokoji Toshimichi takes control of half of Hineno and Iriyamada Villages in opposition to the two Izumi military governors. 1486 Tominokoji Toshimichi appoints two Negoroji monks as deputy contractors for Hine Estate. Negoroji forces defeat the military governors and subdue the province. 1490 Negoroji monks invade Izumi Province, capturing the military governor’s holding, Ihara Village (Sano Estate). 1495 The military governors defect from the Muromachi shogunate and ally with the military governor of Kii (of the Hatakeyama family) and Negoroji. 1496 Kujō Masamoto assassinates his houseman, Hine Estate administrator (bugyō) Karahashi Arikazu, at the Kujō mansion and is censured by Go-Tsuchimikado Tennō. 1497 Negoroji resigns the position of contracting deputy of Hine Estate. One of the Izumi military governors exacts half the income from Izumi Province estates, including Hine. 1500 The two military governors of Izumi—Hosokawa Motoari and Hosokawa Mototsune—are defeated by Negoroji and die in battle. Then their heirs overcome Negoroji and occupy Hine Estate. 1501 Kujō Masamoto comes to Iriyamada Village and assumes direct control of Hine Estate. Hineno Village forces repel the attack of the military governors’ armies. 1502 There is a battle between the two Izumi military governors and Negoroji, and Kujō Masamoto takes refuge at Inunakisan Shippōryūji. Hine Estate in Izumi Province 115

Table 5.1 (continued) 1503 There is a drought in the region including Hine Estate, and a famine results the next year. The two military governors hold the monks of Muhenkōin hostage. 1504 Negoroji subdues Izumi Province and obtains rights to half the income of the entire province. Masamoto appoints the Negoroji monk Akaibō as contracting deputy of Hine Estate and then returns to Kyoto. 1507 Shogunal elder Hosokawa Masamoto is assassinated, and his adopted son Sumiyuki (Kujō Masamoto’s real son) commits suicide. 1518 Hine Estate contracting deputy Akaibō reports to the Kujō that because of war, he cannot forward the estate’s rents. 1533 Hineno village head (bantō) Kita Shōji and contracting deputy Kyūhan pay special levies to the Kujō. Afterward Hineno and Iriyamada Villages fall under the control of Negoroji. 1535 Kujō Tanemichi flees from Kyoto, and the Muromachi shogunate orders the confiscation of Kujō estates. 1560 Kujō Tanemichi takes Sogō Kazumasa, the younger brother of Miyoshi Nagayoshi (the warrior who controlled the Kinki region), as his adopted son-in-law, planning the recovery of Kujō property in Izumi Province. 1585 The armies of Toyotomi Hideyoshi overcome Izumi Province, ending Negoroji’s control of Hineno and Iriyamada Villages.

Archaeological Remains on Hine Estate The area that was once Hineno and Iriyamada Villages on Hine Estate— the modern districts of Ōgi, Tsuchimaru, and Hineno in Izumisano City, Ōsaka Municipal Prefecture—is especially rich in archaeological sites. Historians and local residents have preserved the area’s remains, and research groups and governmental bodies have sponsored archaeological investigations. As a result, sixteen sites from Hine Estate (six in Ōgi, nine in Hineno, and one in Tsuchimaru), including temples, shrines, irrigation canals, reservoirs, and forts, have been initially designated and protected as National Historic Sites. (The numbers of the sites discussed below are those in tables 5.2 and 5.3, and on figures 5.1 through 5.3. National Historic Sites are marked in boldface.) 116 How Do We Know about Estates?

These sixteen, though, are not the only remains of Hine Estate from medieval times. There are many others, and we cannot conduct research on the characteristics of the archaeological remains on Hine Estate if we focus entirely on National Historic Sites while ignoring others. Thus I will present a general overview of the remains of Hine Estate from the present Ōgi, Hineno, and Tsuchimaru districts.9

Table 5.2 Archaeological sites on Hine Estate National Historic Sites are marked in boldface. (See figure 5.1, Ōgi Village; figure 5.2, Tsuchimaru Village; figure 5.3, Hineno Village.)

1 Shippōryūji Shugendō temple in the Katsuragi Mountains, where prayers for rain in Iriyamada were conducted and local residents’ property was stored. 2 Ōi Canal Longest irrigation canal in the Ōgi area; irrigates the eastern bank of the Kashii River. Extant in the Warring States period. 3 Inunaki Canal and Canal that irrigates paddy in Kamiōgi. Higashino Pond There are beautiful terraced paddies in its irrigation area. The Inunaki Canal supplied water for Higashino Pond. 4 Wai Canal Canal that irrigates the right bank of the Kashii River. Constructed before the Warring States period. 5 Shōbui Canal Canal that irrigates the right bank of the Kashii River. It was the canal of Shōbu Village, which existed in the Warring States period. 6 Takashiro Pond Pond at the summit of Mount Takashiro, east of Ōgi. Used only in emergencies. 7 Kōshakuji site Many medieval stone Buddhist artifacts are found at this temple. Its monks transmitted Kujō Masamoto’s orders to the populace. Table 5.2 (continued) 8 Kamiōgi Rengeji Temple in Funabuchi Village (Kamiōgi). Used today as a meeting hall for Kamiōgi. Medieval stone Buddhist artifacts are found here. 9 Kamiōgi stone Buddhist Located along the Inunaki canal in Kamiōgi, image group indicating that there may have once been a religious facility here. 10 Matarin-san Religious facility in Kamiōgi, where medieval stone markers are found. 11 Site of Ōgi Fort On the summit of a mountain behind Saikōji. This may have been a refuge and fortification for Iriyamada Village residents. 12 Saikōji and graveyard Temple in Shōbu Village (Nakaōgi), currently a meeting place for the modern village. In the graveyard are medieval and early modern stone Buddhist artifacts. 13 Small shrine for a At the border of Kamiōgi (Funabuchi) roadside deity (sae no and Nakaōgi (Shōbu) Villages. kami) Medieval stone Buddhist artifacts are found there. 14 Shitaru Pond; Hatakeda Pond and canal that irrigate Shimoōgi Canal from Shitaru Valley. In the Warring States period there were rice paddies in the valley. 15 Hōkyōin stupa at Hatakeda Found in the Shimoōgi terraced rice fields that cross the Shitaru Valley. It is thought to be a memorial stupa for a fallen warrior. 16 Bishamondō; site of A temple in Goshodani, Nakaōgi. Goshodani Ōgi Fort Behind the temple are located medieval stone Buddhist artifacts and the fort remains. 17 Shitadara Canal A canal that irrigates Shimoōgi. In the Warring States period, rice fields were opened along the canal’s banks. Table 5.2 (continued) 18 Hibashiri Shrine Takinomiya, the guardian shrine of Iriyamada. Venue for prayers for rain, festivals, nenbutsu dances, and village assemblies. 19 Endani Pond Called Enbuchi in Kamakura documents. It was an emergency pond connected to the Ōi Canal. 20 Five-part stupas (gorintō) The stupa is north of Endani Pond and the and foundation stones at foundation stones are on leveled ground. Endani This may be the remains of a temple. 21 Tennō Shrine site Remains of a shrine once located within Wakasaki Hamlet, Shimoōgi. Thought to be the medieval Iriyamada shrine to Gozu Tennō, a deity thought to ward off plague. 22 Tachibanadani ponds Located in Tachibanadani (valley) in and canals Shimoōgi; the pond is the same as the Kamakura-period Tachibana Pond (different characters). There is also a smaller pond there. 23 Chōfukuji site Kujō Masamoto resided at this temple, located in Iriyamada, Ōgi Village (present Shimoōgi). Archaeologists have found remains of the structure. 24 Enmanji This temple in Shimoōgi is now used as a public meeting place. In medieval times, Dainichi-dō (hall) and Hachiōji Shrine were adjacent to the temple. 25 Okunoike Pond Pond at the terminus of the Ōi Canal in Ōgi. This canal was dug by cutting through solid rock. 26 Shimoōgi graveyard Extant in the early Edo period, this graveyard is located on a mountaintop behind Enmanji. The Ōi Canal flows nearby. 27 Sefuji and Sakakubo Two adjacent ponds. The latter is identified Ponds with Sakakubo Pond (different characters) of Kamakura times. Table 5.2 (continued) 28 Kamataki Pond; Nagajike Pond and canal in Hiuchigodani, Ōgi. Canal Identified with Hiuchigawai canal and pond of Kamakura times. 29 Giboshi Canal; site of Irrigation canal constructed in the early Edo Ameyama Canal period; a portion of the Ameyama Canal. 30 Tsuchimaru and Mountain fortifications dating from the Ameyama Fort sites Northern and Southern Courts period to the Warring States period and held by the Hineno family and Negoroji. Possibly used as fortifications by residents of Hine Estate. 31 Tsuchimaru Ōi and Irrigation canals of medieval Tsuchimaru Tennō Canals Village, identified with the Kamakura- period Tsuchimarui Canal (different characters). 32 Tsuchimaru Kasuga Guardian shrine of Tsuchimaru Village. Shrine There is a Warring States–period votive lantern on the precincts. 33 Gokurakuji The main worship object of this Tsuchimaru temple is a medieval image of Yakushi. There are medieval stone Buddhist artifacts within the temple precincts. 34 Tsuchimaru graveyard This graveyard is mentioned in medieval documents. In the Warring States period, there were cremation specialists known as sanmai hijiri here. Nearby are remains of the Ameyama Canal. 35 Tsuchimaru Daishidō Shingon temple in Tsuchimaru, containing medieval five-part stupas 36 Tsuchimaru Rengeji Temple near the location of the central hamlet in Tsuchimaru Village in medieval times. Now a branch temple of Shippōryūji. 37 Hashimoto Masataka’s Masataka was a military commander in charge supposed gravesite of Tsuchimaru Fort during the period of Northern and Southern Courts. There are many medieval stone artifacts nearby. Table 5.2 (continued) 38 Inakura River’s irrigation Today’s Arinoki and Soenomae Canals date canals from medieval times. 39 Hine Shrine The medieval Ōizeki Shrine. Guardian shrine of Hine Estate, and one of the five most important shrines of Izumi Province. A Buddhist temple was located on the premises in medieval times. 40 Jigen-in Temple on the grounds of Ōizeki Shrine. The pagoda and Golden Hall (main hall) date from medieval times. Kujō Masamoto’s posthumous name was Jigen-in. 41 Miyuki Road During festivals at Ōizeki Shrine, the sacred carriage traveled along this road. Extant in the Warring States period. 42 Yugawa (irrigation canal) Hineno irrigation canal, dug in medieval times. Several canals branch off from this one as it flows to Jūnidani Pond. 43 Presumed site of Temple that served as the headquarters Muhenkōin of Hine Estate. In medieval times it contained a bathhouse and a place for prohibition placards. 44 Ō Pond Pond constructed in the early Edo period. Largest irrigation pond in Hineno Village at the time. 45 Amazu Pond Corresponds to the Kamakura- period Amazuke Pond (different characters). The most important reservoir in medieval Hineno Village. 46 Amazu Pond’s Nakabi These canals, which flow from Amazu and Sokobi Canals Pond, are thought to have existed in medieval times. 47 Hachiōji Shrine site Shrine near the Yugawa Canal that dates from the Kamakura period. A medieval stone stupa has been discovered there. Table 5.2 (continued) 48 Nonomiya Shrine site Corresponds to Nyū Daimyōjin Shrine from the Kamakura period. Guardian shrine of Hineno Village; assumed the name Nonomiya in the Warring States period. 49 Sōfukuji Tenmangū This shrine was constructed in the Warring States period. There are medieval stone artifacts on its premises. 50 Presumed residence site Thought to be the residence of the Hineno, of Hineno family local proprietary lords. No remains of a fort have been excavated. 51 Ryōgonji Temple in Nakasuji Hamlet, Hineno Village. Called Ryōgon’an in the Warring States period. 52 Presumed site of Iburidō This temple dates from the Kamakura period. The Iburidō Canal flows nearby. 53 Ox deity of Shindōde Religious establishment that enshrines an agricultural ox deity. There was an ox deity shrine in Hineno Village from Kamakura times. 54 Ox deity of Hineno There were many such shrines in Hineno Village. 55 Yaeji Pond Called Yae Pond in the Kamakura period. Canals linked this pond with Amazu and Jūnidani Ponds. 56 Ichimichi Canal; Ōgi A canal that originates in Amazu Pond and Road the road that runs along the canal.

57 Jūnidani Pond; Shimizu The pond was called Jūjidani Pond in the Canal Kamakura period. The canal, extant in medieval times, flows from the pond. 58 Harai Canal A canal that branches off from the Yugawa Canal near Jūnidani Pond and irrigates paddy in the northwest portion of Hineno Village. 59 Shitsu Pond Northwest of Jūnidani Pond; possibly constructed in the Muromachi period. 60 Shiramizu Pond site This pond, dating from Kamakura times, was filled in during a modern construction project. Table 5.2 (continued) 61 Nono Jizō Jizōji, a temple in Nonojizō Hamlet, Hineno Village. Corresponds to No Jizō of Muromachi times. 62 Kokawa Road Road connecting Shippōryūji with Kokawadera and Negoroji in Kii Province. Also called Fudō Road or Inunaki Road. 63 Mizuma Road Pilgrimage road to Mizumadera in Izumi Province. Runs through Hineno and Ōgi Villages and in Ōgi, passes by the Tachibanadani Pond. 64 Gōno Pond and Uwanada This pond and canal are located in Hineno Canal Village but irrigate paddy in the adjacent Kaminogō Village. 65 Presumed site of Temple along the Kumano Road, thought Danbaramitsuji to be the site of the market on Sano Estate. 66 Presumed site of Temple on Sano Estate, a holding of the Manpukuji; Shiokumi Izumi Province military governors. Road Located in Nakashō in Izumi Sano City. 67 Kumano Road; Sano Ōji Pilgrimage road to the Kumano Shrine Shrine site complex in Kii Province. Sano Ōji Shrine is located along this road. 68 Nyoraiji Temple along the Kumano Road. Possibly corresponds to Nyoraidō, a Muromachi-period temple in Hineno Village. 69 Kaino Pond Corresponds to Yabuike in Hineno from the Kamakura period. Now the pond of Nagataki Village adjacent to Hineno. 70 Aritōshi Shrine Guardian shrine of Nagataki Estate. 71 Zenkōji site; Zenkōji Site of a medieval temple near Hineno Jizōdō Village. Hine Estate’s official headquarters were located here. 72 Kuzuwai Spring Spring in existence in Kamakura times that is said to have provided water for Shippōryūji. Table 5.3 Archaeological remains of the Hine site according to type National Historic Sites are marked in boldface.

Ōgi Tsuchimaru Hineno Other Headquarters 23 Chōfukuji site 43 Presumed 71 Zenkōji site site of Muhenkōin Guardian shrines 18 Hibashiri Shrine 39 Hine Shrine (Takinomiya) (Ōizeki Shrine) Temples, shrines, small shrines, etc. 1 Shippōryūji 32 Tsuchimaru 40 Jigen-in 65 Presumed site 7 Kōshakuji site Kasuga Shrine 43 Presumed site of Danbaramitsuji 8 Kamiōji Rengeji 33 Gokurakuji of Muhenkōin 66 Presumed site of Manpukuji 10 Matarin-san 35 Tsuchimaru 47 Hachiōji Daishidō Shrine site 67 Sano Ōji 12 Saikōji 36 Tsuchimaru 48 Nonomiya (shrine) site 13 Small shrine to Rengeji Shrine site 70 Aritōshi Shrine roadside deity 49 Sōfukuji 71 Zenkōji Jizōdō 16 Bishamondō Tenmangū 18 Hibashiri Shrine 51 Ryōgonji 20 Foundation 52 Presumed stones at Endani site of Iburidō 21 Remains of 53 Ox deity of Tennō Shrine Shindōde 23 Chōfukuji site 61 Nono Jizō 24 Enmanji Irrigation facilities 2 Ōi Canal 29 site of 42 Yugawa 64 Gōno Pond and 3 Inunaki Canal & Ameyama Canal (irrigation Uwanada Canal Higashino Pond 31 Tsuchimaru canal) & branch 69 Kaino Pond canals 4 Wai Canal Ōi and Tennō 72 Kuzuwai Spring Canals 44 Ō Pond 5 Shōbui Canal 38 Inakura River’s 45 Amazu Pond 6 Takashiro Pond irrigation canals Table 5.3 (continued) Ōgi Tsuchimaru Hineno Other Irrigation facilities 14 Shitaru Pond; 46 Amazu Hatakeda Canal Pond’s Nakabi 17 Shitadara Canal and Sokobi Canals 19 Endani Pond 55 Yaeji Pond 22 Tachibanadani ponds and canals 57 Jūnidani Pond; Shimizu 25 Okunoike Pond Canal 27 Sefuji and 58 Harai Canal Sakakubo Ponds 59 Shitsu Pond 28 Kamataki Pond; Nagajike Canal 64 Gōno Pond and Uwanada 29 Giboshi Canal; Canal site of Ameyama Canal Fortifications 11 Site of Ōgi Fort 30 Sites of 50 Presumed Tsuchimaru and residence site of Ameyama Forts Hineno family Commercial facilities 43 Presumed site 65 Presumed site of of Muhenkōin Danbaramitsuji Graveyards and stone stupas, tombstones 9 Kamiōgi stone 33 Gokurakuji 49 Sōfukuji Buddhist image 34 Tsuchimaru Tenmangū group graveyard 10 Matarin-san 37 Hashimoto 12 Saikōji Masataka’s graveyard supposed 15 Hōkyōin stupa at gravesite Hatakeda 20 Five-part stupas and foundation stones at Endani 26 Shimoōgi graveyard Hine Estate in Izumi Province 125

Table 5.3 (continued) Ōgi Tsuchimaru Hineno Other Public facilities 8 Kamiōgi Rengeji 32 Tsuchimaru 39 Hine Shrine 65 Presumed site 12 Saikōji Kasuga Shrine (Ōizeki Shrine) of Danbaramitsuji 18 Hibashiri Shrine 43 Presumed site of 23 Chōfukuji site Muhenkōin 24 Enmanji 48 Nonomiya Shrine site Roads 62 Kokawa Road 62 Kokawa Road 41 Miyuki Road 62 Kokawa Road 63 Mizuma Road 63 Mizuma Road 56 Ōgi Road 63 Mizuma Road 62 Kokawa 66 Shiokumi Road Road 67 Kumano Road 63 Mizuma Road 67 Kumano Road Border markers 1 Shippōryūji 30 Tsuchimaru 67 Kumano 66 Shiokumi Road 13 Small shrine to and Ameyama Road roadside deity (sae Fort sites no kami)

The Ōgi District corresponds to three medieval villages within Iri- yamada: Funabuchi, Shōbu, and Ōgi (today’s Kamiōgi, Nakaōgi, and Shimoōgi), otherwise known as the three inner or upper villages. The largest-scale archaeological sites in the district are those of the Shugendō temple Inunakisan Shippōryūji (1) and Hibashiri Shrine (18), called Takinomiya in medieval times, the guardian shrine of Iriyamada. Next comes Chōfukuji (23), where Kujō Masamoto lived during his residence on the estate. Each hamlet also includes temples that were sites of local worship (7, 8, 12, 16, 24). There are also many small shrines, as well as graveyards containing stone stupas or other stone Buddhist artifacts from the medieval age. Archaeological sites include many small-scale irrigation canals and reservoirs constructed in the medieval period as well as 126 How Do We Know about Estates?

Fig. 5.1. Ōgi Village. Drawn by the author.

remains of village fortifications where local cultivators took refuge in time of war. In 2013, the Bureau of National Culture and History designated Ōgi’s historical landscape as a National Important Cultural Landscape. The Tsuchimaru District corresponds to the medieval Tsuchimaru Village within Iriyamada. The largest-scale archaeological site there is the district’s guardian shrine, Tsuchimaru Kasuga Shrine (32). There are also the Buddhist temples Gokurakuji (33) and Tsuchimaru Rengeji (36), as well as remains of irrigation facilities, graveyards, and stone stupas. Other sites include Tsuchimaru and Ameyama Forts (30), where the Hineno family of local proprietary lords and Negoroji monks barri- caded themselves in wartime. These are the largest fortifications on Hine Estate or nearby. The Hineno District corresponds to medieval Hineno Village. Its largest site is that of Hine Shrine (39), formerly called Ōizeki Shrine, Hine Estate in Izumi Province 127

Fig. 5.2. Tsuchimaru Village. Drawn by the author.

the guardian shrine of the estate. The Yugawa Canal (42), the most important irrigation canal in the village, flows through the precincts of Hine Shrine and of the Buddhist temple Jigen-in (40) adjacent to the shrine. There are comparatively large reservoirs on the district’s upland slopes (44, 45, 55, 57, 64), as well as many small-scale irrigation facilities and religious structures and shrines dedicated to Ushigami, the cultivators’ ox deity (53, 54).10 Remains from or related to medieval Hine Estate can be found in surrounding areas (see tables 5.2 and 5.3). Current investigations, based on documents, archaeological remains, and local tradition, verify or suggest the presence of more than seventy sites in and near Ōgi, Tsuchimaru, and Hineno Villages. Most of these sites have existed continuously from the medieval age until the present day, but some were reconstructed on old models in a later age. The sites are of many different types, categorized as follows: 128 How Do We Know about Estates?

1. Offices of estate management and governance (mandokoro) 2. Guardian temples and shrines (central religious facilities) 3. Other temples and shrines on the estate (temples on the grounds of guardian shrines, village temples and shrines, sacred mountain precincts) and small-scale shrines 4. Public facilities on the estate (temples and shrines, village squares, meeting places, notice boards, public steam and hot water baths) 5. Agricultural irrigation facilities (dams, canals, and reservoirs) 6. Commercial facilities (markets, shops, inns and post stations, artisans’ workshops etc.) 7. Graveyards and stone stupas (sites of funerals and memorials)11 8. Estate military and police facilities (fortifications for the estate and for local proprietary lords, mountain forts etc.) 9. Roads (highways, pilgrimage paths, village roads) 10. Markers (bōji) for estate borders, and stones, boulders, rivers, and roads that served as division markers

The classifications above are based on research on the estate system up to now. Various kinds of artifacts and sites from medieval estates, viewed collectively, provide material evidence for the economy, production, and daily life examined in research on the estate system.12 Since these facilities often served several functions, many archaeological remains fall into more than one of the categories above. Offices of estate management and governance might be situated at temples and shrines. For example, the Chōfukuji remains (23) and the area presumed to be the location of Muhenkōin (43) were both managerial and religious establishments. Also, many public facilities occupied offices, guardian shrines, and other shrines and temples, and graveyards and stone stupa collections can be found on temple grounds. In many instances, military and police facilities served as offices or estate guardian shrines. Commercial facilities were often housed on the grounds of temples and shrines, including guardian shrines. Markets and inns were found along roads and at the borders of estates. When we try to map the spatial society of estates onto the actual landscape, documents alone are of limited use. As much as possible, we need to find and examine archaeological remains and use them as historical materials for research on estates. Hine Estate in Izumi Province 129

Fig. 5.3. Hineno Village. Drawn by the author.

Reconstructing the Hine Estate Landscape Reconstruction of landscapes is crucial to research on estates. Up to now this work has focused on rice paddies, but in that the physical landscape of estates was also made up of structures and other facilities—vestiges of which remain at archaeological sites—this reconstruction must be based on the sites in categories 1 through 10 above. Physical evidence— elements that made up the estate landscape—also includes hamlets and topographical features such as mountains, fields, rivers, and the ocean. 130 How Do We Know about Estates?

Along with paddy, such features were the most important venues of production and daily life. Reconstruction of the estate landscape requires the study of all these features. Let us begin with Iriyamada Village, in a valley at the foot of the Katsuragi Mountains and divided into Tsuchimaru and the three inner villages, which in turn were divided into numerous scattered hamlets. These hamlets are centered on Buddhist temples (7, 8, 12, 16, and 24). Paddy fields lie along the banks of the Kashii River that flows through the valley. There are also major irrigation works (2), and scattered throughout the three inner villages are additional irrigation facilities: small-scale canals that take water from the Kashii River, and reservoirs in the folds of the surrounding mountains. Today this makes a beautiful scene of terraced paddy (see plates 10 and 11). The Chōfukuji site (23), the location of Kujō Masamoto’s headquarters, and Takinomiya (present Hibashiri Shrine, 18), the location of the village guardian shrine, are situated in central Iriyamada. These were public places with a broad expanse where village assemblies and festivals were held. Temples at hamlet centers also served similar public functions, and graveyards and stone stupas could be found within temple and hamlet precincts. There were no markets, shops, or post stations in the mountain valleys of Iriyamada. Cultiva- tors who lived in the inner villages took the Kokawa Road (62) to a nearby market, where they exchanged their products. There were no local proprietary lords in the three inner villages, but there are remains of mountain fortifications where village cultivators could take refuge. Fortifications in the Tachibana Valley along the Mizuma Road (63) were probably used as defense against the armies of the two military governors during the Warring States period. The mountain temple Shippōryūji (1) was also a place of refuge where cultivators stored their property. Tsuchimaru Village in Iriyamada also lies in a valley, with arable land on the valley’s southern side and at the foot of the mountains south of the Kashii River. Paddy fields were irrigated by canals that took water from that river (31), by the Inakura River, and by reservoirs (38).13 Today there are hamlets surrounding Kasuga Shrine and Gokurakuji (32, 33) in Tsuchimaru, and in medieval times there was a hamlet to the west of Rengeji and its graveyard (36, 37), as well as hamlets in the Inakura River valley. There were two centers in Tsuchimaru, the aforementioned hamlet and Kasuga, the village Hine Estate in Izumi Province 131 guardian shrine. As a subsection of Iriyamada Village, Tsuchimaru participated in festivals and public assemblies at Takinomiya, as well as in prayers for rain at Takinomiya and Shippōryūji, but there were no official structures of the estate. There were no markets, shops, or post stations, and to exchange their produce, people traveled to nearby markets. Tsuchimaru and Ameyama Forts (30), east of Tsuchimaru Village, were mountaintop fortifications used by local proprietary lords and Negoroji, but it is possible that they were also utilized by cultivators.14 Rather than occupying lowland forts in Tsuchimaru, local lords would retreat temporarily to the mountain fortifications for defense. During the Warring States period, Iriyamada cultivators took up arms to oppose the armies of the two military governors of Izumi, and assembled in Tsuchimaru, at the entrance to Iriyamada (see plate 12). Hineno Village is a large village in the lowlands. According to a map of the village dated 1316, it comprised a number of hamlets. (For the map see plate 9 and the cover and dust jacket photographs.) Their rice paddies were irrigated by reservoirs on slopes to the east (45, 55, 57) and by the Yugawa Canal (42) that took water from the Kashii River. Yugawa water was also sent to Jūnidani Pond (57). The long Yugawa irrigation canal and large reservoirs were special features of the Hineno irrigation system. At the center of Hineno Village were the temple Muhenkōin (43), which served as estate headquarters, and the estate guardian shrine Ōizeki (present Hine Shrine, 39). The Yugawa Canal flowed through the shrine precincts and near the temple. Other temples and shrines that served official functions were situated along the canal (40, 47, 48, 49). The central zone of Hineno Village stretched from this area to the site of the temple Iburidō and the ox deity shrine at Shindōde (52, 53). The zone was under cultivation by 1316, as shown in the map of that year. Wilderness lay to the west, according to this map, but it was opened to rice cultivation in the Muromachi and Warring States periods, and new villages (the modern hamlets of Nishiue, Shindōde, and Nono Jizō in Hineno District) were established along the Kokawa Road that led from Ōizeki Shrine to Shiramizu Pond (60). Thus there developed two centers in Hineno Village: the environs of Ōizeki Shrine and the villages along the Kokawa Road15 (see plates 13 and 14). In the Kamakura period, the heavily traveled pilgrimage route from Kyoto to the sacred site of the three shrines of Kumano in Kii province 132 How Do We Know about Estates? ran through Hineno Village. The 1316 map of the village shows a post station along the road, right in front of the gate to the temple Dan- baramitsuji (65). This was the commercial center of the village and the estate. It is possible that there were also artisans’ workshops in the wide area that made up the precincts of Muhenkōin in the Kamakura age. Later, in the early fifteenth century, both Danbaramitsuji and the post station were invaded by armies of the two Izumi military governors, and the post station was moved to the Sano market on the governors’ holding, Sano Estate. The Hineno family of local proprietary lords probably maintained a residence in the village (50), and they also lived from time to time at Tsuchimaru and Ameyama Forts. In the Muromachi and Warring States periods, when the Kujō and Hineno families were at war, the Hineno residence disappeared. Although some ruins still exist, they are not included in the listing of Hine Estate archaeological remains. The residential areas in the hamlets between the Hineno and the nearby Tsukueba sites, however, are thought to be the remains of warriors’ for- tified residences, probably belonging to retainers of the two military governors.16 In summary, the landscape of Hine Estate in the Kamakura period was centered on political and religious sites such as management facilities and guardian shrines for the estate and its villages. Dwellings and hamlets were scattered throughout, and there were wide stretches of wilderness. In Hineno Village, estate offices, guardian shrines, and local proprietary lords may have maintained important irrigation facilities nearby. In Iriyamada Village, there were no local lords from the beginning, and it is thought that the irrigation canals and reservoirs were probably managed by the small hamlets in the vicinity. In the Muroma- chi and Warring States periods, local lords disappeared from Hineno Village as well, and many villages were governed by the cultivators themselves. By this time, too, residents of both villages had assumed control of the numerous local irrigation facilities, temples and shrines, and graveyards. Post stations along the Kumano Road in Hineno Village became the center of commercial activity. Since there were no markets or shops in Hineno or Iriyamada, cultivators turned to the Sano market to buy and sell products. The rice paddy landscape of the present Ōgi, Tsuchimaru, and Hineno districts has not been transformed by modern agricultural development. Rather, it is a traditional rice field landscape that repre- Hine Estate in Izumi Province 133 sents the history of former times. Reservoirs and larger canals found in archaeological remains of the estate supplied the smaller irrigation canals that flow like capillaries through the fields today. We have not yet researched the nature of all these facilities, but we think the rice paddy landscape formed by these small canals is also an archaeological relic of the estate. This landscape would make a valuable contribution to Japan’s system of cultural heritage preservation.17 Remains of the estate can also be found in mountains and fields and along rivers and the seashore, for example, dams in rivers for canals that existed in the time of the estate. There are remnants of village fortifications in the mountains and fields, and it is possible they were also refuges and hiding places for villagers and their property. There has not yet been any research on the exploitation of mountain areas and moors, but through research on mountain place names and livelihoods (gathering edible wild plants and timber, and ash and charcoal making), we may be able to discover additional traces of the estate.

The Structure of Villages on Hine Estate One by one the villages of Hine Estate in Hineno and Iriyamada freed themselves from control by the Kujō and Hineno families, and in the Muromachi and Sengoku periods they established self-governing communities known as sō. I will examine the structure of these villages, based on evidence from archaeology and from the landscape of the hamlets themselves. In the Kamakura period, Iriyamada Village included not only the three sizeable hamlets of Funabuchi, Ōgi, and Tsuchimaru, but also the smaller hamlets of Wakasaki (present Wakasaki in Shimoōgi) and Goshodani (present Goshodani in Nakaōgi), as well as enclosures where cultivators’ dwellings (zaike) were located. These hamlets and enclosures made up the village landscape in medieval times.18 The physical structure of these villages was determined by the results of paddy field land reclamation, the scattered valleys, hollows, enclosures, and slopes created by the small-scale reservoirs and irrigation canals discussed earlier. In the Muromachi period some changes occurred in the structure of Iriyamada Village, according to records of 1417. Its three large hamlets were then termed villages (mura), and Iriyamada contained a fourth village, Shōbu.19 In addition, by this time Ōgi had absorbed the hamlet 134 How Do We Know about Estates? of Wakasaki. The hamlets of the Kamakura period had developed into the four Iriyamada villages, which themselves were composed of small hamlets. For example, Shōbu Village contained Goshodani Hamlet and the hamlets surrounding the village guardian shrine of Takinomiya. Ōgi Village also included Okuma Hamlet, and Tsuchimaru, as mentioned earlier, was made up of three hamlets. Within these small hamlets were places for public assembly and temples and shrines, such as Bishamondō in Goshodani (16). These hamlets were the smallest local communities in Iriyamada.20 Muromachi-period villages on Hine Estate developed features of self-government, some of which were recognizable in village structures of early modern times. Officials called bantō or satanin (headmen) and shikiji were appointed to posts in each of the four villages. The former, selected from prominent village elders, took charge of village management and collected rents and other levies for the Kujō house.21 Accord- ing to Masamoto-kō tabi hikitsuke, headmen were appointed for limited terms on a rotating basis. The four villages were composed of such prominent elders and the families of other cultivators, as well as age- related groups of elders, the middle-aged, and the young.22 Members of the village, whether elders or ordinary cultivators, passed on the fields they cultivated, as well as other properties, to their descendants. Young men’s groups took charge of village defense and policing.23 Village assemblies were held at temples in the villages: Rengeji in Funabuchi (8), Saikōji in Shōbu (12), Enmanji in Ōgi (24), and Gokurakuji and Rengeji in Tsuchimaru (33, 34). The villages managed their own irrigation facilities. Each of these four villages was an independent governmental entity. There were various occupational and religious groups within Iri- yamada Village; for example, the guilds in Funabuchi and Shōbu that supplied specialized wood ashes for the making of dark blue dye.24 Examples of lay believers’ fraternal organizations were the groups that performed fūryū nenbutsu—a festival, parade, and dances in praise of Amida Buddha—at the Takinomiya Shrine. Performers came from two locations, Funabuchi-Shōbu and Ōgi-Tsuchimaru.25 While there were such various internal structures, the four villages combined to make up the greater Iriyamada Village, or Iriyamada Estate in the parlance of local cultivators.26 The four villages worked together as Iriyamada Village to negotiate with the Kujō house and Negoroji monks, defend themselves against the armies of the two Hine Estate in Izumi Province 135

Izumi military governors, and restore normal operations in time of disaster. The “estate” was a combination of these smaller self- governing villages, managed in concert by their headmen. The result was a layered communal structure: hamlets, villages, and the Iriya- mada “estate” itself. In addition, the monks of Iriyamada temples and hermitages formed communal relationships and led the residents’ religious lives.27 According to the late Kamakura period map, Hineno was also composed of small scattered hamlets. New hamlets emerged in the Muro- machi period when the wilderness in the village’s western section was turned into arable land. This was because the extension of the most important irrigation canal, the Yugawa, connected it with the Jūnidani reservoir, enabling the irrigation of uncultivated land and the opening of paddy fields. Local cultivators, including specialized reservoir managers, oversaw the functioning of the Yugawa Canal and the reservoir28 (see table 5.4). Some of the Muromachi and Sengoku period hamlets in Hineno Village continue to exist today. The eastern sector of Hineno, which was cultivated in the Kamakura period, was divided into four hamlets: Mizonoguchi, Kunoki, Butsugan, and Noguchi, while the newly cultivated western portion was made up of Nishiue, Shindōde, Nono Jizō, and Tsujinohana Hamlets. The hamlet of Tsujinohana grew up along the Kumano Road that ran through Sano Estate, which had become the holding of the two Izumi military governors. This hamlet was the smallest community in Hineno.29 Villages such as Kunoki contained temples and shrines (for example Sōfukuji Tenmangū, 49), but we do not know if any of them served as public assembly places like those in the four villages of Iriyamada (see table 5.5). In Hineno Village, too, headmen were chosen from among the village elders. Masamoto-kō tabi hikitsuke reports that Hineno’s eastern sector had three headmen and the western sector seven; they were appointed not to each hamlet but to each of the two sectors.30 According to this source, there was a tendency for especially powerful individuals to serve as headmen in the eastern sector. The boundaries between the sectors are not certain, and with the exception of Tsujinohana, which was in the upper western sector, and Shindōde, probably in the lower western sector, we do not know the locations of any of the hamlets. Each half of Hineno Village, just as in Iriyamada, was made up of a coalition of hamlets, and the halves combined made up the “entire 136 How Do We Know about Estates? estate” (sōshō) according to a term used in Masamoto’s diary.31 Nono miya (48) served as the guardian shrine for both sectors. In the Warring States period, the communities of Hineno Village were layered, from hamlets to sectors to the entire village. The most important irrigation facility, the Yugawa Canal, was operated jointly by the eastern and western sectors. In order to manage the entire irrigation system, which combined long canals with large reservoirs (45, 55, 57), it was necessary for all communities to collaborate as the “entire estate.” In summary, both Hineno and Iriyamada Villages were composed of three layers of communal structures. Moreover, the two large villages worked together on some occasions, such as defending against the armies of the two Izumi military governors and repairing facilities

Table 5.4 Hine Estate communities: Kamakura period

Iriyamada Village (inner villages) Hineno Village

Tsuchimaru Funabuchi Ōgi Eastern sector Western sector Inakura Wakasaki, Okamoto* wilderness Goshodani Ikejiri† Other hamlets * high terrace area † near pond area

Table 5.5 Hine Estate communities: Muromachi, Sengoku periods

Iriyamada Village (inner villages) Hineno Village Eastern Western Tsuchimaru Funabuchi Ōgi Shōbu sector sector Tsuchimaru Wakasaki Goshodani Mizunoguchi Nishiue Inakura Okuma Other hamlets Kunoki Shindōde Butsugan Nono Jizō Noguchi Tsujinohana Hine Estate in Izumi Province 137 following natural disasters. Iriyamada included several types of organizations in addition to its basic communal structure. On Hine Estate, villages were composites of various kinds of communal organizations.

In this chapter I have emphasized the need for research on estates based on archaeological and other on-site investigations. It is difficult to rely only on documents to reconstruct estates as units of local society or to envision the lives of ordinary villagers. Quite a few concrete examples of estates’ physical features, which documents do not record in detail, can be found in archaeological sites or in local traditions. For the reconstruction of estates and villages, on-site archaeological remains are invaluable historical materials that are not inferior to documents. Needless to say, the physical features of estates are not found only as archaeological sites. It may be impossible to reproduce some features of the landscapes of estates based on archaeological findings, and thus we need to depend on documents and other historical materials. We cannot conduct research on estates without using documents, nor can we do so without using archaeological findings. Thus researchers, universities, and governmental institutions have conducted on-site investigations of many estates, including Hine Estate,32 and as I have shown, these findings have then been used to reconstruct the estates’ physical landscapes. The topic of estates is a difficult one even in the study of history within Japan. When learning about estates, it is important to use documents—often difficult to interpret—to understand the way elite courtiers and warriors controlled their landholdings. I think, however, that it is also beneficial to form an image of local society on the estates based on archaeological materials. Such materials have been preserved in museums on the actual sites of estates and medieval villages. For example, the Izumisano Historical Museum, operated by Izumisano City, was established so that people might learn about Hine Estate. There, a model of Hineno Village is exhibited, research is conducted on the estate’s archaeological remains, and maps of the estate (Ōgi, Tsuchi- maru, and Hineno Villages) are shown as guides for those who want to visit its sites.33 The research and displays at this museum, which focus on archaeological remains and the estate’s landscape, are contributions to the investigation of and education about estate archaeology as a whole. Hine Estate was in the central Kinai, the most advanced economic and cultural region in medieval Japan. For this reason, the estate and its residents maintained their ties with the powerful Kujō house in Kyoto, 138 How Do We Know about Estates? and Kujō control lasted for a long time. Accordingly the Hineno family of local proprietary lords was unable to maintain a permanent foothold, and while Hineno and Iriyamada Villages were often invaded by armies of the two Izumi Province military governors, control by warriors did not develop nor did other local proprietary lords dominate the estate. Its villages were managed by local headmen, elders, and cultivators, and alliances of cultivators evolved within each hamlet and village, as well as alliances of sectors within the entire estate. Among the archaeological remains of the estate are many small temples, shrines, canals, and reservoirs that were operated by the cultivators themselves, but there are few traces of connections with local proprietary lords. Medieval estates were introduced to the English-speaking world many decades ago by the historian Asakawa Kan’ichi, with his study of Iriki-in in Satsuma Province (modern Kagoshima Prefecture in Kyushu). Iriki-in came under the control of local proprietary lords of the Shibuya (Iriki-in) family, and Asakawa argues that village communities comparable to those in Western European feudal society failed to develop.34 Iriki-in was one type of medieval estate, but another is Hine in the central Kinai region: established and opened to cultivation by powerful aristocrats, where local proprietary lords never gained control, and cultivators’ village communities emerged.35 This is the complete opposite of the model of medieval estates such as Iriki-in. It is my privilege to introduce Hine Estate as a new model to the English- speaking world.

Notes

1. There are about three hundred items in the historical materials of the Kujō family pertaining to Hine Estate, including documents, diaries, and illustrations. They are preserved by the Division of Books and Royal Graves of the Imperial Household Ministry. Hine-related documents are primarily found in volume 1 of that division’s Toshoryō sōkan—Kujōke monjo (Collected volumes of the office of books—documents of the Kujō House, 7 volumes; see Kujōke monjo 1971–1977). There are also records of the estate in Toshoryō sōkan—Kujōke rekisei kiroku, the diaries of successive heads of the house, volumes 2 through 4; see Kujōke rekisei kiroku 1989–1999. All records of Hine Estate that do not appear in the Kujō materials are found in Shinshū Izumisano-shi shi, vols. 4 (2004) and 5 (2001). 2. Koyama 1981, 1998; Kudō 2002. Hine Estate in Izumi Province 139

3. There has been considerable research on the estate published in Jap- anese. See Masamoto-kō tabi hikitsuke 1996, and for the most recent results of research on Hine Estate, Shinshū Izumisano-shi shi, vol. 1 (2008). 4. This process, known as risshō (instituting an estate) is discussed by Joan Piggott in chapter 1. 5. Both maps appear in the Kujōke monjo. One, dated about 1310, shows Hineno and Ihara Villages. The second, dated 1316, is a map of Hineno Village alone. Also see Tōkyō Daigaku Shiryō Hensanjo 1988–2003, vol. 4 (Kinai 3, 1999). 6. The Nakahara held simultaneous management posts (gesu and kumon) on the adjacent Nagataki Estate, a hereditary holding of the regents’ house. 7. It is not clear why two military governors were appointed to Izumi Prov- ince, but possibly the shogunate wanted to avoid giving the same individual control over both the region near the capital of Kyoto and the region surrounding the heavily populated city of Sakai. 8. In addition to the 1996 volume cited in note 3, which also includes a pho- tographic reproduction of the diary, printed versions of Hikitsuke are included in Masamoto-kō tabi hikitsuke 1961, and there is an annotated version rendered in Japanese word order (yomikudashi) in Shinshū Izumisano-shi shi, vol. 5 (2001). The latter volume also includes research on the personality and character of Masamoto, his personal chronological history, written by Hirota Kōji (Hirota 2011). 9. Representative examples were cited in my presentation titled “Hine no Shō Remains: the Preservation, Exhibition, and Restoration of the Site,” at the conference Reassessing the Shōen System (University of Southern California, June 2012). 10. Ushigami was the protective deity for oxen, draft animals crucial in the cultivation process. 11. Izumisano Kyōiku Iinkai 2001. 12. Rekishikan Izumisano and Shōen Kenkyūkai 2005; Hirota 2006b. 13. For the latest findings on the history of land reclamation in Ōgi and Tsuchimaru, see Ida 2006. 14. Murata 2012. 15. For the latest findings on the history of land reclamation in Hineno, see Hirota 2006a; Rekishikan Izumisano and Shōen Kenkyūkai 2012. 16. Suzuki 2010. Because such archaeological sites were not preserved after excavation, they were not included in tables 5.2 or 5.3 or on the maps. 17. Rekishikan Izumisano and Shōen Kenkyūkai, 2005 and 2010; Hirota 2006b. 18. From a copy of a report dated 1234 on the villages, paddy, dry fields, and dwellings of Hine Estate, in Kujōke monjo 4:246. The references below to 140 How Do We Know about Estates?

Kujōke monjo and Masamoto-kō tabi hikitsuke are from Shinshū Izumisano-shi shi, vols. 4 (2004) and 5 (2001). 19. A report dated 1417 in Kujōke monjo (Shinshū Izumisano-shi shi, 4:363) lists the cash levies, numbers of paddy fields, and rent payments in rice for four villages in Iriyamada. 20. Masamoto-kō tabi hikitsuke, entries of 1501/6/22, 1503/4/3, and 1504/7/8, in Shinshū Izumisano-shi shi, 5:51, 244, and 378, respectively. 21. Registers listing exempt temple and shrine lands from the four Iriya- mada villages dated 1430, Kujōke monjo (Shinshū Izumisano-shi shi, 4:375); land register listing number of rice paddies in Iriyamada and their assessed rents, dated 1430, ibid., 376. 22. Hirota 2001, 788. 23. Masamoto-kō tabi hikitsuke, entries of 1501/6/22 and 1504/4/5, in Shinshū Izumisano-shi shi, 5:53 and 336. 24. Masamoto-kō tabi hikitsuke, entry of 1503/9/5, in Shinshū Izumisano-shi shi, 5:289. 25. Masamoto-kō tabi hikitsuke, entry of 1501/8/13, in ibid., 86. 26. Masamoto-kō tabi hikitsuke, entry of 1502/8/22, in ibid., 187. 27. See the table in Hirota 2008b, 585. 28. Contract for lower Jūnidani Pond (1446), in Fujitake monjo (Documents of the Fujita House), Shinshū Izumisano-shi shi, 4:410. 29. Record of rent payments from Hine Estate (1481), and copy of records of rent payments from the eastern portion of Hineno Village (1482), in Shinshū Izumisano-shi shi, 4:455–470. 30. Hirota 2008a (table on page 443). 31. Masamoto-kō tabi hikitsuke, entries of 1503/10/20 and 12/13, in Shinshū Izumisano-shi shi, 5:297 and 305. 32. Ishii 1995b and Ōsaka Rekishi Gakkai 2006. 33. Rekishikan Izumisano 1996; Rekishikan Izumisano Historical Museum home page, 2007. 34. Asakawa 1929a. 35. Kawane and Watanabe 1987; Kawane 2010. pa rt i i i

Making the Land Productive

c h a p t e r 6

Agricultural Expansion and Irrigation in the Early Medieval Age

Kimura Shigemitsu Translated by Kristina Buhrman

Since the 1970s, the overwhelming emphasis on wet field rice cultivation in the history of Japanese agriculture—the so-called “paddy-centric” view—has been justly criticized. In its place scholars have argued that the investigation of other forms of grain cultivation and agriculture (including that of nonfood crops), of artisanal production, and of fishing and aquaculture are also important for the study of medieval Japan.1 Research since the 1980s has advanced the field through historical studies focused on (1) dry field (primarily nonrice) agriculture, (2) mountain settlements where agricultural production was secondary to forestry and hunting, and (3) the history of labor. I have previously produced studies of the characteristics and productivity of dry field agriculture in the classical and medieval periods.2 In general, moreover, studies involving dry field and nonrice agriculture have attracted attention in archaeology. For example the Japan Archaeological Association selected the theme “Dry Field Archaeology” for its 2000 annual meeting,3 and they have collected a great deal of data on the remains of nonpaddy fields (hatake). These studies, however, have tended to overemphasize dry field agriculture in opposition to the “paddy-centric” view of Japan’s history, constructing a dichotomy wherein dry fields and nonrice agriculture are seen in opposition to paddies and rice agriculture. This weakens our study of the history of Japanese agriculture. Historical geography, by contrast, has focused on the study of mountain communities during the Warring States and Edo periods. In this

143 144 Making the Land Productive field, unlike in the field of the history of agriculture, there has been not only a focus on a binary between rice and nonrice agriculture but also a polarized dichotomy between “communities on the plain,” which are associated with rice agriculture, and “communities in mountain valleys,” which are associated with production of crops other than rice. While this binary too has been strongly criticized,4 it is nevertheless clear how limited any depiction of agricultural production focused only on wet field rice agriculture in the largely mountainous geography of the Japanese archipelago would be. Following these trends in research, the most promising recent work has appeared in studies of the history of labor in Japan. Here the emphasis has been on a holistic approach that includes not only wet field rice agriculture, dry field agriculture, and nontilled production from the land (such as the harvesting of fruit or lumber), but also nonagricultural production from such sources as fisheries.5 An example is Haruta Naoki’s study of medieval fishing villages (kaison). Recently, by reconstructing the annual labor cycle of communities in Echizen and Wakasa Provinces (modern-day Fukui Prefecture), Haruta has shown that labor and production in such “seaside villages” comprised a complex mix of dry field agriculture and fishing. Considering the balance of production, Haruta has been able to divide such communities into four basic types: those that produced dry field crops (and rice) and manufactured salt, those that manufactured salt and engaged in fishing (along with dry farming and exploiting mountain fields), those that engaged in fishing and salt manufacture (along with rice growing and silkworm raising), and those that exploited mountain fields and raised silkworms while engaging in fishing and salt manufacture.6 It is possible to see from this short introduction that for current research on the history of agriculture in medieval Japan, a focus on agricultural production and techniques of wet field rice agriculture is out of date. What is required is a more holistic approach that incorporates not only dry field agriculture and the use of nontilled land, but also fisheries and other nonagricultural sources of production in the surrounding region. And to continue advancing the field, it is also necessary to evaluate wet rice agriculture, focusing particularly on the irrigation techniques that enabled it. This will correct some preconceptions in the field. Here I will show some of the significant aspects of the history of irrigation while explaining their role in recent research concerning the history of agriculture in medieval Japan. Agricultural Expansion and Irrigation 145

Examining these issues has great significance for understanding the medieval estate system. Most established estates at the time were holdings with a defined territory and borders, and estate holders aimed at possession of all cultivated fields, dwellings, forests, ponds, and open fields and moors within these boundaries. The nucleus, however, was cultivated fields, both paddy and dry fields. On some estates, such as Iketa in Ōmi Province, the proportion of dry fields was fairly high: 164- plus chō of dry fields versus 385-plus chō of paddy. While paddies were the basis for proprietors’ control, dry fields provided cultivators’ livelihood.7 Thus it is important to assess the productivity, management, and expansion of both paddy and dry field agriculture within the context of the medieval estate system. I will begin with a focused look at dry field agriculture.

Characteristics of Dry Field Agriculture in the Classical and Medieval Periods When discussing dry field agriculture in classical and medieval Japan, we must first emphasize the productive capabilities of this form of agriculture to correct misconceptions embedded in earlier historiography. Before the major shift in the historiography of Japanese agriculture discussed above, dry field agriculture was seen as unreliable and less productive than wet field rice agriculture. Therefore it was thought that dry field agriculture was used only to supplement wet rice agriculture, as an emergency source of grain most useful in times of disaster or famine. So let us first consider the reliability of dry field agriculture. Toda Yoshimi’s work on “half-fallow” (kata-arashi) agriculture—a system in which fields were planted only every other year to preserve their productivity—has made the reliable nature of dry field agriculture clear.8 Specifically, Toda analyzed fifteen field surveys dating from 990 (Eiso 2) to 1059 (Kōhei 2). His analysis of these surveys of an estate under the control of Eizanji Temple in Yamato Province shows that all the productive fields there over these many decades could be classified into two types: fields that produced reliably across surveys and fields that did not. The latter could be explained as kata-arashi; they were fields that were left fallow in alternating years. I have produced my own analysis based on Toda’s results, focusing on those fields that were considered reliably productive. Of these, three out of four were marked with an abbreviated character that means it was a dry field.9 Those fields that Toda showed 146 Making the Land Productive to be reliably productive were primarily fields used for dry field agriculture, not paddies used for rice agriculture. The importance of dry field crops for classical and medieval Japa- nese agriculture is further shown by the practice of double-cropping on hatake fields. In field registers (kenchūchō) from estates in the late Heian period, the term “summer dry field” (natsuhata) is found alongside the term for dry field, hatake. While this term has been noted in previous studies, that the term itself indicates the practice of double-cropping has yet to be fully recognized. Specific evidence for this can be found in the 1160 (Eiryaku 1) register for Yuge Estate in Iyo Province.10 In this document there are two entries for dry fields: at one point we see “hatake, nineteen chō five tan 138 bu,” and we also see “natsuhata, nineteen chō five tan 138 bu.” Here the exact same acreage appears for “dry fields” and “summer dry fields,” implying that the same fields were used for crops twice in one year. The document also includes information on assessed levies or rents (jishi), making it clear that “dry field” denoted acreage planted in the summer and harvested in the autumn, while “summer dry field” denoted acreage planted in the winter and harvested in the early summer. In the case of Yuge Estate, the same acreage—the very same fields—produced crops twice a year. Further evidence for double-cropping comes from documents discussing how levies were to be assessed on the land under cultivation on estates, particularly where there are references to “summer rents” (natsu jishi). From examples such as that of Yuge Estate above, what the “summer” of “summer rents” means becomes clearer: it indicates rents levied on a second crop each year. In other words, the term “summer rents” is evidence for the practice of double-cropping in the cultivation of dry fields. The term “summer rents” also appears in a supplementary regulation (tsuikahō) issued after the promulgation of the Kamakura shogunate’s Jōei Formulary (Jōei shikimoku, 1232). Entitled “On whether military stewards should take rents on fields twice in one year (spring and summer),” the regulation begins by noting that the proprietor was already assessing rents twice a year, indicating that both crops from double- cropped fields were being subjected to levies by this time.11 From these documents and others, it is clear that the practice of collecting rents on dry fields twice a year became widespread from the late Heian to the early Kamakura periods. That such twice-a-year rents were based on the practice of double-cropping on dry fields is indisputable, Agricultural Expansion and Irrigation 147 as shown in the case of Yuge Estate. The productivity and importance of dry field agriculture at the beginning of the medieval period in Japan is therefore beyond dispute. Dry field agriculture was marked by its variety. In historical records related to land in the late Heian period, a number of terms referring to various types of dry field agriculture can be found: tabata or hatada, both of which denote a “paddy dry field”; yamahata, or “mountain dry field”; katahata, or “fields not cultivated every year”; and kichihata, or “stable fields.” Although it is difficult to specify the distinctions intended by each term, the variety speaks to the diversity of modes of dry field agriculture. That the growth of dry field agriculture was an important part of medieval Japan’s agricultural expansion is presaged by documents from the classical period, from the eighth to the tenth centuries. From 715 (Reiki 1) to 840 (Jōwa 7) for instance, the Japanese court bureaucracy periodically issued decrees called “Measures to Increase Production of Miscellaneous Grains [i.e., nonrice crops].” Currently, eight such decrees can be confirmed, as seen in table 6.1.

Table 6.1 Extant decrees: “Measures for the Increased Production of Miscellaneous Grains”

Year Japanese Year Content of Decree 1 715 Reiki 1 Each adult male shall additionally sow barley and millet. 2 722 Yōrō 6 For warding off famine, plant late-ripening millet, buckwheat, barley, and wheat. 3 723 Yōrō 7 Sow and harvest barley and wheat. 4 766 Tenpyō-jingo 2 Plant barley and wheat. 5 767 Jingo-keiun 1 Expand the cultivation of mulberry. 6 820 Kōnin 11 Plant barley and wheat. 7 839 Jōwa 6 Sow buckwheat and millet. 8 840 Jōwa 7 Cultivate dry fields. For support in bad years, plant two kinds of millet (kibi and takakibi), barnyard grass, barley, large and small beans, and even sesame. 148 Making the Land Productive

In these decrees we see that barley, millet, and buckwheat were among the “miscellaneous” grains the court encouraged the people to cultivate. The 840 decree (promulgated on 5/2) notes the largest number of nonrice crops. But what is more important for the discussion of agriculture in classical Japan is that subjects were being ordered to cultivate rikuden, another name for dry fields. In other words, this decree encouraged the cultivation of dry field crops in general; that is why a variety of dry field crops, from millet to sesame, are included. The decree of 840 is different from previous decrees, which encouraged the cultivation of specific crops rather than a regime in which dry field agriculture played a prominent role. The innovative nature of the 840 decree is further evidenced by an appended note: “However, this is not a reason for neglecting paddy fields, nor for converting them to dry fields.” The purpose of the decree of 840 was to encourage the cultivation of dry fields, but this addendum makes clear the appeal this form of agriculture had in this period: by forbidding the conversion of paddy fields to dry fields or the abandonment of wet rice agriculture for dry field agriculture, the decree shows precisely that such acts of conversion were being undertaken by contemporary cultivators. Were that not the case, no such addendum would have been required. In short, the 840 decree shows that dry field development had entered a different stage. No longer was it merely meant for the production of nonrice foodstuffs as a buffer against famine in years when the wet field rice crop was poor or failed. By the middle of the ninth century, cultivators themselves were expanding the scale of dry field agriculture as they recognized its productive and reliable nature.12 While it cannot be said that dry field agriculture had precisely the same productive potential as wet field rice agriculture, the above discussion shows that the productive capacity of dry field agriculture was considered high in early medieval Japan and even during the late classical period. The reliable productivity of land under dry field cultivation and the practice of double-cropping attest to this. Furthermore, as the re-evaluation of decrees concerning “Measures for the Increased Produc- tion of Miscellaneous Grains” in table 6.1 shows, the potential of dry field agricultural production was recognized by the cultivators themselves. Were that not true, they would not have been inclined to abandon wet field rice agriculture for dry field agriculture. This point, more than any other, undermines the old “paddy-centric” historiography of Japanese agriculture. And it is why the old preconceptions must be revised. Agricultural Expansion and Irrigation 149

Cultivators’ Self-Management and Dry Field Agriculture Next I will examine the role of dry field farming in cultivators’ management of their own agricultural labor. I will argue here that dry field cultivation was the basis for agricultural management by cultivators themselves. A report of 1129, written by a legal expert, makes the following points concerning dry fields on Tōdaiji’s Kuroda Estate in Iga Province:

These dry fields are not under the authority of the provincial office, because they are [an element of] the residences of the cultivators. And the produce from dry fields cannot be appropriated by the Pro- vincial Office. This is a legal precedent of the province. The Provincial Office is not to take charge of rents from dry fields.13

In short, the dry fields noted above were under the independent control of the Kuroda resident-cultivators. Therefore they were outside the control of the provincial government, which could not assess a levy on them. According to the report, the independent nature of dry field agriculture—not just within Kuroda Estate but elsewhere in Iga Prov- ince—had been previously established. Meanwhile another document from the end of the century shows the cultivators of a new township (ho) called Koinumaru in Harima Province emphasizing the same points in their petition of 1197:

If it should come to pass that dry fields are reduced, where shall the local people live, [and who] shall perform their corvée duties? If the residences here are confiscated, or the people who live here have their livelihoods obstructed, who will plant and harvest the wet rice fields or present tribute goods?14

Herein too the cultivators at Koinumaru insisted that dry fields were equivalent to the very residences of the local cultivators, and that should these residences be confiscated, no one would be left to work the fields for wet rice cultivation or to provide goods and services for the state. This is the same position taken by the cultivators of Kuroda Estate at the beginning of the century. Such documents show that dry field agri culture and the dry fields themselves were under the management of estate residents, who also controlled the harvest from those fields. 150 Making the Land Productive

One final example comes from the middle of the fourteenth century. In this document, which represents a protest by the cultivators of Tara Estate, a Wakasa Province property of the Kyoto temple Tōji, the resident-cultivators clearly assert the critical importance of unhin- dered dry field agriculture. The document states, “All provinces have recognized that cultivation of millet is necessary for the undertaking of agriculture.”15 Here, the cultivation of millet—in other words, dry field agriculture—is seen as a requirement for all agriculture, including wet rice cultivation. The Tara residents assert that this is the case not only for them, but also for the entirety of Japan. As these three historical examples make clear, in early medieval Japan—that is, from the latter part of the Heian period through at least the Northern and Southern Courts (Nanbokuchō) era (1336–1392)— cultivators understood that growing dry field crops was the essential foundation of a system that also included wet rice agriculture. It is therefore clear that it is impossible to discuss the agricultural practices of medieval cultivators while considering only wet rice agriculture. With- out examining the role of dry field agriculture on estates, the nature of medieval Japanese agriculture cannot be fully understood.

Major Characteristics of the Large-Scale Expansion of Agriculture in Japan’s Eleventh Century It is now necessary to re-evaluate previous depictions of the expansion of Japan’s agricultural production that began in the middle of the eleventh century in light of the characteristics of dry field agriculture and the role it played. Previous studies have often uncritically assumed that this expansion of agriculture involved the direct transformation of virgin fields, forests, and mountainsides into rice paddies. This style of development, however, was not typical. A number of important preconditions underlay the expansion of agriculture in medieval Japan. One example is found in a famous document from 1012, a copy of an order from the provincial headquarters of Izumi Province. It orders larger- and smaller-scale agricultural specialists (tato) to develop cultivated land that will be under the control of the provincial government, and it notes, “All agricultural specialists, great and small, are ordered to make ‘old cultivated fields’ (kosaku) as well as ‘wasted fields’ (arata) productive.”16 Agricultural Expansion and Irrigation 151

A close examination of this document shows that the development mandated here did not depend on the opening of virgin lands. Rather, “old cultivated fields” refers to fields that were already developed, that is, fields that were being used contemporarily for wet rice cultivation. Fur- thermore, the “wasted fields” here were paddies that had fallen out of use. Thus it seems that the goal of this provincial order was the renewed cultivation of “wasted fields” that had fallen out of use, together with the continued cultivation of current wet rice fields. These objectives are even more explicit in other statements from the same document, such as, “For example, in developing new wet rice fields, should a cultivator abandon old fields he has been cultivating, he is disobeying the contents of this order.” In addition, the document promises that those who “add the production of [former] wasted fields to that of the old cultivated fields” would be forgiven miscellaneous duties levied on paddy plus five shō of the provincial rice duty. This reward for following the order clearly indicates the provincial government’s goal of preserving the production of currently cultivated wet rice fields while spurring cultivators to bring abandoned fields back into production. Dry field agriculture also played an important role in the development of the new township known as Hisatomi in Harima Province in the mid-eleventh century by Hata no Tametoki. From the historical record we can reconstruct the process whereby Tametoki “opened” these fields:

• The fields inherited by Tametoki that were already being cultivated were producing mulberry and flax. • Tametoki noted that the Hisatomi New Township included “dry fields, passed down by my ancestors as the land of our residence (ryōchi yashiki).” • Several years after Tametoki petitioned for permission to open new fields, about fifty chō of acreage had been developed, including the original five chō two tan. • The “newly developed” fields were opened by rehabilitating an irrigation system that had fallen into disrepair (arai mizo), which “although its traces remained, was difficult to restore.”17

In analyzing this record we should first note that there were originally dry fields in cultivation in Hisatomi, which Tametoki inherited as residential land passed down from his ancestors. As shown above, such dry fields were attached to the residences of local cultivators, and 152 Making the Land Productive they helped to support other aspects of the agricultural production in the area. Second, aside from the fields used for the mulberry and flax crops mentioned as Tametoki’s inheritance, there were additional fields (5.2 chō) that were “presently in cultivation” (tōsaku). These fields are equivalent to the old production fields noted in the 1012 decree from the Izumi Provincial Office; both terms indicate fields that were currently in use for agriculture (whether for wet field rice cultivation or other crops) when the document was written. Furthermore, the fields to be “opened” included an irrigation system that had fallen into disrepair. In other words, the fields had been previously irrigated for wet field rice agriculture, but the ditches and such were not currently in use and existed only as “traces.” Bringing the field anew into cultivation involved restoring the old irrigation system in need of repair. From these two points, the means by which the new township of Hisa- tomi was developed can be summarized as follows. First, the nucleus for development was the existence of dry fields around a homestead and already in cultivation at the site. Second, a prerequisite for developing new fields for wet rice cultivation was a relatively small acreage of wet fields already under cultivation. These pre-existing fields were kept in production, and the fields subsequently brought back into cultivation were abandoned ones that had retained the remnants of an irrigation system. Notable too is the fact that the way in which Hisatomi New Township was developed matches the instructions laid out in the Izumi provincial headquarters order of 1012. Therefore, it appears that the wide-scale expansion of agricultural production in the eleventh century was in fact the redevelopment of previously abandoned fields on the periphery of fields that had remained in production. There are not many historical cases that state as clearly as that for Hisatomi the connection between residence dry fields, fields already in production, and the reopening of rice fields. But the instructions in a 1066 order from the monks of Nara’s Gangōji Temple to one Hase tsukabe no Tamenobe—that he redevelop “paddies that have turned into wilderness” within the township (gō) of Yanase in Iga Province— exhibits an aspect of this pattern of development.18 Here, among fields that had turned into wilderness, there were about seventeen chō of currently productive wet fields among countless waste fields. In order to develop these waste fields, the currently productive fields were a necessary precondition. Agricultural Expansion and Irrigation 153

This precondition is implied in yet another case from the end of the twelfth century. In 1171, Nakahara Yorisada transferred his inherited rights to Shigetomo Village, an area outside the boundaries of Itazu Estate, to one “Lord Sukejirō” (Sukejirō-dono). These fields were currently “all a loss” and “abandoned land.”19 In transferring the property to Sukejirō, Yorisada wrote, “The village in question is near your residence. Invite wanderers to live there without delay, and encourage them to undertake agriculture according to your instructions.”20 To “encourage agriculture” (kannō) meant to develop new fields while continuing existing production. Since Shigetomo Village fields were all “a loss” or “wasted,” what is indicated here must be redevelopment of the land. In other words, Yorisada is advising Sukejirō to invite those without permanent abodes or means of livelihood to reopen fields in the area. This redevelopment was possible because the land in question was close to Sukejirō’s residence and his pre-existing fields. Here too the opening of new fields for agriculture or the restoration of old fields that had gone to waste depended upon the existence of residences and fields already in the area. Actual examples of agricultural land development during the large-scale expansion of agriculture from the eleventh through the twelfth centuries in Japan allow the following conclusions. First, pre- existing residences were the basis for new development. Second, new development depended upon the existence of currently productive dry or wet fields around these residences. There is no evidence showing direct development of virgin plain or wasteland into rice paddies. Instead the grand expansion of agriculture in medieval Japan was the extension of existing agriculture, as cultivation was expanded from the wet and dry fields surrounding residence nuclei to abandoned fields beyond. This leads to the third point, which is that the “opening” of fields during the medieval agricultural expansion meant the restoration of “abandoned paddies,” that is, land that had suffered damage or had been otherwise abandoned. Just as in the case of Hisatomi New Township, the typical form this development took involved the restoration of “traces” of old irrigation systems. In other words, the medieval expansion of agriculture in Japan was more the “reclamation” of former agricultural areas than the opening of virgin land. 154 Making the Land Productive

Irrigation Ponds and Channels: Preconditions for Agricultural Expansion Some previous studies have depicted a tension between resident- proprietors or other management-class individuals who wanted to redevelop rice paddies around their residences (yashiki) and members of the resident-cultivator class whose goal was to develop dry fields around their residences (zaike). While this depiction is generally correct, the two forms of agricultural expansion were not necessarily in opposition. A number of examples, including the case of Koinumaru Township in Harima Province, show that resident-cultivators too could express a strong desire to establish new rice paddies. For instance, in the Koinu- maru petition we read:

Even one drop of water is difficult to obtain on this land. There- fore the people who have lived in this place since time immemo- rial have devised strategies and exhausted their efforts to excavate and construct ponds to irrigate the rice paddies, and so they were able to pay their yearly rent.21

The cultivators of Koinumaru “exhausted their efforts”—they employed both material and labor resources—to create this pond. In other words, cultivators too exhibited a strong desire for the means to expand wet field rice agriculture. Furthermore, the late twelfth-century cultivators of Koinumaru themselves dug ponds. There is a similar example, although it comes from more than a century later, in Higashimura, a corporate village (sō) that was the property of Kokawadera Temple in Kii Province. During the late Kamakura through the Northern and Southern Courts era, the village is known to have constructed a number of irrigation ponds to support wet field rice agriculture.22 Notably, the cases of Koinumaru and Higashimura differ, however, from the pattern exemplified by the opening of Hisatomi in Harima Province. At Hisatomi, Hata no Tametoki’s agricultural expansion involved the rehabilitation of an abandoned irrigation system involving ditches. In other words, the “opening” of areas for wet field agriculture by the local proprietor class largely utilized irrigation systems of this type. It is possible to contrast two styles of land development. Even when the aim Agricultural Expansion and Irrigation 155 of developing wet fields for the cultivation of rice was the same, the village cultivator classes used ponds to develop wet field agriculture while the local proprietor class used irrigation systems that drew from existing waterways. The two styles can be found in the development of fields within the area of Kaseda Estate in Kii Province, a holding of Jingoji Temple in Kyoto. As depicted in a famous map of the estate, the first stage of development drew irrigation water from a cluster of retention ponds created at the foot of the hills at the north side of the property.23 The second stage of development, which was undertaken by the proprietor Jingoji itself, utilized the “Mongaku-i” irrigation system,24 which drew its water from the Shizu River that flowed at the northern edge of the estate. And one more two-stage example of land development is found at Hine Estate in Izumi Province, the property of the Kujō-Fujiwara aristocratic lineage. There the first stage of development also utilized retention ponds constructed in the foothills on the northern side of the property, while the second stage drew water from the Yugawa Canal, which took water from the Kashii River that flowed from the eastern to the southern parts of the estate. Utilizing both methods, the wasted fields that spread across the center of the estate were brought under cultivation.25 There are not yet many studies that cover the development of irrigation, which was itself essential for the development of wet field rice agriculture.26 Yet while further investigation is necessary, the above examples nevertheless show that there were two styles of land development: one that utilized retention ponds and another that utilized ditches and canals to draw water from rivers. This method of land opening by the proprietor class gave birth to new conflicts in local society. For example, in the latter part of the twelfth century, a conflict arose between the Buddhist institution on Mount Kōya and the Kii provincial government over the provincial office’s efforts to restore the Ayai irrigation system in Naka District.27 Kōya officials asserted that the provincial government, in moving the diversion channel for the Ayai irrigation system upstream, had destroyed about two tan of its estate’s dry fields. On the other hand, Kii Province officials counterclaimed that Mount Kōya was threatening some hundreds of chō of rice paddies in order to protect a small number of dry fields. According to the provincial government, Mount Kōya’s “petitions will not yield on this point, but if the restoration [of the irrigation system] is 156 Making the Land Productive not undertaken, the loss to the polity will be unprecedented.” The two positions could not be reconciled: Mount Kōya sought to protect the dry fields under its control, while the provincial office was prioritizing irrigation for rice paddies on which taxes were levied. For the provincial office, the state’s wet field rice production took precedence over a small area used for dry field agriculture. A similar situation, in which a provincial governor sacrificed dry field agriculture for the restoration of wet field rice cultivation, can be found in early twelfth-century Kōzuke Province.28 An irrigation canal about twelve kilometers in length and known as the Ōnabori was found at the base of Mount Akagi. An irrigation channel of that scale could have been excavated only by the provincial government. Recent archaeological excavations have shown that the Ōnabori was damaged by the 1108 eruption of Mount Asama and that it was then re-excavated to bring it back into use. Meanwhile, under the packed-earth bank on the southern end, the remains of many dry fields were discovered. Thus in that area, the restoration of productive agriculture after the eruption damaged the channel began with the conversion of the area to dry field agriculture. Only once the agricultural production from the dry fields had become reliable did state power—represented by the governor of Kōzuke—sacrifice the dry fields to redig the Ōnabori channel to restore wet field rice cultivation. In both cases above, it is difficult to say for certain whether local inhabitants resisted the efforts of provincial authorities. It is clear in both cases, however, that the sacrifice of dry fields for the excavation of irrigation channels and the subsequent rededication of productive agricultural land to wet rice cultivation reflects the thinking of provincial administrators, in other words, the prioritizing of wet field rice agriculture by the ruling class.

Agricultural Conditions at Ōbe Estate and Irrigation Next I will examine the type of cultivation and level of productivity found on Ōbe Estate in Harima Province, an estate of the temple Tōdaiji in Nara. Although this estate has been the subject of excellent studies,29 much about its early agricultural development remains unclear. Using the research of Hashimoto Michinori and Kawabata Shin, however, and combining it with information about customary irrigation practices in Agricultural Expansion and Irrigation 157 the area from the Edo period on, as well as current geographical analysis, we can determine the following elements of the initial agricultural process:

1. Cultivation expanded on areas of lower elevation on the estate after the early classical epoch. The Ōike and Shimonoike irrigation ponds in the vicinity of the present Kumano Shrine were at the center of this development and the main source of water for irrigation. In addition to water from these two sources, it is possible that water was also drawn from the Tōjō River. 2. Agricultural expansion during the early Kamakura period was centered on Jōdoji, which was built on a slope at higher elevation, and it continued from there to the opening of other fields in middle-elevation areas. At this time, the most important sources for irrigation were the northern and southern irrigation ponds, Kita Ike and Minami Ike, respectively. A later goal of field development in middle-elevation areas was connecting the area to land already in cultivation on the lower slopes.

Since these conclusions come from geographical analysis of information on agricultural practices from the early modern period onward rather than from analysis of surviving historical documents, a careful evaluation of these stages is difficult. For the time being, therefore, I will accept this model and use it as a basis for further analysis. It is clear that the geography of Ōbe Estate made it difficult to irrigate fields by drawing water from the Tōjō River, with the exception of some lower elevation areas. Retention ponds, therefore, were absolutely essential for successful cultivation. Despite the difficulty in maintaining a sufficient irrigation level, it is clear upon examining the sources that there are unusually few records concerning dry fields and dry field cultivation. This is not because there were no dry fields to be found on Ōbe Estate or that dry field cultivation did not occur there. It is merely because Tōdaiji, the proprietor of the estate, and the temple’s agent Chōgen, who administered the estate, had a common goal: the increase of cultivated paddy fields. The most important element for the successful cultivation of fields in the middle elevations during the Kamakura period was the construction of irrigation ponds and techniques of dis- tributing water from them, techniques associated with the local proprietor and cultivator classes. 158 Making the Land Productive

To examine how the agricultural situation developed on Ōbe Estate after the early Kamakura period, I turn to an analysis of naikenchō, records of surveys of agricultural land that survive from the Northern and Southern Courts era. In the three years between the surveys noted in table 6.2, the total number of paddy fields, the number of “old cultivated fields,” and the number of “normal cultivation” fields did not change. This shows that the area designated for wet rice cultivation at Ōbe Estate in this period was fixed as far as the documentary record was concerned. Although it is difficult to determine whether these figures truly reflect the agricultural productivity of the land of Ōbe Estate, it is still possible to discern that the number of fields on which the proprietor could assess rents (the tokuden) were few and that the number of fields “repurposed for irrigation” (kawanari)—land that was repurposed for irrigation channels or used for irrigation ponds— and fields that had a failed harvest were many. In terms of rents the proprietor Tōdaiji might receive, the fields of Ōbe Estate presented a low yield on the total registered acreage. Among the roughly 167 chō of “old cultivated fields” in 1337, 56 chō 7 tan, or about 34 percent, had rents assessed on them. Among the normally cultivated fields, the number of rent-assessable fields made up just 28 percent of the total. These numbers did not change much in 1340. So at that time, only approximately 30 percent of the total number of rice paddies at Ōbe Estate could be cultivated. This shows clearly the difficulties in managing Ōbe Estate. Precisely what the difficulties were is revealed by a comparison of the number of fields lost to flooding or needed for irrigation with those that suffered other losses (see table 6.3). So we see that in both surveys those paddy fields that were repurposed for irrigation or otherwise flooded outnumber those that were unproductive for other reasons, and both types of fields further outnumber those fields on which the proprietor could assess rents. Given such information, we can tell the story of the agricultural situation on Ōbe Estate in the fourteenth century. To begin with, many of the fields where the harvest was declared a loss were marked as having suffered from drought. This indicates that the planting that was performed in the spring was damaged by summer drought or insufficient irrigation water, resulting in a failed harvest. In other words, the water needed for irrigating rice paddies was broadly Table 6.2 Cultivation at Ōbe Estate in the Nanbokuchō period

1337 Survey 1340 Survey (Ono-shi shi #393) (Ono-shi shi #409) Total number of 226 chō 8 tan 25 tai 226 chō 8 tan 25 tai wet fields Old Cultivated Fields (kosakuden) Fields repurposed for 11 chō 6 tan 15 tai 11 chō 6 tan 15 tai irrigation Loss that year (1337) 45 chō 9 tan 45 chō 9 tan 30 tai * Flooded (1337) 51 chō [?] tan 57 chō 7 tan 10 tai † Fields upon which a rent 56 chō 7 tan 50 chō 1 tan 30 tai can be assessed Total 167 chō 8 tan 5 tai 167 chō 2 tan 5 tai Normally Cultivated Fields (jōsakuden) Unsuccessful harvest 12 chō [? tan] 40 tai 10 chō 1 tan 30 tai Flooded 15 chō 1 tan 35 tai 12 chō 1 tan 5 tai Fields upon which a rent 10 chō 5 tan 15 tai 8 chō 3 tan 30 tai can be assessed Total 37 chō 7 tan 40 tai 37 chō 7 tan 40 tai Very Productive Fields Loss 4 chō 8 tan 20 tai Flooded 8 chō 3 tan 30 tai Fields upon which a rent 5 chō [? tan] 30 tai can be assessed Total 21 chō 8 tan 30 tai “Rent-exempt” fields, etc., were not included. One bu = 3.95 square yards. The tai, an earlier measure that was no longer standard in medieval times but was used in some records, was equal to five bu. * Loss due to drought in 1340 † Newly repurposed for irrigation in 1340

Table 6.3 Approximate area of “loss” and “flooded” fields at Ōbe Estate

Old Cultivated Fields Normally Cultivated Fields 1337 Repurposed for irrigation and Flooded: 15 chō 1 tan flooded: 62 chō 9 tan Losses: 45 chō 9 tan Losses: 12 chō 1340 Repurposed for irrigation (old and Flooded: 15 chō 5 tan new): 69 chō 3 tan Lost to drought: 45 chō 9 tan Losses: 10 chō 1 tan 160 Making the Land Productive insufficient at the estate. Furthermore, since “repurposed-for-irrigation fields” were so numerous, we know that lack of water remained a problem despite attempts to improve the irrigation situation by creating new waterways and drawing more water from rivers and ponds. These attempts clearly did not work as intended, and the result was a large acreage where wet field rice cultivation was not successful. Although an attempt was made at Ōbe Estate during the Northern and Southern Courts era to create conditions for stable and successful wet field agriculture with river water or retention ponds, the large number of fields taken over for irrigation or in which the harvest failed show that this attempt did not go according to plan. The most important problem facing Ōbe Estate during this period was that of securing a stable supply of water for irrigation. These circumstances also explain why Ōbe Estate had so many fields designated as “rice fields to support irrigation” (iryōden). Such fields provided the capital to support the construction of ponds or ditches to ensure the continued operation of the irrigation system at the estate. In a 1364 document labeled “Calculation of Yearly Rents from Ōbe Estate,”30 the cost of maintaining “the ponds and irrigation systems for the villages” is set at sixteen koku five to of rice. The same document puts the amount of rice produced by a single rent-assessable “old cultivated” field at four to five shō per tan of cultivated land, and the amount of rice produced by a “normally productive” field at two to five shō per tan of cultivated land. Using this as a base for calculations, the acreage of fields set aside for irrigation support would have ranged from three chō seven tan to six chō six tan. If an average of approximately five chō one tan is assumed, then the acreage dedicated to supporting the maintenance and improvement of the irrigation system is greater than that given over to the military steward—about four chō two tan—by almost an entire chō. This calculation shows the extent of the costs consumed by Ōbe Estate’s irrigation system, including for both the pond and irrigation channel components. This information further points to the need for a re-evaluation of how agriculture was expanded on Ōbe Estate. The development of land at middle elevations of the estate centered on Jōdoji Temple at the higher elevation, and at that time, irrigation was largely based on water drawn from the Kita Ike and the Minami Ike ponds. As the surveys from the Northern and Southern Courts era show, however, the acreage Agricultural Expansion and Irrigation 161 of fields that were unproductive or sacrificed for irrigation purposes, as well as the amount of land dedicated to supporting irrigation economically, was the single most important challenge facing Ōbe Estate. In other words, while previously the opening of fields at the middle elevation using Kita Ike and Minami Ike ponds has been seen as a great success on the part of Chōgen and other monks working for Tōdaiji, the truth is that this effort did not proceed smoothly. Even into the Northern and Southern Courts era, the problem of insufficient water for irrigation continued. The end result was that the fields upon which rents could be assessed by Tōdaiji did not exceed 30 percent of the set number of fields within Ōbe Estate. The struggle against insufficient water is attested to by the many retaining ponds still found in the area that was once Ōbe Estate, now part of Hyōgo Prefecture’s Ono City. The problem of securing enough water for irrigation does not seem to have been limited to the medieval period. At the very least, it can be stated that Ōbe Estate’s history was marked by the battle for water and that the fight to secure enough for agriculture continues to this day.

In the first four sections of this chapter, I presented an outline of how the expansion of agriculture in Japan’s medieval period took place. In the last section, I showed how this information helps to analyze the specific agricultural conditions on Ōbe Estate in Harima Province. A great deal of progress has been made in understanding the history of medieval agriculture in Japan. By abandoning a preoccupation with wet field rice agriculture and including a wider variety of agricultural and craft production—such as dry field agriculture and the products of mountain villages—a new picture of medieval agriculture has come to light. We now know that the expansion of medieval agriculture as practiced on Ōbe Estate hardly advanced satisfactorily. The main problem was maintaining a secure source of water for irrigation to create a stable production base. On Ōbe Estate it was particularly difficult to secure the irrigation water needed to open new paddy fields for rice cultivation. The continuing battle to do so was a characteristic of Ōbe Estate’s middle ages. The case study I have presented shows that future research into medieval agriculture will require not only further investigation into general aspects of agricultural production but also detailed research 162 Making the Land Productive into how this general picture relates to and informs specific conditions on individual estates. If my study here makes it clear that only by combining both types of research will the field advance, then I will consider it a success.

Notes

1. The seminal work that called for such a shift is Amino 1980a. 2. Kimura, “Chūsei seiritsu ki ni okeru hatasaku no seikaku to ryōyū kankei.” In Kimura 1992. 3. Papers from this meeting are collected in Nihon Kōkogaku Kyōkai 2000. 4. See Komeie 2002. 5. Kokuritsu Rekishi Minzoku Hakubutsukan 2008. 6. Haruta 2010. 7. Kimura 2013, 128–129. 8. Toda 1967. 9. These fields are marked using phrases such as「乍（作）已白」, 「已白」 and the like. Here, the character for “white” (白) serves as an abbreviation for hatake (畠), as it forms the top part of that character. 10. HI 7:2512–2514 (# 3119). 11. Supplementary regulation (tsuikahō) # 45, in Chūsei hōsei shiryōshū, vol. 1, 1955, 82. In actuality the rents were assessed in summer and autumn. 12. See Kimura, “Nihon kodai no ‘rikuden’ to hatasaku.” In Kimura 1992. 13. See HI 5:1849–1853 (# 2147, 1129/2/3). 14. KI 2:225 (# 912). 15. Tōji hyakugō monjo, ハ 13–20. 16. HI 2:630 (# 462). 17. See HI 3:1076–1077 (# 1059, 1071/6/25), 1126 (# 1113, 1075/4/28). 18. The order can be found in HI 3:1037–1038 (# 1002). 19. HI 7:2791 (# 3570). 20. Ibid. 21. KI 2:225 (# 912). 22. Kuroda 1985. 23. For a copy of this map, see Shōenshi Kenkyūkai 2013, 89. 24. There are a number of theories as to precisely when this system was excavated. See Kaizu 2011. The system takes its name from a famous Jingoji monk, Mongaku, who was active in land reclamation projects. 25. For more on Hine Estate irrigation, see the chapter by Hirota Kōji in this volume. Agricultural Expansion and Irrigation 163

26. Some exceptions are Hōgetsu 1983 and Ōyama Kyōhei, “Chūsei ni okeru kangai to kaihatsu no rōdō hensei,” in Ōyama 1978. 27. HI 7:2531–2532 (# 3153, 1161/5). 28. Minegishi 2001. 29. For example, Hashimoto 1995; Kawabata 1996; and OS 1997 and 2002. 30. OS 4:711–714 (# 443). c h a p t e r 7

Loggers and Cultivators of Nabari Tōdaiji’s Kuroda Estate in Heian Times

Joan R. Piggott

The courtly world pictured so gloriously in the The Tale of Genji dominates our fondest visions of the Heian age (794–1185). In this chapter, however, I shall explore less familiar landscapes in the countryside south and east of the capital district, where during the Heian period cultivators staged an ongoing assault on uncultivated land, in mountain valleys, and on intramontane alluvial plains. Historians agree that in early Heian times as much as 50 percent of arable land was not yet under cultivation, which provided both challenges and opportunities for those at every level of society. As Karl Friday has observed in his study of the Heian military, the countryside was ground zero for fierce competition among court nobles in the capital, middling officials of various sorts, and locals who wanted to secure greater profits from the soil.1 Hundreds of extant records—there are slightly more than 360 dating from Heian times—as well as an extensive historiography, enable a close study of Kuroda Estate, owned by the official temple Tōdaiji (Great Eastern Temple) and located in old Iga Province (in today’s Mie Prefecture).2 Moreover, as will become clear here, Kuroda’s history is not just that of an estate or local place. Kuroda Estate was never isolated, and its stories are as much accounts of the development of court government, provincial government, temple organization, center- periphery relations, local society, and agricultural technology as they are of estate and local history. This study of Kuroda Estate is also informative for ongoing debates about agricultural advance versus stagnation in the Heian period.3 And as Ishimoda Shō wrote in his classical study of Kuroda Estate, the estate’s extensive archive makes possible an expansive and deep investigation of

164 Loggers and Cultivators of Nabari 165 what he thought of as “the birth of Japan’s medieval world.”4 Although contemporary historians have problematized some of Ishimoda’s assumptions and interpretations—for instance, his view that the official temple Tōdaiji and its estate cultivators represented “the classical world and its state-based slavery” compared with the more independent “serf- like cultivators” of the medieval age—his book is still widely read, and it has made Kuroda Estate very well known. I myself read Ishimoda’s account alongside Kuroda’s written record early in my career as a Japan historian, and I have always wanted to write Kuroda’s story. Given that opportunity here, I have gone back to Kuroda’s records for a close reading, following Kimura Shigemitsu’s advice to see the documents as “landscapes in words” from which to gain clues about the formation and operation of one of Heian Japan’s most important estates.5

Nabari District in Ritsuryō Times Some knowledge about Nabari District during the eighth to the tenth centuries helps to illuminate the place where Kuroda Estate came to be in the eleventh century. Nabari was one of four administrative sub- divisions (gun) in Iga Province, and it was traversed east to west by the Eastern Sea Circuit Road (Tōkaidō) that linked the capital district with coastal eastern Honshu. Early court chronicles such as the eighth- century Nihon shoki and Shoku nihongi record royal progresses to Nabari by early monarchs including Tenmu (d. 686, r. 673–686), Jitō (645–702, r. 690–697), and Shōmu (701–756, r. 724–749). Records also indicate that by the early tenth century the district had three additional administrative units: Nabari and Suchi Townships (gō) and Natsumi Village (mura).6 In the eighth century Nabari residents cultivated officially distributed rice fields and paid taxes to the district office, which operated under the supervision of the provincial office in Ueno in Ahe District. Traces of checkerboard-like rice paddies—opened under official auspices with officially produced iron tools—have been identified along the Nabari River (see figure 7.1). The district chieftain had had a role in opening those fields, and he was responsible too for maintaining the irrigation works that watered them. It was also his job to maintain registers of the populace farming those fields, and he forwarded tax revenues to the capital at Heijō (Nara) in neighboring Yamato Province; many wooden documents found at the Heijō palace confirm the key role of district 166 Making the Land Productive chieftains in the collecting and expediting processes.7 The Nabari district chieftain is thought to have resided near Unane Shrine in today’s Nabari City, where he officiated as ritualist. That shrine, which still exists, has a long and venerable history. It is one of two official shrines in Nabari listed in the early tenth-century handbook of official procedures, the Engishiki.8 The place where it stands, called Natsumi in the Heian period, was a central place even in proto-historic times: archaeologists have identified tomb mounds there, as well as a seventh-century temple site and the residence of an eighth-century local notable.9 A tax record dated 730 even names a member of the Natsumi family serving on the Nabari district chieftain’s staff.10 To the west of Nabari the mountains that formed the border between Iga and Yamato Provinces included a logging camp that sent lumber to build Tōdaiji in the mid-eighth century. The area was called Itabae Soma, and its loggers cut and then rafted timbers by river to Heijō.11 In the mid-eighth century, bureaucrats and monks who staffed the temple’s construction agency, Zō-Tōdaijishi, organized and managed such efforts. Records from the operation are preserved in the Shōsōin Archives (Shōsōin hennen monjo).12 To this day, hikers can visit the ruins of a temple complex called Moharadera that likely served as the camp headquarters.13 And according to later records, loggers built homes with gardens and fields at the foot of the mountain forest in Nabari District. We do not know to what extent loggers continued their labors for the temple after Tōdaiji was completed in the 760s. In the mid-tenth century, however, Tōdaiji’s monastic director Kōchi (894–979) began an ambitious construction campaign to halt the temple’s physical decline, and he asserted Tōdaiji’s claim not only to the forest but also to settlements at the foot of the mountain, out to the Nabari River, since loggers had built homes and opened rice fields there.14 And there was another link between Iga and the Nara temple: in 950 a monastic administrator confirmed receipt of benefice rice payments (fumai) from one hundred residence units (ko) in Iga. Such units, called “benefice residence units” (fuko), paid a portion of their taxes to Tōdaiji to support monks, rituals, and temple buildings.15 The arrangement resulted in frequent contacts among Tōdaiji, provincial authorities in Ueno, and local officials.16 Since Nabari was the part of Iga closest to Nara, and a logging forest was located there, it was a likely area to pay benefice rice payments. Indeed a record from 999 cites the existence of a Nabari storehouse at Komō for tax proceeds from the temple’s benefice households.17 Fig. 7.1. Kuroda Estate. Drawn by Arnie Olds. 168 Making the Land Productive

Nabari in the Ninth and Tenth Centuries, and Komō(fu) Pasture Demographic and administrative landscapes across the archipelago were changing in the ninth and tenth centuries. For instance, archaeologists have uncovered evidence that townships and villages organized around official fields in alluvial river valleys lost population, perhaps because it proved difficult for local officials to maintain the needed irrigation works.18 In the written record we also find descriptions of “vagrants” who left their places of official registration on the plain to open new fields in mountain valleys. In such places, which were essentially off the old administrative grids, small valley fields were watered by springs or by ponds dug by the land openers themselves.19 These developments were not illegal: earlier in the eighth century the New Fields Law of 743 had encouraged just such development when it directed that rice paddies opened by such individual efforts, once approved by provincial authorities, should be partially tax-exempt and eternally heritable.20 In the mid-ninth century too, court orders like that of 852 advised provincial officials to do all possible to encourage land opening.21 The new mountain valley populations were dispersed, so cultivators of the valley fields needed to cooperate to marshal needed labor for annual planting and harvesting. A result was complex family units and new communal bonds.22 A place where such entrepreneurial land openers thrived was in Yamashiro Province, near the Heian capital. The record states that over the course of a century more than three thousand acres had been successfully opened there by 896.23 The cultivators lived in partially fenced homesteads with gardens and occasional paddies in areas of three to four tan (about an acre or more), including both dry and wet fields.24 Monks at official temples in the old Heijō capital were trying to charge rent on the newly opened fields, which they claimed were located within the temples’ logging forest. The homesteaders resisted, however, and as a union of residents they protested to the provincial governor, who sent their complaints on to the court for resolution. Happily for the homesteaders, the court denied the temples’ claims, but the cultivators were ordered nonetheless to pay a share of the provincial land tax on their fields. So were old townships on the plain being replaced by new valley settlements whose headmen represented them to higher officialdom. In mid-tenth century Nabari, wet rice paddies were appearing within the private holding (shoryō) of a Council of State advisor, Fuji- Loggers and Cultivators of Nabari 169 wara Asanari, at Komō.25 This was also a busy time in Nabari forest land, where loggers were cutting and sending lumber to Tōdaiji for temple repairs, and the temple director Kōchi was increasing his claims to land at the foot of the mountains. In this environment a quarrel with Asanari broke out in 964. Tōdaiji insisted that houses and new fields at the base of the mountain forest belonged to the temple, as its loggers’ homes.26 Local officials, ordered to investigate, reported that given the conflict with Asanari they could not confirm either side’s assertions.27 Unlike in Yamashiro in 896, in Nabari in 964 the temple director and Advisor Asanari had to negotiate their own resolution. Specifically, in 965 Kōchi bowed to Asanari, likely due to the latter’s high status at court. Tōdaiji’s director wrote to local authorities that he recognized the difference in functions between the logging forest and pasture land and that therefore he had no objection to their erecting official markers to identify Asanari’s holding, which included rice fields, uncultivated open land, and chestnut orchards. For Tōdaiji’s side of the settlement, however, logger cultivators at Komō were to be exempted from any special labor taxes that might be levied by the court, since they were the temple’s laborers.28 Nonetheless, three decades later Kōchi’s successor, Hyōsū, who was an intimate of the powerful court leader Fuji- wara Michinaga (966–1027), was able to go further in 999: he secured a clear deed to “Komō no shō,” Komō Estate, which was then recognized as being part of the temple’s forest.29

Getting the Nabari Plain Recultivated in the Eleventh Century Although land at the foot of the mountains and under the protection of private holders like the minister Fujiwara Asanari and Tōdaiji was being opened gradually in the mid-tenth century, by the early eleventh century Iga’s provincial governors were also busy developing strategies to get the old checkerboard fields east of the river recultivated. Specifically, they recruited middling officials to manage the reopening of fallow fields.30 In return for fixed tax responsibilities that included the promise of profits, field contractors (tato) were engaged to oversee the organization of needed resources and labor, cultivation, and tax payment. In Iga a mid-ranking official from the Left Stable Bureau in the capital, Fujiwara Sanetō (d. 1061), became a contractor on the Nabari plain. We do not know much about Sanetō himself, but there is an anecdote 170 Making the Land Productive in the late Heian compilation Konjaku monogatarishū about his father, Kiyokado, who worked in the Ministry of the Treasury around the year 1000 and contracted to manage fields in the three provinces of Yamato, Yamashiro, and Iga. Of particular note, Kiyokado had a house in Nabari District. When he failed to pay taxes on his Yamato fields, the governor there reportedly locked Kiyokado up with a bunch of feral cats, of which Kiyokado had great fear, to get the tax bill collected.31 The son’s relations with Iga authorities is not mentioned in the record, but when questioned one village elder reportedly called him “a big shot” in the province.32 Indeed, when Sanetō made a list of his property holdings in the mid-1050s, he claimed twenty-eight properties across Iga’s four districts. As for his management technique, Sanetō reportedly recruited cultivators; provided them with tools, seed, food, and temporary lodgings; and then collected the entire harvest as rent, as had early tenth-century governors who rented out extra, undistrib- uted fields. Like his father, Sanetō would have been expected to pay the provincial field tax from his harvested fields to provincial storehouses, from which benefice rice likely went to Tōdaiji.33 In the 1020s, however, Sanetō claimed to be suffering failure in his attempts to restore old dikes and ditches on the Nabari plain. Two decades later, in 1041, he nevertheless requested a new deed for what he termed “hereditary property,” including wet and dry fields in Yagawa, where the river split. He explained that “all the residents have died or run off, no one has lived there for decades, and there is nothing but brambles and fallow fields.”34 He needed the deed to sell the property to none other than the current monastic director of Tōdaiji, Shinkan (1003–1050). When called in to investigate and confirm the request, the local headman and the Nabari district chieftain confirmed Sanetō’s claims, including the property’s borders.35 The process was slow, however; it took five years for Shinkan’s proprietorship to be established, with the provision that Shinkan would pay the basic land tax but no special levies on newly opened (or reopened) fields, conditions similar to those for Asanari’s property in Komō. Just a few days later, however, a second document certified Shinkan’s exemption from even provincial land taxes at the newly recognized Yagawa Estate36 (see figure 7.1). It seems likely that the provincial governor and the temple had reached an agreement to the effect that the exempted taxes at Yagawa would cover some portion of Iga’s benefice rice obligation.37 Shinkan enjoyed a special advantage in achieving this favorable settle- Loggers and Cultivators of Nabari 171 ment: he was not only the respected head of the temple of the Great Buddha but also a prince, a son of the deceased monarch Kazan Tennō (968–1008; r. 984–986).38 Nevertheless, monastic directors and provincial governors came and went, and every change of office necessitated new negotiations. Shinkan himself died in 1050; his influence may have preserved the peace in Nabari until then. After that, however, disputed issues included Tōdaiji’s ongoing relationship with Yagawa Estate and whether provincial agents should be able to enter the property to survey and track the expanse of its fields as cultivation increased and the river’s course shifted. Accord- ing to the deed, Shinkan and his heirs—whether disciples or relatives— were to be recognized as private proprietors who would receive “added rent” (kajishi) from cultivators.39 As for Tōdaiji’s interests in the area, a report by Sanetō to Tōdaiji dated 1043 stated that “outside fields” (dezukuri) across the Nabari River boundary from “Kuroda Estate” had long been accepted as Tōdaiji’s property.40 According to Sanetō, then, an entity called “Kuroda Estate” already existed in 1043, and it included fields across the Nabari River such as those in Yagawa. On the other hand, the extant archive of Kuroda documents includes not a single official record that mentions “Kuroda Estate” until 1053, when provincial officials were aggressively disputing the estate’s expansion beyond the river.41 The gap between the two views resulted in a great deal of contention, judicial and military. Around this time provincial governors all over the archipelago were gearing up to enforce new estate-regulating laws promulgated in 1045. In addition, increasing numbers of special tax levies were being ordered to support the rebuilding of the royal palace and official temples, as well as to finance the every-twenty-year reconstructions of Ise Shrine.42 So in Iga in 1053 provincial officials announced that they would henceforth consider the Nabari River as the irrefutable boundary between the fully tax-exempt Itabae Forest and the official domain where fields and cultivators were to be fully subject to provincial authority, taxes, and extraordinary labor levies. On this basis they proposed to remove boundary markers that had been erected beyond the original riverbed, including some set in place by order of the previous governor, whom they excoriated as foolish for ignoring crimes by the temple.43 Against these actions, Tōdaiji—whose director was then Yūkei, himself a governor’s son—contended that the Nabari River had been changing course over time and thus squeezing the living 172 Making the Land Productive and cultivating space of temple loggers. Provincial officials nonetheless rejected the argument and were determined to collect provincial taxes beyond the river using whatever force was needed. Unfortunately for the temple, the officials of the Board of Control- lers (Benkankyoku) at court agreed with the province’s views, and the provincial headquarters—the governor, his administrative team, and several resident provincial officials (zaichōkanjin)—received a controller’s order (senji) in the spring.44 By harvest time, however, and despite the fact that agents previously sent to Nabari by the controllers and the provincial office had met with violence—agents’ clothes, cattle, and horse gear were all stolen, while estate residents (jūnin) beyond the old riverbed had had their homes (yashiki) and belongings burned— the temple was able to reverse the earlier order against its interests.45 There was more violence in 1055, when the provincial governor himself tried to enter fields between the old and new riverbeds that cultivators insisted were within the “Kuroda Forest.” And while controllers at court inveighed against the violence, they were unable to decide which fields were new or old or to implement specifics for collecting the provincial land tax. The quarrel was complicated still further by the temple’s plaint that Iga’s benefice rice had not been paid since 1053, leading the temple’s monastic council, known as the three deans (sangō), to insist that “the outside fields” opened by loggers be exempted from all taxes to make up for the unpaid benefice rice.46 In its own defense the province argued that there were numerous nonloggers living across the river whose fields and homesteads, about two thousand chō, should pay taxes. Such complex issues and complicated politics gave the principals—in the old and new capitals of Heijō and Heian-kyō (Kyoto) and in Iga—a lot to argue about. Various methods to resolve the dispute were tried: particular units of named fields across the Nabari River were assigned to pay benefice rice to Tōdaiji, with the proviso that those units should be surveyed yearly by provincial authorities and that the land tax would be collected and paid to Tōdaiji from provincial stores, not by cultivators themselves. There was still the issue, however, of aggrieved cultivators whose crops had been harvested and carted away during the fighting of 1053, as well as of those who had lost their homes in Yanase, Kuroda, and Yagawa.47 Who were these cultivator-residents of the Nabari plain? Clues come from a description of a successful agricultural specialist from around 1050. Loggers and Cultivators of Nabari 173

Tanaka no Toyomasu makes farming his livelihood and has no other occupation. He is a patriarch, a large-scale field boss whose authority covers several chō of land. In advance he ponders the possibility of drought and keeps his plows and hoes in good condition. With- out fanfare he considers the needs of both fertile and infertile soils and provides himself with horse- and ox-driven plows. He looks after field hands and cultivators when they are busy building dams, dikes, ditches, and paddy ridges. And he excels at overseeing the sowing and care of seedlings, preparing paddies, and assembling youths and maidens for planting in the fifth month. The rice he grows— whether early or late strains, whether glutinous or nonglutinous— and his harvesting techniques are superior. And the rice he polishes increases year by year. Moreover, his upland crops—barley, soybeans, cowpeas, adzuki beans, foxtail millet, millet, barnyard millet, buckwheat, and sesame—ripen to the last seed. When he plants a single seed in spring, it has multiplied ten-thousand-fold in the autumn. From planting until harvest he never makes a mistake, and he constantly knows the joy of ripened crops and bounteous harvests. He has yet to face losses from drought, flood, locusts, or failed fields. He never fails to provide sustenance for officials who come to survey his land or collect taxes; and he has never reneged on official levies, whether rent or taxes on his fields, special paddy imposts, produce taxes, dues in lieu of rice, special assessments, agents’ fees, provisions for the governor, levies on local products, rice-wine and seed- grain imposts, operating expenses, rice for exchange, rent on the governor’s fields, interest on official rice loans, or levies on officially assigned paddies. Although his is not the wealthiest of tax-paying households, it is far better than that of the usual poor farmer who is always pleading impoverishment.48

We can assume that the prosperous land-opening cultivators targeted by provincial officials at Kuroda Estate in the 1050s would have resembled Tanaka. They would have owned draft animals and iron tools: plows, hoes, spades, sickles, and axes.49 Archaeological evidence from the time suggests that their homesteads were dispersed, although the remains of four to five dwellings, each housing four or five persons, are sometimes found in clusters.50 As for their crops, we know that mulberry, flax, and fruit trees were planted between the old and new river beds and beyond the river.51 In fact, according to Kuroda Hideo 174 Making the Land Productive and Kimura Shigemitsu, who have studied the agricultural technology of the time, cultivating such dry field crops was the first stage of opening an area to wet rice cultivation.52 We also know that land openers at Kuroda used the labor of servants, relatives, and sharecroppers, and that fifteen cultivators in Nabari, likely using fields beyond the river, were paying benefice rice from thirty-nine chō to Tōdaiji by 1057.53 Their losses in 1053 and 1055, however, led outer field residents to claim they were unable to pay taxes, rent, or timber supplies to anyone. It was likely an attempt to avoid further violence that led the Board of Controllers at the Kyoto court to send an order in late 1056 directing that estate markers should be replaced by its own agents and that the restored Kuroda Estate would be exempt from entry by provincial agents. Cultivators who resided therein were to pay all their dues to the temple to support its repairs and for protection of the realm.54 Against this the provincial governor and his staff insisted on the need for annual field surveys even on the mountain side of the river, since loggers living there were cultivating fields and storing the harvest from beyond the river. This proved a successful argument, which enabled Governor Ono Moritsune to have estate markers removed again in 1058 and which resulted in increased payment of levies by Kuroda cultivators.55 We need to put Ono’s success in historical perspective: at this point in the mid-eleventh century, the court was relying more on martial provincial governors to “pursue and destroy” tax rebels; for instance, an army of “pacification” led by the military governor Mina- moto Yoriyoshi was sent against the Abe of Mutsu for nonpayment of taxes in 1051, just as an earlier force had been led by Yoriyoshi’s father Yorinobu against Taira Tadatsune (967–1032) of Hitachi in 1028. Ono, like these Minamoto military governors, came from a family known for martial skills. By appointing him as governor in Iga, the court showed itself to be serious about enforcing tax collection even in cases involving great official temples like Tōdaiji. Still, land openers in Nabari continued hard at work, and generally benefited from the temple’s backing. A record of 1066, for instance, provides a window on how locals cooperated with temple authorities to get fields in Yanase just across the Nabari River opened. That year the Tōdaiji director notified his monastic council that a local opener named Hase(tsukabe) Tamenobu was quickly to recultivate fourteen chō of fallow fields that had once belonged to Sanetō. This name, “Hase,” was one that had been associated with the district chieftain’s office Loggers and Cultivators of Nabari 175 going back to the eighth century. As for the terms of this deal, in the first three years he would pay no rent. Later, however, he would pay the provincial land tax, likely as part of the temple’s benefice rice, plus an additional one to per tan as rent to a private holder who was none other than the monastic director himself. For his part, Tamenobu was to be permitted to pass down cultivators’ rights (sakute) for the property to his heirs.56 How did Tamenobu open these fourteen chō ? There are no details in the record but another land opener in Harima Province reportedly utilized five thousand corvée laborers to repair a well and irrigation ditches with earthenware conduits to open thirty chō there in 1075.57 We can be sure that if Tamenobu could command the needed 2,500 laborers to get the Yanase land cultivated, he must have been a local notable with significant wealth, connections, and authority, just the sort of man Tōdaiji’s director needed to get the job done. Notable too is the complexity of land holding on the Nabari plain at this time: while the province claimed it as official domain where fields and cultivators must pay taxes, nevertheless part of those taxes were to go to the temple as benefice payments, while private holders also collected rent. Surely such circumstances led to quarrels, which principals would have hoped to avoid. That is likely why in 1080, when units in Yagawa and Nakamura neglected to pay both taxes and rent to private proprietors, the temple itself ordered those responsible to pay up. And in 1087 temple authorities ordered logger residents in Yagawa and Nakamura to pay the levy demanded by the private proprietor in Kyoto.58 Behind the issues may have been the new regime of taxation emerging at this time: in addition to the standard land tax and rent for private proprietors, landlords and provincial authorities were trying to collect a new levy on produce from dwellings (zaike), which often included gardens growing dry field crops and rice paddies. It is in the records of the 1080s that we gain our first glimpse of organized local leadership by the estate management staff (shōkan) at Kuroda. For example, in 1086 the temple’s monastic council sent orders to two senior managers called gesu and several residents.59 The basic unit of the estate at the time was the “named field unit” (myō) with its boss. By the twelfth century, more staff members had been added, including a registrar (kumon) and a sentō, whose role is unclear.60 And in the early 1100s, a document from the staff of the temple’s forest in Kuroda Township indicates that loggers working there had their own headman, the tōryō.61 176 Making the Land Productive

There were other changes on the ground in the 1080s as well. For one, the Nabari District chieftain, Hasetsukabe Chikakuni, likely a kinsman of the field boss Tamenobu, allied himself with the provincial office and turned against the temple. As a first step, Chikakuni dispatched gifts of lumber, carpenters, and silk to the provincial office. In 1088 he received a gift in return: an order from the provincial governor presented him with two holdings in the Yagawa area, including a portage that gave him strong influence over the shipping of timber along the river.62 Empowering Chikakuni was no doubt a strategy by the governor and his team to battle Tōdaiji’s ever-increasing influence in Nabari. Indicative too was Chikakuni’s promise of 1092 to enforce collection of provincial taxes in his jurisdictions of Yagawa and Naka- mura.63 And yet the promise proved hard to keep: in 1097 authorities complained that 380 chō of outside fields were being cultivated by temple dependents, from which 1140 koku of taxes in rice (equivalent to 3,800 bolts of silk) were unpaid.64 Given the debt, the province wanted the temple to share in its payment: it could not, it let the temple know, pay benefice rice under such circumstances. Still another issue was a heated dispute between the heirs of Sanetō in which monks from three different monasteries were involved, with some violence, that resulted in the confiscation of a unit that paid benefice rice to Tōdaiji.65 Such was the worrisome local situation in Nabari as the twelfth century dawned.

Hearings, Land Surveys, and Increased Contestation in the Twelfth Century The twelfth century witnessed great changes politically and socially across the archipelago. With court leadership increasingly dominated by retired monarchs who had ambitious palace- and temple-building plans, and whose intimates included provincial governors eager to enforce estate controls, the century brought renewed campaigns in many provinces to map cultivated fields and register estate dwellers. Iga was no exception. At the same time the temple ordered its own surveys and tightened relations with its logger-cultivators dispersed over the Nabari plain. Both sides prepared detailed documentation and arguments to win political and judicial contests in Kyoto, where provincial agents constantly challenged the temple’s proprietary claims to Kuro- da’s outer fields and cultivators. Loggers and Cultivators of Nabari 177

To provide efficient defense of its interests, temple administrators in Nara formed the new Shōen Office (Shōji). It was headed by a capable monk named Raishin, who acted as a sort of vice president for estate affairs. At the time Tōdaiji claimed twenty-one properties in the five provinces of Yamato, Yamashiro, Iga, Settsu, and Mino, with those in Iga and Mino the largest and most valuable.66 Another strategy embraced by Tōdaiji’s monks to safeguard its estates was the compilation of the temple’s history and key documents in its fundamental records, the Tōdaiji yōroku.67 Such materials were critical for helping the temple’s leaders make persuasive arguments about the importance of Kuroda and other estates to fulfill the vow of Tōdaiji’s eighth-century founder, Shōmu Tennō. The temple’s mantra for the era, which first appears in a petition to the court dated 1106, was “the decay of Tōdaiji means the decay of the realm.”68 Indeed that mantra fit contemporary circumstances well. In 1109 a significant judicial battle pitted the Iga governor against temple authorities over the issue of “added fields” (kanōden)—that is, the “outside fields” across the Nabari River. To prepare for hearings, sheaves of documents dating from the 1070s circulated among the Nara temple, courtier officials of the Controllers’ Board, Tōdaiji’s director (who frequently resided at Tōji or some other temple in the capital), and the two offices of the Iga provincial governor, one at his residence in the capital and the other at the provincial headquarters in Ueno. The sub- stance of the governor’s complaint was that Tōdaiji had stolen official fields all over Nabari District, that since 1104 at least two thousand koku of taxes remained unpaid, and that an annual survey of all outside fields was necessary to correct the situation. Tōdaiji’s arguments were that if its logger-cultivators worked fields only on the mountain side of the old riverbed, they could not support themselves or pay lumber to the temple as well as benefice rice, and that temple dependents should be allowed to pay a lower provincial land tax of two to per tan (or the equivalent of one hiki of silk for every five to of rice) rather than the usual tax rate (three to per tan) since they also supported the temple with their logging efforts. Legal experts at court charged with evaluating these arguments and the evidentiary materials submitted with them opined that Tōdaiji was due 180 koku of benefice rice from Kuroda Village, but that those cultivating fields across the river should pay the provincial land tax.69 After this report, which was only one step in the judicial process, more documents were called up, scrutinized, and evaluated late 178 Making the Land Productive into 1112 in preparation for the trial, known as a taiketsu, in which both sides would face each other as well as expert questioners. It was about this time too that Tōdaiji and Kōfukuji—the latter being the Fujiwara family temple in Nara and long a rival to Tōdaiji in Iga and elsewhere— were embroiled in a bitter quarrel over the boundaries of their respec- tive shares of the mountain forest above Yagawa and Nakamura. Given that Nara’s Kōfukuji was managed by the Regents’ Line (Sekkanke), the regent’s household tried to settle the argument, but dissension continued because Tōdaiji and its dependents rejected what they called the prejudices and ineptitude of regental agents.70 Here is a sign of the weakened ability of the Regents’ Line to lead the court by the second decade of the twelfth century. One can only imagine the worries of cultivators in Nabari as the ceaseless disputation went on and on. In 1114 and 1115, the temple encouraged them to be at ease while resisting illegitimate demands by provincial authorities. In late 1115, however, Nabari loggers submitted their own petition to the temple asking for assurance of exemption from the provincial labor levy and special levies they were currently being forced to pay. Without such exemption, they warned, they could not supply the thousands of boards demanded from them for ongoing repairs of the Great Buddha Hall in Nara.71 Since a subsequent record indicates that those repairs were actually completed in 1116, the temple must have negotiated that exemption with court leaders. The temple director at the time was Shōkaku, the son of a high-ranking courtier, and his successor, Kanjo, was the son of a controller who was an intimate of the retired Shirakawa. We know too that Shirakawa visited Tōdaiji on his way to Mount Kōya in 1124, indicating his royal interest in the temple of the Great Buddha.72 Indeed the record suggests that Tōdaiji and its Kuroda cultivators were enjoying reasonably good times. We know, for instance, that provincial authorities were anxious to collect taxes in 1121 from the one hundred temple-affiliated settlers who had built homes east of the river at Yanase, Yagawa, Nakamura, and Natsumi, suggesting their prosperity and sense of security beyond the riverbed. On the other side, provincial officials—their documents now being signed by the provincial governor’s representative (mokudai) and a few resident provincial officials—were in a dark mood. They could make little progress against the expanded Kuroda Estate: in 1122 they calculated that three thousand koku were in arrears.73 They insisted that they needed to be able to survey Loggers and Cultivators of Nabari 179 all fields regularly, collect taxes, assess taxes on all fields and house lots across the river at the rate of three to per tan, and pay out the temple’s benefice rice directly from provincial stores. When the court ordered the temple to have its logger-cultivators send in their defense against such proposals, the estate manager responded that he had orders in hand sealed by former governors who had confirmed the status quo: twenty logger-cultivators working for the temple were exempt from provincial labor taxes, wherever they lived in Nabari, while other units of fields across the river paid the temple’s benefice rice.74 Those units, the manager insisted too, had been long exempted from extraordinary levies because of their services to the temple in Nara. In 1129 the dispute between the two sides again resulted in violence, in which provincial agents were said to act “like barbarians” while estate dwellers “swarmed like wasps” whenever officials tried to survey fields.75 Estate boundary markers were again torn out, outer fields were confiscated, and benefice rice was withheld. Notably it was at this point that the temple took to calling its estate dwellers “slaves of the Buddha” (Hotoke no nuhi) and “temple slaves” (jinu) to strengthen its argument for exempting them from taxes.76 Still, in 1132 the court ordered special tax collectors to survey fields and assess taxes even on Tōdaiji’s holdings in Iga to support the rebuilding of the royal tutelary Ise Shrine.77 This was the historical moment when Tōdaiji’s “Kuroda New Estate” (Kuroda shinshō) appears in the record as a new designation for outer fields in Yagawa, Nakamura, and Natsumi. The temple’s objective was doubtless to establish a new level of protection for the outer fields by transforming them into a fully recognized estate, and we must suppose too that Tōdaiji had support at court for this new territorial entity. A survey done in 1134 by estate managers listed fifty-three chō, of which twenty-three chō were located between the old and new riverbeds and planted mainly with mulberry bushes. It showed too that the New Estate had been divided into thirty-nine myō of modest size.78 Barley as well as rice was cultivated, with the former predominant.79 The largest of the New Estate units was Naka Village, with thirty-three myō. That may be why it was the target of violent confiscation by the Nabari district chieftain that very year.80 Even as Tōdaiji and its clients in Iga launched the New Estate strategy, other challenges to Tōdaiji’s interests were emerging from within the estate itself. In 1135, for instance, there was a report that twenty Kuroda residents were sending lumber to Kasuga Shrine in Nara. Given that 180 Making the Land Productive

Kōfukuji and Kasuga, both entities belonging to the Regents’ Line, had frequently quarreled with Tōdaiji, this was seen as a treasonous act for Tōdaiji dependents. Furthermore, there were complaints to the temple by fourteen Kuroda cultivators (hyakushō) about unfair exactions and confiscations by the temple’s monastic representative. Accusers wrote that the representative was behaving just like the “barbarian” provincial agents, utilizing undue force to collect dues that he had judged unilat- erally to be in arrears.81 To all this dissension was added the crushing failure of the New Estate strategy. In 1144, at the request of the Iga provincial gover- nor, who reported that even a royal police (kebiishi) agent had been unable to settle trouble given the temple’s “extreme greed,” the court promulgated a royal order that “Tōdaiji’s usurpation” of Yanase and other parts of Iga Province should cease. Moreover, the governor’s complaints—about cultivators’ resistance to field surveys by provincial agents, nonpayment of provincial land taxes for outer fields, and nonpayment of labor taxes by those living beyond the old riverbed— reached the household office (in-no-chō) of the retired monarch Toba (1103–1156; r. 1107–1123).82 So that high office added its own order confirming the court’s directive, commanding that all of Tōdaiji’s properties in Iga, including Kuroda, abide by the orders of the Iga provincial office.83 An order sent by the Board of Controllers in the first month of 1145 indicated that an official would be sent to Iga to join with provincial and temple agents to collect the land tax from every village of the official domain. Of those taxes, 360 koku were to go to the temple as benefice rice, while the balance, about six hundred koku, would go to the province. All Kuroda Estate residents living across the river were to pay the provincial labor tax, and Yanase, the sector of the estate that lay furthest out from the river, was declared to be in the provincial domain.84 This series of orders of 1145 meant that the status quo favoring the temple had ended; historians have called it “Tōdaiji’s defeat.”85 Why did Tōdaiji suffer this defeat? A likely possibility is that the provincial governor at the time, Fujiwara Nobutsune, was an intimate of the retired monarch Toba and that his arguments carried more influence than did those of the Nara temple.86 To cope with the crisis, the temple turned to a capable monastic administrator named Kakunin, who took charge of the Estate Office around 1149. The historian Kudō Keiichi has dubbed him “the armed Loggers and Cultivators of Nabari 181 monk (akusō) from Nara who brought Tōdaiji’s holdings in Iga to their mature form.”87 Kakunin first appears in the written record as a participant in a hearing at the Regent’s Household Office in the fifth month of 1149, defending the temple’s rejection of the standard land tax rate (three to per tan) then being enforced across Iga by the governor’s representative, Nakahara Toshimune.88 Amazingly we have a transcript of testimony by both Kakunin and Nakahara that was probably made for review by the then-regent, Fujiwara Tadamichi (1097–1164; regent/ chancellor 1123–1158) to help Tadamichi decide the case.89 In lengthy exchanges in response to examiners, Kakunin called Nakahara’s esti- mate of unpaid taxes wildly inflated, while Nakahara responded that Kakunin could not prove that cultivators beyond the river had ever been allowed to pay a tax rate of only two to per tan. One Tōdaiji cleric wrote at the time, “The outcome of these hearings will mean life or death for the temple of the Great Buddha.”90 In the highly factionalized environment of the Heian court in the mid-twelfth century, ending the dispute proved very difficult.91 Kakunin had other challenges to deal with as well. In the autumn of 1149, for instance, he himself led estate dwellers in the harvesting and confiscation of crops from fields belonging to a Kōfukuji monk in the Yagawa area.92 In Nara he and other temple administrators created rules for using and filing records to make sure they were accessible when needed. And a decade later Kakunin presented Tōdaiji’s successful request for exemption when, after the victory of Go-Shirakawa (1127– 1192, r. 1155–1158) in the attempted Hōgen coup of 1156,93 the monarch issued his famous order condemning all “outer fields” and “added fields” on the estates of shrines and temples. Kakunin also succeeded in nullifying the earlier confiscation of Yanase.94 In 1160 and 1161, in the transition between the terms of two provincial governors, he reportedly led a force of three hundred lower-ranking monks (dōshū) to take over (ōryō) Yanase, with surprisingly no repercussions from the court.95 He then continued on as the presiding head of Tōdaiji’s monastic council, and in 1168 was again accused of committing arson and murder in the Yanase area, clearly a hot zone in the estate, where an impressive area of some seventy-four chō was under cultivation.96 And finally Kakunin was involved in a border dispute with a neighboring estate, Fujii-no-shō in Yamato, which was charging a toll on the lumber Kuroda loggers dispatched to Tōdaiji. This proved a difficult fight given that the estate was overseen by a persistently hostile manager.97 182 Making the Land Productive

During the subsequent 1170s Tōdaiji was clearly fortunate in the favor it enjoyed from the then retired monarch, Go-Shirakawa In. In 1173, for instance, when the court proclaimed that all estates of fifteen great temples in Nara should be confiscated, Tōdaiji gained an exemption, as it did from the special levy on estates to rebuild the royal palace in 1174. And late in that same year Go-Shirakawa ordered that Kuroda Estate, its outer fields, and the Kuroda New Estate should all be exempt from provincial taxes.98 What was the reason for Go-Shirakawa In’s generosity? The most likely answer is that Tōdaiji was benefiting from the monarch’s antipa- thy toward Tōdaiji’s rival in Nara, Kōfukuji, whose unruliness and violence the monarch criticized frequently. Perhaps too, Go-Shirakawa saw himself as the protector-successor of the legacy left by the eighth- century builder of Tōdaiji, Shōmu Tennō, as Janet Goodwin has argued.99 In any event Tōdaiji’s monks understood the importance of their win and they moved quickly to assure the retired monarch that his favor would be well used. Several of the temple’s elite monks proclaimed in 1175 that they would devote all proceeds from Kuroda’s various sectors to support scholarly and ritual activities by one hundred scholar monks.100 Provincial officials responded angrily, blam- ing the provincial governor for “selling out the [provincial] domain.” They warned too that were the situation not reversed, and especially if Yanase were not returned to the provincial domain (kokugaryō), it would destroy Iga Province.101 Even while the temple enjoyed Go-Shirakawa’s benevolence, circumstances on the ground in Nabari were challenging. In the late 1170s complaints in the record show that some Kuroda dwellers were allying themselves with monastic factions at Tōdaiji. One of these groups at the temple was associated with the Middle Gate, and another was associated with the Hokkedō, Tōdaiji’s oldest chapel. Perhaps these groups were dismayed that Kuroda’s support was to be used exclusively for scholarly activities rather than for other temple needs. When he learned of this turn of events, the temple director ordered the five masters (goshi), the elders of the monastic assembly (shuto), to prohibit the recruiting activities of such monks within Kuroda Estate. Doubtless he feared that such activities could result in nonpayment of rents and other dues. At the same time Tōdaiji’s monastic council sent down its own order that Kuroda’s managerial staff should submit oaths (kishōmon) to the effect that they were not sympathizers with such factions in Nara.102 According Loggers and Cultivators of Nabari 183 to the extant record, however, only two of the estate managers submitted such oaths. Meanwhile the Nabari district chieftain, Minamoto (Hasetsukabe) Toshikata, who was surely Chikakuni’s progeny, provided another cause for worry by temple monks and their clients. He was said to be working closely with provincial authorities and Kōfukuji to recruit followers in the New Estate area.103 The situation was so serious that at the end of 1175 the assembled monks of Tōdaiji sent their own petition to the court reporting an invasion of the New Estate at harvest time by thirty men armed with bows and arrows. The miscreants had chased residents, killed a young monk, and stolen supplies, goods, and several head of cattle. The monks’ petition points out how the estate was then considered critical for support of the temple, and how a failure by the court to deal with Toshikata’s crimes was sure to activate Shōmu Tennō’s curse against any who would disdain the temple of the Great Buddha.104 In response, Go-Shirakawa’s office sent its own orders against Kōfukuji and Toshikata, commanding that the latter appear in the capital with his three sons to face charges.105 Just on the verge of the Genpei War era (1180–1185), when warrior followers of the great commanders Taira Kiyomori and Minamoto Yori tomo were beginning to face off on battlefields, a survey of outer fields at Kuroda Estate was completed, likely by provincial officials preparing to assert their case one more time. They reported 289 chō of fields under cultivation, surprisingly just ten more than in 1108. It would seem then that land opening had not progressed at a great rate during the twelfth century, perhaps due to all the dissension.106 Extant too is a defense statement by Tōdaiji’s monastic council that lists eight villages in the estate, from Komō in the east to Nakamura in the southwest. Yagawa is missing; perhaps troubles caused by the Nabari district chief and Kōfukuji had attenuated Tōdaiji’s hold there.107 For all their successes in recovering from their crushing defeat in 1145, Tōdaiji’s monks and Kuroda cultivators faced yet another crisis in late 1180: the political situation at court had shifted significantly, with great impact on temple and estate. Around harvest time of that year the Council of State sent its order supporting earlier complaints by Iga authorities: (1) Tōdaiji had wrongly taken over official lands in the province and called them “support for scholar monks,” (2) all the fields beyond the river, numbering 289 chō according to the 1176 survey, were to pay full provincial taxes and (3) in the future, provincial authorities 184 Making the Land Productive themselves would pay the temple’s annual allotment of benefice rice, 311 koku, from provincial stores.108 Why this full stop in Retired Monarch Go-Shirakawa’s generosity to the temple? The fact is that since 1177 Taira Kiyomori had essentially taken control at court. Early in 1180 he put his grandson on the throne (Antoku Tennō, 1178–1185; r. 1180–1185). The decision against Tōdaiji and Kuroda Estate in the ninth month of 1180 represented Kiyomori’s policy on the Nara temples and their land holdings.109 Indeed 1180 was the year when he sent a force to Nara to chastise Kōfukuji, and it ended up burning down some of Tōdaiji’s structures as well, including the Great Buddha Hall and parts of its image. Tōdaiji had lost the favor of the court it had enjoyed under Go-Shirakawa’s leadership. And it was some time before the temple’s fortunes improved. Although the record shows that the legal struggle over the temple’s property began in 1181, it was not until after the end of the war in 1185 that Tōdaiji won a new royal decision establishing the privilege of “no entry” by provincial personnel for the Kuroda New Estate. It was even later, in 1189, when the Estate Records Office canceled various military stewardships that had been created in Iga, including one at Yanase. Doubtless the Hasetsukabe chieftain had been considered a Taira ally, and his property there had been confiscated and turned over to a Mina- moto follower as military steward. The story of Kamakura-period Tōdaiji must, however, await a future opportunity for its telling.110

We have seen that Kuroda Estate took a long time to develop in Nabari District of Iga Province. First, Nabari was the site of a logging forest that provided wood products of all sorts for construction in the early capitals, especially for Heijō and its official temples, and most particularly the temple of the Great Buddha, Tōdaiji. Loggers who associated themselves with Tōdaiji lived in Nabari, and local officials there maintained close relations with the temple. Without that strong foundation, it is hard to imagine there would ever have been a Kuroda Estate. In addition, perhaps because of its logging forest and proximity to Nara, Iga became a province that paid taxes from benefice households to Tōdaiji by the tenth century. Then too, as Iga provincial governors recruited contractors to help them recultivate fields on the Nabari plain in the late tenth and early eleventh centuries, some of those contractors, like Fujiwara Sanetō, had relations with Tōdaiji. Over time, Tōdaiji monastic authorities became proprietors of Nabari properties Loggers and Cultivators of Nabari 185 east of the Nabari River, where dynamic land opening was proceeding in the eleventh century. Sanetō referred to “Kuroda Estate” in one of his missives from the early 1040s, implying that many fields east of the river were then regarded by some as Tōdaiji’s land. The situation was complex, however. There were other absentee holders, some connected with Tōdaiji and others not; the temple claimed the mountain forest, its logger-cultivators with their homes and fields, and its share of provincial taxes as benefice dues. And the provincial headquarters claimed the right to tax official fields and those cultivating them beyond a moving riverbed. Only in the early 1050s do we see references to “Kuroda Estate” in the official record, and those references are denials. As far as Iga provincial authorities were concerned, Kuroda Estate included only the narrow band of land at the foot of the logging camp on the mountain side of the river. The “outer fields” and residences of cultivators should be taxed, a point that the authorities made by fire and blood, with forceful attacks on those residences, harvesting of fields, and confiscation. The provincial governor Ono Moritsune had success collecting provincial taxes on the Nabari plain. Nonetheless the temple’s logger-cultivators continued to open more land across the river, and the extant documents show that provincial authorities had a tough time collecting taxes from them, even after the promulgation of estate-control laws. Until the early 1130s, Tōdaiji and its dependents mostly held their own against provincial authorities on the Nabari plain. And clients of the temple, such as Hasetsukabe Tamenobu, gained hereditary rights for his progeny as managers of fourteen chō (about forty-two acres) and as managers of Kuroda Estate. For their part, Iga provincial authorities recruited a champion of their own, the Nabari district chieftain Hasetsukabe Chikakuni, who promised provincial officials to do his best to advance provincial interests in Nabari. In return he gained hereditary control of land in the Yagawa area. As locals chose their sides in this struggle, Tōdaiji had to cope too with disputes over rents between locals and the heirs of private proprietors, some of whom had ceded land and rights to important temples such as Kōfukuji and Kinpusen. Such quarrels were exploited by the Nabari district chief and provincial authorities, who continued to complain to court authorities that Kuroda cultivators owed thousands of koku in back taxes on “stolen” fields on the Nabari plain. As violence continued to flare in the early twelfth century, Tōdaiji appointed specialized 186 Making the Land Productive administrators to write and archive documents, compile the temple’s history, and liaison with Nabari clients more closely. An important moment came in 1132, when despite a long history of exemptions from such levies, all Kuroda residents were forced to pay a levy to rebuild Ise Shrine. That decision surely weakened Tōdaiji’s stature as a strong patron of Nabari cultivators. It was then that Tōdaiji renamed its outer fields “the Kuroda New Estate,” in hopes of gaining more protection for them. We also find Tōdaiji calling Kuroda residents “temple slaves” and “servants of the Great Buddha” to strengthen temple-client ties. But the monks’ hopes were dashed in 1144 and 1145, when everyone at court—then led by the retired monarch’s (Toba’s) household—reached the consensus that Iga authorities were right and Tōdaiji wrong, which resulted in Tōdaiji’s “great defeat.” This was when the multitalented Kakunin appeared, hero to some and “evil monk” to others. He argued the case of Kuroda Estate in hearings in the capital, kept close control over the documentation critical for the temple’s interests in Nara, and led armed followers to enforce the temple’s interests in Nabari. He presided over Tōdaiji’s monastic council through 1168. His aggressive leadership got the temple and its estate to the cusp of the 1170s, when Go-Shirakawa gave full recognition to the expanded Kuroda Estate. Ultimately, though, Go-Shirakawa’s favor was only as secure as Go- Shirakawa’s rule itself. In 1180, on the verge of six years of civil war in the realm, a court dominated by the Ise Taira took Nabari back to the 1145 status quo. That year was also when the Great Buddha Hall and much of its icon were destroyed by fire at the hand of an Ise Taira commander, Shigehira. So does this story of Heian-period Kuroda Estate, teased out from an archive of more than 360 documents, end on a low point for Tōdaiji monks and their Kuroda cultivators. Of course, the story of Kuroda Estate did not end there. Taking the study forward into the thirteenth and fourteenth centuries is high on my list of future projects. I want to ask how the hundreds of remaining records in the Kuroda archive inform us about ongoing developments at the estate in what was now a dual-centered polity, and why in the fourteenth century cultivators on the estate ultimately joined “evil bands” (akutō) that turned against the Nara temple. That project will provide us with important perspectives for studying other Tōdaiji estates, including the Akanabe and Ōi properties in Mino as well as Ōbe Estate in Harima (on which see various chapters in this volume). Loggers and Cultivators of Nabari 187

I agree with Ishimoda Shō that the Kuroda Estate archive gives us remarkable perspectives on the historical development of Japan’s countryside and the emergence of estates and the estate system through its early stages, from the classical into the medieval age. Rather than seeing Kuroda Estate as a small world where the struggle between classical and medieval worlds was waged, however, the story I have written here emphasizes developments in court government, provincial government, temple organization, center-periphery relations, temple-client ties, local society, and agricultural technology. We need many more such studies, written from a variety of perspectives and asking as many different questions as our documentary “landscapes in words” can inspire us to answer. Fortunately such studies are underway, with some results right here in this volume.

Notes

1. Friday 1992, 76. 2. Documents have been compiled by Takeuchi Rizō: KS, Iga no kuni Kuroda- no-shō shiryō. Other sources are Tōdaiji monjo (the archives of Tōdaiji, DNK Sec- tion 18), and HI. I have prepared a chronological and annotated database that makes these documents easily searchable. Piggott, forthcoming, Shōen Docu- ments in English Translation. For a relatively recent and helpful study of Kuroda Estate in Heian times, see Arai 2001, 2–261. 3. Historians including Furushima Toshio, Toda Yoshimi, Inagaki Yasuhiko, Yoshie Akio, Kawane Yoshiyasu, Kuroda Hideo, Ōyama Kyōhei, and Kimura Shi- gemitsu have argued that there was expansion of arable land in Heian times, especially in terms of dry fields. For a good discussion see Kimura Shigemitsu 2000, esp. 30–47. In contrast, Nagahara Keiji, Ishii Susumu, and, in English, William Wayne Farris, see arable land in Heian times stagnating due to demographic constraints. Charlotte von Verschuer (2007b, 2009) has also argued that the arable land stagnated but that dry fields were nonetheless extremely important for feeding people. Farris (2006) attributes stagnation to demographic stagnation. For facets of the debate, see von Verschuer 2009, 337–362. An English translation of her monographic study of rice and agriculture in premodern Japan was published in 2016. 4. Ishimoda 1946. For an English abstract of Ishimoda’s famous book, and a translation of one section, see Piggott 2006, 326–329. A short account of Ishi- moda’s oeuvre and methodology is Ishii 1991, 131–146. The contents of Ishi- moda’s collected works are discussed in Kitō Kiyoaki’s review, 1991, 62–71. 5. Kimura 2000, 232. In English Jeffrey Mass too has made a comparable argument for what he calls “history through documents.” See Mass 1976, 3–13. 188 Making the Land Productive

6. Ikebe 1981, 281–282. The Suchi area later came to be called Yagawa. 7. Yoshida 1980 (192–198) and 1998, and Kobayashi 2000 are good guides to research on eighth-century settlements, and, in English, Piggott 1997, esp. 199–208. Also see Baba 2010 and most recently, Asano 2014 for a good state- of-the-field report that incorporates documentary analysis and comparisons between the mandates of the Tang and Japanese law codes on which administration was based. On the district chieftain, see the Yōrō Shikiinryō (Laws on Official Personnel) in Ritsuryō 1976, 194–195; Aida 1964, 281–285; Mori 2000, 118–163; and for archaeological evidence, Tanaka 2008. For translations of documents concerning the eighth-century district chieftain, see Piggott, forthcoming. 8. Compiled in the early tenth century, the Engishiki (Procedures of the Engi Era) lists the names of all shrines and deities throughout Japan that received official offerings from the Bureau of Shrine Affairs (Jingiryō). For an English translation of the relevant passage in the Engishiki see Engishiki 1970, 2:127. The location of the other shrine listed in Nabari, Nai Shrine, is not known. 9. Nabari-shi Kyōiku Iinkai 1985, esp. 3–4. On the Natsumi abandoned temple, see Gekkan bunkazai hakkutsu shutsudo jōhō 1991 (8:59; 12:221), 1992 (3:149; 4:135). The site has been preserved, with an exhibit hall. 10. Hayashi and Suzuki 1985, 58. 11. On portages in the Kinai, which were critical for the movement of Nabari lumber to Nara, see Matsubara 1976, 464–504. Such geography, and maps that make it visible (see his map therein), are critical to understanding developments in Nabari. 12. SHM, vols. 1–25; Piggott 1987, chap. IV, esp. 130–133. For the actual donation of Itabae Soma to the temple by Kōken Tennō (718–770; r. 749–758, 764–770) see SHM 4:84; and Takahashi 1987. Maruyama 2001, 502–545, presents a good study of Itabae Soma and other early Tōdaiji properties, including the so-called shoki shōen, “early estates.” 13. On Moharadera, see Naoki 1987, 191–194. 14. See for example, KS 1, # 20 (967), which mentions Kōchi, who served as monastic director (bettō) from 950 to 964. See Arai 2001, 56–59; Kuroda 1978, 79–80; and Hiraoka 1980, 136–137, 267. On his role in Iga, see Kuroda 1978, 79–80. On Tōdaiji’s physical decline by the tenth century, see Arai 2001, 45–86. 15. KS 1, # 12. For an overview of research on Tōdaiji benefice units, see Itō Keisuke 2010. 16. Wakita 1969, 24–45, contains a good discussion. In English, see Piggott 1987, 188–206. 17. KS 1, # 21 18. Kawane 1984, 27–45. Loggers and Cultivators of Nabari 189

19. Takahashi 1987, 51–52. Takahashi thinks that the Unane Shrine was likely an important focal point for those opening new fields in Yanase and Komō from the 850s on. Its importance to the court is confirmed by the fact that the shrine’s rank was raised twice in the 860s and 870s. He also thinks local land-opening cultivators (hyakushō) made contacts with powerful patrons in the capital to protect newly opened fields. 20. Shoku nihongi, 743/5/27; for a translation, see Piggott, forthcoming. 21. Ruijū sandai kyaku, 852/3/13, on which see Toda 1967, 60–61. There is a French translation of this compendium of supplementary laws from the eighth through the early tenth centuries, by Francine Herail (2008–2011). 22. A well-known source forbidding recruitment of laborers by wealthy openers who offered food and drink to entice them is found in Ruijū sandai kyaku 806/4/16. Good discussions are Toda 1967, 51; Yoshida 1980, 118–121. Historians see herein evidence of more patriarchal family units, including bondsmen and slaves in the household, given the need for more substantial labor in these mountain valleys (for instance, Yoshida 1980, 118–140). 23. Ruijū sandai kyaku 896/4/2, 502; and Arai 2001, 45–61. For a translation of the Ruijū sandai kyaku entry, see Piggott, forthcoming. 24. It is thought that the size of a plot for many cultivators of the time varied from 0.8 to 1.2 acres. On these homesteads see Toda 1967, 87–91; Yoshie 1970, 109; Yoshida 1983, esp. 123–198; Nishiyama 1984, 141–150; Takahashi 1987, 51. Nishiyama describes what he thinks was a similar property sold in 788 that included seven buildings within an area of four tan. For images of such homesteads, see Yagi 1998, 114–117. Sakuma (1993) reports on archaeological finds and analysis of excavated village sites from classical to medieval times. 25. Yoshie 1974, 118. Takahashi notes that the Kojidan compilation of courtly anecdotes identifies Asanari as one of the heads of the Royal Secretariat of Murakami Tennō (926–967; r. 946–967), which means that he was particularly close to that monarch. See Takahashi 1987, 61l, note 91; and Kuroda 1984, 319–327. 26. KS 1, #s 14, 15. For the Komō dispute, its documentary record, and what it shows about the important role of the district chieftain with regard to certifying proprietorship at the time, see Takahashi 1987, 37–62, especially the chart of players, 43. 27. For a good discussion of these local officials and their duties, see Kuroda 1984, 324–326. 28. KS 1, #s 17, 18. Subsequently Asanari’s property was recognized and set up by local authorities who erected border markers in 965. The mid-tenth century was a time when monastic communities at the great temples were becoming increasingly conscious of and assertive about their property inter- 190 Making the Land Productive ests. See Kuroda 1980, 22–38; and in English, concerning Ryōgen’s leadership and its challenges at Mount Hiei at this time, see Groner 2002. 29. KS 1, # 21. 30. Ogawa 2007 discusses this contracting system, known as fumyō taisei, and its use in Iga. See also chapter 1 of this volume. 31. Konjaku monogatarishū 28:31. This collection of tales dates from the latter Heian period. The governor’s action here reminds us that in 988 there was a complaint by Owari cultivators regarding thuggish behavior by provincial authorities there to collect taxes. See von Verschuer 2007a, 305–328. 32. KS 1, # 87. 33. KS 1, # 50, dated 1056. Also KS 1, # 87. For a recent and very good discussion of Sanetō and other contractors like him, see Ogawa 2007. 34. KS 1, # 26. 35. KS 1, # 27. 36. KS 1, #s 32, 35. 37. KS 1, #s 35, 36; Ogawa 2007, 10–11. Ogawa sees a policy in the 1040s of governors privileging the interests of the great temples. 38. Ogawa 2007. 39. Shinkan himself, while the highest-ranking monk of Tōdaiji, actually lived at Zenrinji in Kyoto, not at Tōdaiji. 40. KS 1, # 31. 41. The earliest is KS 1, # 42. There we see storehouses for the estate at Kuroda Village. 42. On these estate-regulating laws, see Maki 1993, 1–38. It has been argued that a particular objective of these laws was to increase the number of fields on which the special levies could be collected. In English, see Sato 1974, 91–108. 43. KS 1, #s 40, 43, 44. Ōtsu Tōru (2001, 90–91) has pointed out that provincial governors, overseen by the Council of State, constituted the pillar of the eleventh-century realm, a fact well-demonstrated by Kuroda’s history. 44. KS 1, # 40. A senji is a transmission of a royal order by the Board of Controllers. 45. KS 1, # 42. Specifically on this “Tengi incident,” see Kawashima 1981 and Ogawa 2007. 46. KS 1, # 58. This is an early example of a temple chancellery order (mandokoro kudashibumi) signed by the monastic director rather than the monastic council. As Tōdaiji had to negotiate with the court concerning its estates, it had to reorganize and develop more specialized posts and offices while also nurturing unity in the temple community. See Nagamura 1989. 47. KS 1, #s 47, 50. Such residences, argues Kuroda Hideo (1984, 43–45), were evidence of developing communities. Kimura (2000, 21) notes that the Loggers and Cultivators of Nabari 191 term “resident” (jūnin) appears regularly after 1053 and signifies a sense of community, especially when groups submitted complaints to a proprietor (ex. HI #s 702, 834–835, dated 1053, from Tōdaiji’s Akanabe Estate in Mino). On the tools possessed by such residences, and a sign of increasing stratification among cultivators, see Kuroda Hideo 1984, 45–46. 48. This is a selection from Shinsarugakuki (New monkey music), probably written about this time (1050) by the literatus Fujiwara Akihira and translated here by myself, based on the annotated text by Ōsone Shōsuke. Its historicity is discussed in Kuroda 1984, 42. I have published a brief introduction to the Shinsarugakuki as well as other translated selections in Shirane 2007, 491–497, titled “The New Monkey Music.” 49. Researchers agree that iron tools were broadly distributed throughout western and eastern Japan by the tenth century, enabling increasingly dynamic land opening by local elites in the mid- to late eleventh century. 50. KS 1, #s 72, 93, 119, 122. Also see Kuroda 1984, 130. He connects the development of homesteads with more secure rights of cultivation, called sakute, by the eleventh century. 51. KS 1, #s 72, 84. 52. On processes of land opening and the importance of dry fields, see Kuroda 1979 and 1984, esp. 97–121; and Kimura 1982 and 2000. Also see Kimu- ra’s chapter in this volume. 53. KS 1, #s 60, 61. 54. KS 1, # 57. Arai Takashige notes that a significant amount of construction was going on at the temple at this time, so assuring lumber payments from the Nabari forest would have been a particularly important objective for temple administrators (2001, 62–79). 55. KS 1, # 63. As seen here, this Tengi era from 1053 to 1057, when Go- Reizei Tennō (1025–1068; r. 1045–1068) ruled with the help of Michinaga’s son, Yorimichi, is well known for violence between proprietors and cultivators clashing with provincial authorities. See Kawashima 1981. 56. KS 1, # 66. A to is one-tenth of a koku, a measurement of volume that is difficult to determine for the Heian period. See the introduction to this volume. Notable too is that Tamura Noriyoshi (1994) sees this as the time when the medieval village (chūsei sonraku) was emerging. 57. HI 3:1126 (# 1113), dated 1075/4/28. The opener in Harima’s Akō District was the district chieftain, Hata Tametatsu. The newly opened area was called Hisatomi Township (gō). See Hiroshima-ken shi vol. 3 Kodai-chūsei hen, 157–159; and HI 3, # 1113. For a detailed analysis of the new townships over which these openers gained control, see Yoshie 1974. For translations see Piggott, forthcoming. An earlier document confirming the land opening by 192 Making the Land Productive a district chieftain is HI 9:3560 (# 4614), dated 1031/6/3, at “Mita gō” and “Shigeyuki myō.” Hisatomi Township is also discussed in Kimura Shigemitsu’s chapter in this volume. 58. KS 1, #s 77, 78, 85, 120. 59. KS 1, #s 83, 85. 60. KS 1, #s 83, 84. Arguments concerning this property, its proper holder, and the terms of the holding vis-à-vis cultivators and the province continued into the 1090s. 61. KS 1, # 115. 62. KS 1, # 86. 63. KS 1, #s 91, 95, 103, 292, 321, 335. 64. In a later record from the Tax Office of Iga dated 1122 (KS 1, # 185), we also see that rice or silk could be used to pay taxes, based on exchange rates set by the province, which were sometimes a cause of complaint. See too KS 1, #s 213–219 and #s 221–225. 65. KS 1, #s 104, 106. 66. KS 1, # 118. Raishin’s titles were azukaridokoro and kutō, both of which denote “the person in charge.” On the twenty-one properties in 1128, see KS 1, # 205. 67. Horiike 1980. Daigoji was also compiling its history at this time, for the same reasons. 68. HI 4:1518–1519 (# 1662), from a document authored by temple authorities and cited in a controller’s order. Also see Arai 2001. Interestingly, there is no reference to the synergy of “King’s Law and the Buddha’s Law” (ōhō buppō) in the Heian Kuroda record, although it occurs frequently in the record of other Tōdaiji estates such as Akanabe and Ōi in Mino Province. 69. KS 1, # 121. 70. KS 1, #s 126, 127, 128. 71. KS 1, # 135, 138, 139, 140. The Great Buddha Hall project was an official one, since it was superintended by a project management unit (gyōjisho) formed at court. Its manager was likely a member of the Council of State, part- nered by a controller. 72. Tōdaiji bettō shidai (List of successive directors of Tōdaiji), Kanjo’s section. 73. KS 1, # 184. 74. For instance, KS 1, # 148–179. A number of receipts from this time indicate that “outer field” units were paying benefice rice payments; perhaps such receipts served as proof that given units—which varied from large (nine chō) to small (five tan)—were inside the temple’s domain and not liable for provincial tax assessments. Loggers and Cultivators of Nabari 193

75. KS 1, # 193. 76. KS 1, #s 207, 208, 335. 77. KS 1, # 226. 78. KS 1, # 235. In KS 1, # 212 we also learn from a temple missive to the provincial office that several new settlement names had come into use. One was Mizoguchi, likely named for a water outlet for irrigation. 79. For instance, see KS 1, #s 232 and 235, and detailed discussion of same in Kuroda 1979. Kuroda agrees with Toda and Kimura on the importance of dry field crops like barley for increasing and stabilizing grain cultivation in the later Heian age. 80. KS 1, # 235–236. Notable too is that Nakamura was sufficiently prosperous to include three temples; in this respect it is unique in Kuroda records. Also, on the importance of dry field food crops see, in English, von Verschuer 2009 and the 2016 translation of von Verschuer 2007b. 81. KS 1, #s 237, 239. An influential backer of the project was the Regents’ Line, as well as Taikenmon’in, whose queenly household weighed in on behalf of raftsmen transporting lumber to the project. In other words, the project had powerful political support to which Kuroda Estate logger-cultivators may have been drawn for all sorts of reasons. See also the chapter by Sachiko Kawai in this volume. 82. KS 1, #s 250–251. 83. KS 1, # 251. 84. KS 1, # 252. 85. Koyama 1973, 112. 86. The name of the provincial governor appears in KS 1, # 250. He was newly appointed in the first month of 1144. 87. Kudō 1978, 69–85. See also Gomi 1984, esp. 273–284. On Kakunin as an “armed monk,” see, in English, Adolphson 2007, 100–103. 88. KS 1, #s 256, 258. 89. For instance, per KS 1, #258, Tadamichi sent his order (migyōsho) to the head monk at Tōdaiji in mid-1149, telling him to “make your orders based on proof documents.” Beyond that we don’t know exactly what Tadamichi thought, since he writes that “my messenger will speak with you.” 90. KS 2, # 257. 91. For insights into the trial system of the late Heian era, which is well illuminated in the Kuroda record, see Kiley 1982, 29–44. 92. KS 2, #s 261, 264, 267. 93. It was an armed conflict over succession to the throne between court factions and supporting military factions. 94. KS 2, #s 268–269. For the broader picture see Kon 1994. For an English translation of Go-Shirakawa’s edict, see Piggott, forthcoming. 194 Making the Land Productive

95. KS 2, #s 288, 292. 96. KS 2, #s 310, 315, 317. The keeper of documents, the registrar or kumon, signed # 315. 97. KS 2, # 305, although an earlier record is # 277, from 1158; and there are later records from the 1170s: KS 2, #s 354, 355. Fujii Estate had been confiscated from its earlier holder after the Hōgen coup of 1156, and then it was subject to provincial authority. See Gomi 1984, 273–284, which notes that the instability at Fujii and other estates that resulted from confiscations after the Hōgen coup was one aspect of what the Hiei abbot Jien meant when he wrote in his Gukanshō that Hōgen initiated “the age of warriors.” 98. KS 2, #s 319–321. 99. Goodwin 1990. 100. KS 2, # 323. 101. KS 2, # 334, a report (ge) by Iga’s local provincial officials (zaichōkanjin). 102. KS 2, #s 324, 328, 329. The oath in the Kuroda record had not yet developed to its more mature Kamakura form, in which numerous buddhas and kami were called upon to witness the oath and punish dishonesty. But the tradition dates back to monastic settings, especially on Mount Hiei, in the early tenth century, according to Ogino Minahiko (Kokushi daijiten 4:85–87). See Chūsei hōsei shiryōshū 1955, 1:94–95. For an English translation see Piggott, forthcoming. 103. KS 2, #s 335, 339, 340, 341, 344. 104. KS 2, # 348. 105. KS 2, # 340. 106. KS 2, # 352. 107. KS 2, # 355. 108. KS 2, # 361. 109. KS 2, # 362. 110. For an outline of Kamakura developments at Kuroda see Ishii 2000, 9–28, and Arai 2005. In English see Piggott 1982, 45–91, and my introduction to Ishimoda Shō, “Japan’s Medieval World,” in Piggott 2006. pa rt i v

Secular and Sacred

c h a p t e r 8

Hijiri and Temple Monks Contrasting Styles of Estate Management

Nagamura Makoto Translated by Janet R. Goodwin

In medieval times, temple-owned estates provided the primary economic support in the form of rents and miscellaneous dues for monastic institutions such as Tōdaiji and Kōfukuji in Nara and Enryakuji and Onjōji near Kyoto. These estates were managed by their proprietor temples in collaboration with—and sometimes in opposition to—local administrators.1 Focusing on Tōdaiji, this chapter will examine personnel who managed these estates, in some cases regular temple monks, in other cases religious practitioners called hijiri, who operated in various ways outside orthodox temple structures. I will point out some of the differences in their management styles that stemmed from disparities in their religious outlooks and in their status vis-à-vis the proprietor temples. At the end of the Heian period, a wartime crisis forced Tōdaiji to reconstitute its land-based support system. On 1180/12/28, an army led by Taira no Shigehira set both Tōdaiji and Kōfukuji on fire, and almost all the important halls and pagodas at both temples went up in flames. The destruction was an early event in a civil war between two military forces, followers of Taira no Kiyomori and Minamoto no Yoritomo, that would rage through the entire realm for the next four years. With its buildings ruined and its estates confiscated by the Taira as leaders at court, Tōdaiji restored itself over several decades. This was possible partly because of its monks’ tenacious devotion to the Buddhist law, but also because of the involvement of a monk from outside the corporate temple structure, Shunjōbō Chōgen (1121–1206). Appointed to the

197 198 Secular and Sacred

Chief Alms Collector’s Office (Daikanjinshiki) at Tōdaiji, he exhausted his efforts in collecting donations and performing other tasks to aid in the temple’s reconstruction. (For sculptural portraits of Chōgen, see plates 1 and 17.) Chōgen (also known as “Namuamidabutsu”) had begun his religious life with ascetic practice at Daigoji, a temple just outside the Kyoto capital. According to his autobiography Namuamidabutsu sazenshū (The collected good works of Namuamidabutsu), “When he was seventeen years old, he performed austerities at various places on the island of Shikoku. At the age of nineteen, he ascended Ōmine (a mountain in the Yoshino area of Yamato Province) for the first time for ascetic practice.”2 Thus Chōgen undertook religious practice at various sacred locations on Shikoku and Honshu, including Mount Kumano and Ōmine. After he took the tonsure he traveled to Song China, and after that he acted as a kanjin hijiri, a religious practitioner who collected donations for temple projects, based at Mount Kōya in Kii Province.3 During the last ten days of the second month of 1181, Chōgen traveled to the demolished Tōdaiji. Shocked and moved by the temple’s condition, he then called on Fujiwara no Yukitaka,4 who had inspected the ruins of Tōdaiji as a royal emissary. Through Yukitaka, Chōgen sought the monarch’s permission to assist in Tōdaiji’s reconstruction.5 Yukitaka added his support, and in the sixth month of that year, the retired monarch Go-Shirakawa (1127–1192, r. 1155–1158) agreed. An edict in the name of the child sovereign Antoku (1178–1185, r. 1180– 1185) ordered the reconstruction, and in the eighth month Chōgen drafted an appeal for donations and inaugurated reconstruction activities.6 As head of the Alms Collector’s Office, Chōgen thus promoted a campaign for contributions to rebuild Tōdaiji under the authority of Go-Shirakawa’s religious vow. When considering Tōdaiji as it was rebuilt in the Kamakura period, we should not overlook the role of hijiri like Chōgen in the management of temple estates. These hijiri acted under the authority of the Alms Collector’s Office, which notably operated independently of the corporate temple itself. One of Tōdaiji’s estates managed by Chōgen was Ōbe Estate in Harima Province, discussed in several other chapters in this volume.7 In what follows I will examine the management of such a temple property in detail. Hijiri and Temple Monks 199

Temple Monks and Hijiri Among the monastic groups that made up medieval Tōdaiji were scholar monks known as gakuryo, as well as dōshū, service monks who assisted at rituals for the main images at the temple and with the upkeep of temple buildings. Scholar monks lived at Tōdaiji’s three original residences and its distinct cloisters (in), while service monks lived at residences attached to the Hokkedō (Lotus Hall) and the Chūmondō (Middle Gate Hall). These groups formed the nucleus of monastic organization in the Heian and Insei periods (794–1185). Then in the early Kamakura period (1185–1333), two new groups emerged within the temple organization. Esoteric monks (misshū) were based at the newly reconstructed Shingon’in and Shinzen’in, and they devoted themselves to esoteric practices. In addition, ritsu monks, who as world-renouncing ascetics (tonseisha or tonseisō) devoted themselves to the monastic precepts (vinaya, Jp. ritsu), were intimately involved in management duties, such as the construction of buildings within the temple. They were based at the newly reconstructed Ordination Platform (Kaidan’in). Ritsu monks such as Gyōnen (1240–1321) left much evidence of their scholarly activities that differed from those of the scholar monks (see table 8.1). Tōdaiji’s regular monks had rights and duties appropriate to their station. Specifically, their dharma names—the names they had taken as monks—were entered in the temple register, and they resided in residence halls within the temple. At religious services, they chanted sutras and prepared offerings for the altar, and they also collected donations to support such services. Such rights and duties defined their status as Tōdaiji monks. In addition, temple monks supported the functioning of the temple by filling various internal offices. In other words, a regular temple monk could claim membership in Tōdaiji, serve at religious rituals inside and outside his temple as a Tōdaiji monk, and receive appointments to positions within the temple. Materials for his support and religious activities were guaranteed. Positions a temple monk might assume and for which he could receive remuneration included those of the temple’s director (bettō), the monastic supervisor (shigyō), and one of the three deans (sangō), who oversaw temple operations, as well as one of the five masters (goshi), beginning with the master in charge for the year (nen’yo goshi). These five masters were the elders that represented the corporate organization of Tōdaiji monks. Temple monks also served as estate custodians, proprietor’s deputies, and rent collectors who managed temple landholdings. 200 Secular and Sacred

Table 8.1 Strata within Tōdaiji in the Kamakura period

Group Based at Scholar monks (gakuryo) Three original monastic residences (sōbō), and the Tōnan’in, Sonshōin, and Sainan’in cloisters Service monks (dōshū: including Hokkedō and Chūmondō monastic precept masters called risshū or residences rikke, and meditation masters called zenshū) Esoteric monks (misshū 密衆; Shingon’in, Shinzen’in sometimes called misshū 密宗 or yuimitsushū) Precept masters (rissō or risshū) Ordination Platform, Chisokuin, Ryōgain, Aburakura (Oil Storehouse)

Scholar monks, service monks, and esoteric monks were classified as regular temple monks, but precept masters who had taken the tonsure and lived within the temple as world-renouncing ascetics were, strictly speaking, not included in that category.

Scholar monks, who were engaged in doctrinal studies, were at the center of the monastic organization at Kamakura-period Tōdaiji. Some three to four hundred of them lived at the temple, along with one hundred or so service monks. Not only did scholar monks outnumber service monks but they also occupied higher ranks and had more exalted responsibilities. They were the nucleus of day-to-day management at Tōdaiji, and they were sometimes dispatched to temple properties in their management capacity. For example, the scholar monk Shōgen (c. 1200–1250) issued the following order in 1240, in connection with his duties as the custodian of the Kuroda New Estate, a Tōdaiji holding in Iga Province:

Orders to Kuroda New Estate: You shall immediately exempt Seiraku name-fields (myō)8 at Hiyamizuda from paying designated miscellaneous dues on one tan and one dai of paddy [about 1860 square meters]. Hijiri and Temple Monks 201

Regarding the above: It is certain that in recent years the miscellaneous dues from the paddy of this myō have not been paid to Tōdaiji but have been diverted elsewhere. However, it was announced that from this year, miscellaneous dues would now be levied on these fields. Nevertheless it is ordered that from this time on, the exemption will continue as before, in accord with precedent. Ennō 2 (1240), Fifth Month Custodian, Eye of the Dharma (seal) (Shōgen)9

As a Tōdaiji regular monk and its custodial agent (azukaridokoro), Shōgen had personally traveled to the Kuroda New Estate, where he oversaw the management of fields by locally based estate officials. He was therefore responsible for collecting rents and miscellaneous dues and forwarding them to Tōdaiji. In the context of these duties, he issued many documents. In the one above, Shōgen acknowledges the extraordinary rights to an exemption from miscellaneous dues held by a particular agricultural unit. That the original of this document was preserved by Tōdaiji suggests that most probably the persons who acknowledged these exemptions—those who received the order—were temple monks who themselves held official positions and associated fields on the estate. We also know that in connection with his management duties, Shōgen himself accumulated personal landholdings—a temple monk’s holdings—within Kuroda New Estate. In other words, Tōdaiji’s cloisters and regular monks held lands scattered throughout the temple’s estates. Hijiri also played a significant role in the management of Tōdaiji’s estates during the Kamakura period. While hijiri were included within monastic society, they lacked the standing of regular temple monks. One reason why was that they did not necessarily follow the procedures of ordination and conferral of the precepts that had been required of regular monks since before the Nara period. Hijiri included unordained “private” religious practitioners who assumed the guise of monks and spread Bud- dhist teachings among the people. Then there were the world-renouncing ascetics who appeared from the Heian and Insei periods. These were regular monks who resigned their official positions, abandoning the associated duties and security, to devote themselves solely to Buddhist ascetic practice. They are the type of hijiri I am discussing here. They did not belong to any particular temple as regular monks, but through building temples, making images, copying sutras, and the like, they formed various 202 Secular and Sacred religious connections (kechien) that enabled them to lead many people along the Buddhist path. Moreover, lacking the duties and security that went with positions in a temple, hijiri maintained distance from their own personal interests and profit. Thus both inside and outside of temples, they were valued as honest individuals who did not pursue their own advantage. Based on this reputation in society, hijiri—especially those known as kanjin hijiri, who collected donations for religious purposes— pursued wide-ranging economic projects such as construction and repair of public facilities, beginning with Buddhist temples. Furthermore, kanjin hijiri were actually given positions within temples corresponding to those of regular monks. At Tōdaiji they were also close to the ritsu monks, who had abandoned the world under the auspices of the Ordination Platform and served as its monks. Indeed, by the late Kamakura period, “hijiri ” were closely associated with ritsu monks. The following inscription on a placard attached to the ridgepole at the Hokkedō at Tōdaiji provides a view of the activities of hijiri in the repair of that hall during the Shōji era (1199–1201).

This hall was first constructed in 733, ten years before the Great Buddha Hall. But after the passage of many years and months, the building fell into ruin. Accordingly the Dharma Master (daihōshi) Shūkei wrote an appeal to supporters within Tōdaiji. However, even though he solicited help from many sources, from the chancellery (mandokoro) down to all of the monks, he could not obtain enough funding to accomplish his aims. Shūkei then appealed to Namuamidabutsu (Chōgen), who immediately presented the request to Shin’amidabutsu (Bengyō) and had the repairs accomplished. The building was really in terrible shape, and had to be just about totally reconstructed. [Work began on] the eighth day of the eighth month of 1199, under the supervision of the alms collectors En’amidabutsu and Gakuamidabutsu. Five people took charge of various management duties. The chief carpenter was Gonnokami Kunimune (Sakurajima), and assistant carpenters were Sadanobu Taifu and Yukikiyo Taifu. There were seventeen helpers. The roofer was the carpenter Yukisada.10

Thus Chōgen had his kanjin hijiri group repair the Hokkedō, a feat that the Tōdaiji headquarters and monastic groups had been unable to accomplish. This was possible because the group, with Chōgen at its Hijiri and Temple Monks 203 head, had its own revenues with which to rebuild the temple and also had organized artisans necessary for the task, such as carpenters and roofers. The group headed by Chōgen was based at Jōdodō, an autonomous temple (bessho) within the Tōdaiji precincts. From there they launched their campaign for donations, with the backing of the court and the Kamakura shogunate. Hijiri also pursued their aims at several other autonomous temples that Chōgen established across Honshu, such as Jōdoji on Ōbe Estate—which provided revenues for the rebuilding of Tōdaiji after the 1180 conflagration—and Amidaji in Suō. In fact the entire province of Suō was designated to provide materials, especially lumber, for rebuilding Tōdaiji. Hijiri thus established autonomous temples as the headquarters for their daily lives, which they spent promoting the management of temple landholdings in the provinces where they settled. In short, in the Kamakura period, both temple monks and hijiri supported the management of Tōdaiji’s estates. In the next sections I will consider just how each group carried out its duties, focusing first on the case of Ōbe Estate.

The Management of Ōbe Estate Having received Chōgen’s appeal, the retired sovereign Go-Shirakawa ordered the donation of Ōbe Estate to the Chinese artisan Chen He-qing, who had performed meritorious service in recasting the Tōdaiji Great Buddha image destroyed in the 1180 fire. Since He-qing then donated the estate to the Great Buddha, in effect its management reverted to Chōgen’s hands.11 Later still, Chōgen designated the head monk (inju) of the Tōdaiji cloister Tōnan’in to succeed to his management position. Chōgen’s choice of a successor was no doubt related to his own initial temple affiliation at Daigoji. In 1190 Shōken, the abbot (zasu) of Daigoji, had simultaneously taken the position of head monk at the recently reconstructed Tōnan’in.12 Both Tōnan’in and Daigoji traced their founding to the Shingon monk Shōbō (832–909), and the resulting close relationship between the two temples allowed Shōken to assume the two positions. Although Chōgen originally chose Shōken as his heir to the management rights of Ōbe and several other Tōdaiji estates, the Tōnan’in head monk predeceased Chōgen. The latter’s will, dated 1197, then 204 Secular and Sacred

designated Jōhan (Gan’amidabutsu), Shōken’s successor at Tōnan’in, to inherit these rights.13 The pertinent section of the will reads:

The Tōdaiji preceptor Namuamidabutsu [Chōgen] wills temple holdings and structures to the head monk of Tōnan’in, Gan’amidabutsu [Jōhan], including: Temple estates in Iga Province, Awa, Hirose, Yamada Arimaru Estates; in Harima Province, Ōbe Estate. . . . Temple structures & autonomous temples [including] the autonomous temple on Ōbe Estate in Harima Province, with a Jōdodō, three bays on each side, with a tile roof. [There] we installed a colossal [jōroku] standing Amida triad, completely gilded. . . . Namuamidabutsu [Chōgen] has established the above-mentioned temple estates, chapels, and autonomous temples solely to propagate the Buddhist law and benefit sentient beings. . . . Through handing over these precious assets solely to the lineage of the honorable master Shōbō, I place them entirely under the control of a single temple, Tōnan’in. However, affairs must be conducted so that, although generations may pass [and head monks will change], those in charge of this cloister shall inherit these assets. Absolutely do not dare to divide up these holdings and hand them over to another temple or lineage! Also it will not do to place [these holdings] under the jurisdiction of the director or the three deans of the main temple [Tōdaiji]. Such things will cause the decline of revenue in the future. Revenues from the estates are allotted for various Buddhist services, and for no other purpose, so even in the far future the payment of rents must not be neglected. The distribution of rents for Buddhist services shall be as follows: for the performance of ceremonies on the two worlds at the Great Buddha Hall, for altar implements for those ceremonies, and for provisions for twelve monks to perform the services, 234 koku; for services on the Saishōō sutra, also at the Great Buddha Hall, and provisions for thirty lecturers, 352 koku 8 to; for the continuous presentation of flowers at the same place, provisions for two hundred meditation monks, 360 koku . . .

As “the holding of the Great Buddha,” Ōbe Estate had been ear- marked to supply revenue for ritual and maintenance expenses at Tōdaiji. When Chōgen passed on its management not to regular tem- Hijiri and Temple Monks 205 ple monks but to the cloister Tōnan’in, it was because he feared that Tōdaiji monks might use the revenue selfishly. To understand these concerns we have to remember that Chōgen was a world-renouncing hijiri who took the restoration of the Nara temple, beginning with the Great Buddha and the Great Buddha Hall, as his own Buddhist practice. As mentioned earlier, since hijiri like Chōgen did not seek personal secular profit, they were regarded as honest and pure. Based on such a reputation, they were chosen to manage sources of revenue for Tōdaiji reconstruction, such as Ōbe Estate and the public lands of Suō Province, and they supervised carpenters and other artisans who built temple halls and pagodas. After the first stage of work at Tōdaiji had been completed—the reconstruction of important buildings such as the Great Buddha Hall—it was necessary to maintain and repair these buildings. The function of kanjin hijiri groups that took on this maintenance work was thus passed on to an office, the Kanjin- sho, formed by these hijiri. Until the Muromachi period, this office at Tōdaiji had charge not only of the management of building resources and materials but also of repair and construction activities. Chōgen, who had led the Kamakura reconstruction activities, was personally involved in management of estates such as Ōbe. After he handed over his rights to Shōken and then to Jōhan at the end of the last decade of the twelfth century, Ōbe Estate was then managed by regular temple monks who lived at Tōnan’in, rather than by kanjin hijiri. Revenue from the estate was used for religious services at Tōdaiji’s Great Buddha Hall, the Hachiman Shrine within Tōdaiji, the Ordina- tion Platform, and Jōdodō, in other words, for a variety of religious services within the main temple. In summary, in the period when Tōdaiji’s Ōbe Estate was “the holding of the Great Buddha,” management was carried out by alms-collecting hijiri. Rents and special levies were used for construction expenses at the main temple. But after the estate came under the authority of Tōnan’in and its regular monks took charge, these revenues were devoted to various religious rituals. There does not seem to be a big gap between the management methods of alms collectors and regular temple monks, but the use of revenues differed in each case. It is important to note, however, that the Harima autonomous temple Jōdoji, which Chōgen had established as his base of operations on Ōbe Estate, was a temple that worked vigorously to meet the religious needs of local estate residents. It is notable that hijiri there, in the process of 206 Secular and Sacred estate management, exercised influential power through the medium of devotions at Jōdoji, especially through rituals such as the Mukaekō (or Raigō-e; Welcoming Assembly) first held there in 1200. Its purpose was to allow participants to visualize for themselves their salvation (ōjō) in Amida’s Western Paradise. In other words, Jōdoji was not only the operational base for estate management but also the religious center for estate residents, lay and monastic. Similarly, local officials of the provincial government were parishioners of the autonomous temple Ami- daji also established by Chōgen in Suō Province, and Chōgen intended that that temple would continue as a sacred protective institution for its congregants.14 In comparison to regular temple monks, who were bound by temple organization and hierarchy, hijiri—who had few such restraints—were able to carry out estate management through different means.

Management of Temple Estates: The Power of Religious Coercion From the Kamakura period through the Sengoku age, hijiri and regular monks employed various methods to manage temple estates. Many were skilled in accounting, many were useful for day-to-day management duties, and some had computational skills such as knowledge of the abacus. Moreover, temple proprietors possessed great influence in society as well as coercive power that took advantage of people’s religious beliefs. The following section, citing examples from the fourteenth through sixteenth centuries, will examine how temples and their regular monks used that power to oppose lawsuits and the withholding of estate rents. One well-known technique was the “forcible petition” (gōso), in which monks would set down a sacred carriage, bearing a temple’s guardian deity (kami), near the headquarters of their opponents. Tōdaiji, which advanced its own interests by filing lawsuits with court and shogunate in Kamakura times, also employed this method, using the carriage of the kami of the temple’s Hachiman Shrine. In 1375, for example, the warrior Toyofuku Norimitsu seized fields under the control of the Ōbe Estate reeve, and officials at Tōdaiji thought this would interfere with revenues for construction work at the Hachiman Shrine. So the temple successfully filed suit with the Muromachi shogunate to stop the seizure, but the military governor (shugo) refused to enforce the shogunate’s order and Norimitsu’s violation could not be stopped. At that point Tōdaiji’s Hijiri and Temple Monks 207 monks convened and decided on “the resolute action of setting down a sacred carriage in three places [at the spot of contention], causing the kami himself to set forth in battle.”15 Such forcible petitions had been common in the Heian and Insei periods: monks from Nara and Mount Hiei would often set down their sacred palanquins in the capital, sometimes at the palace gates, an action that halted palace ceremonies and compelled cooperation from the court. From the late Kamakura through the Nanbokuchō periods, too, Tōdaiji sometimes relied on the fearsome majesty of the kami to obtain desired results. Another effective method temple proprietors used against recalcitrant estate officials or residents who declined to pay rents or who violated the holdings of others was called rōmyō, literally, “imprisoning the name” of a culprit within sacred precincts. The offender’s name would be inscribed on a piece of paper, which was then put before the main image at a temple. Presumably the deity would punish the individual if he or she did not behave properly. This method of coercion had been used since the late Kamakura period. The following example, from Nigatsudō-e engi (The illustrated traditions of Tōdaiji’s Nigatsudō), dates from late Muromachi (Sengoku) times.

In the winter of 1536, the Tōdaiji monk Seal of the Law Eikun was sent to Mino Province to check on some temple holdings there. In the countryside, he made various efforts to achieve his aim. After collecting rents for the temple, he quickly hopped on a palanquin and headed for the Southern Capital (Nara). But when he reached the area of paddy owned by the temple, he was surrounded by hateful brigands [i.e., the local peasantry] who pillaged everything he was carrying! Then Tōdaiji monks assembled to impose a punishment on the village [where the culprits lived]. The names [of the culprits’ fields inscribed on a piece of paper] were locked up inside the Nigatsudō, [and curses were called down upon them]. Then in 1546, all the cultivators of that temple paddy, with devout hearts, brought many donations for illumination by votive torches, saying that they wanted to join the Nigatsudō [Shūnie] ceremony (on the second day of the second month, the high point of the ritual).16 They said that after the names of their fields had been confined within the temple, a plague broke out in which many people lost their lives, and several hundred homes had no residents. Now there are but fifty-six [inhabited] houses. Realizing that this had all happened because 208 Secular and Sacred

of the curse, they repented their actions and wrote an oath on the reverse side of an ox-king amulet, swearing that they would abandon wickedness and return to righteousness so that they might obtain the protection of Tōdaiji. The assembled monks thought that this was a wondrous thing, and the culprits were granted pardons.17

The practice of “imprisoning a name” and invoking curses on it was a ritual known as the chōbukuhō. From the Kamakura period on, such practices were frequently employed by Tōdaiji and Kōfukuji to punish local strongmen who interfered with the temples’ management of their holdings. The practice of chōbukuhō (sometimes called gōfukuhō) was employed especially within Shingon esotericism. A similar practice, called tenpōrinhō, was used by Daigoji; the name of the individual to be cursed was inscribed on a plaque, which was placed in a long narrow tube called a “tenpōrin” tube. The tube was then set before the temple’s main worship object, where esoteric rituals designed to defeat foes and overcome calamities were performed before it.18 It is not clear how effective such rituals were, but estate officials and residents, knowing that they might be cursed, must have felt indescribable fear. Thus it is certain that when religious authority and secular society clashed over issues such as estate management, “imprisoning a name” was an effective method to demonstrate the power of temples to estate officials and residents.

For regular monks who belonged to Tōdaiji, activities related to estate management were not directly related to Buddhist practice. They were no more than secular activities that supported the monks as members of the temple organization, guaranteeing revenue for Tōdaiji. In other words, although temple monks did participate in estate management, they did not find any particular religious meaning in such activities. For hijiri, on the other hand, any activity that formed ties with Buddhism was in and of itself a “good deed.” Secular activities such as estate management, which was connected to temple construction and the funding of ceremonies, were no exception. This work, which in the case of Tōdaiji funded its reconstruction and the resumption of ceremonies there, was seen as a religious activity that truly supported the transmission of the Buddhist law, and it provided hijiri with an opportunity to accumulate meritorious actions. Thus there is no question that there was a big difference in the way temple monks and hijiri regarded their estate-management duties. In Hijiri and Temple Monks 209 large-scale projects such as the reconstruction of Tōdaiji, alms-collecting hijiri —who had few institutional restraints—used their abilities much as they wished. Moreover, hijiri were concerned with the religious beliefs of estate residents. They established and managed autonomous temples in order to deepen the faith of such people, while temple monks, on the other hand, were inclined to employ more coercive methods to deal with disputes regarding estate holdings or rent collections, such as forcible petitions and “imprisoning names.” At Tōdaiji in the Kamakura period, alms-collecting hijiri accomplished the reconstruction of temple buildings, and once those buildings were in place, regular temple monks conducted rituals therein while at the same time assuming the burden of estate management. In this we can see the actual situation of the division of duties regarding the reconstruction of Tōdaiji. The importance of this project transcended the differences in status and practice of the two groups.

Notes

1. In examining historical sources for temples such as Tōdaiji in Nara and Daigoji in Kyoto, I have focused on texts known as shōgyō, which were prepared for the educational activities of temple monks, while investigating the form of Buddhist teachings in each time period. I have also inquired into the consciousness of temple monks, how they supported religious teachings, and the levels of their concern with supporting their temples. In addition, I have examined the significance of secular activities such as estate management in which temple monks were involved. See Nagamura 1989 and Nagamura 2000. 2. The Collected Good Works of Namuamidabutsu is available in several printed editions, including OS 4:537–543 (# 227). A complete English translation, along with a photograph of one page of the text, can be found in Rosenfield 2011, 209–231. 3. Tōdaiji zoku yōroku, Kuyō hen (Fundamental records of Tōdaiji, continued: section on religious services), 1907, 206–248; 2013, 27–114. 4. Yukitaka was an official of the office of the retired monarch Go-Shirakawa and concurrently of the Kurōdodokoro, the court secretariat. 5. Tōdaiji zōryū kuyōki (Record of the commemoration service for Tōdaiji’s reconstruction), 47. 6. Tōdaiji zoku yōroku, zōbutsu hen (Section on constructing the Buddha image), 1907, 195–206; 2013, 3–26. 210 Secular and Sacred

7. See the chapters by Endō Motoo, Yoshiko Kainuma, Noda Taizō, Kimura Shigemitsu, Dan Sherer, and Janet Goodwin. 8. The term myō or myōden is often translated “name-field.” It was a unit of land that took the name of the individual who supervised its cultivation and was responsible for paying its rents or other dues. It is one of the harder terms to translate in medieval Japanese history. 9. Tōdaiji monjo, vol. 8:144–145 (# 609). 10. Gonnokami is a title often translated as “provisional governor,” but in this case it was probably an honorary title. 11. For details on these actions, along with some attendant difficulties, see OS 4:543–545 (# 228). Also see the chapters by Endō Motoo and Janet Goodwin in this volume. 12. For more on the Tōnan’in and its relationship with the main temple, Tōdaiji, see the chapter by Endō Motoo in this volume. In addition, see the essays in the special issue of Nihon Komonjo Kenkyūkai 2017 devoted to Tōdaiji’s records and its various constituent groups. 13. For Chōgen’s will, see OS 4:532–535 (# 224). 14. KI 2:405–410 (# 1163), 1200/11. Also see Yoshiko Kainuma’s chapter in this volume. 15. Tōdaiji monjo 1‑12‑122 (DNK 19, # 1206). 16. The Shūnie ceremony is still held annually at Tōdaiji’s Nigatsudō. Originally scheduled for the second lunar month, it is now held for two weeks beginning on March 1. The high point of this ritual of repentance today is the late-night Omizutori, in which monks carry sacred water to present to the Nigatsudō main image, the Eleven-Headed Kannon. The Omizutori is preceded by the spectacular burning of torches. For details see http://www.todaiji .or.jp/contents/function/02–03syunie1.html, accessed October 1, 2013. 17. The text of Nigatsudō-e engi can be found in Zoku gunsho ruijū 27.2, 128–134. 18. The term tenpōrin means “turning the wheel of the Buddhist law,” in other words, promoting Buddhist teachings. c h a p t e r 9

Beyond the Secular Villages, Estates, and the Ideology behind Chōgen’s Land Reclamation Projects

Ōyama Kyōhei Translated by Janet R. Goodwin

I reverently address all buddhas and bodhisattvas of the three worlds and ten directions. Regarding the repair of Sayama Pond: Long ago in the third year of the Tenpyō era [731], when Gyōki Bosatsu was sixty-four years old, he constructed the original dikes and laid the pipes for this reservoir. But as the years went by, [the facilities] were destroyed. And so, the residents of more than fifty villages (gō) downstream in Settsu, Kawachi, and Izumi Provinces requested that [the facilities be repaired]. From the spring of 1202, when the master of precepts Namuamidabutsu was eighty- two years old, he planned to make the repairs. Thus on the seventh day of the second month, ground was first broken, and on the eighth day of the fourth month, the first stone pipes were laid. On the twenty-fourth day of the same month, the work was completed. During this time, monks and laity, men and women, novices and children, beggars and outcasts, hauled stones and constructed the dikes with their own hands. They did this not for their own glory, but solely for the benefit of all. I pray that, through the ties thus formed [with the buddhas and one another], that the one Bud- dha’s blessings benefit all creatures in the universe equally and forever.1

211 212 Secular and Sacred

Thus the distinguished monk Shunjōbō Chōgen (1121–1206)—who took the phrase Namuamidabutsu (“Hail to Amida Buddha”) as an alternate name—described his efforts to repair the dikes and aqueducts of Sayama Pond, a major irrigation facility in Settsu Province. The account is inscribed on a stone plaque discovered in the central aqueduct of the pond’s northern dike2 (see plate 15). While Chōgen is best known for the restoration of the Nara temple Tōdaiji’s Great Buddha and Great Buddha Hall after their destruction in 1181 in the early days of the Gen- pei War, he was also involved in many construction projects that, strictly speaking, might be termed “secular.” In addition to restoring Sayama Pond, he managed the revenue from Tōdaiji estates to support the temple’s rebuilding and ceremonies, constructed irrigation facilities that opened estate land to cultivation, and repaired transportation facilities such as bridges, roads, and anchorages. Rather than seeing these projects as secular, however, we need to understand how they were driven by Chōgen’s religious beliefs, in particular the teachings of Shingon esotericism to which he was particularly devoted. (For sculptural portraits of Chōgen, see plates 1 and 17.) Chōgen’s activities involved both estates and villages. It is his work on the village level, however, that offers the most salient example of his connections with ordinary people. People considered the village, not the estate, to be the setting for their daily lives. In the case of the Sayama inscription, the appeal to Chōgen to restore the pond was issued by the fifty villages that used the pond for irrigation water, not by estate officials. Both generating the project and providing the labor to complete it, villagers reached across class divides and estate and provincial boundaries to accomplish a project that benefited them all. An important factor in Chōgen’s work was the extent to which he took part in such activities closely tied to people’s daily lives. Chōgen’s footprints are figuratively incised on the shore of Sayama Pond, and his religious ideology also contributed to his land reclamation work on Tōdaiji estates in several provinces. The discovery of the Sayama Pond plaque has raised our awareness of religiously motivated economic activities and the wider popular trends in the background. Landholders and other members of the lay community, of course, promoted economic activities such as land reclamation and the repair of irrigation and transportation facilities, but one important phenomenon of the early medieval age was the extent to which such activities were initiated by religious figures such as Buddhist monks. Beyond the Secular 213

In this chapter I will explore how Chōgen used religious ideology to propel and justify projects such as the restoration of Sayama Pond and land reclamation on estates, projects that transcended their seemingly secular nature.3 The development and management of estates can thus be seen as part of a religious agenda that also embraced other projects. Furthermore, this perspective allows us to see estates within a broad socioeconomic context that includes both smaller units of habitation and production such as villages, as well as lateral connections that cut across estate borders.

Sources In addition to fortuitous discoveries such as the Sayama plaque, evidence for Chōgen’s activities comes primarily from two texts, his testament and his autobiography Namuamidabutsu sazenshū (The collected good works of Namuamidabutsu [Chōgen], hereafter Sazenshū).4 The testament was written in 1197. In this document, Chōgen handed down his assets to the head monk of the Tōdaiji cloister Tōnan’in, “so that the Buddhist projects [he] had initiated would continue into the future, and that rents [from Tōdaiji holdings] might be distributed without neglect.” Like Chōgen, the head monk Gan’amidabutsu Jōhan had strong ties to the Shingon temple Daigoji in the capital.5 Although the extant version of the testament is a later copy and is not entirely reliable, we can use it to reconstruct the material underpinnings of the system that Chōgen organized, but we must be aware that his plans were not fully realized in the years after his death. Nevertheless, the testament provides a glimpse of Chōgen’s religious community in the form that he intended it should take.6 Chōgen divided his assets into two categories: landholdings he managed for Tōdaiji and the four small autonomous temples (bessho) he had established by 1197. The landholdings included Ōbe Estate in Harima, to be discussed in detail below, as well as three estates in Iga Province, two in Suō, and four in Bizen. The autonomous temples were Jōdoji on Ōbe Estate, the Senjū Ōjōin on Mount Kōya, the Jōdodō at Tōdaiji, and the Watanabe bessho in Settsu. Along with the human assets that were Chōgen’s followers, these landholdings and temples provided support for his activities. Chōgen launched his religious and economic activities from the autonomous temples, backing these activities with revenues from the estates. Sazenshū, a rambling text with neither date nor signature, was written on the reverse side of a Bizen Province document dated in the sev- 214 Secular and Sacred enth month of 1203. The calligraphy is in the style of the late Heian or early Kamakura period and the text is thought to be in Chōgen’s own hand. It was finished sometime after the date of the Bizen document. Since that document seems to have been compiled over a period of time beginning in 1202, however, Chōgen may have actually begun drafting Sazenshū in that year, using the document as a source of paper. The full title of the autobiographical text, Namuamidabutsu sazenshū, was added later.7 Sazenshū relates Chōgen’s activities in greater detail than does the testament and, of course, adds his “good deeds” performed after the testament was written in 1197. The account begins with a listing of the fifty- three colossal Buddhist images Chōgen had made or repaired—first of all the Great Buddha at Tōdaiji and also including the Amida triad at Jōdoji on Ōbe Estate. Then the text goes on to enumerate Chōgen’s religious activities at Tōdaiji, Daigoji, and the autonomous temples. The “good deeds” enumerated in this text can be divided into three categories. The first includes activities connected to the estates, the autonomous temples, and other temples noted earlier in his testament. These were properties that he could manage freely; perhaps we can call them his private holdings. These holdings were organized under his control and thus relatively extensive religious activities took place there. A second category relates to activities Chōgen initiated or supervised at temples and other facilities that were not under his control. Major examples are the restoration of Tōdaiji, including the reconstruction of the Great Buddha Hall. Temples and images in this category, such as the Great Buddha, could not be listed among Chōgen’s heritable assets in his testament. Sazenshū lists many “good deeds,” such as the construction of images and pagodas at temples that were not under his authority. Most of his good deeds fall into this category, which also includes the restoration of Sayama Pond in 1202. A third category includes those religious activities others initiated but to which Chōgen contributed; in Sazenshū’s terms, Chōgen formed religious ties (kechien) with such- and-such temple or image. For example, Chōgen donated three pillars to rebuild the pagoda at Kōfukuji in Nara. The term “kechien” appears often in Sazenshū, where it seems to indicate a comparatively low level of participation in the project, such as the provision of raw materials. One of Chōgen’s major concerns was to pray for the peace and prosperity of the realm, and the centerpiece of this effort was the reconstruction of the Great Buddha. Thus Sazenshū begins with Tōdaiji and then moves on to other projects. While his work at Tōdaiji was clearly Beyond the Secular 215 his most important project, some of his activities were aimed not at powerful people or institutions but at ordinary people. One example is the restoration of Sayama Pond, which is mentioned in Sazenshū along with the repair of bridges and the construction of roads so that people could travel back and forth safely.

Sayama Pond and Its Plaque Sayama Pond is in an area with a gentle south-to-north slope, where the old Amano River formed an alluvial delta. The pond was used to irrigate fields in Settsu, Kawachi, and Izumi Provinces. It was first constructed at the beginning of the seventh century; its northern dike, which has been extended twelve times, has survived to this day.8 The original purpose of the pond seems to have been to irrigate the alluvial valley below. In the eighth century the pond was repaired twice, and its original embankment was heightened, enabling the irrigation and cultivation of higher-terraced land. The first repair is attributed to Gyōki (668–749), the famous monk to whom Chōgen referred in the inscription on the plaque. Gyōki, who collected donations for the original construction of Tōdaiji by the sovereign Shōmu (r. 724–749), was often cited as a model in Chōgen’s writings. Another repair was undertaken in 762. These repairs made the dikes even higher and added a new intake gate and additional aqueducts that extended to the center of the pond. Chōgen’s repairs, which were concentrated on the middle aqueduct, extended the height of the dikes even more, to 79.2 meters.9 Chōgen’s plaque was discovered in the central aqueduct of the pond’s northern dike, mixed in with material from stone coffins from the Tumulus age that Chōgen used to repair the facilities. The plaque is made of Izumi sandstone, and its largest dimensions are 58.5 centimeters in height, 192 centimeters in width, and 79.5 centimeters in thick- ness. Incised on it are twenty-six lines of text, with most but not all of the characters legible. The last thirteen lines are signatures and Sanskrit seed characters. Also in Sanskrit is the kōmyō shingon (mantra of light), an esoteric Buddhist mantra thought to extinguish evil karma.10 There are two groups of signatures: those of Chōgen and his religious associates, and those of artisans. Identified as part of Chōgen’s cohort by the “amidabutsu” suffix he granted his followers, Ban’amidabutsu (Banna), Jōamidabutsu, and Jun’amidabutsu helped to manage the project. Chōgen himself is identified as the chief 216 Secular and Sacred alms collector (daikanjin), while Banna is the assistant alms collector (shōkanjin), indicating that it was they who solicited financial support for the repair of the pond. All three of Chōgen’s associates participated with him in other projects, including some related to the Tōdaiji reconstruction. The names of Banna and Jun’a appear in inscriptions dated 1203 within images of the Benevolent Kings (Niō or Kongō Riki- shi) that guard Tōdaiji’s Great South Gate (Nandaimon). The inscriptions identify the two monks as members of a group that managed the project.11 Banna, who appears prominently in such listings, seems to have been Chōgen’s chief assistant on the Sayama project. He was active at the Shingon Buddhist complex on Mount Kōya, with which Chōgen also had close ties.12 As for Jō’a, although he was a native of Yamato Province, he seems to have moved to Kawachi—one of the provinces that bordered Sayama Pond—and perhaps lived at the provincial headquarters there. Thus he had local connections with the repair project. According to Sazenshū, Jō’a had assisted Chōgen in the installation of a Chinese wooden image of Amida at a Kawachi temple.13 Jō’a’s association with Chōgen was not limited to local projects: his name also appears on a manuscript of the Hōkyōin darani sutra found within one of the images at Tōdaiji’s Great South Gate.14 According to the plaque, twenty Japanese and three Chinese artisans participated in the Sayama project. Some of them also had previous associations with Chōgen. For example, Mononobe no Tamesato, head of the Japanese group, was affiliated with the Tōdaiji Reconstruction Agency and had assisted Chōgen some years before in obtaining lumber from Suō Province to rebuild the great Nara temple. At the time, he traveled to Suō along with Chōgen and Chen He-qing, the Chinese casting master who reconstructed the Great Buddha.15 Along with Chi- nese artisans, metal casters from Kawachi Province had assisted in that monumental project,16 a compelling reason, perhaps, for Chōgen to restore the pond that helped to irrigate Kawachi paddy. The Chinese artisans include one named Shoubao, who is identified as a carpenter, and a stonemason named Yi Shichuan (Jp: I Seiken), also associated with the Tōdaiji Reconstruction Agency. The latter was a member of an immigrant group from Mingzhou, an East China sea- port with which Chōgen had connections: according to Sazenshū, he sent logs from Suō to construct a relics hall at the temple complex on A-yu-wang Mountain there. The Yi stonemasons were also active in producing stone sculpture in Yamato, such as the two lions at the Tōdaiji Beyond the Secular 217

Great South Gate. In other words, they undoubtedly had close ties with Chōgen from some years in the past. As for the people who lived in the fifty villages of the Sayama basin and personally hauled rocks and constructed dikes to repair the pond, the plaque tells us that they did so not only for worldly profits, but also so that the Buddha’s blessings might benefit all creatures in the universe forever. According to the inscription, people from all social classes and various walks of life—“monks and laity, men and women, novices and children, beggars and outcasts”—supplied labor for repairing the dikes and aqueducts. This points to an important socioeconomic phenomenon of medieval times: the relationship between evangelical Buddhist monks and low-status groups, such as those outcasts called hinin. The latter—literally termed nonpersons—were a loosely defined group that included beggars, those who suffered from leprosy and other disfiguring diseases, and specialists in kiyome, the cleansing of polluted precincts and removal of polluting objects.17 While we know that monks from later in the period, such as Eison (1201–1290) of Saidaiji and his disciple Ninshō (1217–1303), gave charity and administered the Buddhist precepts to outcasts,18 such activities are not well documented for the early Kamakura period.19 The plaque, however, documents Chōgen’s use of outcast labor for the construction of Sayama Pond and thus provides an early example of the relationship between evangelical monks and the outcast community, filling a gap in our historical understanding of this relationship. We know from later evidence that outcasts were quite helpful in projects sponsored by monks and temples. In 1266, for example, Eison held a religious service (kuyō) at Shinpukuji in Kawachi, headquarters of a group of metal casters, and he gave alms there to more than one thousand outcasts from Izumi, Kawachi, and Settsu, the same groups that must have provided labor for the Sayama repair project some sixty years earlier. The outcasts, who seem to have developed technical skills, probably participated in the construction of a building at the Shinpukuji site, for which archaeological remains have been discovered.20 Another example of outcast participation in construction projects comes from Hine Estate in Izumi Province. According to a map that provides information on a land-reclamation project dated 1316, monks from the Sai daiji branch temple Kumedadera oversaw the project, using labor from “people of the hill” (saka no mono) believed to be outcasts.21 218 Secular and Sacred

Chōgen’s Shingon Beliefs and the Sayama Pond Restoration Like many religious figures of his time, Chōgen was inspired by an eclectic Buddhist ideology that included Pure Land beliefs focused on the Bud- dha Amida, devotion to Buddhist relics, and Shingon esotericism. When Chōgen was young he studied at Daigoji near the capital, a temple established by Shōbō (832–909), founder of the Ono branch of Shingon and the first head monk of Tōnan’in at Tōdaiji. Although details of Chōgen’s early life are not clear, he tells us in Sazenshū that he performed ascetic regimens at Ōmine, Kumano, Ontake, and Katsuragi, well-known centers of mountain Buddhist practice based on esoteric concepts.22 He also founded an autonomous temple at the Shingon center of Mount Kōya. Recent research, particularly by Ishida Hisatoyo, further demonstrates the centrality of Shingon esotericism in Chōgen’s thought, as well as the equivalence of Amida and Dainichi Nyorai, central buddha of the Shingon school. Beginning with an analysis of Chōgen’s inscription on the bell at Kōya Enjuin, Ishida inquires into the significance of the five-tiered “triangular” stupas23 erected at each of Chōgen’s autonomous temples and examines his practice of placing Buddhist relics within the stupas. Ishida emphasizes the influence of the thought of the Shingon master Kakuban (1095–1141) over Chōgen’s own religious ideas. The essence of Kakuban’s philosophy is expressed by his treatise Gorin kuji myō himitsu shaku, which argues for the unity of Dainichi, represented by the five tiers, and Amida, represented by the kuji or nine characters of the Amida mantra.24 Kakuban’s ideas were represented in Chōgen’s thought not only by the stupas but also by the images of Amida he installed at each autonomous temple. Shingon concepts are represented on the Sayama plaque by the San- skrit seed characters for the five elements (gorin)—sky, wind, fire, water, and earth—and for the mantra of light. The latter, which expresses the idea that the light of the Buddha extinguishes all sins, represents faith in Dainichi, and it is natural that the gorin seed characters and the Sanskrit for the mantra of light appear together on the plaque. Taken as a whole, the Sayama plaque inscription represents a fusion of practical with religious concerns, specifically Shingon, a motivation that can be seen in the practices of other religious figures as well as those of Chōgen. In fact, a common thread linking Chōgen with Eison, who also devoted himself to helping people in their daily lives, is their mutual devotion to Beyond the Secular 219

Shingon teachings: when they were young both studied at Daigoji, and at Saidaiji, Eison launched a movement combining Shingon esotericism with adherence to the precepts. It seems that these concepts inspired a concern for people’s welfare in this world. Moreover, the materials that Chōgen used to construct the aqueducts for the pond suggest not only his technological innovativeness, but also the centrality of Shingon in his religious thought. Archaeological excavations have uncovered the remains of fifteen stone coffins near the pond’s central aqueduct. The coffins used were of a type called iekata, in which the lid resembles the roof of a house. The coffins were lined up end to end and holes were drilled in each end, resulting in a U-shaped pipe that followed the pattern of medieval wooden aqueducts. The laborers who repaired the Sayama Pond must have hauled the coffins from nearby burial mounds. Rather than fearing the wrath of the dead whose burial places had been dismantled, Chōgen and his followers appropriated their charisma and put it to work for a project that had both secular and religious dimensions. They must have based their action on a practical spirit within Shingon, which sought supernatural means to improve people’s daily lives. Faced with the illuminating truth of Shingon, those buried in seventh-century mounded tombs had no choice but to render service!

Medieval Villages and Irrigation Projects Another significant factor in the Sayama Pond restoration is that the repairs were requested not by estate or provincial officials, but by more than fifty villages (see figure 9.1). That so many villages in three provinces shared Sayama Pond’s irrigation water is an extremely important finding for the socioeconomic history of the early medieval age, since it indicates the extent to which villages from different estates and provinces cooperated with one another. The figure is most likely accurate. After all, Chōgen did address his remarks, a report on accomplished work, to “all buddhas and bodhisattvas of the three worlds and ten directions.” Thus the plaque is unlikely to contain a falsehood. Moreover, Chōgen seems to have been particularly careful with figures: the plaque records the very days that ground was broken, pipes were laid, and the project was brought to completion. The meticulous nature demonstrated in Sazenshū as well indicates that the details in the plaque are indeed reliable. The villages that divided the water from Sayama Pond did not do so randomly, and it goes without saying that allocation must have been strictly managed. We can 220 Secular and Sacred imagine that on the day the repair work was completed, representatives from each village gathered at the dike, to gaze at the newly inscribed plaque as water began to flow from the pond to their fields. The term used for villages in the Sayama plaque is gō, which strictly speaking indicates a political unit: a subdivision of the district, which itself was a subdivision of the province. Unlike estates, which could sprawl across district or provincial borders, gō were units located within

Fig. 9.1. Villages drawing irrigation water from Sayama Pond. From Ōyama Kyōhei, Nihon chūsei no mura to kamigami, Iwanami 2011. Used with permission. Beyond the Secular 221 an official territorial hierarchy: province, then district, then gō, and they were sometimes divided into smaller villages called mura.25 Some of the villages referred to in the inscription on the plaque must have been located in the alluvial valley of the Amano River, but the height of the northern dike at Sayama Pond indicates that others were spread across the upper terrace. They overlapped territorially with a number of estates and provincial landholdings (kokugaryō), and the irrigation water from the pond crossed the borders that marked these holdings. By the thirteenth century, through sharing the pond’s water, the villages had formed a loose league of communities that transcended the political boundaries of province, district, or estate. At Sayama, Chōgen ignored the estate framework and worked instead with villages, which were the setting for people’s daily lives, in contrast to estates, established as units of land possession by powerful aristocrats and religious institutions.26 Medieval gō were equivalent to communal villages or incorporated these units. We might more properly call them hitosato or murazato, emphasizing their nature as communities rather than as administrative units, as places where people lived and travelers could take refuge. This is probably how Chōgen thought of them. The nucleus of such villages was the water that irrigated their rice fields. By Chōgen’s time the number of villages dependent on Sayama Pond water was just about triple that recorded in Wamyō ruijushō, a tenth-century encyclopedic work that lists geographical data, among many other items.27 At the start of the thirteenth century, then, the villages must have suffered from insufficient water supplies, and Chōgen’s work was designed to resuscitate the pond and thus preserve these centers of human habitation. Along with other projects, such as the repair of Uozumi Anchorage along the Seto Inland Sea, and of roads and bridges in Iga and Bizen Provinces, the restoration of Sayama Pond was a project that enhanced the popular welfare.

Land Reclamation and Other Projects on Tōdaiji Holdings Chōgen was also involved in land reclamation and irrigation projects on estates that Tōdaiji held as proprietor. He was given management rights over a number of these estates, which were designed to provide revenues for Tōdaiji reconstruction. In his will, Chōgen allocated revenues from these holdings for Buddhist services at the Great Buddha Hall, the Hachiman Shrine, Jōdodō, and Kaidan-in (the Ordination Platform) at Tōdaiji, and the Watanabe autonomous temple. Almost half of the 222 Secular and Sacred total was reserved for the Great Buddha Hall. Not only was the nucleus of Chōgen’s religious activities the restoration of this structure, but the preservation and development of regular Buddhist ceremonies there was an important goal. While Chōgen’s responsibilities included the collection of donations for the reconstruction of Tōdaiji, the list in his testament includes no appropriations for rebuilding the Great Buddha Hall. These were probably calculated and funded separately, since he had been assigned revenues from Suō Province in 1186 and Bizen in 1193 to provide for reconstruction expenses.28 Concerning the allocation outlined in the testament, he wrote, “Allotments should be made without neglect, taking account of all the rent rice from the estates and placing it in the [Tōnan’in] storehouse. Without dividing it up according to its source, it should be all stored together as provisions for whatever Buddhist services [require it].” In other words, these allotments were considered a kind of general fund on which the various subunits within Tōdaiji, and the Watanabe autonomous temple, could draw when needed.29 This indicates Chōgen’s intention to avoid intervention by the monastic authorities at corporate Tōdaiji, while placing management rights and control of assets solely in the hands of the head monk of Tōnan’in after Chōgen’s retirement. To exploit the resources of Tōdaiji estates, Chōgen had to involve himself in the “secular” work of land reclamation. His work on Ōbe Estate, a Tōdaiji holding in Harima Province, is a case in point.30 The actual history of that holding begins with Chōgen—Tōdaiji owned the estate but did little to develop the land before his involvement in 1192—and religious authority was used there to drive the land reclamation process. In medieval times the terrain of the estate was divided between lowlands along the Kako River and terraced higher ground to the east.31 The lowlands had long been cultivated using water from the river, and rents from these fields went directly to Tōdaiji. In the northeastern corner of the terrace, Chōgen established Jōdoji as an autonomous, financially independent temple charged with managing the estate. Kanohara, the area just west of the temple, was still uncultivated, and Chōgen opened the land to cultivation, designating the harvest to support Jōdoji. It is possible that he worked on the lower terrace as well: a canal called Tera-yu (“Temple Canal”) that runs through the lower terrace may have been connected to Chōgen’s development efforts, but a final determination of that matter depends on archaeological evidence that has not yet been found. Beyond the Secular 223

Evidence for Chōgen’s activities on Ōbe Estate comes from several sources, including a document dated 1192 as well as his aforementioned writings. The document, in the form of an order issued to the officials of Ōbe Estate, lays out Jōdoji’s claim to revenue from Kanohara paddies, and his testament notes the establishment of Jōdoji, the construction of its two halls (Jōdodō and Yakushidō), and the dedication of revenues from Kanohara to the new temple.32 Development of Kano- hara involved the construction of irrigation facilities such as the North Pond, a reservoir immediately to the west of Jōdoji. In short, the newly constructed temple had an independent economic base, as Chōgen’s testament notes as well.

1. An excellent place in the northeast corner of the estate was chosen by divination, and a new autonomous temple, called Namuamida- butsu bessho, was established. 2. A three-bay square tile-roofed chapel was constructed and called Jōdodō. A colossal gilded Amida triad was installed there.33 3. Another three-bay chapel, Yakushidō, was also constructed, and some eight hundred images collected from abandoned temples on the estate were installed therein. 4. Fields of Kanohara, on the northeast edge of the estate, were dedicated to Jōdoji. Revenues were to be forever devoted to the Jōdodō Amida, enabling the efforts of temple monks. The land reclamation efforts would provide revenue for lamp oil and nenbutsu services (rituals to chant the name of Amida Buddha).

Regarding the fields dedicated to the Harima autonomous temple, the testament continues: “Previous levies exacted by estate officials and the main temple should not be assessed. If Kanohara were left permanently undeveloped, it would be unprofitable land.”34 Thus Chōgen affirmed the independence of the Harima autonomous temple and firmly laid claim to the revenues from reclaimed land in Kanohara to support its religious services.

Functions of the Autonomous Temples Chōgen was also involved in development projects on or near several other Tōdaiji holdings. In most cases autonomous temples like Harima Jōdoji were involved in the supervision of these projects. Chōgen’s 224 Secular and Sacred

testament makes it clear that the estates and autonomous temples formed the material base that supported his religious and economic activities. Not only land reclamation was involved. For example, Amidaji in Suō, a province under Tōdaiji’s control, oversaw logging operations that provided lumber for the Nara temple’s reconstruction, and the Watanabe autonomous temple in Settsu Province oversaw the storage of lumber as it was shipped up the Seto Inland Sea to Nara. The autonomous temples also served as bases for Chōgen’s religious activities, which included the establishment of baths where nenbutsu rituals were performed. The following document, signed by local officials of Suō Province in the eleventh month of 1200, describes the establishment and functions of Amidaji:

The performance of the nenbutsu and meritorious activities in the bathhouse are things all the buddhas long for, and [these rituals establish] especially good roots. Thus Namuamidabutsu establishes such facilities in every suitable location. . . . He divined a place where water and timber were accessible at the foot of a mountain east of the provincial capital. . . . There he established ceaseless nenbutsu rites and an all-day bathhouse [ceremony]—in other words, the most meritorious of activities.35

At the four autonomous temples in existence when he wrote his testament, Chōgen built a Jōdodō (Pure Land Hall) that enshrined images of Amida, as well as a bathhouse, and centered his activities on nenbutsu and bathing practices. It seems clear that his intention in establishing the autonomous temples was to realize Amida’s Pure Land in this world. In the testament, the four autonomous temples are listed in their order of establishment, but the central one was obviously the one at Tōdaiji, in an area between the Nigatsudō and the Great Buddha Hall (see figure 4.1 in the chapter by Endō Motoo). If one climbs the stone steps on the slope from the north side of the east corridor of the Great Buddha Hall, one may see the bell tower reconstructed in the Kamakura period by Eisai, Chōgen’s successor as chief collector of funds for Tōdaiji building projects. The bathhouse Chōgen built at this autonomous temple is still standing, and inside one can find the iron cauldron made in 1197 by Kusakabe Koresuke, a member of a Kawachi metal casters group. The Watanabe autonomous temple is an interesting example of the way Chōgen combined religious and economic activities at his facili- Beyond the Secular 225 ties. It was established in Nanba (present-day Osaka) along the Inland Sea. Timber from Suō designated for Tōdaiji rebuilding projects would be stored temporarily at this temple, then transferred for shipment up the Yodo and Kizu Rivers to Kizu, still some distance from Nara.36 The structures at Watanabe and Kizu were substantial in size; both were two stories in height, nine bays wide, and two bays deep. As was the Tōdaiji autonomous temple, the Watanabe temple was managed directly by Tōnan’in, but it also had an independent economic base after 1201, when Hachijō-in, half-sister of the retired sovereign Go-Shirakawa, donated Zunashi Estate in Settsu Province to provide support for nenbutsu practitioners, lamp oil for rituals, and offerings to the Ōji deity (kami) at the temple’s Jōdodō.37 To summarize, Chōgen accumulated assets such as estates and autonomous temples, and these formed the material support for his religious and economic activities. Large lumber storehouses attached to the Tōdaiji and Watanabe temples were bases for his economic activities and distribution network. The assets were passed on to Jōhan, the head monk of Tōnan’in. Some of the autonomous temples, as noted above, were under Tōnan’in’s direct control, but the Harima temple was established with an independent economic base, and the Watanabe temple acquired one long before Chōgen’s death. But the wish expressed in Chōgen’s testament, that Tōnan’in continue to oversee these assets, was in the end not fulfilled, as corporate Tōdaiji moved to control them. The interests of Tōdaiji and Chōgen differed fundamentally: the great Nara temple saw estates and autonomous temples as tools to promote the practice of Buddhism at Tōdaiji and the prosperity of its own monks, while Chōgen looked beyond the Tōdaiji framework to the restoration of the Buddhist law overall.38 Chōgen’s “good deeds” supported the central pillar of royal authority, which fused the Buddhist law (buppō) with secular rule by the court (ōbō). For him, the restoration of the Tōdaiji Great Buddha was central to the existence of this religious ideology. Not only were his seven autonomous temples the physical bases for his particular religious activities, they can also be seen as centers from which to realize the goal of a peaceful and prosperous realm overall. In different ways, his activities in several provinces show how he pursued this goal. Perhaps the most representative are his actions in Suō, a province turned over to Chōgen for his use in the Tōdaiji reconstruction project. There he established close ties between the autonomous temple Amidaji and provincial and local officials, some dozen of whom 226 Secular and Sacred inscribed their names on a small iron stupa, dated 1197, that remains at that temple today. Moreover, at Chōgen’s behest, forty-one local officials from eleven lineages cosigned the document dated 1200 allotting public paddy and dry fields to Amidaji. According to the document, “local officials who for generations had acted in the place of the provincial governor” have become “patrons (danna)” of Amidaji, and “until far in the future, their descendants will take this temple [Amidaji] as their family temple (ujidera).” Such officials bore major responsibility for government operations in late Heian and Kamakura times, and the Suō autonomous temple, not far from provincial headquarters, was at the center of their activity. Chōgen, moreover, did not limit his activities to Amidaji, but contributed to the construction of buildings at several important local shrines. Like Suō, the province of Bizen was also handed over to Tōdaiji and placed under Chōgen’s control. He developed some 260 chō of land in scattered places in the province and negotiated to exchange them for a contiguous parcel that would form Noda Estate.39 Chōgen also constructed the temple Jōgyōdō, established a large bathhouse at the provincial headquarters, built another temple, Bukōji, on Toyohara Estate, and restored twenty-two additional temples.40 Such activities were aimed at enhancing the stature of the provinces he controlled.

Fixing the Roads and Helping the People The purpose of transportation policy in Heian and Kamakura times was to guarantee safe travel within and beyond each province. Chōgen’s work to improve roads and bridges, however, should be seen in the context of a worldview that extended well beyond the archipelago to embrace all of East Asia. In particular, Chōgen had made great efforts to import advanced culture and technology through contact with the human and material resources of Song dynasty China. His public works projects, therefore, should be seen as efforts not only to improve transportation routes within Japan but also ultimately to link those routes with the continent. Sazenshū records how he cleared a road in Bizen Province:

From times of old, people traveled back and forth in the shade of Mount Funasaka in Bizen Province. Some suffered great distress, Beyond the Secular 227

while others lost their lives. So I collected donations from noble and base, had the brush cut back, and made a safe path, bringing a final end to the danger of bandits.

In Iga Province as well, Chōgen repaired roads to make them safe. According to Sazenshū:

I cleared paths in various mountain locations in Iga Province, enabling people to travel in safety. Because the roads were very dangerous in that province, people and horses traveling back and forth met with many calamities—they were either injured or killed. Thus in order to help them, I repaired all these dangerous places, putting an end to the laments of man and beast.

These projects combined service to the people with service to Tōdaiji, since from ancient times the province had provided timber for Tōdaiji and one of the Nara temple’s most important holdings, Kuroda Estate, was located there.41 In addition, Chōgen contributed to the repair of bridges such as the one near Kiyomizudera that spanned the Kamo River in the capital, the Seta Bridge along the Tōkaidō Highway, and the Watanabe and Nagara Bridges near the mouth of the Yodo River. Thus he helped to preserve roads that linked the capital regions of Kyoto and Nara to the periphery. Another important project was his repair of Uozumi Anchorage along the Inland Sea, an important way station for the shipment of goods. Sazenshū tells us:

As for the Uozumi Anchorage, that island was constructed long ago by Gyōki Bosatsu in order to help people. But as the years and months passed, it was destroyed by the billows of the ocean, and thus ships heading east or west encountered wind and waves. We do not know how many thousands of persons drifted to their deaths. Thus I reached this place sacred to the bodhisattva [Gyōki] and endeav- ored to restore it to its original condition.

In 1196 the Council of State issued an order to Settsu Province officials to exact levies for the restoration of Uozumi and Ōwada Anchor- ages and a small island at Kawajiri, all of which provided refuge for boats traveling along the Inland Sea. The order quoted a request from 228 Secular and Sacred

Chōgen for a levy of one shō per koku of rice42 shipped from the southern and western circuits of San’yō, Nankai, and Saikai to pay for the repairs. He also asked each estate and public holding in these circuits to supply a boat, to be disassembled and used as construction materials. Holdings in Yamashiro, Kawachi, Settsu, Harima, and Awaji were to supply oak and bamboo to be used as posts, and estate proprietors in those provinces were to furnish laborers from their own personal holdings. In addition, Chōgen requested an inventory of damaged boats from ten provinces, as well as from the local ports of Kawajiri and Yodo. This was a large-scale plan to overhaul the Inland Sea route. Judging from what was recorded in Sazenshū, only Uozumi was actually restored. But the document gives an idea of the grandeur of Chōgen’s intended project.

What was the significance of Chōgen’s activities for the formative process of Japanese medieval society? Both religious and economic in nature, his projects included land reclamation and the restoration of irrigation facilities, the construction of transportation network facilities such as ports, bridges, and roads, and the securing of provincial highways. These projects achieved a scale and scope that could not be attained by the local landholder class in early medieval times. Such activities were based on an ideology of development inspired by a practical spirit within Shingon esotericism. Unlike Hōnen or Shinran, Chōgen did not form a religious group that persisted as an identifiable school of thought, but the influence of his religious practice endured nonetheless. While Chōgen’s religion may not show introspective depth, he had an unshakeable will no doubt developed through mountain asceticism, and he used his strong body freely. He concerned himself with worldly affairs and aimed at the realization of a tranquil realm for all classes in society, including ordinary people. His projects, focused on that aim, manifest the characteristics of an enlightened ideology at the dawn of the medieval age.

Notes

1. From the inscription on the plaque found in a 1989 archaeological investigation at Sayama Pond. In Japanese, see Ōyama 2012, 82–83. 2. Ōyama 2012, 81. 3. For research on Chōgen, see Kobayashi 1965, 1971; Nara Kokuritsu Bunkazai Kenkyūjo 1955; Horiike 1955; Nakao and Imai 1983; Karikome 1998. Beyond the Secular 229

4. For printed versions of the testament, see Hyōgo-ken shi, chūsei 5, 90–93 and OS 4:532–535 (# 224). The Historiographical Institute of the University of Tokyo has a copy of the manuscript. For printed versions of Sazenshū, see the version compiled by Nara Kokuritsu Bunkazai Kenkyūjo 1955. The text also appears in OS 4:537–545 (# 227). For a full English translation with commentary, see Rosenfield 2011, 209–231. 5. Jōhan’s predecessor Nyoamidabutsu Shōken was actually Chōgen’s first choice as heir, but he predeceased Chōgen. 6. For an analysis of the testament, see Nakao 1977. 7. For a discussion of the dating of Sazenshū and its handwriting, see the explanation by Tazawa Yutaka in Nara Kokuritsu Bunkazai Kenkyūjo 1955. For a discussion of the use of the reverse sides of documents to record other texts, see the chapter by Endō Motoo in this volume. 8. Remains of an original aqueduct have been found beneath a current one. A tree-ring analysis of its wood indicates that it came from a tree logged in 616. 9. Details of these repairs can be found in Ichikawa 1999, sec. 2, part 6. 10. For an English-language summary of the inscription, see Rosenfield 2011, 235–236. The plaque, along with artifacts from various stages of construction and repair of the pond facilities, can be viewed at the Sayama Pond Museum (Saya- maike Hakubutsukan) in Ōsaka-Sayama City, Ōsaka Prefecture. 11. Matsushima 1992, 106; photograph on page 139. 12. See Ichikawa 1999. 13. Sazenshū, item 143. OS 4:537–545 (# 227). 14. Matsushima 1992, 109. 15. Tōdaiji zōryū kuyōki (Record of the commemoration service for Tōdaiji’s reconstruction). 16. Gyokuyō (Jeweled leaves), Juei 3 (1186)/1/5). 17. The precise nature of hinin, and the degree to which they suffered dis- crimination, has long been a topic of scholarly debate. See “Chūsei no mibunsei to kokka,” in Ōyama 1978. 18. Hosokawa 1987. In English, see Quinter 2015. 19. “Chūsei no mibunsei to kokka,” in Ōyama 1978. 20. Miura 1990, 288–294. 21. Miura 1990, 344–347; Miura 1993. For a study of Hine Estate, see the chapter by Hirota Kōji in this volume. 22. Ōmine is in Yamato Province, Kumano in Kii Province, Katsuragi on the border between Yamato and Kawachi Provinces, and Ontake on the border between Shinano and Mino Provinces. 23. This is a special form of the five-tiered pagoda in which the third tier, representing the element fire, takes a triangular form. 230 Secular and Sacred

24. Ishida 1988, 354–355. For a discussion of Kakuban’s text in English, see Waterhouse 1996, 8–11. 25. Translator’s note: The usual translation of gō, used elsewhere in this volume, is township; here, I have translated it as village, to emphasize Professor Ōyama’s characterization of the gō as a community rather than as a political unit. 26. See Ōyama 2012, 380–394. 27. Wamyō ruijushō (Topical collection of Japanese terms) 1992, 27オ (Izumi), 27ウ–28ウ (Settsu), 28ウ (Iga). 28. See KI 1:87–89 (# 133, 1185/7) for instructions from the Tōdaiji cabinet and monks to Suō provincial officials to allocate revenues for expenses related to the Great Buddha. 29. For the complex structure of medieval Tōdaiji, see the chapter by Endō Motoo in this volume. 30. For another view of Chōgen’s land reclamation work on Ōbe Estate, see the chapter by Janet Goodwin in this volume. 31. See the chapter by Nishida Takeshi in this volume. 32. The document, a kudashibumi, is reproduced in OS 4:527–529 (# 219). Also see the chapter by Janet Goodwin in this volume. 33. For a discussion of the Jōdoji Amida triad, see the chapter by Yoshiko Kainuma in this volume. 34. Hyōgo-ken shi, chūsei 5, 90–93; OS 4:532–535 (# 224). 35. KI 2:405–410 (# 1163) (from the records of the paddy and dry fields held by Suō Province Amidaji). 36. Hiraoka 1980, 77. 37. Sazenshū. OS 4:537–545 (# 227). For a discussion of the estate holdings and strategies of royal ladies such as Hachijō-in, see the chapter by Sachiko Kawai in this volume. 38. Kohara 2009, 21. Also see the chapter by Nagamura Makoto in this volume. 39. OS 4:531–532 (# 223). 40. Sazenshū. OS 4:537–545 (# 227). 41. For a discussion of Kuroda Estate as a Tōdaiji holding, see chapter 7 in this volume, by Joan Piggott. 42. In other words, this was a 1 percent tax. For more on this project see the chapter by Janet Goodwin in this volume. c h a p t e r 10

Claiming the Land Chōgen and the Development of Ōbe Estate

Janet R. Goodwin

For many years, this land has been wilderness without cultivators, and has gone to waste as the unproductive haunts of boar and deer. But now, donating the lands [to provide for religious offerings and services] opens them to cultivation for the first time.”1

So wrote Chōgen, the Buddhist monk in charge of efforts to rebuild the great Nara temple Tōdaiji, in a document dated 1192 and addressed to officials of Harima Province. (For sculptural portraits of Chōgen, see plates 1 and 17.) Chōgen thus linked the opening of agricultural land— in this case on Ōbe Estate, a Tōdaiji holding in Harima—with a fundamental religious purpose. Locally, the development project had two aspects: the enlargement of the estate’s arable paddy, and the establishment of a temple, Jōdoji, which depended upon the new paddy, supervised its cultivation, and served as the center of a religious community based on Chōgen’s Pure Land beliefs.2 The project is one example of the way Chōgen linked religious proselytizing, secular public works such as land development, and the formation of local communities, all in the service of a realm-wide project, the reconstruction of Tōdaiji, a temple originally built by order of the throne. The early development of Ōbe Estate depended ultimately on religious authority, a common thread in an institutional arrangement in which temples and shrines managed landholdings as major players in the economy and politics. (For a map of Harima Province showing the location of Ōbe Estate, see figure 16.1.)

231 232 Secular and Sacred

The incineration of many structures at Tōdaiji at the outset of the Genpei civil war (1181–1185) was a tremendous blow to the throne and court, and even while the war was still raging, central authorities set about restoring the venerable temple. Especially important was its central image, a gigantic figure of Rushana Buddha—the Great Bud- dha (Daibutsu)—and the hall in which it was located. Built in the mid- eighth century in accord with a vow by the ruler Shōmu (701–756, r. 724–749), Tōdaiji symbolized royal authority and the mutual support of the throne and the Buddhist establishment. The great temple had a role in popular history as well: its construction had been funded in part by a campaign to collect donations from the public, headed by the monk Gyōki (668–749), who began his career by preaching to ordinary people. By the late twelfth century, Gyōki was depicted as a builder of roads, bridges, anchorages, and irrigation systems as well as of temples.3 Thus he was an apt model for Chōgen and his multifaceted endeavors. In Chōgen’s time, sacred and secular projects were not viewed differ- ently. Constructing an anchorage, for example, reduced the impact of storms at sea and thus served a compassionate Buddhist purpose. In addition, secular projects won the hearts of potential donors to religious causes. Beneficiaries of a monk’s public works efforts might form a community of believers, or a lay community would turn to Chōgen because of his fame and religious standing and ask for his help in a project conceived by the community. Thus in addition to accomplish- ing the restoration of Tōdaiji’s Great Buddha and the reconstruction of the Great Buddha Hall (the Daibutsuden), Chōgen helped to open land for cultivation and to develop the transportation infrastructure. Ostensibly this was all for the sake of Tōdaiji, but in fact it benefited many local groups as well. Such activities took place under the shadow of powerful authorities: throne, court, military government, and Tōdaiji itself. Chōgen badgered the throne and court for edicts granting land to Tōdaiji and assessing fees to rebuild anchorages along the Inland Sea. He badgered Tōdaiji to devote one portion of its own lands to support a local, virtually autonomous temple, and he secured support from military over- lord Minamoto no Yoritomo for logging operations in Suō Province to rebuild the Great Buddha Hall.4 Chōgen both served and manipulated authority, and often the beneficiaries of his activities included ordinary people. Yet unlike the early Gyōki, accused in the eighth-century Claiming the Land 233 chronicle Shoku nihongi of subversive activities, Chōgen did not foment dissent. Like Gyōki in later years, he served Tōdaiji well, and through the temple he also served the throne that used Tōdaiji to manifest its own glory. As the hands-on director of the campaign to rebuild Tōdaiji, Chōgen had many jobs. He collected donations for the project; he selected master sculptors and builders; he investigated architectural models; he sought sources of materials such as tile and lumber; and he acquired, developed, and managed landed estates to add to Tōdaiji’s store of wealth in the form of rice, the prime medium of exchange in Japan at the time. One of his most important projects was the revival and expansion of Ōbe Estate, a moribund Tōdaiji holding. The work of Chōgen and his disciple and probable nephew Kan’amidabutsu (Kan’a; probably d. 1243) to cultivate the wilderness and establish and manage Jōdoji demonstrates some of the linkages among religion, public works projects, and local communities in early medieval Japan.

Before Chōgen: Tōdaiji’s Efforts to Develop and Control Ōbe Estate Located in the interior of Harima Province along the Kako River, Ōbe Estate was established in 1147 as a Tōdaiji holding. According to a document dated 1162, Tōdaiji had exchanged three disconnected parcels for the land that would become Ōbe Estate. That same document, however, demonstrates that superior tenure to Ōbe was already in dispute: the provincial government was demanding the return to its jurisdiction of the new holding, and Tōdaiji was to take back its three original parcels. Tōdaiji appealed to the throne for an edict confirming its rights to Ōbe Estate.5 Tōdaiji’s intention was to make the land into an estate that paid dues to the temple as rents rather than to the government as taxes. The 1162 document does not explain why the provincial government preferred Ōbe to Tōdaiji’s former holdings, but the estate’s inclusion of flat, productive land along a major river was no doubt a major factor. More- over, the Tōdaiji representative claimed that “just as we were making the land productive,” the provincial government tried to get it back. With- out more documentation, of course—we have nothing from the other side—it is impossible to assess the dispute properly. Ōbe Estate receives little attention in the records for some thirty years thereafter. Lists of 234 Secular and Sacred

Tōdaiji records from the 1160s to 1180s mention those pertaining to Ōbe, however, indicating that the temple had won its case. Later documents suggest that before Chōgen’s involvement, Tōdaiji did little more than set the holding’s boundaries, a task that had to be done all over again in 1192, when development began in earnest. The estate was divided topographically into three terraces, rising in elevation from west to east (see plate 2). The section near the river—the flattest and easiest to cultivate—was developed long before Tōdaiji took over. It was inhabited in Yayoi and Tumulus (Kofun) times and includes remains of a seventh-century Buddhist temple. On the west side of the estate, along the Kako River, traces of the jōri grid system used to appor- tion land in earlier times indicate that this area was under cultivation well before Chōgen’s arrival.6 That already developed area would later be known as the satokata, the place of villages. According to Kawabata Shin, the land further east was uncultivated when Chōgen took charge of the estate; as Nishida Takeshi points out in this volume, Tumulus graves and temples were situated there. Chōgen’s job was to renew and reorganize cultivation efforts on the lower terrace and to develop the middle terrace and part of the upper terrace.7

Chōgen Revitalizes Ōbe Estate In 1181 Chōgen was appointed chief alms collector (daikanjin) for Tōdaiji’s reconstruction, through an order formally promulgated by the three-year-old Antoku Tennō (1178–1185, r. 1180–1185) but in reality emanating from the retired monarch Go-Shirakawa (1127–1192, r. 1155–1158).8 Chōgen was to supervise the rebuilding project and to acquire and manage resources to support it. In 1186, public revenues from Suō Province were designated for Tōdaiji, and Chōgen was named master of the province (chigyōkokushu), giving him broad authority over the disposition of these revenues. Similar authority over Bizen Province was granted to him in 1193.9 In Harima, he seems to have focused his efforts specifically on Ōbe Estate. In a petition dated 1196/2/7, Chōgen requested that revenues from several Tōdaiji holdings, including Ōbe, be devoted to services centered on the Great Buddha, the restoration of which had been completed in 1185.10 The efforts of Tōdaiji and Chōgen to revitalize Ōbe Estate began in 1192, when several documents were issued to confirm the temple’s authority over the holding. Two of them, one written on 8/15 and the Claiming the Land 235 other probably on 8/23, were fortuitously discovered on the back of a folding screen held by the Beinecke Library at Yale University. Both concern a deed that granted the township (gō) of Ōbe to Tōdaiji as an estate, an order that ultimately originated with the office of Go- Shirakawa. Appended to the second one is Chōgen’s kao, a stylized handwritten signature that functioned as a seal.11 In a third document, dated 8/25, the court directed provincial officials “to re-establish as of old the four decayed border markers of Tōdaiji’s holding, Ōbe Estate in Kamo District.” A fourth document, issued on 9/2 by the on-site proxy for Harima’s absentee governor, ordered local residents to cease violating temple territory, as commanded by the throne.12 These documents signaled Tōdaiji’s intent to develop the land and its recourse to the highest authority, the throne, to support its efforts. Chōgen was put in charge, and he asked Go-Shirakawa to allot income from the estate to Chen He-qing, the Chinese artisan who had supervised the casting of the Great Buddha and was now managing subsequent rebuilding efforts at Tōdaiji. This request, which must have been made before Go-Shirakawa’s death in the third month of 1192, is documented in Chōgen’s will dated 1197.13 On 1192/9/27, Chōgen issued an order to officials of the estate. It does not deal with revenues for Tōdaiji reconstruction; rather, it indicates Chōgen’s intention to establish a base independent of the Nara temple. Chōgen demanded that the wilderness of Kanohara—at the eastern edge of the estate—“be donated immediately [for support of] the Namuamidabutsu autonomous temple [Jōdoji]; that this land then be opened for cultivation under temple authority; and that rents [from these fields] be used to provide sacred lamp oil for Jōdodō and Yakushidō [within Jōdoji], and to support monks’ recitation of the continuous nenbutsu [there].”14 In other words he wanted to establish and support a bessho, an autonomous temple not under Tōdaiji control, on the Nara temple’s land. At the Jōdodō, one chapel within this temple, thirty monks would continuously chant the sacred name of Amida Buddha (the nenbutsu), “in order to pray for the peace of our sacred realm, the fulfillment of [Shōmu’s original] vow [to construct Tōdaiji], [the benefit of] the entire cosmos, and the extinction of sin and fostering of virtue for the sake of salvation for all.” While reconstructing Tōdaiji was an important aim, it was only part of Chōgen’s religious agenda. His political agenda, in fact, was to keep the Tōdaiji holdings he supervised out of 236 Secular and Sacred the control of the Tōdaiji monks’ assembly, with which he had some serious issues. For example, after the Great Buddha Hall was finished, the monks wanted to build a residence for themselves and a lecture hall, but Chōgen favored a pagoda, and he got his wish.15 In 1192, Chōgen specified that in the end—in other words, after his death—the holdings, including Ōbe Estate, should be placed under the control of the Tōdaiji cloister Tōnan’in, not under that of the corporate temple. Tōnan’in was founded in 904 by Shōbō of the Shingon temple Daigoji, where Chōgen had practiced early in his life; the two Tōnan’in head monks successively designated as Chōgen’s heirs, Shōken and Jōhan, were of Shōbō’s lineage.16 The deep connections Chōgen felt with that lineage overrode his relatively new ties with Tōdaiji. Chōgen, however, did not place absolute trust in Tōnan’in: he asked the throne to guarantee that Tōnan’in would not obstruct the supply of lamp oil and other needs for nenbutsu services at Jōdodō. As ultimate protection for Jōdoji, he invoked the curse of the gods on any temple director, resident monk, or estate manager who might violate his directive: “In the present life may he suffer white and black leprosy, and in the next life may he fall to the bottom of endless hell without any escape!”17 The year 1192 was thus a crucial one for Ōbe Estate. The court reaf- firmed Tōdaiji’s rights, Chōgen founded Jōdoji, Tōnan’in was given ultimate authority over the estate, and, according to Jōdoji engi, a running account of Jōdoji history probably compiled in 1372, Kan’a entered the scene.18 It is not clear how much time Chōgen spent on the estate, and he seems to have turned over day-to-day operations to his disciple. Chōgen’s order of 1192/9/27, however, indicates his personal presence and deep emotional involvement with religious affairs on the estate:

Within this estate there are a number of old ruined temples. Although it would be best to restore them all, I do not have time to take care of such matters while Tōdaiji’s construction is still underway. Since I know that [the old temples] have not been repaired, I fear that it will be hard for me to escape retribution for failing to do so! There- fore, we have divined [a location for a temple] at the northeastern corner of Kanohara, gathered building materials [from the ruined temples], and constructed a hall in which to install a number of [rescued] Buddhist images. The autonomous temple shall be called Namuamidabutsuji,19 and its [first] chapel shall be Yakushidō. We Claiming the Land 237

are also constructing a new Jōdodō, in which to install a standing gilded Amida triad, colossal in size.

According to Jōdoji engi, construction of Jōdodō was begun in 1194, and its dedication ceremony took place in 1197. The structure exists today, as does its magnificent Amida triad, produced during this interval by the noted sculptor Kaikei20 (see plates 16 and 18).

The Autonomous Temple Jōdoji If the estates Chōgen managed on Tōdaiji’s behalf were the economic foundation for his religious activities, temples such as Jōdoji were bases for these efforts.21 Jōdoji was one of seven temples that Chōgen established as bessho, literally, “separate places.” All but the one on the grounds of the Shingon monastery at Mount Kōya seem to have played a part in Tōdaiji’s reconstruction. Alms-collecting activities were based at Jōdodō within Tōdaiji precincts, logging operations in Suō Province were supervised from Amidaji, the Watanabe bessho in Settsu served as interim storage for lumber shipped from Suō to Tōdaiji, Jōdodō in Bitchū may have been used to oversee tile making, and Shindaibut- suji in Iga and Jōdoji in Harima were bases to oversee Tōdaiji estates. According to Chōgen’s autobiography Namuamidabutsu sazenshū (the collected good works of Namuamidabutsu, probably written in 1203),22 six of these seven temples enshrined an image of Amida as its central object of worship, and six also had a bath, which served as a location for ritual recitations of the nenbutsu.23 What were bessho, and why did Chōgen so diligently establish them? Although it was on Tōdaiji’s holding, Jōdoji was generally not considered a branch temple (matsuji) of Tōdaiji. Had it been so, it would have been required to pay dues to Tōdaiji, which also would have had authority over appointments to its monastic staff. It was probably also not considered a betsuin (detached cloister) of Tōdaiji, a looser and less restrictive form of affiliation between a local temple and a major institution.24 The defining feature of a bessho, according to Ōyama Kyōhei, was its financial independence from the main temple; hence the designation of Kanohara lands to support Jōdoji.25 Yet its status was curiously ambiguous, and Chōgen himself seems uncertain. The 1192 document in which Chōgen established Jōdoji refers to Tōdaiji as the “main temple” or honji, leading Karikome Hitoshi to conclude that Jōdoji was 238 Secular and Sacred actually a Tōdaiji branch temple.26 But in this instance Chōgen was simply demanding that efforts of another temple to take over Jōdoji as its branch, or to cheat Jōdoji out of due revenues, be reported to Tōdaiji. Chōgen, speaking as the head of the Tōdaiji reconstruction project, perhaps quite naturally referred to Tōdaiji as the “main temple.” More telling, perhaps, is an inscription on the shaft of a scroll signed by Chōgen that refers to Ōbe Jōdodō—not Jōdoji—as a matsuji of Tōdaiji.27 Perhaps at this point, in 1194, Chōgen wanted for some reason to emphasize strong ties between the great Nara temple and the major Pure Land chapel at Jōdoji, then under construction. In any case, there is no indication anywhere else that anyone at Jōdoji sought to subordinate that temple to Tōdaiji. The specific designation of bessho, in fact, suggests that Chōgen wanted to keep these temples—and by extension the entire Tōdaiji rebuilding process—out of the control of the Tōdaiji monks’ assembly.28 Perhaps it was for this reason that in 1200, Chōgen successfully petitioned retired monarch Go-Toba (1180–1239, r. 1183–1198) to designate Jōdodō as a royal prayer chapel,29 adding a layer of prestige to claims that at least this one temple building was beyond Tōdaiji authority. Sometime after 1192, when Chōgen requested the donation of wilderness lands to support Jōdoji,30 the Kanohara area was developed into paddy. “Through the efforts of monks who performed Buddhist services at Jōdodō and devotedly served Amida,” Chōgen wrote in his testament of 1197, “this land was opened to cultivation.”31 What might this process have entailed? There are no records describing the land reclamation efforts in any detail, but archaeological remains indicate that once forest had been cleared, reservoirs were constructed on high land to irrigate fields below.32 Pertinent to the Jōdoji area are the North Pond (Kita Ike), immediately west of Jōdodō, and the South Pond (Minami Ike), less than a thousand meters to Jōdoji’s southwest. According to Kawa- bata Shin, the area first developed by Chōgen was probably that to the west of the temple where another pond, the Jizō (or Nigori) Pond, was also located; although the latter pond was small, it could easily collect runoff and irrigate large areas farther to the west.33 The South Pond, mentioned in medieval records,34 may have been part of the efforts by Chōgen or Kan’a to develop the area, although it is not certain when the pond was originally constructed. In any case, water from reservoirs on the Jōdoji elevation seems to have irrigated fields on the lower (middle) terrace to the west. Claiming the Land 239

The North Pond may well have been conceived by Chōgen, if not actually dug under his supervision. It is possible to raise the back (western) wall of the Jōdodō, allowing rays of the sun reflected by the pond to illuminate the heads of the Amida triad within. If ceremonies in the hall were conducted in the late afternoon, the reflection from the North Pond would have been especially glorious. I believe Kawabata Shin is correct in suggesting that the pond may have had a ritual as well as a practical purpose and that its construction dates from around the same time as that of the Jōdodō.35 (For Jōdodō, see plate 18; for the Amida triad, see plate 16 and figure 11.3.) Additional archaeological findings support the thesis that Kanohara lands were developed in the early thirteenth century, when Chōgen or Kan’a was managing the estate. According to Kawabata, remains of a road and dwelling sites dating from that time have been found at the Kiyotani site in Kanohara, immediately to Jōdoji’s west. The road runs from south to north and then veers to the east, perhaps skirting the temple. Implements found at the site include bowls, plates, and kettles, evidence of daily life, suggesting the presence of a farming village. The structures at Kiyotani seem to have been abandoned sometime in the fourteenth century.36 Along with the documents cited above, archaeological evidence suggests a strong role for Jōdoji in developing the eastern portion of Ōbe Estate. While we don’t know for certain the structures and methods Chōgen used to accomplish this, evidence from other projects, which I shall discuss below, can help to produce some tentative conclusions.

Chōgen and Local Communities The key to Chōgen’s success in the Tōdaiji building project lay in his organizational skills. No single person could have collected donations, established temples containing fine sculpture and paintings, developed transportation systems, and acquired, opened, and managed agricultural land. Chōgen accomplished his tasks by flattering and cajoling nobility, handsomely rewarding artisans, and alternately using and serving local communities. In the back country of Suō Province, where Chōgen acquired timber for rebuilding the Great Buddha Hall, he established baths for the loggers. In Settsu Province, he restored an old irrigation pond originally attributed to Gyōki.37 Along the Inland Sea coast, he repaired anchorages that had gone to ruin, providing refuge for mariners 240 Secular and Sacred in storms and way stations to ship lumber as well as rice from Tōdaiji’s far- flung holdings. And, of course, he promised spiritual benefits to those who helped with his projects, just as had Shōmu Tennō when he sought broad assistance for Tōdaiji’s initial construction. Chōgen, in short, used a variety of methods to build a network of support. Chōgen justified his projects in practical, religious, and historical terms, as typified by his argument for the repair of Uozumi Anchorage in Harima Province. In 1196, his request for official funding to restore the anchorage opened thus: “Uozumi Anchorage was established long ago in the Tenpyō era (729–749) by Gyōki Bosatsu.”38 After laying claim to the prestige of the famous monk who helped Shōmu build Tōdaiji, Chōgen then went on to lament the ninth-century destruction of the anchorage and unsuccessful attempts to repair it. The anchorage’s sad state was a life-threatening condition:

The banks along the shore have eroded more and more, and people speedily go to their graves. Whenever clouds thicken, the moon grows dark, the wind howls, and the stars disappear, [seafarers] have no choice but to lower their sails and abandon their oars, calling [for help] from the east and the west. Because of this situation, in recent years eighty or ninety percent of the public and private vessels from the San’yō, Nankai, and Saikai39 regions have been stranded and have perished.

Those who dwell near the anchorage and monks nearby lamented this, saying:

“That this port has come to such a state! Not only do we deplore the loss of profits from shipping, but we also grieve because people have even lost their lives in vain. If not the holy man [you, Chōgen], who will rescue us from this disaster? We implore you to take charge of the ruined facilities and quickly make plans for new construction. Moreover, [just as] Gyōki Bosatsu sought the support of ordinary people to build Tōdaiji, wise and virtuous men [can] restore this anchorage, closely scrutinizing its current condition. Why not tread in the footsteps of the Tenpyō era?”

Members of the local community, in other words, were imploring Chōgen to solicit donations to repair the anchorage. Chōgen objected, Claiming the Land 241 however, citing his unfinished work in reconstructing Tōdaiji. Yet pressed by the “fervent requests” of monks and lay believers, he asked himself, “If I violate their wishes, how can I achieve Buddhahood?” Thus he decided to help. Adding repair efforts at two other anchorages in adjacent Settsu Province, Chōgen proposed not an alms-collection campaign, but levies of rice, lumber, and laborers imposed by the throne on public and private holdings. Those assessed were the most likely users of Inland Sea shipping routes: holdings in the San’yō, Nan- kai, and Saikai circuits, with additional levies imposed on those in provinces nearby the anchorages. Chōgen concludes, “To gratefully receive the sage ruler’s beneficence will surely ease the tear-soaked grief of the common people.”40 The local community turned to Chōgen because of his success in soliciting help from all socioeconomic classes. But rather than collecting donations from the people for a project he was promoting, Chōgen proposed to tax those who controlled the land for the benefit of a project requested by the people, in a sense, a kind of reverse alms-collection campaign, or perhaps a user fee for wealthy shippers. The repair work seems to have been successful. Archaeologists have found materials at the Uozumi site that might well be remnants of Chōgen’s work: large pine logs incised to fit together and arranged as a two-tiered square support with large stones in the center.41 The reconstruction of the anchorage was not only a benefit to the local community, but also an aid to both Ōbe Estate and Tōdaiji, since Uozumi was used for shipping rice from the estate to the Nara temple. Another project in which Chōgen involved himself with the local community—and engaged in construction work at the same time— was the restoration of Sayama Pond in Settsu Province, another site linked with the ubiquitous Gyōki. Chōgen described his efforts in detail in an inscription dated 1202 on a stone plaque at the site (translated and discussed in the chapter by Ōyama Kyōhei in this volume) (see plate 15). The inscription reveals valuable details about the organization of Chōgen’s project and his relationship with local communities. Like the Uozumi project, the Sayama Pond repair was requested by the communities themselves. Since Chōgen is identified as “chief alms collector,” he solicited donations for the project; unlike the Uozumi repair, this does not seem to have been funded through government levies. Chōgen was assisted on the project by three monks with “amidabutsu” names, an appellation Chōgen often granted to his followers. Crafts- 242 Secular and Sacred men included two of Chinese descent, attesting to Chōgen’s strong connections with Song artisans. The main body of the inscription lists all those who participated in the actual labor of restoring the pond: “monks and laity, men and women, novices and children, beggars and outcasts.” While some of these people were probably paid—religious institutions often hired outcasts, for instance, for construction work—the inscription suggests that the labor was at least partly voluntary and was performed by the very persons who had urged Chōgen to help. Both the project itself and the ties formed among laborers and between laborers and buddhas would redound, in Chōgen’s hopes, to the benefit of all. The Uozumi and Sayama projects differed in significant ways, but both were efforts that involved local communities, and Chōgen justified both in religious terms. Moreover, in both projects he cited the historical precedent of Gyōki, in whose footsteps he presumed to follow. While Gyōki’s involvement in public works projects may have been the stuff of legend, Chōgen’s was certainly real. The two efforts also suggest ways he may have organized labor and support in his construction of Jōdoji and his reclamation of land on Ōbe Estate. Yokouchi Hiroto characterizes Chōgen’s alms collection and development efforts as “half free-will donations, half taxation.”42 Indeed, there were many coercive aspects to Chōgen’s activities, and government authorities were deeply involved. Both the court and the Kamak- ura shogunate had reasons to support the reconstruction of Tōdaiji: the court because the great Nara temple was a symbol of the throne’s authority, and the shogunate, in the person of Minamoto no Yoritomo, because rebuilding the temple allied him with the throne. In 1188, according to Azuma kagami, Yoritomo ordered estates’ military stewards, who were under his control, to cooperate with Chōgen’s campaign to rebuild Tōdaiji. Then in 1193, with the Great Buddha Hall still unfinished, Go-Toba Tennō ordered levies on western Japan to support continued transport of lumber from Suō Province, enjoining the shogunate’s cooperation to keep the roads open.43 On the other hand, Chōgen undertook the Uozumi project at the behest of local people; it benefited them perhaps even more than it benefited Tōdaiji as a port for shipment of goods from temple holdings. And while Sayama Pond helped to irrigate land in areas of Kawachi Province inhabited by metal casters’ groups on whom Chōgen had depended for work on the Tōdaiji Great Buddha image,44 these groups were only one por- Claiming the Land 243 tion of the residents of three provinces who profited from Sayama’s restoration. How might this web of cooperation and mutual benefit have played out on Ōbe Estate? Two factors need to be considered: relationships between the estate proprietor and local managerial entities, including Jōdoji, and the role of local estate officials and cultivators. According to Chōgen’s 1192 order to estate officials, the Kanohara area on the estate’s eastern edge was to be opened to cultivation not for the sake of the proprietor Tōdaiji, but for Jōdoji. It also appears that Jōdoji was intended to wield religious authority over the entire estate. Chōgen declared in his 1192 order, “Jōdoji monks recite prayers for the sake of the estate officials, and the estate officials support the temple monks. They should think kindly of one another, just as if this were a family temple (ujidera).”45 His 1196 request that proceeds from Ōbe and several other estates be set aside to support offerings to the Tōdaiji Great Buddha implies that Jōdoji was to act as a middleman, forwarding Tōdaiji its due; as specified in 1192, however, Jōdoji would be allowed to exploit newly opened Kanohara fields for its own benefit. The chief beneficiaries of land opening must have included local people. Did they cooperate with the land reclamation effort? It is hard to believe that they did not; labor, after all, had to come from them, and some of the capital to support both land reclamation and the construction of Jōdoji may have come from wealthy local landholders. But can we say that residents of Ōbe Estate formed a community bound by religious ties to Chōgen and Jōdoji? Chōgen was no doubt fully aware of the power of religious ties in forming communities that produced results at once sacred and secular. Promises of religious benefit underlay Chōgen’s recruitment of labor and solicitation of funds, and the worldly benefit of an anchorage or an irrigation pond could be couched in religious terms. To local communities, the value of religious connections could be immediate and practical, and it is by no means clear which was more significant to a participant in a project, the blessings of the Buddha or the opening of paddy that one could then exploit. In one sense this was a natural outcome of a worldview that did not isolate religion but incorporated it into economics, politics, and the fabric of daily life.46 When Chōgen invoked the medieval model of Gyōki, moreover, he combined two forms of religious authority described in this volume by Nagamura Makoto: that of the independent 244 Secular and Sacred and virtuous hijiri who benefited the people both practically and spiritually, and that of the official monk as delegate of institutionalized authority.47 It may be that he saw no contradiction between them. In his study of Kuroda Estate in Iga Province, another Tōdaiji holding, Arai Takashige argues that in the early and middle Kamakura period, local estate residents and the proprietor in some cases formed a type of community, or kyōdōtai. Although later in the period villagers would sometimes band together to resist a proprietor’s orders and imposts, Tōdaiji was sometimes known to defend residents against local warriors or misbehaving monks. Examples of the close relationship between small landholders and Tōdaiji include the former’s donation of land rights to the Great Buddha.48 Chōgen, however, formed communal ties with residents of Ōbe Estate without regard to Tōdaiji. The central religious focus was Jōdoji, which not coincidentally had been granted its own land to support its construction and that of its very expensive triad. The connections formed in this particular case linked local landholders, an autonomous local temple, and the opening of the land itself. At this point the Ōbe Estate community was not opposed to Tōdaiji; far from it, the bulk of estate revenues supported the Great Buddha and its rituals. But unlike the situation that Arai describes at Kuroda, Tōdaiji lay outside the Ōbe communal circle. How did Chōgen create the religious ties that supported communities? One way was through naming, a technique used both inside and outside Ōbe Estate. Chōgen called himself Namuamidabutsu, and according to Sazenshū, “In Kennin 2 (1202) I began conferring the name amidabutsu on noble and base, high and low, throughout the realm of Japan.”49 Those who bore the amidabutsu suffix—some had adopted it before 1202—included the master sculptor Kaikei (An’amidabutsu); Chōgen’s disciple Kan’amidabutsu, and Chōgen’s heir to control of his autonomous temples and several estates, the Tōnan’in head monk Jōhan (Gan’amidabutsu). The suffix was also borne by others who helped Chōgen solicit funds for both religious and public works projects; examples include three persons involved in the Sayama Pond reconstruction, as mentioned above. Taking the name of Amida not only signified a person’s devotion to the Buddha but also presented an opportunity to invoke Amida every time one uttered one’s own name or that of one’s colleague. Claiming the Land 245

Another way religious ties might be created was through ceremonies that welcomed the public. Yoshiko Kainuma has explained how the open structure of the Jōdodō welcomed crowds of lay believers.50 According to Sazenshū, a raigō ceremony, which enacted Amida’s welcome of the dying faithful to paradise, was performed in 1198, and Chōgen also records that in 1200 he installed “a [second] standing image of Amida coming to welcome the dead.”51 (For that image of Amida, see plate 22.) This suggests that he intended to hold raigō ceremonies regularly, and we know from later evidence that they were continued throughout the Edo period.52 Underlying both “sacred” and “secular” efforts, moreover, was the concept of kechien: forming connections, creating ties. Throughout his autobiography, Sazenshū, Chōgen repeatedly refers to this concept. One formed connections with a buddha or bodhisattva, with a temple, with a sacred image, with other like-minded people. Ties might be created by worship, or by donations of money or labor. In Sazenshū, Chōgen reports that he made ties with various temples and images, as well as with two bridges. In one instance, Chōgen established connections not only between a shrine and himself as an individual donor, but between two Buddhist institutions, a temple on shrine grounds (jingūji) and one that he himself had founded:

When Kibitsu Shrine [in Bitchū Province] was reconstructed, I established religious ties with it. I had a bell cast and donated it to the shrine, establishing ties between the jingūji and the Buddhist temple [Jōdodō, Chōgen’s autonomous temple in the province].

It may be that Chōgen envisioned a series of connections snaking through the Buddhist world, binding local communities and temples to distant ones, even those in China to which he had donated Japanese lumber or from which he had obtained images or scriptures.53 While on the one hand Chōgen formed local communities for specific purposes, on the other hand community might transcend the local and become a cosmic concept. At the most basic level, religious ties were often rooted in local communal groups. It was common for lay donors to form groups called kechienshū to achieve a religious purpose that a single individual or family could not accomplish. Such groups included those with other ties as well, of family, of neighborhood, of mutual economic interest. The 246 Secular and Sacred formation of a Jōdoji-based community must have been a mutual process, and we cannot assume that local participants were simply passive followers of Chōgen’s initiative. Evidence for such a community on Ōbe Estate can be found, in fact, within and on two images of Amida at Jōdoji: the central figure of the Jōdodō triad and the second image made for the raigō ceremony. More than two hundred names were inked on the inside surface of the first Amida, plus thirty-five or so on the skirt of the second. Who were all these people, and could they have included the very ones who supplied funding and labor to turn the Kanohara forest into paddy? Names include “amidabutsu” appellations associated with Chōgen, eleven on the inside surface of the Amida in the triad and at least twenty-five on the second image carved in 1200 or 1201.54 Some of these people have other known connections with Chōgen: the names of Hōamidabutsu and Kyōamidabutsu also appear on a 1203 document recording receipts and expenditures of rents forwarded from Bizen to Tōdaiji. Chōgen used the back of this document for his autobiography Namuamidabutsu sazenshū. Kai’amidabutsu’s name appears on vows written in 1202 at Chōgen’s autonomous temple in Iga Province, Shindaibutsuji.55 Most signatures within the early Amida image are undated, but one cluster bears the date 1195/4/15. Since so many characters are illegible, it is impossible to count the names accurately or to tell what percentage belongs to a given category. Nor is it always clear whether a particular name represents one person or two; thus all figures given here are approximate. Female names are fairly common; I have identified forty- seven laywomen and two nuns. Besides those with the amidabutsu suffix—who could be either clergy or lay believers—there are thirty-some men who have taken orders of some sort and identify themselves as resident monks (jūsō), monks (sō), or novices (shami), along with four lay monks (nyūdō). Some signatures are clustered together and appear to belong to persons of the same family; sometimes these clusters include names of the deceased. Many but not all of the signatures include surnames, and some persons specifically identify themselves with Ōbe Estate, using Ōbe as a surname. An individual who appears in later documents is Ō Hisakiyo, who was appointed kumon (local manager or reeve) of the estate in 1207.56 One deceased man’s name is preceded by the characters san’i, indicating a person with court rank but no official appointment. He may well Claiming the Land 247 have been an elite local resident, like a similarly designated “Lord Fuji- wara (Fujiwara ason)” who received a shogunal order of 1203 addressed to estate officials.57 The frequent appearance of surnames indicates that many others were local notables of some sort and may have held prominent positions on the estate; Karikome Hitoshi proposes that they included cultivators’ headmen known as myōshu.58 Others, of course, could have been allies of Chōgen who lived elsewhere. Unfortunately there is no explicit evidence linking donors to the Jōdoji images to land reclamation work on Ōbe Estate. Yet it is clear that Jōdoji was meant to be a fulcrum for the Ōbe community. Karikome argues that as the estate’s ujidera, Jōdoji was intended to organize the entire Ōbe population.59 This may explain why Chōgen decided not to restore any of the abandoned local temples but instead to appropriate their images—and the images’ sacrality—for the new temple. Knowing the way Chōgen worked, moreover, it seems likely that many local individuals were involved in both land reclamation and temple construction. Then too, in a time when funds were scarce and much capital was diverted for the reconstruction of Tōdaiji, estate officials and prosperous cultivators may well have invested in both land reclamation and the construction of Jōdoji. Local residents would have profited from a share of the proceeds from newly reclaimed land, while expecting religious benefits from their assistance in constructing temple buildings and holy images. Jōdoji was also given a protective function on Ōbe Estate. Chōgen stipulated that the temple’s monks should pray for the welfare of local managers and, by implication, for the prosperity of the entire estate. The location of Jōdoji, on a terrace overlooking the estate’s eastern expanse, suggests that very function. Jōdoji engi claims that the temple’s location at the estate’s northeast corner protected the land from the incursion of demons. While we do not really know if this was on Chōgen’s mind, locating a religious institution at the northeast “demon’s gate” of a secular establishment had a famous precedent: the Tendai complex at Mount Hiei occupied such a position in relation to the capital at Kyoto. Furthermore, as Karikome Hitoshi points out, large-scale land reclamation that changed a natural landscape was feared to offend local deities, a concern that might be overcome through the authority of a more powerful religious institution as well as buddhas and more prestigious deities.60 Although ultimately this institution was Tōdaiji, for local residents it must have appeared that Jōdoji—which not coincidentally incorpo- 248 Secular and Sacred rated a shrine to the centrally important deity Hachiman—really took on this role. Of course, Jōdoji’s supervision of irrigation facilities indicates that its protective function was practical as well as symbolic. Chōgen not only urged the mutual cooperation of Jōdoji monks and estate officials,61 he also maintained that devoting revenues from Ōbe and other estates to services for the Great Buddha should “lead to tranquility at court, the monarch’s long life, peace in the realm, and happiness for our subjects.”62 While this might be dismissed as hyper- bole, the development of new paddies and the repair of transportation and irrigation systems were practical side effects of the Tōdaiji reconstruction project that greatly benefited local communities. In Chōgen’s view, good works—including both opening lands to cultivation and constructing images and temples—would yield bountiful treasures to all.

Community Undone? Ōbe Estate after Chōgen The proprietorship and management of Ōbe Estate was complicated by the means Tōdaiji used to make the land productive. First, the temple handed over control to Chōgen, who was not a member of the monks’ assembly, then it acquiesced in the formation of an autonomous temple and its support by newly opened fields within estate borders. In the end, Chōgen willed control over the properties that he managed for Tōdaiji to the head monk of the Tōdaiji cloister Tōnan’in, rather than to the monks’ assembly of corporate Tōdaiji. Eventually, however, proprietor’s rights reverted to the main temple and throughout the Kamakura period toggled back and forth between the two monastic entities. As Endō Motoo points out in this volume, this appears to have been mostly a cooperative arrangement. Control over the estate did not always proceed smoothly, however. Just before Chōgen died, trouble arose with management of several Tōdaiji holdings under his jurisdiction, including Ōbe Estate. In the fourth month of 1206, the retired sovereign Go-Toba ordered officials in all provinces to put an end to the violations of the Great Buddha casting master Chen He-qing. The order is in response to a complaint from the Tōdaiji monastic cabinet. To support his work, He-qing had been allotted income from Ōbe and several other Tōdaiji estates, but the document states that he had officially donated this income back to Tōdaiji and that then, instead of actually forwarding the income, he had illegitimately commandeered it for his own benefit.63 The complaint points out how difficult it was for even powerful institutions to guaran- Claiming the Land 249 tee proceeds from estates and suggests the conflict over absentee holders’ rights that in fact characterized the whole system. As both Japanese and Western scholars have argued, the estate system played an important role in binding center to province and tying together the interests of various classes; it was the central factor in the development of a realm-wide structure that may be called the medieval polity.64 The structure’s connective sinews, however, could not prevent quarrels among claimants to income rights or over borders with public land or other estates, disputes that became quite common in the middle and late Kamakura period. In Chōgen’s early work to establish Ōbe Estate as a profitable Tōdaiji holding, communal ties are emphasized, but the conflicts lie just below the surface and in fact would emerge at the very end of Chōgen’s life, as the situation with Chen He-qing amply demonstrates. By the end of the thirteenth century the situation had deteriorated even further. Beginning in the 1290s, for example, quarrels broke out between rival claimants to Ōbe Estate resources, setting Tōdaiji and estate reeves against Jōdoji monks, and estate cultivators against the proprietor’s representative.65 Even within one family, conflict over the reeve’s position and income generated a long series of lawsuits claiming theft and forgery of documents, a dispute that stretched well into the era of Northern and Southern Courts.66 Chōgen’s community based on religious ties and mutual economic interests was actually quite ephemeral, as was, I think, any role of the estate system in creating a harmonious structure that bound center to province and capital elites to local farmers. And it seemed Chōgen knew the delicate nature of his undertaking, as he jockeyed to keep control over Ōbe Estate in trustworthy hands and urged estate managers and Tōdaiji to protect the rights of the small temple he had established as an independent entity. Those who violated these rights were, in Chōgen’s terms, sworn enemies of Jōdoji and obstructions to the Buddhist path.67 Thus Chōgen invoked the wrath of the gods and buddhas on those who would destroy the community he had forged. The troubled history of Ōbe Estate in later Kamakura times suggests that such punishment was a long time coming.

Notes

1. OS 4:527–528 (# 219). 2. For a treatment emphasizing Chōgen’s equally important Shingon beliefs, see the chapter by Ōyama Kyōhei in this volume. 250 Secular and Sacred

3. See OS 4:535–536 (# 225); Ōsaka-furitsu Sayama Ike Hakubutsukan 2002, 5. 4. KI 1:129–130 (# 210); 133–134 (# 219). 5. HI 7:2565 (# 3218); OS 4:506–507 (# 209). 6. Ōsaka-furitsu Sayama Ike Hakubutsukan 2002, 54–55. 7. Kawabata 1999, 15–20. Also see the chapter by Nishida Takeshi in this volume. 8. Details of the Tōdaiji rebuilding campaign can be found in Kujō Kane- zane’s diary Gyokuyō (1181/3/21, 1181/7/14–16, 1181/10/6, 1181/10/9) and in Tōdaiji zoku yōroku (1181/4/9, 1181/6, 1181/8). 9. Yokouchi 2008, 545. For Suō, see Gyokuyō 1186/3/23. 10. OS 4:530 (# 222). 11. Kondō 2008. 12. OS 4:526–527 (#s 217, 218). 13. OS 4:532–535 (# 224). 14. OS 4:527–528 (# 219). 15. See Nagamura 1981, 72; Ōyama 1999, 4; and Goodwin 1994, 99. 16. Ōyama 1999, 9. Also see the chapter by Nagamura Makoto in this volume. 17. OS 4:527–528 (# 219). Leprosy is a loose translation for rai, a term that probably includes other disfiguring diseases. See Goble 2011, 68–77, for a discussion of rai, including some descriptions of the symptoms of white and black rai. 18. See Nakamura 1978a, 204. 19. This was an alternate name for Jōdoji. 20. For a discussion of Jōdoji and the Amida triad from an art historical perspective, see the chapter by Yoshiko Kainuma in this volume. Also see Kai- numa 2014. 21. Ōyama 1999, 3. 22. For a discussion of the contents and dating of this text, see the chapter by Ōyama Kyōhei in this volume. 23. Goodwin 1994, 91; also see the chapter by Yoshiko Kainuma in this volume. The Kōya bessho contained two images of Amida but neither seems to have been a central worship object. No bath is listed for the Bitchū temple. 24. See Adolphson 2000, 57, 61, for matsuji and betsuin. 25. Ōyama 1999, 8. 26. Karikome 2004, 157–158. 27. OS 4:529 (# 221). 28. See Nagamura 1981, 72; Goodwin 1994, 99. 29. OS 4:535–537 (# 225). Usually, prayer chapels were constructed by a patron hoping to obtain a good rebirth or good fortune. Here, Chōgen is offer- Claiming the Land 251 ing an extant building to the retired monarch, perhaps in return for financial support. 30. OS 4:532–535 (# 224). 31. OS 4:537–543 (# 227). 32. For an archaeological overview of Ōbe Estate, see the chapter by Nishida Takeshi in this volume. 33. Kawabata 1999, 16–17. 34. Kawabata 1999, 18. 35. Kawabata 1999, 17. 36. Kawabata 1999, 16; Ōsaka-furitsu Sayama Ike Hakubutsukan 2002, 35, 54–55. 37. For a detailed discussion of this project, see the chapter by Ōyama Kyōhei in this volume. 38. KI 2:188–189 (# 847). Gyōki was often called a bosatsu, or bodhisattva. 39. The San’yō region is the westernmost area of Honshu bordering the Inland Sea; the Nankai region includes Shikoku as well as Awaji Island and current Wakayama Prefecture in Honshu; the Saikai region refers to Kyushu. See the chart of “Provinces and Prefectures” at the beginning of this volume. 40. KI 2:188–189 (# 847). For another account of the Uozumi project, in Chōgen’s autobiography Namuamidabutsu sazenshū, see the chapter by Ōyama Kyōhei in this volume. 41. Ōsaka-furitsu Sayama Ike Hakubutsukan 2002, 38. 42. Yokouchi 2008, 537. 43. Yokouchi 2008, 546; Azuma kagami (1976–1979) 2:38 (1188/3/10); KI 2:73–74 (# 670; 1193/5/8). 44. Yokouchi 2008, 552–553; Ōyama 2012, 85. 45. OS 4:527–528 (# 219). This argument is made in Karikome 2004, 165. 46. See Yokouchi 2008, 538–539. 47. See the chapter by Nagamura Makoto in this volume. 48. Arai 2001, 194–195, 221–229, 242–248. 49. OS 4:537–543 (# 227). 50. See her chapter in this volume; also see Kainuma 2014. 51. OS 4:537–543 (# 227). The document gives a date of Shōji 2, without a month, for the second Amida image, while Jōdoji engi dates it Kennin 1 (1201). The image, still extant, is 266.5 centimeters (eight shaku) tall. Like the earlier triad, it is the work of Kaikei. 52. OS 2:558–561. 53. OS 4:537–543 (# 227). 252 Secular and Sacred

54. See Hyōgo-ken shi, Shiryōhen, chūsei 4; OS Bekkan (Supplementary volume), 168–174; Mōri 1976, 66–70. Also see the chapter by Yoshiko Kainuma in this volume. 55. Mizuno 1997, 370. 56. OS 4:546 (# 229). 57. OS 4:536–537 (# 226), translated by Michelle Damian; also translated in Mass 1976, 162. 58. Karikome 2004, 159. 59. Karikome 2004, 157–158. 60. Karikome 2004, 173–176. 61. OS 4:527–528 (# 219). 62. OS 4:530 (# 222). 63. OS 4:543–545 (# 228). For further discussion of this document, see the chapter by Endō Motoo in this volume. 64. See Kiley 1974, 109–124; Inaba 2012, 54–79. 65. OS 4:618–620 (#s 313, 314, 315); OS 4:575–577 (# 273). For the dispute between Tōdaiji and Jōdoji at the end of the thirteenth century, see Goodwin and Wilson 2015. 66. OS 4:626–629, 652–653, 660–661 (#s 329, 330, 348, 357). 67. OS 4:527–528 (# 219). c h a p t e r 11

The Jōdoji Amida Triad How Its Iconography Advanced Chōgen’s Mission

Yoshiko Kainuma

A monumental, golden-colored Amida triad seems to be floating in the air, as if the three figures, standing on the clouds, were descending to save their wor- shippers. The dramatic effect of this raigō form must have strongly captured the hearts of local people in a remote corner of medieval Japan.

The Amida triad at Jōdoji, a small temple in Harima Province, was made at the end of the twelfth century by the famous sculptor Kaikei (d. before 1227). Unusual for the time, all three figures of the triad—the Buddha Amida and the flanking bodhisattvas Seishi and Kannon—are standing in a posture that seems to indicate an active effort to welcome the faithful to paradise, in a ritual known as raigō (see plate 16). This posture, along with the design of the hall in which they were enshrined, suggests that both images and hall were meant to appeal to a wide audience of lay believers, especially those residents of Ōbe Estate, the Tōdaiji landholding on which Jōdoji was located. Jōdoji, in present Kiyotani-chō, Ono City, was constructed at the beginning of the Kamakura period by Shunjōbō Chōgen (1121–1206), who had undertaken the collection of donations and materials for the reconstruction of Tōdaiji, the great Nara temple destroyed in 1181 at the outset of the Genpei War (see plates 1 and 17). One of Chōgen’s tasks in this effort was to oversee several Tōdaiji estates, including Ōbe, and to make sure that estate revenues were directed toward the rebuilding project and rituals at Tōdaiji. This task required local estate managers and cultivators to support both Jōdoji itself and land reclamation efforts nearby. At this temple, Chōgen incorporated architectural and

253 254 Secular and Sacred

iconographic features that served as means for obtaining such support. The design of the Jōdo Hall (Jōdodō), the form—especially the posture—of its Amida triad, and the original layout of the temple were intimately connected with the management of Ōbe Estate. Also called the Harima Bessho, Jōdoji was an autonomous temple, one among several established by Chōgen as a base for managing the Tōdaiji project.1 Although remotely located, Jōdoji’s main images and its Jōdodō, where they are enshrined, are very well known (see plate 18). The hall itself dates to the early Kamakura period, around the same time as the images. Both Jōdodō and the triad have been designated National Treasures by the Japanese government. Provincial temples such as Jōdoji, many of which were established in the early medieval period, had diverse origins and functions. In contrast to Heian-period temples that mostly served the aristocracy, many early medieval temples on estates and elsewhere in the provinces were established and maintained through connections with the local populace. Along with shrines, some temples served as religious protectors of estates. For example, when a fishing port and land were developed along Tagarasu Bay on Nishizu Estate (Wakasa Province, Onyū District) in the early Kamakura period, both a Tenman Shrine and a temple, Daifukuji, were established there to perform a guardian function.2 Such provincial temples also include ujidera (family or lineage temples) founded by estate officials, small chapels built by local residents, or those restored by kanjin hijiri, who collected donations for religious projects while preaching to the populace.3 Fund-raising campaigns were conducted widely for small and medium temples in medieval times, and such temples played an important role in both the lives of ordinary people and the management of estates. The types and iconography of Buddhist images vary according to the locales or traditions of the temples in which they were enshrined, but inscriptions within the bodies of images at this time frequently indicate connections with ordinary people. Such inscriptions testify to people’s enthusiastic involvement in building temples and making images. For example, ink inscriptions inside the Amida Nyorai (1154) at Kannonji in Kawachi Province demonstrate that the image was donated by the temple’s parishioners, and “clergy and lay, men and women” in Kawachi. The inscription was signed by a manager (geshi) of the estate on which the temple was located.4 Also, a list of names Jōdoji Amida Triad 255 inside the body of the Kannon image (1159) at Heishōji, Toyota City, Aichi Prefecture, identifies people who had formed religious ties (kechien) with the sacred icon. Heishōji was probably a temple of the local elite Wakao family, and the list of names includes those of the Wakao and other local residents. The list suggests that residents in the area cooperated to have the image made. Similarly, more than one hundred kechien names inside the seated Yakushi figure (1171) now at Rinkōji in Shinshiro City, Aichi Prefecture, indicate that the image was funded by the local elite community, possibly including some estate officials, and led by an alms-soliciting monk.5 Cooperation between such solicitors and local residents was a hallmark of medieval temples, including those on estates. Temples on estates often served as management headquarters or as meeting places,6 and temple monks sometimes helped residents craft petitions or other documents.7 In other words, temples had functions to play in the life of an estate that were not strictly religious ones. One example of an estate temple that fulfilled both religious and secular roles, in this case on behalf of the proprietor, is Kongōfukuji on Hata Estate in Tosa Province. Established as a possession of the Kujō family during the early Kamakura period, shortly afterwards Hata became a hereditary holding of the Ichijō, another branch of the powerful Northern Fujiwara. Kongōfukuji, a well-known center of the worship of Kannon, was restored by Kamakura-period kanjin monks, who raised donations to rebuild the temple. Centered on an Eleven-Headed Kannon as its main icon, the temple functioned as the prayer temple for the Ichijō family. At Kongōfukuji, prayers were said for peace on the estate, abundant harvests, and the defeat of the Mon- gols who invaded Japan at the end of the thirteenth century. Thus the temple provided religious support for Ichijō control over the estate. Kongōfukuji’s three subtemples also managed transportation along the Shimanto River, sustaining the Ichijō proprietors economically and socially as well.8 Temples on estates performed their functions in part through ties with local people. Focusing on one such temple, Jōdoji on Ōbe Estate, this chapter will explore these questions: What were the religious aspirations and secular benefits that people expected of temples and their monks, and how did temples and monks respond to ordinary people’s hopes? What rituals and prayers were conducted to help people, and how did people respond? And finally, how was 256 Secular and Sacred the relationship between residents and estate temples reflected in the business of the estate? This relationship, particularly involving the images on which people’s religious beliefs were focused, is a significant topic in estate studies that has not been thoroughly explored. Jōdoji and its images provide a concrete example of the connections between icon and estate residents that reflects new conditions in the early medieval age.

Ōbe Estate and Jōdoji Jōdoji’s initial function was to oversee land reclamation operations on Ōbe Estate, a neglected Tōdaiji holding that Chōgen was charged with restoring. After the temple was established by Chōgen, it was supervised by his nephew Kan Amidabutsu (1165–1242) and by Nyo Amidabutsu, thought to have been a member of the local elite. Both served as custodians, an office that in this case was held by an individual residing on site who reported to the proprietor. Jōdoji also functioned as the religious guardian of the estate. In addition, as one of the autonomous temples Chōgen established throughout western Japan, Jōdoji served as a base not only for his efforts to raise funds and gather materials for Tōdaiji’s reconstruction, but also for the spread of religious practices such as chanting the nenbutsu (praising the name of Amida Buddha). In addition to establishing such temples in several provinces, Chōgen also opened land to cultivation, exploited local resources, and managed Tōdaiji’s holdings, projects that not only supported the temple’s restoration but also enabled him to establish ties with local people. For example, the autonomous temple Amidaji was established in Suō Province, revenues from which had been assigned in 1186 to Chōgen for use in the Tōdaiji project. Chōgen assumed authority over the province and had timber cut from its forests for the reconstruction work. Land reclamation to support Amidaji began in the following year, and a good portion of revenue thus obtained from mountain dry fields was devoted to all-day bathing rituals and uninterrupted recitation of the nenbutsu at the temple.9 Remains of nearby baths that were probably used by the loggers can be seen today.10 Likewise, at Ōbe Estate rents and other levies supported the reconstruction of Tōdaiji, but the economic foundation of Jōdoji was provided by the newly opened fields of Kanohara, revenues from which were devoted to lamp Jōdoji Amida Triad 257 oil for Buddhist services and food and clothing for monks who recited the nenbutsu there. A portion of these revenues must have been spent for bathing rituals at Jōdoji as well, and the bath—with its perpetually heated water—may have been one of the great attractions for people who visited the temple. In other words, Jōdoji was not simply a center for religious affairs on the estate, but through its supervision of land reclamation and estate management, it helped to develop local society. Just as in Suō, Jōdoji monks took control over the land opened under their supervision. Of course, land reclamation also served the interests of estate residents, many of whom became parishioners of Jōdoji, like the local officials of the Suō provincial government who supported Amidaji. Prosperous cultivators on Ōbe Estate must have expected that the Jōdoji monks would value their hopes, desires, and religious beliefs. The images at Chōgen’s autonomous temples reflect his devotion to Amida. Each temple except the one at Mount Kōya had a Jōdodō where one or several images of that buddha were installed. Most of these images, like those at the Harima Bessho, were noted as “jōroku” in size (standing figures were about sixteen feet tall and seated ones about eight feet) and gilded, and they were enshrined in the center of their hall.11 According to Chōgen’s autobiography, Namu Amidabutsu sazenshū (The good deeds of Namu Amidabutsu, hereafter Sazenshū), he sponsored or was otherwise involved with the installation of the images at his bessho listed in table 11.1. Only the Jōdoji triad is extant in entirety, but one can easily imagine the great impact on worshipers when they encountered these colossal gilded statues, many probably in a standing posture. This posture—an iconographic feature unique at the time—represents concrete aspects of Chōgen’s Amidist beliefs. It appears that at Jōdoji, the posture, the hall that accommodated a large number of visitors, and the raigō ceremony were intended to captivate lay worshipers, such as the residents of Ōbe Estate. Here I will introduce the form of the triad from both religious and art historical viewpoints and will examine how its unusual nature was entwined with the social role of the local temple on the estate. Chōgen’s establishment of Jōdoji in 1192 is documented in an order (kudashibumi) addressed to Harima Province officials.12 Sazenshū records the following about this temple:13 258 Secular and Sacred

Table 11.1 Sculptures at Chōgen’s Bessho (Namu Amidabutsu Sazenshū)

Temples/ Figures Tōdaiji Kōya Shin Watanabe Harima Bitchū Suō Iga Bizen Total

Jōroku 1 1 1 1 1 5 Amida Kannon 1 1 1 1* 4 Seishi 1 1 1 1* 4 Sanjaku‡ 1 1 Amida Raigō Amida 1 1 1† 3 Jōroku 10§ 1 11 Shaka 1 1 2 Yakushi 1 1 Tenbu 4 4 Others 8 8

* Jōroku † Perhaps a jōroku Amida ‡ Sanjaku: about three feet high § One (probably Amida) donated by the Rokujō-dono nun Gozen; nine transferred from Awa Province, probably Amida of nine grades

Harima Bessho One Jōdodō. One jōroku golden standing Amida, and Kannon and Seishi, are enshrined [in the hall]. One one-bay four-sided Yakushidō with one shu 竪 jōroku [image].14 One bathhouse. Hot water is always available . . . I dedicated one bell. Two years after the beginning of the mukaekō [ceremony to welcome the dead, also called raigō-e] in Shōji 2 (1200), [I installed] one raigō standing Amida and one bell. One painting of an Amida triad, made in China, which became the model for the jōroku [images] of Harima and Iga.15

Like most of Chōgen’s other autonomous temples, Jōdoji had a Jōdodō with a large main icon, as well as a bell and a bathhouse. And yet Jōdoji Amida Triad 259 only at Jōdoji, according to Sazenshū, is there a Yakushidō, a hall built in the same form and size as Jōdodō.16 While Sazenshū does not list the dates when Jōdodō and Yakushidō were constructed, such details can be found in other texts: Chōgen’s 1192 order; his testament, dated 1197; and Jōdoji engi, a later account of temple legends and history.17 These materials tell us that Jōdoji was founded when Chōgen built a small hall in the northeast corner of Ōbe Estate and installed several images from ruined temples nearby. The Jōdodō was built around 1197, and then the small temporary hall was replaced by a larger one, the Yakushidō, completed in 1200.18 Its main image, a seated jōroku Yakushi,19 was brought from Kōdoji, one of the old temples on the estate, along with seven hundred images from nine other temples.20 Jōhan (1165–1225) of the Tōdaiji cloister Tōnan’in presided over the Yakushidō’s opening ceremony on 1200/4/13. Chōgen had a close relationship with Jōhan, and willed him the rights to several estates in Iga, Suō, and Bizen, as well as to four autonomous temples: Kōya Shin, Tōdaiji, Watanabe, and Harima.21 For local people, the spiritual value of Yakushidō must have been high. The central image was of Yakushi Nyorai, the “healing Buddha” associated with overcoming illness and calamity and bringing about peace and tranquility in the present world. People probably also thought that such this-worldly benefits were guaranteed by the many old images gathered from ruined temples on the estate. When Chōgen installed these images in the Yakushidō, this must have both allayed concerns about them on the part of local people and enhanced the position of Jōdoji as the estate’s guardian temple. Opening new land to cultivation, too, benefited not only Jōdoji monks but also estate residents, helping to ensure the “peace and tranquility” of their present lives.

Jōdoji Jōdodō and the Amida Triad References to Jōdodō and its main icon may be found in all the documents related to Jōdoji mentioned above.22 According to the Engi:

Regarding the hall of Jōdojidō,23 [it is] a one bay four-sided structure with a tile roof. On 1194/10/15 a celebration of the completion of the framework of the hall [was held] with the carpenter, Bungo no suke Ki no Kiyonaga. One jōroku sized golden Amida Nyorai and han [half]-jōroku sized Kannon and Seishi standing bodhisattvas made by 260 Secular and Sacred

Daibusshi Tanba Hōgen Kaikei24 were installed [in the hall]. A ceremony of completion of construction was held on 1197/8/23 with Kasagi Gedatsu Shōnin Jōkei as the leading priest.25

The text notes that the completion of Jōdodō was earlier than that of the Yakushidō in 1200 and makes it clear that the Amida triad was made before the completion ceremony of Jōdodō in 1197. Jōkei (1155–1213) was a high-ranking priest of the Hossō school in Nara who resided at Kasagidera in the mountains east of Nara and was closely associated with Chōgen and Kaikei. The structure of the Jōdoji Jōdodō enables it to accommodate a large audience, such as residents of the estate, whose support was crucial to Chōgen’s management function. The hall is of simple but sturdy construction, built in a Song Chinese architectural style known as “daibutsuyō ” (Great Buddha style).26 The Kamakura-period Tōdaiji Daibutsuden (Great Buddha Hall), which housed the Great Buddha, was built in this style—hence its name—but that hall is no longer extant. The one other extant example of the style besides Jōdodō is the Tōdaiji South Gate, built in 1199. Chōgen and the Chinese metal worker Chen He-qing, who restored the head of the Great Buddha, introduced this architectural style to Japan, but it was popular only from the end of the twelfth to the beginning of the thirteenth century.27 Chōgen seems to have specifically chosen this style for an appropriate venue for the main icon at Jōdoji. The square hall, with a pyramid-shaped roof, has three bays on each side. Each side is almost sixty feet long, with spaces between pillars that are almost twenty feet in length. Unlike Japanese-style halls, the roof does not curve up at the corners, and the eaves are straight without showing any rafters, which produces a monumental effect.28 Entering from the main entrance on the east side, one notices various daibutsu- style characteristics inside the spacious hall, simple and bold, especially with its high roof structure and exposed rafters and beams. The thick, red-colored pillars and the sturdy bracket system create an overwhelmingly voluminous impression. According to architecture specialist Suzuki Kakichi, the most important feature of the daibutsu style is the nuki (piercing tie rail), a horizontal member that connects pillars by running through them. Because of this technique, which provided additional structural support, the space surrounding the inner sanctuary where the images were displayed could be made unusually large. In fact, Jōdoji Amida Triad 261 the outer aisles are close to twenty feet wide, the same as the width of the sanctuary. Although the one-bay, four-sided Amida hall became popular from the end of the Heian period, those halls that were constructed in the ordinary Japanese style have limited space around the inner sanctuary29 (see figures 11.1 and 11.2). Comparing Jōdoji Jōdodō with Pure Land halls of the late Heian period verifies its unique position in the art and architecture of Pure Land Buddhism. A key late Heian example is the Byōdōin Phoenix Hall (Hōōdō) and its icon, an Amida made in 1053 by Jōchō.30 The Phoenix Hall was built to manifest Amida’s Western Paradise on earth. The hall is composed of the center chapel (46.71 feet × 38.77 feet), two flanking galleries, and the tail corridor that connects to the back of the center chapel. The entire building is surrounded by a pond. The name Phoenix Hall may be derived from its structure, reminiscent of an imaginary bird, or perhaps from the pair of phoenixes that embel- lish the roof of the chapel. Inside the center chapel, a solitary Amida figure, over nine feet tall, is seated on the altar. The figure presents the jōin, the mudra (hand position) that signifies meditation. The surrounding environment is fully decorated with paintings and relief sculptures with Pure Land motifs, including a lavish canopy and dais. The Phoenix Hall was a sanctuary for the Buddha, and people were not generally

Fig. 11.1. Jōdodō section diagram. Jōdoji, Hyōgo Prefecture. Photo courtesy of Ono-shi Kyōiku Iinkai. 262 Secular and Sacred allowed to enter. Only a few people, such as the donor, Fujiwara no Yori michi (992–1074), might have entered the hall. Other devotees in the Heian aristocracy usually worshiped and contemplated the Amida and its paradise through the grids of its hall, from a smaller structure (kogosho) across the pond. When viewed from there, the Phoenix Hall must have seemed like the representation of paradise depicted by a Pure Land mandala.31 The remarkable physical differences between the Phoenix Hall and the Jōdoji Jōdodō seem to reflect Chōgen’s aims along with the development of Pure Land ideas through the years. In contrast to the Phoenix

Fig. 11.2. Jōdodō ground plan. Jōdoji, Hyōgo Prefecture. Photo courtesy of Ono-shi Kyōiku Iinkai. Jōdoji Amida Triad 263

Hall, the Jōdoji Jōdodō is designed in such a way that the holy triad and human nenbutsu believers can share the same space. Unlike the lavish setting of the Phoenix Hall, the Jōdodō structure is wide open with almost no decor. It creates “space” where the Amida triad seems to descend through the air as depicted in many later raigō paintings. The Jōdoji triad is not hidden or isolated from worshipers but rather stands in the center of the unusually wide hall, as if it were occupying open space shared with nenbutsu devotees. The monumental triad is fixed to the dais and the ground below by extensions of the main component of each figure under the dais. Then the extended parts are joined with one another with a piercing tie rail, and finally connected to stone foundations on the ground.32 Since this arrangement dates from Chōgen’s time without any major repair work, we can use the present setting to draw conclusions about his religious intentions and the ways worshipers may have responded. The triad is mounted on a circular dais at the center of the Jōdodō, facing east.33 Similarly, the Amida triad at Shin-Daibutsuji, Chōgen’s Iga Bessho on Awa Estate, also stood on a huge circular stone dais (which remains at the temple today). The circular shape must have had meaning for Chōgen’s nenbutsu practice, in which monks and believers circumam- bulated the dais while chanting the nenbutsu. As described in Chōgen’s order of 1192 establishing Jōdoji, “thirty monks will meet together and practice chanting the nenbutsu ceaselessly in sonorous voices, in order to pray for the peace of our sacred realm.”34 For Chōgen’s Amidism, reciting the nenbutsu ceaselessly was one of the most important practices and he seems to have encouraged it to the public at the Jōdodō.35

Visual Effects of the Jōdoji Amida Triad An additional special feature of the hall is its overhung latticed doors (shitomi-do), which no doubt were opened during ceremonies that crowds of estate residents attended. These doors can be lifted to allow sunlight to illuminate the images, producing a spectacular effect. Ritu- als focused on the triad also advanced Chōgen’s efforts to help believers visualize Amida and his Pure Land. The most impressive of these uses the visual effect of the late afternoon sun shining on the images in the hall. When the latticed door is lifted, the strong sunlight from the west illuminates the triad and, especially on a summer afternoon, creates a special effect that enhances the impression that the triad, standing on 264 Secular and Sacred the clouds, is descending from the sky to save worshipers.36 This special visual effect peaks at 4:30 p.m. and can be observed for about one hour after that.37 This dramatic effect seems truly to have been planned at the beginning of Jōdodō’s construction. In the spacious hall, nenbutsu believers can participate in this raigō scene. According to some interpretations, the entire illuminated hall manifests the scene of Amida’s appearance (genzen) as described in Kan muryōju kyō (Sutra on the Meditation of the Buddha of Infinite Life): Amida and holy attendants flying down from the Western Paradise to welcome the deceased.38 All figures lean slightly forward and the clouds on the pedestals seem to drift behind each figure, expressing Amida’s aggressive pursuit of his original vow to save all those who invoke his name. The standing posture of all three figures illustrates the phrase, “Muryōju (Buddha Amida) stands in the air and the bodhisattvas flank him in standing position,” in the seventh Kezakan (“to contemplate Amida’s lotus pedestal”) of the sutra. This illustrates Amida’s original vow to save humankind. What Chōgen wanted to convey in his images is advanced Pure Land ideas in which Amida literally stands up to bring salvation. The three upright figures represent more immediate movement than would images depicted as seated in the Western Paradise, since the former expresses a more forceful attempt to bring salvation to the dying. The clouds depicted on their pedestals, moreover, make it seem as if the triad is even flying to the deceased. The wide-open, almost undecorated space in the surrounding hall also enhances the impression that the triad is floating in space (see plate 16 and figure 11.3). The Jōdoji Amida triad is well known for other distinct characteristics in addition to its standing posture. The images are colossal. The height of the central Amida is almost 17.5 feet and each of the flanking Kannon and Seishi bodhisattvas is just over twelve feet tall. The gold-colored finish is a gold foil applied with lacquer on the cypress wood of the figures. The figures’ large, strongly curved eyes add to the impressive appearance of the sculpture. The Amida’s hands form the raigō mudra that signifies the welcome of the deceased to the West- ern Paradise. The image’s left hand, however, has thumb and finger touching while the right hand is pendant with palm out, opposite from the usual configuration. The rising clouds surrounding the pedestals on which each figure stands also identify the sculpture as a raigō Amida triad. Fig. 11.3. Amida triad. Side view of Jōdoji Amida triad. Photo: Nagano Rokumeisō. 266 Secular and Sacred

From the late Heian period on, as Pure Land Buddhism spread throughout the archipelago, many paintings and sculptures depicted the raigō Amida triad. The Jōdoji triad, however, manifests quite different iconographical features from the standard raigō Amida style. As time passed, in fact, the depictions of the raigō Amida triad in Japan underwent several changes, mainly in their mudra and posture. The first change was in the mudra formed by the central figure, a seated Amida. While earlier figures had still formed the meditation mudra as seen in the Amida at the Phoenix Hall, in the late Heian period the raigō hand position became more common. For an example, see the Amida triad of 1148 at Sanzen’in (Kyoto) (plate 19). Secondly, the postures of the attendants varied from fully seated to fully standing, with some depicted in intermediate positions. All three figures of the Sanzen’in triad are seated, but in the Heian-period Ganjōji triad (Yamanashi Pre- fecture), while the central Amida forming the raigō mudra is seated, the attendants are both standing. Subsequently, at the beginning of the Kamakura period, Kaikei’s contemporary Unkei (d. 1223) made an Amida triad at Jōrakuji in Kanagawa Prefecture (1189) with standing bodhisattvas, while the Amida is still seated. Finally, as we see at Jōdoji, during the Kamakura period Amida and both attendants are all shown standing. While Kaikei had produced solitary standing raigō Amida figures about three feet in height before he made the Jōdoji sculpture, the latter was probably the first colossal triad in which all three figures assumed the standing posture. The Jōdoji triad’s standing posture was one further development in the Japanese tradition of raigō Amida triads. That posture is also depicted in another raigō triad by Kaikei at Kōdaiin (Wakayama, c. 1221) (see figure 11.4). Kaikei added a new feature: both bodhisattvas bend their knees slightly, displaying more eloquently the triad’s willingness to welcome the deceased. The soft, subtle gesture of the bodhisattvas with the standing Amida became standard for raigō triads thereafter in Japan. While it helped to provide a new posture for Japanese Amida triads, the iconography of the Jōdoji triad occupies a rare position among extant raigō sculpture in Japan as exemplified by the “mainstream” from the Sanzen’in to the Kōdaiin triads. In addition to Amida’s “reverse” mudra, the Jōdoji triad’s distinctive features include the treatment of Amida’s costume, which adds an undergarment and surplice to the usual simple robe. The front of the surplice is decorated with U-shape drapery folds. Other distinctive features are the figures’ long finger- Fig. 11.4. Amida triad by Kaikei, c. 1221. Wood with gold pigment and cut gold leaf. Amida, height 79.2 cm; Kannon, height 43 cm; Seishi, height 48.1 cm. Kōdaiin, Wakayama Prefecture. Important Cultural Property. Photo courtesy of the National Research Institute for Cultural Properties, Tokyo. 268 Secular and Sacred nails and, especially, the hand positions of the flanking bodhisattvas, Kannon and Seishi. The bodhisattvas of the Jōdoji triad were quite different in form than those in most contemporary raigō triads in Japan. Although there are some variations, in most raigō Amida triads from the late Heian period and afterwards, Kannon holds a lotus pedestal and Seishi prays with folded hands, regardless of their postures (see plate 19 and figure 11.4). The function of the lotus pedestal is to carry the soul of the deceased, and Seishi’s folded hands show respect for the dead. But the Jōdoji bodhisattva on the left (probably Kannon) holds a jar and the bodhisattva on the right (probably Seishi) holds a lotus stem with a bound scripture-like object on top of the flower.39 To understand why the Jōdoji bodhisattvas differed from the more common form, I will examine the models on which Chōgen seems to have based the Jōdoji triad, models that reflect his strong interest in Song Chinese culture. In Sazenshū, Chōgen cites “one painting of an Amida triad, made in China, which became the model for the jōroku [images] of Harima [Jōdoji] and Iga [Shin-Daibutsuji].”40 Thus the Amida triads of Jōdoji and Shin-Daibutsuji are based on the same Chinese painting and are similar in style. Kaikei made the Iga triad under Chōgen’s commission around 120241 (see plate 20). Unfortunately, only the head of the central Amida of the Iga temple remains for certain today, so we cannot compare the two Amida sculptures in detail.42 The painting that served as the basis for the triads is not specified, but a similar form is depicted in other paintings from China. Among them, the figures in a Chinese raigō Amida triad painting at Saihōji, Aichi Prefecture, are almost identical in form to those of the Jōdoji triad, including the Seishi figure’s kebutsu (miniature Buddha image) on the crown43 (see plate 21). The Saihōji painting is similar in iconography to the Jōdoji Amida triad, but it probably dates from the Ming dynasty (1368–1644) and thus suggests only that similar iconography may have existed in paintings of Chōgen’s time.44 It should be noted that the above iconography seems not so unusual in Song raigō paintings or in those in Goryeo-era Korea that were influenced by Song work.45 In an earlier article I demonstrated that Chōgen, not the sculptor Kaikei, dictated the style of the Jōdoji triad.46 His preferences produced a work that differed from contemporary raigō Amida triads or Kaikei’s other Amida images. In fact, Chōgen actively promoted Chinese culture in Japan, and the Song style of this triad is eloquent proof of his efforts to bring an eye-catching, exotic icon into his bessho. Jōdoji Amida Triad 269

Icons, Rituals, and Estate Residents Standing before the Jōdoji triad today, one can only imagine how impressive it must have seemed to the officials and cultivators who lived on Ōbe Estate. For one thing, the triad was certainly much larger than any of the older Buddhist images in temples on the estate. Even more important was the welcoming, even aggressive posture of the standing figures, encouraging viewers to expect their own rebirth in paradise. The triad may symbolize Amida’s Western Paradise but it is not shown residing there, nor does it float in space with all three figures in a seated mode, as often depicted in the raigō painting of the time. The iconography of the Jōdoji Amida triad, especially its standing posture, represents Amida’s active efforts to bring salvation to the devotee in his or her final moment. To represent this Pure Land Buddhist idea even more con- cretely, Chōgen seems to have designed the monumental triad to match its daibutsu-style hall, and the totality of the hall and the icon produces the specific visual effects discussed above. Chōgen also presented the raigō theme in the mukaekō (raigō) ceremony instituted in 1200 at Jōdodō, according to Sazenshū. The overall layout of Jōdoji, in which the Jōdodō is opposite the Yakushidō across an open courtyard, played a significant role in this ritual. Jōdoji engi describes the raigō equipment at Jōdoji as follows:

Objects for the raigō ceremony [instituted on] 1200/10/3: An eight- shaku standing Amida Nyorai made by An Amidabutsu, twenty-seven bodhisattava masks also made by An Amidabutsu. Also twenty-seven bodhisattva costumes.47

A standing Amida statue (about 8.75 feet tall) with a bare torso and various masks were made for the ceremony by Kaikei and his studio (see plates 22 and 23). At the ceremony the Amida statue was dressed, placed on a cart, and carried back and forth between Jōdodō and Yakushidō, the two halls built by Chōgen in the same form and size. The Yakushidō is to the east of Jōdodō and the halls face each other across an open courtyard. Mukaekō performances were held at only one other autonomous temple, Watanabe in Settsu Province, which, in addition to a Jōdodō (Pure Land Hall), also had a Raigōdō (hall of Amida’s descent) and a Shabadō (hall that represents “this world”), the latter 270 Secular and Sacred being where the mukaekō performance took place. At Jōdoji, Yakushidō functioned as the Shabadō when the performance was held. During the ritual, people wore masks and costumes representing bodhisattvas and heavenly beings and marched in procession across the courtyard that separates the two halls. Banners and musical instru- ments livened up the atmosphere, creating the illusion of a raigō scene. It seems that with the support of local residents, the ceremony was held every year until around 1591 and then was resumed in 1656.48 This raigō performance seems to manifest Chōgen’s great concern for local residents. While high-ranking visitors from Kyoto might be expected to visit Watanabe Bessho in Settsu (present-day Osaka),49 ceremonies at Jōdoji, far from the capital, must have been aimed at local people. Moreover, the performance was held there repeatedly. It seems certain that Chōgen wished many local people to participate actively in the events and to learn about Pure Land Buddhism. Those religious teachings depicted in such a concrete form could be easily understood through their dramatic presentation. Both the raigō Amida triad and the raigō performance must have been effective in impressing nenbutsu believers and the community in general, especially important for spreading Chōgen’s Amidist beliefs, encouraging donations to his kanjin campaigns, and attracting support for his land reclamation projects on Ōbe Estate. The daibutsu style hall, the large Song-style Amida triad, and the mukaekō ceremony were Chōgen’s means to appeal to local people. He also installed a bathhouse at Jōdoji (as well as at the six other bessho), where monks recited the nenbutsu while bathing. I believe that Chōgen aimed to open these facilities to the public and share with them the blessings from the Bud- dhist images in various forms. I think Chōgen’s attempts to disseminate his Pure Land beliefs in such ways were inspired by his experiences as a kanjin hijiri, which required him to associate actively and aggressively with potential donors to the Tōdaiji reconstruction effort. Just as at other provincial temples noted earlier in this chapter, the residents of Ōbe Estate recompensed Chōgen for his services. Evidence for local support for Jōdoji comes from a list of names of some two hundred supporters inscribed inside the body of the Amida image in the Jōdodō triad.50 The inscriptions date from 1195, at the early stage of the making of the image. From 1985 to 1988, the central cavity of the Amida image was investigated using a fiberscope and a video image scope. The inscriptions Jōdoji Amida Triad 271 discovered in this investigation include the donors’ names (kechien names), along with the date 1195/4/15. This date supports the description in Jōdoji engi, which indicates that the Amida triad was made before the ceremony marking the completion of Jōdodō in 1197.51 Donors’ names can be divided roughly into four categories: names with the amidabutsu suffix, who were probably followers of Chōgen, and monks, nuns, and lay believers. Among the amidabutsu-suffix names are some that also appear on other works made by Kaikei or commissioned by Chōgen, indicating that they belonged to the same religious group.52 These include the name of En Amidabutsu, thought to be Kai- kei’s disciple Shinkai.53 Curiously, neither Chōgen’s nor Kaikei’s name appears, and also missing are the names of others closely associated with Jōdoji or Ōbe Estate: Kan Amidabutsu, Nyo Amidabutsu, and Gan Amidabutsu. Some of the missing names, of course, may appear within the head or the left side of the body, which could not be examined due to technical difficulties. Most of the names, however, seem to be those of lay believers. Sur- names appear to include those of local residents, such as Ōbe and Ō. One Ō Hisakiyo, in fact, was appointed reeve (kumon) of Ōbe Estate in 1206. Toponymic surnames such as Tanba, Takamuko, and Ki probably belonged to those who lived elsewhere. Lay surnames include Naka- hara, Kiyohara, Takashina, and Mononobe, and they are also often found on other images connected with Chōgen and Kaikei. Karikome Hitoshi suggests that many of the lay believers on the list belonged to the myōshu class that supervised cultivation on Ōbe Estate.54 Whether or not they were myōshu, they were no doubt prominent members of the cultivating class who had a role in managing the estate. There are also a number of short prayers that appear among the inscriptions, such as “May I be reborn in the Gokuraku [Amida’s] paradise,” “May the distant deceased soul set out on the road to achieve Bud- dhahood,” and “May we have peace in this life and attain Buddhahood in the next.” These prayers appear often in other contexts and might be regarded as formulaic, but they also reflect the fervent hopes of the Buddhist faithful. The economic benefits accompanying land reclamation and social development through the establishment of hamlets and communal groups might ensure tranquility in the present, but people were also focused on rebirth in paradise for their deceased relatives and for their own next lives. Therein lay the significance of Jōdoji and its Amida triad. Although happiness in both this world and in the afterlife 272 Secular and Sacred is always a human desire, the turn of the thirteenth century was a cha- otic time not long after the end of a civil war, and people must have felt particularly anxious and eager for rebirth in paradise. As a center of nenbutsu practice—one practice thought to guarantee that rebirth—Jōdoji could fulfill that function. For Chōgen himself, on the other hand, the temple could also be used as a headquarters to oversee the estate that provided for Tōdaiji’s own rebirth. To do so successfully required that he establish connections with the local residents, a task facilitated by the iconography of the Amida triad, the design of the Jōdodō, and the overall layout of the temple itself.

Chōgen, the main solicitor for the Tōdaiji reconstruction project, established smaller temples called bessho in various areas of Japan to assist in various aspects of the project and to spread his Pure Land beliefs. In this chapter, I focused on his activities at Jōdoji on Ōbe Estate, which is the only bessho that still maintains an original hall and an Amida triad from some eight hundred years in the past. With special emphasis on the icon from an art historical perspective, I clarified Chōgen’s intense efforts to capture local people’s hearts, which assisted both his religious propagation and land development on Ōbe Estate. The Amida triad has specific iconographical features that were not common at this time in Japan. Among these, the standing posture of the three figures was innovative for raigō Amida triads of the day. The posture reflects specific Pure Land ideas and represents Amida’s aggressive form of salvation. The Jōdoji triad may be the earliest example of such a raigō Amida triad, and it is certainly the oldest extant example we have today. Along with the innovative design of the hall that shel- tered it, the triad must have been welcomed by the local public, some of whom had their names inscribed inside the Amida image and probably donated goods or labor to the temple. The layout of the temple buildings enabled the enactment of raigō ceremonies, which also must have appealed to local residents. It is likely that when Chōgen made such an unusual and impressive icon, he hoped to enhance the lives of estate residents by offering them both rebirth in Paradise and worldly happiness. In practical terms, of course, this must have encouraged people to open new land to cultivation for the sake of the temple, and to cooperate with the temple both as a religious center and as the managing entity of Ōbe Estate. Jōdoji Amida Triad 273

Notes

I would like to express my deep gratitude to Professor Joan Piggott for giving me the opportunity to participate in the shōen project. The project expanded an art historian’s knowledge and views to a great degree and this chapter is a result. I would like to thank Dr. Janet Goodwin for her thorough review of this chapter. Without her comments and suggestions, this chapter would not have been possible. I would also like to thank the reviewers and the editors of this volume and all the relevant temples and museums of Japan that provided permission to reproduce photographs of their precious images. 1. There are various definitions of bessho, but Chōgen saw them as economically independent of Tōdaiji even though they were for the most part on Tōdaiji landholdings. See the chapter by Ōyama Kyōhei in this volume. 2. Fukui-ken 1994, vol. 2, 243. Along with the development of the Taga- rasu area, additional halls such as the Amidadō, the Yakushidō, and the Izu- miji Kannondō were built by the mid-thirteenth century. See KI 14:340–341 (# 10949, 1272/12/29). 3. For the role of hijiri in estate management, see the chapter by Nagamura Makoto in this volume. 4. HI 12 (Kinsekibun hen, 1965): 316–317 (# 348). 5. See Kamikawa 2010, 4–10. 6. See the chapter by Hirota Kōji in this volume. 7. Tanaka 1984, 6. 8. See Tōchika 2009. 9. KI 2: 402–403 (# 1161). 10. Ōsaka-furitsu Sayama Ike Hakubutsukan 2002, 17; Hiraoka 1988, 347. 11. Sazenshū, 42–47． 12. Chōgen’s order establishing Jōdoji (1192/9/27) is contained in the Jōdoji monjo document collection. Both Jōdoji and Tōdaiji maintain copies of the document, but the latter is missing its first half. See Hyōgo-ken shi (His- tory of Hyōgo Prefecture) chūsei 2, 356–357 and OS 4:527–528 (# 219) for the complete version. Almost all Jōdoji related documents are now included in the section of Ōbe Estate related documents in OS 4:506–1040. Some researchers, such as Yamamoto Eigo and Noji Shūza, doubt the reliability of some of these documents. See Tanaka 1973, 84–102, on this issue. 13. Sazenshū, 45 for Harima Bessho. Also see OS 4:537–543 (# 227). 14. “One-bay four-sided” describes the size of a hall according to Heian- period measurements. Dimensions are based on the size of the central altar 274 Secular and Sacred and do not include the rest of the building, as was the case in later periods. In the case of Jōdoji Jōdodō, the facade of the square altar is one bay in width and it is surrounded by a one-bay-wide space. According to Tanaka Tan, the character for shu 竪 in Sazenshū is an error for the character 周 used for the Chinese Zhou dynasty (c. 1122–770 BC). A jōroku image is about 12.86 feet tall in the Zhou-period scale; in the more common Tang scale, a jōroku standing image is 15.75 feet tall. See Tanaka 1973, 89. Note, however, that jōroku images can be larger than that; the Jōdodō image is almost 17.5 feet tall. 15. Sazenshū, 45. 16. Although there is no description in Sazenshū and therefore it might not have been part of Chōgen’s original plan, a document dated 1200 mentions that a Yakushidō of the same size as Jōdodō was built at Suō Amidaji along with a relics hall, a sutra repository, a bell tower, and a dining hall. See KI 2:402–403 (# 1161). An inscription dated 1197 on the iron stupa in the relics hall there mentions the Yakushidō, which was probably completed sometime between this date and 1200. 17. For Chōgen’s testament see OS 4:532–535 (# 224). For the version of Jōdoji engi thought to be the most reliable, the one presently owned by Kobe University, see ibid., # 446, 716–719. 18. The entry for Yakushidō in Jōdoji engi states that the framework of the hall was built on 1197/3/28 by the carpenter who constructed the Jōdodō. 19. While Sazenshū indicates that the image was measured using the Zhou scale, the Engi makes no mention of this. Tanaka Tan thinks this could have been an error in the original or in its 1614 copy. Tanaka 1973, 91. 20. According to Chōgen’s testament, the images numbered eight hundred. 21. OS 4:532–533 (# 224). 22. This discussion of the features of the triad and the hall is based on a fuller account in Kainuma 2014. 23. This is what the document calls the hall. 24. Kaikei’s name is written in different characters in the Engi than in other standard sources, but in premodern Japan names were often written with different characters, so the Jōdodō sculptor was undoubtedly the famous Kaikei, who in fact occasionally used the title “Tanba Hōgen.” In 1197 Kaikei had not yet received the Buddhist rank of “hōgen” (eye of the law), which was awarded to him between c. 1208–1210; thus, this title was probably used anachronistically by the writer of the Engi. For a discussion of questions regarding the triad’s sculptor, see Kainuma 2014. 25. OS 4:717–718 (# 446). Jōdoji Amida Triad 275

26. Song dynasty dates are 960–1126 (Northern Song) and 1127–1279 (Southern Song). The style is also called “tenjikuyō,” which literally means a style from India, but actually indicates a foreign (continental) style. 27. For an explanation of the short life of the daibutsu style in Japan, see Itō 1983, 137–149. The Jōdodō is designated a National Treasure because it displays this short-lived architectural style. 28. This interval between the pillars is unusually large. In the Byōdōin Phoenix Hall, one bay is just under ten feet. 29. For instance, the space around the inner sanctuary of Hōkaiji Amidadō (Kyoto) is just over thirteen feet wide, and the sanctuary itself is about eighteen square feet. See Ono-shi Kyōiku Iinkai 1997, 7–8. Figures 11.1 and 11.2 are reproduced from OS Bekkan: Bunkazai hen (Special Volume, Cultural Materials Collection). 30. For the Byōdōin Phoenix Hall, see Nishikawa et al. 1987–1992. Also see the exhibition catalogue, Visions, from Tokyo National Museum 2000, Kainuma 2014, figs. 11–12, and Yiengpruksawan 1995. 31. See Kyoto National Museum 1975, figs. 7–18; Nara National Museum 2013 catalogue (for various versions of the Pure Land mandala); Fujisawa 2013, 25–78. 32. For more information concerning this structure and the device, see Kurata 1971, 6–7; Tanabe 1972, 30; Yamazaki 1982, 73–74. 33. The present dais might be a later addition. Tanabe Saburōsuke suggests, however, that the original one was also in a circular shape. Tanabe 1976, note 13. 34. OS 4:528 (# 219). 35. For more detail concerning Chōgen’s Pure Land faith, see Kainuma 1994, 236–252. For speculations concerning public participation in this ceremony, see Hiraoka 1988, 336, and Ono-shi Kyōiku Iinkai 1997, 7–8. 36. Today the shitomi-do is opened for part of the summer afternoon (late July to early August) in order to present Amida’s raigō scene to the public. 37. According to the investigation by a Tokyo Geijutsu Daigaku group, in the afternoon the sunlight from the west comes into the hall, hits the floor, and then shines on the red rafters and beams of the uncovered under- surface of the roof. The reflected light next shines on the triad. The result is that the images turn reddish in color from their heads to their knees, and the triad appears to be floating in air. See Nishimura et al. 1982, 11–14, 29–36. In July 2004, Sakata Akihiro made a simulation of this sunlight phenomenon at Jōdodō, based on the above study. He reported that the incoming light hits the floor, causing the diffusion of the reflection. The rays reach the lower part of the western ceiling and then shine on the upper part of the 276 Secular and Sacred eastern ceiling. The diffused light rays finally shine on the triad. See Sakata et al. 2006, 7–23. 38. For example, see Hamada 1979 and Nakano 1985, 190–193. 39. Jōdoji priests identify the left-hand bodhisattva as Seishi and the right- hand one as Kannon. 40. Sazenshū, 45. 41. Sazenshū, 46. The Iga Bessho was founded during the Kennin era (1201–1204). The most likely date for its founding is 1202, the date on a copy of a vow that was probably related to the temple’s opening ceremony. See Mōri 1961, 40. There are ink inscriptions inside the head of the Amida image, which include the names of “Daiwajō Namu Amida” (Chōgen), and “Daibusshi An’a” (Kai- kei). In fact, the original head of the Amida is more in the style of Kaikei than is the Jōdoji Amida figure. See Kainuma 2014 for Kaikei’s style. 42. One ear and one hand have been found and identified as possibly belonging to this Amida triad. For the ear, see Yamamoto 2002, 39–44, and for the hand, see Iwata 2004, 1–4. During the Edo period the head was joined to a new body to form a seated Rushana Buddha. Tanabe 1976, 76; Tanaka 1976, 41. 43. The exceptions are minor features such as Kannon holding a willow or the figures of the triad standing on a fumiwari pedestal (which consists of two lotus pedestals, one for each foot). For more detail, see Yamamoto 1990, 343–362. 44. Although Yamamoto Yasukazu originally introduced the Saihōji raigō Amida painting as a Song period work in his article (1990), a Ming period date is widely accepted in today’s Japanese scholarship. See Yamaguchi 2010, 104. 45. For instance, the Amida Jōdozu (dated 1183) at Chion’in in Kyoto is a Song painting (see Nara National Museum 2006, 198–199; Kainuma 2014, fig. 8) and the Amida triad of Senshūji (Mie Prefecture) is a Korean painting probably dated in the thirteenth or fourteenth century (Goryeo period). See the illustrations of the latter in Kikutake and Yoshida 1981, e.g., plates 15–19, 28. 46. Kainuma 2014. 47. OS 4:718 (# 446). A shaku is approximately equivalent to a foot. 48. OS 2:560. 49. The retired emperor Go-Toba attended the mukaekō at the Watanabe Bessho on 1201/9/21. See Hyakurenshō 1901, 172. 50. Devotees’ names were also found in the Jōdoji Amida statue made for the mukaekō. See Tanabe 1976, 89–90, and Hyōgo-ken shi, chūsei 4:480–481. Many of Kaikei’s works contain inscriptions including donors’ names inside them. Jōdoji Amida Triad 277

51. The research was carried out three times during the years 1985–1988 by Mizuno Keizaburō and others. See Mizuno and Soejima, 1996, 367–376. 52. Among over two hundred names, there are a limited number of Amida- butsu names (e.g., Hō, Kyō, Kai, En, Zen, Myō, Ji, Dai, Mon). 53. His name can be found in Kaikei’s Amida at Shinkyōji (Tochigi), Dainichi at Tokyo Geijutsu Daigaku (Tokyo), Dainichi at Ishiyamadera (Shiga), Monju at Monjuin (Nara), Fudō at Daigoji (Kyoto), and Jizō in the Burke Col- lection (New York). 54. Karikome 2004, 159. pa rt v

Power, Space, and Trade

c h a p t e r 12

Nyoin Power, Estates, and the Taira Influence Trading Networks within and beyond the Archipelago

Sachiko Kawai

In the fourth month of Jishō 4 (1180), Minamoto no Yukiie1 was making his way to the east, having been entrusted with Prince Mochihito’s directive calling on the Minamoto to rebel against the Taira, the military house that then dominated the throne and court.2 His mission was to “inform the previous vice-director of the Military Guards of the Right [Minamoto no Yoritomo] first, and then to spread the order to other Genji [Minamoto] families.”3 Just before his departure, Hachijō-in (1137–1211), an adoptive mother of the prince and a half sister of the retired sovereign Go-Shirakawa (1127–1192, r. 1155–1158), appointed Yukiie as her close attendant or kurōdo (secretary-provisioner).4 Kurōdo were in charge of various tasks such as document management, event preparation, and other miscellaneous work, but they were also important for delivering messages. Because Hachijō-in’s private estates were widely distributed in the eastern part of the archipelago, Yukiie’s appointment served to hide his real mission. As the largest landholder of the realm, Hachijō-in also provided Yukiie substantial assistance and protection while he traveled through the eastern provinces. In the fifth month of the same year, just after the failure of Prince Mochihito’s rebellion, Taira no Kiyomori (1118–1181) took decisive action to move the capital to Fukuhara (Harima Province) by sending his own grandson, Antoku Tennō (1178–1185, r. 1180–1185), and his son-in-law, the retired sovereign Takakura (1161–1181, r. 1168–1180), to live there. Takakura’s father, Go-Shirakawa, who was under house arrest by Kiyomori’s command, also went along. The rumor that

281 282 Power, Space, and Trade

Go-Shirakawa’s full sister Jōsaimon-in (1126–1189) was also moving to Fukuhara with the Taira stunned court society.5 Jōsaimon-in possessed many estates, especially in and near the Kinai region, and with her estates in Ōmi, Yamato, and Settsu, she controlled many strategic points along transportation and trade routes. Moreover, she occupied an important political position within the royal family and court society as a senior daughter of Toba (1103–1156, r. 1107–1123) and Taikenmon- in (1101–1145). After her parents’ deaths, Jōsaimon-in had become the equivalent mother (junbo)6 of Go-Shirakawa and helped him strengthen his claim over the throne. Taking such a politically important royal to Fukuhara would have been beneficial for both Antoku and Kiyomori (see figure 12.1). For both Hachijō-in and Jōsaimon-in, political influence derived in part from their vast holdings at strategic points in the archipelago. In fact, estate proprietorship by a number of royal women at the end of the twelfth century had great economic, military, and political importance. Although specialists on the late Heian and Kamakura periods traditionally emphasize the roles of warriors and powerful retired sovereigns, another important but too often forgotten aspect of this era is that royal

Fig. 12.1. Royal family genealogy. Nyoin Power, Estates, and the Taira Influence 283 daughters and wives accumulated massive numbers of estates. These female landowners often held the title of nyoin, the female equivalent of a male retired monarch. Due to its long and complex history, the term nyoin is difficult to define. The criteria for receiving the title underwent dramatic changes between the emergence of the first nyoin in 991 and the late Heian age, the period covered by this chapter. To receive the title, a woman originally had to meet two criteria: first, she must have already been promoted to the position of queen consort; second, she had to be the biological mother of a crown prince or a sovereign. These qualifications started loosening within a century, as in the case of the fourth nyoin, who received the title in 1074 without being the biological mother of a sovereign. In 1093, an unmarried royal princess became the fifth nyoin without ever having been a wife of a monarch.7 In this chapter I will examine how three late Heian nyoin—Hachijō-in (became nyoin [hereafter n.] in 1161), Jōsaimon-in (n. 1159), and Kōkamon-in (1122–1181, n. 1150)—struggled to maintain and benefit from their estates. Jōsaimon-in and Hachijō-in were unmarried royal daughters. Kōkamon-in was a Fujiwara daughter who, although she was married to a monarch, never gave birth to a future occupant of the throne. With “authority” over their estates, nyoin and their associates gained opportunities to attain substantial political, economic, and even military power. I define “authority” here as sociopolitically sanctioned rights, and “power” as the ability to exercise those rights effectively to get things done. Instead of assuming that power always comes first and authority follows, I see a two-way relationship that permits us to analyze how women used their authority to wield power, while asking whether the authority of nyoin actually did guarantee power. One fundamental component of the complex relations between authority and power in medieval Japan is the medieval estate system, characterized by its vertical hierarchy and by shiki, shared rights to income from the land.8 In the late Heian and Kamakura periods, nyoin were among those who held the highest authority, the honke shiki or supreme proprietorship. Nonetheless they had to share the products of the land with many others. Furthermore, supreme proprietors did not directly manage ground-level operations and thus could face serious challenges in securing a share of the estate’s material wealth. Nagahara Keiji explains that upper-class shiki holders residing in Kyoto “possessed the most lucrative proprietary rights” while lower shiki 284 Power, Space, and Trade holders took on day-to-day managing tasks at the actual site.9 He concludes, however, that it was not necessarily the supreme proprietor who was “the dominant proprietary lord”; sometimes the estate custodian (azukaridokoro, who managed the estate as the proprietor’s representative) benefited most from the estate’s wealth. For example, Nagahara argued that Taira no Shigehira (1157–1185), the custodian of Ōta Estate in Bingo Province, enjoyed a larger share of estate dues than did the supreme proprietor, the retired sovereign Go-Shirakawa.10 Although the situations (or ways in which shiki holders divided income) differed from one estate to another, Nagahara’s research nonetheless suggests the danger of correlating the hierarchical order of estate proprietorships with economic gain.11 The above example also suggests that the supreme proprietorship of a nyoin did not always promise her economic prosperity despite her large landholdings. Close investigation of nyoin estate management therefore helps us understand royal women’s land-based power as well as the complex power-sharing mechanism of medieval Japan. Early twentieth-century Japanese historians such as Nakamura Nao- katsu and Yashiro Kuniji recognized that royal women, especially nyoin, possessed sizable estates, but they considered royal women to have been landholders in name only and that their estates were under the control of male retired monarchs.12 Furthermore, these historians centered on the chain of royal inheritance rather than on the royal heiresses themselves. Instead of discussing the ways nyoin inherited and managed their estates, scholars focused on the ways late thirteenth- and fourteenth- century male royals vied for properties transferred from their female ancestors when the royal house split up into two competing branches. Fortunately, historical scholarship concerning nyoin and their property has been increasing of late.13 By challenging the older view that royal women were mere titular landholders, post-1990 studies also argue that nyoin held authority to manage their estates independently from male royals. For example, Gomi Fumihiko and Banse Akemi have disputed the view that retired sovereigns were dominant over nyoin. By suggesting that nyoin enjoyed political and economic independence at court, these studies have encouraged historians to reconsider nyoin as historical figures in their own right.14 Although newer scholarship has made a significant contribution, it tends to emphasize nyoin’s successful and independent control over their estates and unwittingly has created the simplistic idea that the more estates a royal woman owned, the greater was her financial and Nyoin Power, Estates, and the Taira Influence 285 political influence. Careful analyses of individual cases have provided a more nuanced image of nyoin power. For example, Noguchi Hanayo challenges the equation of nyoin’s estates to their economic wealth by examining the revenue and expenditures of Hachijō-in’s vow temple, Anrakuju’in. Through this case study, Noguchi reveals that all the resources submitted from the estates were consumed by the temple to maintain its livelihood and ceremonies. By pointing out that medieval nyoin usually owned two different types of estates—those that were managed by their administrative headquarters (chōbun) and those that supported vow temples (goganjiryō)—she argues that the former covered a nyoin’s living expenses while the latter were used only to finance memorial services for the repose of royal ancestors at the vow temples. Nogu- chi’s study also underscores that nyoin estates did not always indicate their wealth but rather reflected their responsibility for memorializing their deceased family members.15 In this chapter, I want to further existing scholarship by considering additional factors such as estate type, geographical location, and the experiences of individual nyoin as landowners. Through several case studies, I argue that two conditions were necessary for a nyoin to wield maximum influence and power through her estates. First, she had to own estates in areas important to transportation and trade, and second, during times of crisis, she had to seek assistance from other powerful authorities rather than maintaining absolute independent control over her land.

Estate Locations and Taira Connections: The Case of Jōsaimon-in In the late Heian period, domestic and international trading networks that linked Japan with other parts of Asia produced cultural prestige and wealth for estate holders. Leading warriors and courtiers not only participated in commerce within the country but were also interested in international opportunities as well. In particular, it was advantageous to own estates in regions that were convenient for trading with the Asian continent and the Korean peninsula. For example, with the support of the retired sovereign Go-Shirakawa, Taira no Kiyomori accumulated estates in the Seto Inland Sea and built a trading port to promote Japan–Song China trade. The Fujiwara branch that ruled the northeastern region during the twelfth and thirteenth centuries acquired wealth through supporting different trading routes, including the Pacific 286 Power, Space, and Trade

Ocean up to the Tsugaru Strait.16 Despite the limited evidence concerning nyoin involvement in international trade, we can establish that our three royal ladies in fact also held estates in strategic locations that were likely to have facilitated domestic transportation and commerce and were also necessary to enhance international trade. Recent research suggests the meshing of international and domestic trade. For instance, Watanabe Makoto argues that importing foreign goods required more than a single transaction and involved numerous individuals including shrine associates, monks, and Song Chinese sea traders. Even alluvial gold from the northern province of Ōshū (Mutsu), the item most famously exported to China, was not entirely sent abroad: merchants used it within Japan to finance the transport of other exports such as sulfur, mercury, and pearls.17 Royals and aristocrats residing in the capital also utilized gold they received from their estates. For example, Sen’yōmon-in (1181–1252, n. 1191), a contemporary of the three royal women under discussion, collected gold from her estates to support her secular and religious activities at her Rokujō mansion. She ordered a majority of her estates to submit gold to prepare for New Year’s Day, to renew the décor of her residence twice a year, and to sponsor Buddhist events. Since production areas for gold were limited, her estates would have acquired it through trade. This suggests that export items such as gold were sold at venues between their place of production and the final destination for foreign trade, demonstrating that domestic and international trading networks were inseparable. A nyoin who owned estates close to such trading routes could gain economic power to realize her secular and religious goals. Jōsaimon-in owned estates around the capital that played a significant part in connecting trade routes from the eastern, western, and Japan Sea regions. (For an overview of all of Jōsaimon’in’s estates, see figure 12.2.) Her Matsui and Ichinobe estates in Yamashiro Province were in an area that had developed as a strategic transportation zone even before the Heian period (see figure 12.3).18 Both were adjacent to roads and waterways. Furthermore, rivers such as the Kizu played an important role in tying these estates not only to Kyoto but also to other trade nodes on the shore of Lake Biwa in Ōmi Province.19 There Jōsaimon-in possessed six estates, at least four of which—Yoshida Estate, Ikadachi Estate, Seri Estate, and Funaki Estate—were adjacent to Lake Biwa (see figure 12.4).20 Lake Biwa connected the transportation networks around Kyoto with networks on the north side of the lake, and in Nyoin Power, Estates, and the Taira Influence 287 turn with the coast of the Japan Sea.21 All these properties contained or were close to rivers, suggesting that they played a significant role in facilitating transportation to and around Lake Biwa.22

Fig. 12.2. Jōsaimon-in’s estates (all). Template from Jeffrey P. Mass, ed., The Origins of Japan’s Medieval World, Stanford University Press, 1997. Originally drawn by Arnie Olds. Used with permission. Fig. 12.3. Jōsaimon-in’s estates, Yamashiro Province. Drawn by Philip Garrett. Fig. 12.4. Jōsaimon-in’s estates, Ōmi Province. Drawn by Philip Garrett. 290 Power, Space, and Trade

Jōsaimon-in’s estate residents not only used rivers to transport resources domestically, but also joined the broader trading networks tied into Lake Biwa. Once residents arrived at the lake, they had access to other waterways. For example, the Seta River flowed out from the lake, merged into the Yodo River, and then reached all the way to Osaka Bay. The Hino and Mano Rivers also belonged to the same drainage system as the Yodo River. Such connections brought Funaki and Ika- dachi estates “closer” to the Seto Inland Sea. As commerce between Heian Japan and Song China became increasingly active between the late Heian and mid-Kamakura periods, the Seto Inland Sea played an important role in spreading trade-related wealth.23 The names of Jōsaimon-in’s estates indicate their industries and products. One such example is Ikadachi Estate. Its name denotes raft building or raft launching; logs, after being harvested from local timber, were likely tied together there to transport people and material resources from the estate. According to thirteenth-century sources, Ikadachi Estate submitted timber, fishing boats, and charcoal to its proprietor.24 Although these sources postdate Jōsaimon-in’s death, they suggest that the estate had provided her with wood products. The name of Funaki Estate also denotes logs for shipbuilding. Like Ikadachi, Funaki Estate was probably associated with the manufacture of boats and goods made of wood. Estates also provided metal products for their proprietors. The name of Jōsaimon-in’s Dezu Estate in Yamashiro Province literally means the Coinage Office Estate (see figure 12.3). This estate contained the minting site for early royal coins, such as the Wadō kaichin that were issued in 708. Even after the abolition of the Minting Office, copper mines at Dezu Estate continued to operate.25 Being rich in copper-bearing ore, this estate is likely to have produced copper-related materials for its proprietors. An important characteristic of Jōsaimon-in’s estates in Ōmi was their close ties to powerful religious institutions. Medieval religious institutions often had shrine associates (jinin) who served as merchants, traders, and moneylenders.26 Located on the north side of Mount Hiei, Ikadachi Estate had close connections with Hiei temples and shrines (see figure 12.4). In some sources, Ikadachi Estate appears as one of the estates that supported Mudōji, a Buddhist temple on Mount Hiei.27 Since no evidence indicates that Mount Hiei and Jōsaimon-in vied for possession of the estate, they may have owned different parts of it, or Nyoin Power, Estates, and the Taira Influence 291 they could have had a profit-sharing agreement. Mount Hiei’s shrine associates, the Hie jinin, played a key role in transporting goods across provinces and conducting lending businesses.28 If Jōsaimon-in were able to bring these mercantile experts onto her estate and enhance commercial activities without conflict with Mount Hiei, she would have increased her chances to gain both material and human resources. In the latter scenario, the relationship between Jōsaimon-in and Mudōji would have been one between a supreme proprietor (honke) and a managing proprietor (ryōke). Powerful royals often held supreme proprietorship over estates supporting a branch temple of a large religious institution. A supreme proprietor held the highest authority and responsibility to manage estates, to maintain the condition of the land and the people living there, and to receive agreed-upon profits. Right under the supreme proprietor, a managing proprietor usually managed the estate and ensured that dues were submitted. Funaki Estate was also closely associated with another religious institution, the Kami (Upper) Kamo Shrine in the Heian capital. One list of Jōsaimon-in’s estates indicates that Funaki supported Hōkongō’in, the vow temple of Jōsaimon-in’s mother.29 Late twelfth-century evidence, however, indicates that Funaki Estate was also sending dues to the Upper Kamo Shrine.30 Jōsaimon-in and the shrine might have owned different parts of the same estate, but again, it is possible that the latter held the managing proprietorship under Jōsaimon-in, the supreme proprietor. Associates of Kamo Shrine may have participated in the estate’s shipbuilding business, its fishing industry, and the transport of goods around the lakeshore, all activities in which such associates were often involved.31 Jōsaimon-in was once a Kamo priestess; she was nominated for the post at the age of one year (in 1127), and she served until illness made her resign in 1132. But the reason for her close connection with the shrine appears to have been more complex. She had inherited a number of other estates, indicating that she occupied a prominent status within the royal family. For example, she had inherited royal properties (mikuriya) that provided food to the royal family or supported Ise Shrine.32 She also held Tamasaki Shrine and its estates in Kazusa and Shimōsa provinces. According to Zenkoku jinja meikan, Tamasaki Shrine was established by the legendary prince Yamatotakeru no Mikoto to pray for his success in eastern expeditions.33 Albeit a mythical one, this account nonetheless suggests the significance of the shrine for the royal family and the state. 292 Power, Space, and Trade

Jōsaimon-in’s important position within the royal family and her estate holdings suggest why the rumor in 1180 that she was moving to Fukuhara with the Taira shocked court society. Unmarried royal prin- cesses often strengthened the claims of a brother or nephew to the throne. By adopting a male relative or living with him, royal daughters indicated to court society that he was the legitimate royal successor.34 In this instance, the Taira needed to strengthen the legitimacy of the one- year-old monarch Antoku Tennō, grandson of the Taira chief Kiyomori and the major element of their political capital at that point. Jōsaimon- in’s move with Antoku to the new capital would have made her support for him quite visible. Moreover, the locations of Jōsaimon-in’s estates would have benefited the Taira. Exchanging goods and information is critical to the exercise of power and thus makes firm control over transportation important.35 As we have seen, Jōsaimon-in possessed estates in Ōmi, Yamashiro, Kawachi, and Settsu, where the Taira were not provincial proprietors (chigyōkokushu).36 In these areas, shrine associates of powerful religious institutions had formidable influence over transportation systems. It is said that the reason the Taira moved the court to Fukuhara was their unfavorable relationship with these religious institutions. Nonetheless, the Taira could not give up the Kinai and Ōmi regions, since the area was not only a commercial and agricultural center but also held historical significance as the site of older capitals. Knowing that their influence over this region would become even weaker after moving the capital to Fukuhara, the Taira likely thought that cooperating with Jōsaimon-in, who owned many estates in these areas, would be crucial for their success in both domestic and international trade. For her part, Jōsaimon-in was empowered through her ties with the Taira. Kiyomori’s sister-in-law, Taira no Jishi (1142–1176; became Kenshunmon-in in 1169), served Jōsaimon-in as her attendant. Later, Jishi married Jōsaimon-in’s full brother Go-Shirakawa and gave birth to Takakura Tennō, father of Antoku. Helping Kiyomori and his wife’s family to build the foundation for their political power, Jōsaimon-in was very much a respected benefactor of the Taira family. Also, a noticeable number of Jōsaimon-in’s female attendants and several Taira daughters came to serve Kenshunmon-in after the latter became a nyoin.37 This shows that the two nyoin shared networks of their female attendants.38 Since attendants of a late Heian nyoin often managed Nyoin Power, Estates, and the Taira Influence 293 estates for their mistresses as estate managers or custodians (ryōke or azukaridokoro), Jōsaimon-in’s close ties with Kiyomori’s kinswomen and Kenshunmon-in’s attendants would have encouraged members of the Taira clique to be supportive of Jōsaimon-in’s estates. In contrast, Hachijō-in’s Arakawa Estate faced continual challenges from Taira followers who tried to harass the estate residents and confiscate their products.39 Although more comprehensive research is necessary, Kiyo- mori and his followers seem to have been less aggressive concerning Jōsaimon-in’s estates. Jōsaimon-in’s case also strongly suggests how the location of estates might affect the political and economic influence of royal ladies. But owning estates did not automatically guarantee one’s power or wealth, as seen some years earlier in the experience of Jōsaimon-in’s mother, Taikenmon-in. In 1135 Taikenmon-in ordered agents (satanin) at three ports within Kuroda Estate in Iga Province to cease “illegally” impos- ing tariffs on lumber needed to repair Tōendō, the Buddhist hall built upon her request within the monastery complex Kōfukuji.40 The agents in question served another powerful Nara monastery, Tōdaiji, the proprietor of Kuroda Estate. Taikenmon-in’s administrative headquarters issued an order decrying her loss:

During the previous year these agents of the three landings acted against precedent for the first time. In attempting to levy an illegal port toll, they seized our lumber destined for hall repair and prevented [the transporters] from passing through [the check point]. The Izumi Lumber Landing, in its age-old history, has never confiscated a single log from our dependents (yoriudo).41 This is the first time that the agents of Natsumi, Yakawa [alt. reading Yagawa], and Nakamura docks have attempted to take [our lumber], and such an act has no legitimate basis.42

After listing the names of twenty of the dependents, Taikenmon-in’s order added:

[Nevertheless] if these dependents have followed others’ words and claimed different logs [from lands belonging to others] as our repair lumber [to escape the tariff], then [such immoral acts] should cease from this point forward.43 294 Power, Space, and Trade

This quote helps to reveal the complexity of the problem, suggesting Taikenmon-in’s concern over the possible wrongdoings of her own dependents. From this incident we can deduce three things. First, Taikenmon- in had the authority to issue this order concerning the transport of her goods and problems with them outside her estates. Second, despite such authority, she did not have the power to resolve the problem immediately. This incident had occurred in the “previous year (1134),” but she did not issue the order until the fifth month of 1135. No conclusive source is available to explain why it took so long for Taikenmon-in to issue the order, but some primary documents suggest that the agents of Natsumi, Yakawa, and Nakamura landings not only held control over the ports but were also aligned with neighboring areas. According to a land survey conducted in the seventh month of 1134, Natsumi, Yakawa, and Nakamura became a part of the “newly established [Kuroda] estate” of Nara’s Tōdaiji. It also encompassed dry fields, ponds, buildings, and local Buddhist temples such as Shōrenji, indicating a large-scale community surrounded the three landings.44 Moreover, rivalry with other power holders at court likely affected Taikenmon-in’s ability to control the situation at the three landings. For example, a previous head of the Council of State, Fujiwara no Tada- zane (1078–1162), had interests in this area, including showing support for Tōdaiji. Three months after the aforementioned land survey, Tada- zane’s household office issued an order to prevent assaults from the Iga provincial governor against the residents of Yakawa and Nakamura.45 Taikenmon-in’s case is an example of conflicts characterizing the medieval estate system in which various estate proprietors and local people had to share profits from the lands and waterways. Third, she had to keep a careful eye on the dependents sending timber to her hall, because they could take advantage of their privileges. As her order indicates, they might have tried to avoid port tolls when transporting items for other estate owners or for themselves. Being supreme proprietors, nyoin had to protect these dependents and empower them. As Taikenmon-in’s case shows, however, a nyoin also needed to protect the status and influence of her household by avoiding false claims. Jōsaimon-in, a primary heir to Taikenmon-in’s temples and properties, must have faced similar challenges while managing her estates. Nyoin Power, Estates, and the Taira Influence 295

Close analysis of the geographical and historical significance of Jōsaimon-in’s estates uncovers the ways in which she gained power from her estates. A noticeable number of her estates had good access to circuits important for both domestic and international trade, from which she increased opportunities to obtain wealth. She also held supreme proprietorship over estates tied to religious institutions through which shrine associates and estate dependents worked not only for religious causes but also for their own economic benefit. Her authority and responsibility over estates therefore broadened her economic, political, and religious influence. As seen in her mother’s case, however, having such authority did not always guarantee a nyoin actual power. An early medieval nyoin had the authority to issue an order to protect her estates against any opponents, but without using that authority in a timely and effective manner, she could lose her chance to wield power.

The Production Center and Maritime Trade Networks of Hachijō-in and Kōkamon-in We have already seen that political and economic influence resulted from the movement of people and goods, and that estate locations affected the ways nyoin exercised power. Focusing on Hachijō-in and Kōkamon-in’s estates within the capital and along maritime trading routes, this section further investigates the relationship between nyoin power and estate locations. Hachijō-in, who owned one hundred estates in 1176 and accumulated even more later, had her main residence with her administrative headquarters and storehouses on Eighth Avenue in the capital (see figure 12.5).46 Archaeological evidence shows that late Heian Kyoto’s Seventh and Eighth Avenues were a manufacturing center for items including paintings, ornaments, weapons, and Buddhist images. Crafts- men such as metal workers, sculptors, cap makers, and lacquer paint- ers made the area an artisanal center.47 Seventh Avenue was home to metal work specialists and a production center of quartz sand, a crucial ingredient in crafting metals. Moreover, living close to these production centers of war-related materials including armor, weapons, and stirrups was also strategically important for Hachijō-in, who needed to sustain both political and military influence when warriors such as Kiyomori were expanding their military power in the capital and provinces.48 Fig. 12.5. Properties of Hachijō-in and Kōkamon-in in the capital. Drawn by the author. Nyoin Power, Estates, and the Taira Influence 297

An older nyoin who lived close to Hachijō-in and the Taira was Kōkamon-in. Unlike Jōsaimon-in and Hachijō-in, who were unmarried royal daughters, Kōkamon-in was a Fujiwara daughter who had married Sutoku Tennō (1119–1164, r. 1123–1141). She became a nyoin in 1150, twenty years after her marriage, but her husband was exiled to Sanuki when his faction lost the Hōgen War of 1156. Kōkamon-in, however, remained in the capital and maintained her nyoin status. Being the only child of the primary wife of Fujiwara no Tadamichi, she inherited quite a few estates, including properties within the capital around Ninth Ave- nue and in other provinces.49 In comparison to Hachijō-in, the number of estates Kōkamon-in accumulated during her lifetime was small. According to the “List of Kōkamon-in’s Estates” provided by Kokushi daijiten, she owned around thirty-five properties.50 Given the difficulty of determining the exact number of her estates due to a lack of comprehensive records, it is highly possible that Kōkamon-in owned others in addition to the listed estates. The total number, however, probably did not exceed one hundred as it did in Hachijō-in’s case. Kōkamon-in’s estates, however, were in good regions for commercial and trading activities. Living in her Kujō mansion and owning several Ninth Avenue properties, she had easy access to the Seventh and Eighth Avenue production center (see figure 12.5). She also owned Misaki Estate in Shimōsa Province, which included important ports facing the Pacific Ocean (see plate 24). Scholars have emphasized that Pacific Ocean shipping routes played a significant role in connecting Kyoto and Hiraizumi, the power base of the wealthy Ōshū Fujiwara warlords in northeastern Honshu.51 In this regard Noguchi Minoru, by analyzing Kamo no Chōmei’s poem about the Misaki Port, points out a possible tie between Settsu sea traders and Misaki Estate. The poem—“How brightly shining! The moon rising on the waves of Misaki Bay”—was based on information given by marine traders working at Settsu in 1180. It suggests that even if Settsu tradesmen did not physically travel to the Misaki Port, they learned about the port while exchanging goods with the easterners.52 Noguchi also posits that northeastern warrior families, such as the Kataoka, Satake, and Iwaki, supported routes connecting the Kinai region to Hiraizumi through the Pacific Ocean lanes. Interestingly, the late twelfth-century local manager of Misaki Estate was Kataoka Tsuneharu, whose toponymic surname originated from a northern province, Mutsu.53 Here 298 Power, Space, and Trade

Kōkamon-in’s supreme proprietorship over Misaki and other estates in northeastern provinces likely enhanced her chance of tapping the wealth produced in that region. Her estates in western provinces such as Settsu, Izumi, and Ōmi, where traders from different regions com- municated with each other, further helped her achieve economic gain through access to a broad trading network. For example, she owned Northern Kozuru Estate and Southern Kozuru Estate in Hitachi, near the Pacific Ocean shipping routes. She also owned Kuragaki Estate in Settsu, Ōizumi Estate in Izumi, and a place called Ōhata in Ōmi.54 In addition, her estates in northern Kyushu were advantageous for international trade. Previous scholarship emphasizes the roles played by regental family heads in acquiring rare materials from southern islands

Fig. 12.6. Hachijō-in’s estates. Template from Jeffrey P. Mass, ed., The Origins of Japan’s Medieval World, Stanford University Press, 1997. Originally drawn by Arnie Olds. Used with permission. Nyoin Power, Estates, and the Taira Influence 299 and the Song through their ownership of Shimazu Estate in southern Kyushu.55 Noguchi claims that this trading network extended farther north as well; for instance, he thinks that marbled turban shells used to decorate the Ōshū Fujiwara’s Golden Hall at Chūsonji in Hiraizumi were first imported to Shimazu Estate from the southern islands.56 Fujiwara daughters also owned estates along these trading networks. Certainly Kōkamon-in’s estates in northern Kyushu were conveniently placed to take advantage of international trade. Her Tsumori Estate in Bungo, for example, was developed at the mouth of the Ōita River near today’s Ōita Port (plate 24). Facing the Seto Inland Sea, this estate was a strategic zone for domestic and international trading routes. Kōkamon-in also held two other estates: Yoka and Ukiu in Hizen Province. Yoka Estate was further inland than Tsumori Estate, but it was close to the Chikugo River in Hizen and had access to Ariake Bay. The actual location of Ukiu (or Ukyū) is unknown, but there are areas called Uki and Uku in today’s Nagasaki Prefecture. While Uki is on the coastline of Tachibana Bay adjacent to the Amakusa Sea, Uku is on Uku Island, just east of Hirado Island. If this were the location of Ukiu Estate, it would have provided Kōkamon-in access to the Korean peninsula and Song China. Despite her husband’s political downfall due to losing the Hōgen War in 1156, Kōkamon-in retained her proprietorship of these estates. This was probably because prior to the war, she had built a political alliance with Bifukumon-in (1117–1160, n. 1149), who was on the winning side.57 Kōkamon-in’s father, Fujiwara no Tadamichi, also supported Bifu- kumon-in. Kōkamon-in’s political ties helped her keep her estates in geographically advantageous locations that supported her livelihood in Kyoto. By owning Ninth Avenue properties and provincial estates along important transportation nexuses, she had the potential to generate economic power through domestic and international trading networks. Hachijō-in also owned property at strategic transportation points connecting Kyushu and the Kinai region (see figure 12.6). Facing the Genkai Sea, her northern Kyushu estates—Noke and Munakata—were close to trading activities centered around Dazaifu, the Kyushu government headquarters. Facing the Suō Sea, her Denbōji Estate was on the western edge of the Seto Inland Sea network. From there, Hachijō-in’s estates were scattered along the coastlines of Shikoku (Iyo, Sanuki, and Awa) and Honshu (Aki, Harima, and Settsu). Some of her estates, on islands such as Shōdo-shima, Nōmi-jima, and Kurahashi-jima, served as steppingstones for boats traveling across the sea. When Kiyomori invited 300 Power, Space, and Trade

Song and Korean merchants into the Seto Inland Sea, these traders may well have stopped at Hachijō-in’s estates. The Japan Sea also connected the southern and northern parts of the archipelago. Late Heian and early Kamakura nyoin, as seen in the cases of Hachijō-in and Kōkamon-in, owned many estates in the San’in and Hokuriku regions, along the Japan Sea side of Honshu. Thus their estates often stood at convenient locations for river and sea transportation within Izumo, Hōki, Kaga, and Noto Provinces. For example, Hachijō-in owned Sada Shrine, Kimachi Estate, and Yamuya Estate in Izumo, which were all close to the Japan Sea coast. Kōkamon-in also owned two large-scale estates—Ōe Estate and Masuda Estate—in Iwami, encompassing a large river flowing into the Japan Sea (see plate 24 and figure 12.6).58 Previous studies argue that Japan Sea trading routes fell apart by the mid-twelfth century due to the fall of the Northern Song Dynasty and restrictions put on foreign trade in Wakasa by the provincial gover- nor, but I disagree.59 When the power of the Song was radically decen- tralized, trade might have actually become easier due to reductions in regulations and tariffs. Also, records of officially unsanctioned trade are likely to be limited or nonexistent. Because Japan Sea ports such as Tsu- ruga were centers of international trade, teeming with merchants from the continent and the Korean peninsula during the eleventh and early twelfth centuries, it is hard to believe that the ports completely ceased to function. Furthermore, the northeastward Tsushima ocean current helped merchants from the Korean peninsula and the Ryukyu and Nan- sei Islands to reach the Japan Sea coastlines. Of course, further research is necessary to fully determine whether foreign merchants ceased using the Japan Sea routes during the late Heian period.

Kōkamon-in’s Wakayama Estate and Its Black Wares One of Kōkamon-in’s estates facing the Japan Sea—Wakayama Estate in Noto Province—provides new insight into how nyoin participated in twelfth-century trading networks as estate proprietors.60 Wakayama Estate comprised a vast area around the tip of the Noto Peninsula and produced stoneware that modern researchers call Suzuyaki.61 I have found no document directly explaining how Kōkamon-in supported Suzuyaki production, but archaeological evidence and some written sources suggest the nyoin’s involvement in the industry. Excavation reports of Suzuyaki kilns indicate that the industry began with the estab- Nyoin Power, Estates, and the Taira Influence 301 lishment of Wakayama Estate in the mid-twelfth century and ended with its decline in the early sixteenth century.62 Rising and falling along with the estate, it seems that the Suzuyaki industry reflected both the needs of estate proprietors and their ability to support production.63 As the first supreme proprietor of Wakayama Estate, Kōkamon-in must have played a crucial role in assisting Suzuyaki makers to start kiln operations and expand the industry at its earliest stage. Furthermore, archaeological evidence indicates that in terms of its distribution, Suzuyaki ware was used all over the Hokuriku area. It was also in high demand in other areas such as the eastern coastline of the Seto Inland Sea, the northern edge of Honshu, and the southern part of Hokkaido.64 For example, Suzu-style cylindrical vessels containing copied sutras and Suzu-style urns for cremation have been excavated in the Tsugaru region of the northeast (today’s Aomori Prefecture). Archaeologists have dated these shards to the twelfth century.65 And since they have found no late Heian and Kamakura kilns in the Tsugaru region, it is possible that these sutra vessels and urns were traded from Wakayama Estate.66 The close tie between Suzu and eastern Harima (Tōban) wares also suggests a possible route along which Wakayama Estate residents interacted with the other side of Honshu. Suzuyaki specialists Yoshioka Yasu nobu and Moriuchi Shūzō argue that the Suzuyaki techniques were actually brought from the Tōban region on the Seto Inland coast. The argument goes as follows. First, Buddhist stoneware reliefs (senbutsu) excavated from Shingū in present Tatsuno City, Hyōgo Prefecture (i.e., the Tōban region), were made from the same materials as those excavated from Suzu in Noto. A wooden mold used to shape the Shingū reliefs was identical to those for the Suzuyaki reliefs, and chemical analysis shows that the clay used to bake the Shingū reliefs was likely made from Suzu soil.67 Second, because Suzuyaki craftsmen used decorating and baking techniques similar to those used for eastern Harima ceramics, eastern Harima craftsmen seem to have initiated the Suzu style.68 Technological and artistic exchanges between Tōban and Suzu craftsmen likely went both ways, but the influence of the former over the latter was undeniable because the Tōban kilns had already been active during the tenth century and were further developed well into the twelfth century. Moriuchi Shūzō claims that the Tōban techniques were transferred along the route from Harima through Tanba to Echizen and then influenced Suzuyaki development. Based on his analysis of the geographical 302 Power, Space, and Trade features connecting the Suzu and Tōban areas, Moriuchi points out the existence of a dividing ridge, that of the Yura River (flowing into the Japan Sea) and the Kako River (flowing into the Seto Inland Sea). Given this watershed one hundred meters above sea level, early medieval people and goods could travel between the Japan Sea and the Seto Inland Sea without crossing the high mountains. The Suzu and Tōban locals probably used this route to exchange knowledge and materials. As a result, Kōkamon-in’s Wakayama residents played crucial roles in initiating Suzuyaki production and creating its distribution networks. We have already learned that Taikenmon-in’s dependents traveled across provinces and transported annual rent or dues. Kōkamon-in’s dependents likely also exchanged knowledge and skills, expanding Suzuyaki culture from the other side of the archipelago. Kōkamon-in’s estate officials would have assisted Suzuyaki production as this local industry grew. Archaeological reports indicate that a number of Suzuyaki wares traveled across the Japan Sea throughout medieval times. Late Heian and Kamakura boats, retrieved from the sea- bed within twelve kilometers from the Noto Peninsula coastline, carried Suzuyaki pots, bottles, and vases. Early Kamakura evidence indicates that Suzuyaki traders headed off to different provinces such as Kaga, Echizen, and Echigo, carrying smaller Suzuyaki containers inside larger ones.69 In order to satisfy local and interregional demands, Wakayama Estate produced quite a few Suzuyaki wares. Such large-scale manufacturing processes required the support and supervision of local notables who would also have served as estate officials. Medieval managing proprietors and estate custodians who resided in the capital regularly visited or sent deputies to the estate, dealt with local problems, and checked productivity.70 Kōkamon-in’s Wakayama Estate custodian must have informed her about the growing Suzuyaki industry and asked her for patronage. During the visit of such custodians, local officials and residents needed to support them by throwing welcome and farewell parties, preparing accommodations and necessary materials for their stay, and covering the costs for their return trip to the capital.71 The custodian had opportunities to receive Suzuyaki wares and trade them on the way or brought them back to Kyoto as gifts. Estate officials, residents, and dependents thus supported Suzuyaki production in different ways. Previous studies, however, claim that estate proprietors in Kyoto never requested Suzuyaki as annual tribute, based on the following evidence.72 First, few Suzuyaki pieces have been excavated in Kyoto. We have only one Nyoin Power, Estates, and the Taira Influence 303 example of early medieval Suzuyaki found at the Heian capital site, a late twelfth-century pot excavated from the remains of Prince Mochihito’s residence.73 Second, the Tōkai and Seto Inland Sea regions also produced stoneware. Prior scholarship therefore argues that estate proprietors must not have commissioned heavy dishes from the Hokuriku region. Despite these arguments, I nevertheless see the possibility that Kōkamon-in might well have requested Suzuyaki ware from Wakayama Estate. To begin, Kōkamon-in’s contemporary, Sen’yōmon-in, annually received ceramics and cooking appliances as miscellaneous dues from her estates in Hokuriku. According to her estate list, although Sen’yōmon-in previously ordered her estates in Tōkai to submit jars and bowls, she later exempted them from paying such dues. In contrast, she requested her estates in Hokuriku to submit dishes and cooking implements.74 Sen’yōmon-in’s example refutes the argument that medieval estate proprietors did not systematically order their Hokuriku estates to submit ceramics. It is also important to note that Sen’yōmon-in annually ordered only a few pots and bowls (nabe and hachi). Most common Suzuyaki items excavated from the Hokuriku area were pots, bowls, and jars. If Kōkamon-in requested Wakayama Estate to send a limited amount of Suzuyaki each year, it would be difficult to find their remains in large numbers at any given site. Second, Kōkamon-in’s half brother Kujō (Fujiwara) Kanezane recorded that at the memorial service held on the twenty-seventh day after her death in 1181, he prepared “a blackish earthen cup with a long stem” (kuroki takatsuki no doki) rather than “a black-lacquered dish (koki).”75 His comment that he chose the black earthen cup in place of a black-lacquered dish strengthens the possibility that the cup was Suzuyaki.76 Suzuyaki specialist Yoshioka Yasunobu claims that Suzuyaki makers produced sake bottles by emulating lacquerware.77 Supporting this view, Hirata Tenshū assumes that a twelfth-century sake bottle excavated from Yōzenji, a Buddhist temple in Suzu, was an imitation of lacquerware. This temple was in the area that was once part of Wakayama Estate. The estate’s twelfth-century Kamegatan kilns also produced “dark black and hard” stoneware that was like lacquerware.78 Kane- zane and his son Yoshimichi (who became Kōkamon-in’s adopted son) closely served the nyoin, and as a result they inherited her properties including Wakayama Estate. Wakayama Estate officials, therefore, likely sent Suzuyaki to these new owners if the nyoin did not already own the cup before her death. 304 Power, Space, and Trade

In addition to its relationship with Kōkamon-in, the Suzuyaki industry developed close ties with local religious authorities at the temple Hōjūji and the regional religious center, Hakusan Shrine.79 Since these were branches of the great religious centers, Mount Kōya and Mount Hiei respectively, Suzuyaki may well have been brought by shrine associates to the capital as specialty goods. Few archaeological records show that Suzuyaki were traded on a large scale in Kyoto, however. Even if Suzuyaki seems not to have become a large commercial item in Kyoto, Kōkamon-in gained economic benefits from the growing industry on Wakayama Estate. Excavations of the Jike Kurobatake kiln site in the mid-1990s indicated that thirteen kilns were in operation from the mid-twelfth through mid-thirteenth centuries. This discovery challenges the previous notion that a small group of workers operated only one kiln at a time.80 Since Suzuyaki production started on a relatively large scale and continued to expand in the following centuries, some substantial support consolidated its industrial foundation during Kōkamon-in’s time. It is doubtful that estate proprietors in Kyoto knew nothing about such an important and well-supported project. Wakayama Estate residents did not send a large number of Suzuyaki wares to Kyoto, but they could have exchanged Suzuyaki to obtain other items that Kōkamon-in requested as estate dues. Suzuyaki became a popular commodity in the Hokuriku and northeastern Honshu regions. Archaeological reports show that Suzuyaki wares made their way even to Hiraizumi in Ōshū, which was the best producing area for alluvial gold.81 Traders who brought Suzuyaki to Ōshū must have received something in return, presumably gold. Contemporaries used gold to support different events such as seasonal refurbishments of their mansions (koromogae) and religious ceremonies.82 For example, royals and aristocrats often provided gold as offerings at Buddhist events. Since Kōkamon-in held a number of Buddhist ceremonies after she became a nyoin, she must have needed gold and would have obtained it by trading goods from her estates.83 Furthermore, as Yoshioka points out, Suzuyaki incorporated styles and techniques from multiple East Asian cultural zones such as the Ryukyu Islands and the continent. For example, tree-branch patterns inscribed on Suzuyaki show influence from the Korean peninsula. Suzu makers and traders actively interacted with broader cultural networks when developing the Suzu industry on Wakayama Estate. This shows that the world of Kōkamon-in was not confined to her capital residence. By supporting the economy of Wakayama Estate and Suzuyaki production for domes- Nyoin Power, Estates, and the Taira Influence 305 tic trade and international cultural exchange, Kōkamon-in expanded opportunities to enhance her religious, cultural, and economic power.

Empowerment through Strategic Alliances Although late Heian and early Kamakura nyoin owned estates deeply involved in trade and commercial activities, these royal ladies benefited fully from their economic and political advantages only when they successfully managed their land. As mentioned previously, newer scholarship emphasizes that nyoin maintained firm and independent control over their estates, but it downplays the difficulties these royal women faced in developing the material and human resources from their lands. His- torians have tended to assume that “independent” nyoin wielded strong power. Gomi Fumihiko, for instance, underscores Hachijō-in’s independent (and perhaps transcendent) position at court as the source of her political power. He argues that Hachijō-in held “sacred power” and increased her political influence through such power and the ability to function as a mediator due to her familial connections.84 Likewise, Yamada Akiko emphasizes Kōkamon-in’s independent position in the Fujiwara regental family as the source of her ability to support Kanezane’s lineage, Kujō Fujiwara. Yamada argues that among the children of Fuji- wara no Tadamichi, the nyoin was the only one who independently held his memorial services instead of jointly sponsoring such rituals with her siblings.85 Meanwhile, Banse Akemi argues that during this era a nyoin could independently expand her estates.86 Cooperating with landowners who asked for her protection, Hachijō-in included their lands within her private estates and used this tactic to enhance her political and economic power. These examples of nyoin’s authority and influence over their estates problematize the idea that powerful retired male sovereigns had absolute control over a nyoin’s lands. But it is also true that the nyoin often willingly relied on male family members, powerful warriors, or religious institutions, and they sometimes even gave up their rights and responsibilities in land management when they saw that such strategies were advantageous. Consider the case of Hachijō-in, who clearly chose not to exclude the influence of retired sovereigns. Rather, she willingly involved them in her estate affairs when disputes occurred between her properties and those of other powerful households. During the Genpei War (1180–1185), for instance, vassals of the Taira attempted to expand their influence over her estates.87 According to an order from Hachijō-in’s household office 306 Power, Space, and Trade in 1193, Taira no Shigehira’s retainer Narisuke had illegally acquired the position of gesu (local estate manager) at Kagato Estate in Bizen Prov- ince.88 Because of his “immoral behavior and delinquency in payment,” Hachijō-in asked Go-Shirakawa to issue a directive informing Narisuke that he would be stripped of his title if he refused to obey. This, however, neither denoted Go-Shirakawa’s control over Hachijō-in’s estate nor symbolized a loss of Hachijō-in’s power. For example, Go-Shirakawa consulted her when he made important political decisions, such as the time for choosing the next sovereign to replace Antoku. She also let him stay at her residence when he was in a precarious political situation after being released from house arrest by Kiyomori.89 This suggests that Hachijō-in and Go-Shirakawa politically and financially supported each other during the war. Although Go-Shirakawa could not entirely stop Narisuke’s illegal behavior, Hachijō-in enhanced her protective influence over the estate by cooperating with her male relative.90 Another good example can be found in the records of Hachijō- in’s Arakawa Estate in Kii Province. Long-lasting threats from adjacent estates—Tanaka Estate of the regental family and Yoshinaka Estate, whose proprietor was Hōjōji—occasionally involved armed conflicts at the border of Arakawa.91 Beginning in the 1130s, just after Hachijō-in’s father Toba, a retired sovereign, established Arakawa Estate, the Tanaka estate custodian Satō Nakakiyo attempted to seize a portion of it. And with Toba’s death in 1156, Yoshinaka Estate officials also intruded on Arakawa’s borders.92 During the height of these incursions—and following the recent death of her mother Bifukumon-in—Hachijō-in inherited Arakawa in 1160. Protecting it from the aggression of Nakakiyo, who was from a warrior family with connections at court, was a challenging task.93 And in addition to Nakakiyo’s military threat, assaults on Arakawa Estate by the provincial governor, Minamoto no Tamenaga, were also a serious danger. On 1161/10/19, Tamenaga led a large military force into the estate, arrested residents, extorted various resources, and burned religious buildings and residences.94 Although the court sent emissaries to investigate this incident, Tamenaga and Nakakiyo bribed them and then removed the markers that had defined long-established estate borders.95 Hachijō-in sought various ways to resolve this unfavorable situation. She first attempted to entrust estate affairs at Arakawa to another powerful authority, Kongōbuji of Mount Kōya. In 1159 her mother Bifukumon-in had already commended the estate to this regionally influential Buddhist monastery. In contrast to Bifukumon-in, however, who never gave up con- Nyoin Power, Estates, and the Taira Influence 307 trol over the estate even after the commendation, Hachijō-in was more than happy to delegate control of the estate to Kongōbuji.96 We know this because when Kii Province needed to pay a construction tax to repair Hinokuma and Kunikakasu Shrines in 1170, the retired sovereign Go-Shi- rakawa sent his official to discuss levying this tax on Arakawa Estate with Hachijō-in. She responded that “after commending the estate to Mount Kōya, [I] had absolutely no involvement in its management.”97 Nonetheless, Hachijō-in actually had been involved in estate affairs in the early 1160s. When Tamenaga and his deputy collaborated with Tanaka Estate and attacked Arakawa in 1161 and 1162, Hachijō-in sent her emissaries to conduct an on-site investigation of the quarrel between the two properties.98 If this was the case, why did she decide to with- draw from management of the estate in later years? One possible reason was that it was ultimately more harmful than beneficial for her to get involved in such hostile circumstances. As mentioned above, Nakakiyo of Tanaka Estate was from a military family, and the provincial governor Tamenaga, who was not at all cooperative, also wielded great military power, being a close retainer of Kiyomori. Seeing the rising influence of Kiyomori and his family through marital politics from the early 1160s, when Kiyomori’s sister-in-law Jishi married Go-Shirakawa and gave birth to his son, Hachijō-in may have sensed that the Arakawa matter would further deteriorate. She had many other estates to look after, and it would have been wise to let Kongōbuji take charge of Arakawa Estate problems rather than dealing with them directly. In any event, the case of Arakawa Estate shows that Hachijō-in did not always try to maintain firm control over her estates; the degree of her involvement in estate management changed depending on her interests and the prevailing situation. By maintaining good terms with different political actors, Hachijō-in established a wide circle of political alliances. She was politically astute and willing to cooperate even with her own enemies until she had a chance to strike back. Specifically, Hachijō-in’s cooperation with Kiyo- mori helped her sustain political power and control over her estates. Con- sider the instance of 1173 when Kiyomori’s grandson Taira no Sukemori prevented Hachijō-in’s officials from expanding the Ippon Royal Grant Fields in Echizen. Hachijō-in’s mother Bifukumon-in had been the provincial proprietor of Echizen, but right after her death the Taira took the province under their control.99 And in 1179 came Kiyomori’s well-known coup: he took the unprecedented action of marching into Kyoto with all his warriors, removing high-ranked courtiers from their posts, and 308 Power, Space, and Trade depriving Go-Shirakawa of political power. Although Hachijō-in likely felt animosity toward Kiyomori, she cooperated with him. For example, when Kiyomori’s daughter Tokushi (1155–1213; became Kenreimon-in in 1181) gave birth to the future Antoku Tennō, Hachijō-in sent clothes for the newborn on the Seventh-day Birth Celebration.100 Hachijō-in also offered her residence to Kiyomori’s son-in-law, Fujiwara no Motomichi, while secretly helping Prince Mochihito to rebel against the Taira.101 By cooperating with the relatives of Kiyomori, Hachijō-in not only concealed her support for Prince Mochihito but also gained support from the retired sovereign Takakura. In 1180 Arakawa Estate was still struggling to prevent attacks from Tanaka Estate and Yoshinaka Estate. Although Hachijō-in claimed to be willing to delegate Arakawa affairs to Kongōbuji, she needed to protect the estate that was used to support rituals that consoled the spirits of her deceased parents. Go-Shirakawa, whom Hachijō-in asked for help in other cases, had been under house arrest by Kiyomori since 1179.102 In 1180/12, the new retired sovereign (Takakura) ordered incursions at Arakawa to cease.103 Takakura, a son- in-law of Kiyomori, seems to have been more effective in this matter than Go-Shirakawa had been. Because of his close ties with the Taira, Takakura had more influence over Kiyomori’s retainer Tamenaga.104 To understand how economic growth on her estates affected Hachijō- in’s ability to accomplish her goals, we need to analyze the mechanism by which medieval estate proprietors received necessary materials and services in Kyoto. They wielded power not only through collecting estate dues on a regular basis but also through making extraordinary demands outside of predetermined agreements. As an example, Fujiwara no Teika and his family closely served Hachijō-in while managing Yoshi- tomi Estate as her custodian. According to Murayama Shūichi, Teika and his family had commended this estate to either Hachijō-in or her mother in exchange for support whenever their rights were challenged by proprietors or managers of neighboring holdings.105 Later in 1196, Hachijō-in sponsored a ceremony to celebrate her adopted daughter’s promotion to quasi-queen-consort and used her relationship with Teika to levy “various dues” over Yoshitomi Estate to support the ceremony.106 Similarly, earlier in 1143, one of Kōkamon-in’s officials, Minamoto no Suekane, was appointed custodian of Wakayama Estate after he commended it to her with the expectation of receiving her support and protection.107 In order to guarantee timely receipts of extra goods and services, a nyoin had to provide her officials with adequate rights and Nyoin Power, Estates, and the Taira Influence 309 responsibilities to manage her estates and to ensure their own economic prosperity.108 Although no specific sources exist, Hachijō-in’s case suggests that Kōkamon-in may well have used Suekane and Wakayama Estates to support her extraordinary needs upon request. Compared to Hachijō-in, however, Kōkamon-in had weaker connections with powerful male royals and her natal family heads when her estates came under attack by other powerful households and their associates. Although previous scholarship underscores that Kōkamon-in was “independent” from the regental family heads, she sometimes suffered because of that independence. Her half-brother and chosen heir Kujō Kanezane served Kōkamon-in as her trusted follower, but not all her relatives were so trustworthy. For example, her other brother Fujiwara no Motofusa was a potential threat to her planned legacy. Just after her death, for instance, he questioned her decision regarding the heir to her properties. Likely sensing the possibility that this situation should arise, two and a half months before her death Kōkamon-in asked Go-Shirakawa to witness her will.109 This was a rare occasion when she allowed a retired sovereign to engage in her estate affairs. Despite her proactive measure, however, Motofusa was able to convince Go-Shirakawa to order Kane- zane to submit a copy of Kōkamon-in’s testament for an evaluation of its credibility. Calling this incident her “disgrace after demise,” Kanezane was unable to hide his disappointment with the outcome. In fact when Kōkamon-in was still alive, the Taira were even successful in taking lands away from her. On the twenty-seventh day of the first month of 1181, Kanezane expressed his anger and disappointment in witnessing “unlawful violence” committed at his sister’s estates.110 That year Kōkamon-in lost some of her properties on Ninth Avenue because they were near a newly built temple belonging to Kiyomori’s son Mun- emori. Kiyomori and Munemori were “planning to live at the eastern side of this temple and considered the adjacent areas extremely important.”111 Close to both her main residence and the commercial center within the capital, these properties must have played an important role in her daily life. In Kōkamon-in’s case, we cannot tell whether she ever regained her lost estates. In contrast to Hachijō-in, who relied on all sorts of aid, we do not know if Kōkamon-in asked other power holders—retired sovereigns, the regental family heads, or religious institutions—to assist her in protecting her properties. Kōkamon-in’s example indicates that a nyoin faced greater challenges in protecting her estates 310 Power, Space, and Trade when her enemies were other powerful figures and that at such times independence could make the situation worse. This does not mean, however, that Kōkamon-in failed in managing her estates; far from it. She maintained estates such as Wakayama and Misaki and supported their production and trading activities. She was capable of collecting human and material resources to build her residential palaces in a relatively short time. She lost her main residence to fires three times between 1174 and 1181, but in all these cases, she was able to rebuild within seven to ten months.112 The residence of a late Heian nyoin had to be grand enough to reflect her authority and to accommodate her attendants and officials. Her ability to build large- scale accommodations with dispatch indicates that her lands gave her more than an adequate livelihood.

All three nyoin discussed in this chapter were supreme proprietors of estates that were advantageously located for transportation between the Kinai region and more distant provinces. By incorporating the study of nyoin and their estates into the discussion of domestic and international commerce, we can begin to appreciate the increasingly complex trade networks—as well as the politics of factions and family disputes—of later Heian times. In order to maximize their chances to amass economic and political power at a time of increasing violence and warfare, late twelfth-century nyoin tried different strategies. Hachijō-in maintained a delicate balance among political factions including those of her potential enemies, while securing powerful supporters who would assist her with her estates in an emergency. She did not hesitate to ask retired sovereigns and religious institutions for help in fending off threats to her land from other power holders. In Jōsaimon-in’s case, she avoided any confrontations with the Taira when her royal siblings began moving against Kiyomori’s family from the late 1170s. By maintaining ties with Kiyomori’s kinswomen and having estates in locations beneficial to transportation, Jōsaimon-in seems to have ensured her own prosperity. When she lost her residence within a Buddhist temple, Hōkongō’in, to fire in 1181, she moved in with her brother Go-Shirakawa and stayed there for several years.113 This was different from Kōkamon-in, who rebuilt her residence soon after each time it was lost to fire. In comparison to Hachijō-in and Jōsaimon-in, Kōkamon-in was more independent from other power holders. Her independence, however, cost her the Ninth Avenue properties when the Taira decided to appropriate them. She faced a greater challenge in establishing alliances with Nyoin Power, Estates, and the Taira Influence 311 other powerful authorities, perhaps because she was a childless widow who had lost her father-regent in 1164. Jōsaimon-in and Hachijō-in, however, never married and had lost their parents even earlier. These nyoin used different strategies due to their individual circumstances, such as Kōkamon-in being a royal wife from the regental family or Hachijō-in and Jōsaimon-in being unmarried royal daughters. It is important to examine how the life cycles of nyoin—age, marital status, having children, or taking a tonsure, etc.—affected the degree to which nyoin maintained their independence. Such examination helps us understand why Kōkamon-in did or could not depend more on her natal family head and other royals. Independence was not always a negative factor for nyoin power, but dependence served as a strong weapon to protect estates in times of crisis. This study focuses on nyoin’s experiences, but it nonetheless helps us understand early medieval estate management and the estate system itself more broadly. In contrast to the stereotypical image of estates as rice-yielding lands, this chapter has shown that medieval estates supplied a diversity of resources including lumber, gold, and ceramics. As supreme estate proprietors, nyoin gained opportunities to utilize and participate in transportation networks, local industries, and human resources. Estates were clearly important tools in medieval society to obtain wealth, political influence, and military power. Nyoin, however, could not always gain the land-based power that they needed, nor could they always do so as quickly as necessary. Such challenges were not unique to royal women; other supreme proprietors would have shared common problems with nyoin in obtaining resources from their estates, and they would have used similar strategies to cope with these problems. To further study the relationship between power and authority, it is important to compare nyoin estate management with that of other estate proprietors, including royal males and regents, and to explore possible gender disparities in techniques for managing estates and gaining land-based power.

Notes

1. Minamoto no Yukiie, an uncle of Minamoto no Yoritomo, held great military power. In 1181, as the leader of warriors from Mikami and Owari, he fought against Taira no Kiyomori’s son (Shigehira) and grandson (Koremori). 2. “Taira” here refers to the Ise Taira family, descendants of Taira no Masamori. 3. Azuma kagami (Mirror of the east) 1976, 1:2–3, 1180/4/9. 312 Power, Space, and Trade

4. Hikoyoshi Mieko claims that Hachijō-in was behind the rebellion based on the fact that her kurōdo, Yukiie, was given the important task of nurturing it. See Hikoyoshi 1973. 5. Gyokuyō (Jeweled leaves), 1:159, Jishō 4 (1180)/5/30. 6. As a sister or aunt of the monarch, an unmarried royal daughter who received the status of junbo often strengthened his lineage. See Kuriyama 2001. 7. For further information on nyoin, see Hashimoto 1986; Nomura 1989a, 1992, and 2006; Takamatsu 2004. 8. See the introduction and chapter 1 in this volume. 9. Nagahara 1975, 279. 10. Ibid., 280. 11. Nagahara investigated the situation of the Kawabara area, which was only a part of Ōta Estate. He made this conclusion by comparing the cloth sent from this part to the supreme proprietor with rice sent to the estate custodian. 12. For a summary of this scholarship, see Nomura 2006, 15; Banse 1993, 47. 13. For example, see Hashimoto 1986; Nomura 1989a and 1992. 14. For example, see Gomi 1990; Banse 1993; and Nagai 2005. 15. Noguchi 2005. Kondō Shigekazu also claims that late Heian and Kamakura nyoin often inherited estates supporting Buddhist temples and took responsibility for holding religious ceremonies. See Kondō 1992. 16. See Noguchi 2002; Watanabe 2010. 17. Watanabe 2010. 18. Teishitsu Rinya-kyoku 1937, 135. 19. For further geographical and historical information on these areas, see histories compiled by local governmental units. They explain how rivers around the lake had played important roles in transporting goods and people since premodern times. For example, see Shiga-ken 1927–1928, vol. 1; Ōtsu- shi Shiritsu Kyōikukai 1911; Ōtsu-shi 1963, 1978, and 1979. Also see Nihon rekishi chimei taikei 26:133, 212–213, on Matsui Village and Ichinobe Village. 20. Teishitsu Rinya-kyoku 1937, 136. 21. Takahashi 2011. 22. For example, both the Hino River (adjacent to Jōsaimon-in’s Funaki Estate) and the Mano River (running across her Ikadachi Estate) ran into the lake. These rivers are large enough to be defined as “first-grade rivers” under the modern Japanese river law. Although characteristics of these rivers could have changed over time, I assume that the Hino and Mano Rivers were at least close to Funaki and Ikadachi Estates in the premodern period and that they were large enough to serve as a transportation tool between the estates and Lake Biwa. Nyoin Power, Estates, and the Taira Influence 313

23. Some exported items such as alluvial gold must have been traded at Lake Biwa because transporters from northern and eastern parts of the archipelago used the lake to get to Kyoto and other western regions. 24. Matsuyama 2003, 96–99. 25. Nagare-ga-oka, just west of the area where the provincial Mining Office existed, is still rich in copper ore. See Nihon rekishi chimei taikei 26:77, on Dezu Village. 26. For a discussion of shrine associates, see Toda 2006. 27. The Chōshūki entry of Hōen 1 (1135)/8/20 suggests a close relationship between Ikadachi Estate and the Mudōji by recording that a monk of the temple requested tax exemptions for the estate. Chōshūki (Record of a long autumn) 1975, 2:301. Also, a late medieval source called Hiesha shintō himitsuki (The secret record of Hie Shrine) explains that the mother of Seiwa Tennō (830–880) commended the estate to the Mudōji. See Nihon rekishi chimei taikei 25:236–237. 28. Wakita 1969. For various activities of shrine associates, see Amino 2008a. 29. Teishitsu Rinya-kyoku 1937, 136. 30. Upon the request of the Upper Kamo Shrine in 1184, Go-Shirakawa and Minamoto no Yoritomo issued orders to suppress violence committed by warriors and allow the shrine estates to submit dues to the shrine. These orders listed Funaki Estate as one of the properties of the Upper Kamo Shrine. See HI 8:3128 (# 4155). 31. Nihon rekishi chimei taikei 25:593–594, on Funaki Estate, also suggests this possibility. 32. For example, Jōsaimon-in inherited Kurobe no Mikuriya, Kōshi no Mikuriya, and Kukiimafuku no Mikuriya. 33. See Miura et al. 1977, 274. Nihon rekishi chimei taikei 12:787–788 also explains that the deities that Tamasaki Shrine worshiped were Yamatotakeru and Jinmu Tennō’s mother, the Goddess Tamayori. 34. Yamada 2003, Kuriyama 2001. 35. For a good discussion see Oka 2010, 203–204. 36. Chigyōkokushu: According to the glossary in Hall and Mass 1974 (255), a chigyōkoku was a province held in “proprietorship” by a ranking court noble or religious institution during late Heian and Kamakura times. Provincial proprietors (chigyōkokushu) enjoyed the right to nominate governors, as well as the right to receive income from the cultivated “public domain.” For the Taira’s chigyōkokushu in the late 1170s to early 1180s, see “Heike-ryō to chigyōkoku,” in Gomi 1991a, 125–138. Also see Fukui-ken shi 1993–1996, vol. 2. Motoki Yasuo argues that the Taira maintained tenuous control over provinces and local warriors. See Motoki 1996, 357–358. 314 Power, Space, and Trade

37. Tamakiwaru 1994, 257–263. 38. As for male officials, not so many Taira kinsmen or their followers served at Jōsaimon-in’s administrative headquarters. Her officials and close attendants rather had close familial relations with those who served other nyoin such as Hachijō-in and Takamatsu-in (Hachijō-in’s full sister, 1141–1171). See Sonpi bunmyaku (Lin- eages of high and low), vols. 1–3. 39. The later section of this chapter provides a close examination of Hachijō- in’s Arakawa Estate. 40. See Naka 1974, 161. Also see Joan Piggott’s chapter 7 in this volume. 41. The writer of Taikenmon-in’s order referred to the historical Izumi Land- ing to emphasize that her lumber had never been taken for toll. 42. HI 5:1962 (# 2324). 43. Ibid. 44. See HI 5:1945–1951 (# 2303). The estate was, of course, Kuroda Estate, discussed by Joan Piggott in chapter 7 of this volume. The “new estate” was Kuroda New Estate. 45. See HI 5:1954–1955 (# 2306). 46. Her estate list, which was recorded a decade after her death (1221), contained 221 entries that were spread all across the archipelago. The 1176 estate list survived in two different locations: the first half as part of the Kōzanji komonjo and the second half as part of the Yamashinake komonjo. For the printed version of the second half (Yamashinake komonjo), see HI 10:3904–3905 (# 5060). Other document collections such as Kogake monjo, Kujōke monjo, Saionjike monjo, Anrakujuin komonjo, Tōji hyakugō monjo, and Jingoji monjo also provide further information about Hachijō-in’s estates. Some of these are accessible as printed versions. For example, see HI 6:2124–2127 (# 2519) and 2470–2471 (# 3029); and HI 8:3126–3127 (#s 4151 & 4153). Secondary sources presenting useful information about Hachijō-in’s estates are Ishii 1988 and 2005, 251–401; Gomi 1991b and 1991c; Endō 2007. 47. Wakita 1969; Noguchi 2010, 187–190. 48. Noguchi 2010, 189. 49. See Hyōhanki, Nin’an 2 (1167)/5/23. 50. See Kokushi daijiten 1985, 5:297–298. 51. Sukigara 1999; Ōmura 2000; Noguchi 2002, 48–50. 52. Ōsone and Kubota 2000, 339. Also see Noguchi 2002, 48. 53. Noguchi 2002, 49–50. 54. According to Kokushi daijiten 5:297–298, moreover, her estate list included her right to collect human resources from these three provinces as guards and laborers. These laborers, who were called ōban toneri (attendants chosen through a grand rotation system), received pieces of land to support themselves. For studies on ōban toneri, see Watanabe 1970; Shimizu 1974; Kojima 1932; and Maki 1932. Nyoin Power, Estates, and the Taira Influence 315

55. Noguchi 2002, 45–46. 56. Noguchi 2002, 46. This shell is the source for mother-of-pearl inlay. 57. Kōkamon-in supported Bifukumon-in’s son, the monarch Konoe, as his equivalent mother. See Nyoinki 1977, 40. 58. Kōkamon-in also owned Hayashigi Estate in Izumo, which became known as an important strategic transportation spot by the Sengoku Period. It is important to examine the roles of later nyoin—Gishūmon-in (1173–1238) and Muromachi-in (1228–1300)—who inherited this estate. Due to a lack of sources, we are unable to determine when Hachijō-in established Yamuya Estate, but it is included in the estate group named after her. See Hayashigi-no-shō in Nihon rekishi chimei taikei 33:403. 59. The source on Wakasa survived because the back of the paper was recycled to record Tōdaiwajōtōseiden. See Yamauchi 2003; Fukui-ken shi 1993–1996, vol. 2, 133–136. 60. See Kunaichō Shoryōbu 1972, 45–46; Ishikawa Kenritsu Rekishi Hakubut- sukan 2000. 61. Wakayama Estate was in present-day Suzu City. In 1961, local archaeologists met with a Japanese antique specialist from Tokyo and decided to name the stoneware after the city. See Hirata 2002, 12. 62. Nishiyama 1997; Yoshioka 2010. 63. Several historians and archaeologists agree that powerful royals such as retired monarchs and nyoin were behind the late Heian development of kiln productions. See Zauhō Kankōkai and Narasaki 1977, 59–60, 152; Hotate 1984, 188–189. 64. Kentani 1981, 30; Suzu-shi Kyōiku Iinkai 2010; Hirata 2002. 65. Sekine 2004. 66. Sekine 2004, 72. 67. Moriuchi 2010, 4. 68. Moriuchi 2010, 4–5. Also see map #9 on page 10 in his article. 69. Nishiyama 1997; Sekine 2004. 70. The two terms, “managing proprietor (ryōke)” and “estate custodian (azukaridokoro),” could refer to the same person in medieval sources. For example, Hachijō-in’s female attendant, Ben-no-tsubone, appears in primary documents as both ryōke and azukaridokoro of Yoshii Estate. The interchangeable usage of these two terms occurred because within the layered relations in the estate system, the azukaridokoro was the person who “was entrusted” with a certain job by his or her superior. Ben-no-tsubone, who was managing Hachijō-in’s estate right under the supreme proprietor, should be a managing proprietor, but she was also named as estate custodian because she “was entrusted” the task by the nyoin. See “azukaridokoro” in Kokushi daijiten 1:206. 316 Power, Space, and Trade

71. Morimoto 2008. 72. Yoshioka 1994, 606, and 1989, 294. 73. Prince Mochihito possibly received Suzuyaki through his connection with his adoptive mother, Hachijō-in, who also owned an estate in Noto. For the information about the excavated pot, see Yoshioka 1994, 605–606. To learn more about the history of medieval ceramics excavated in the capital, see Hirose 1986 and Uno 1984. 74. See the list of miscellaneous dues supporting Sen’yōmon-in’s Rokujō mansion and the Chōkō Temple in Nihon Engyō Taikei Henshū Iinkai 1975, 29–82; Hyōgo-ken Shi Henshū Senmon Iinkai 1997, 535–561. 75. Kanezane prepared this cup for his son, Yoshitsune, who was an adopted son of Kōkamon-in. Kanezane explained that Yoshitsune’s cup should not be black- lacquered because Kōkamon-in was not his biological parent. See Gyokuyō 8:74 (Yōwa 1 (1181)/12/18). 76. Although successive regental family heads owned an estate called Kusuha- no-maki that also produced black wares, neither he nor Kōkamon-in held proprietorship over this estate before 1181. See Ujita 1991; Umekawa 1997. 77. Yoshioka 1990, 295–296. 78. Hirata 2002, 11. 79. Yoshioka 2010, 25. 80. See Ishikawa Kenritsu Maizō Bunka Sentā 1994. In addition to this report, the Suzu City education board and the Toyama University archaeological research group have published their reports on this kiln site. See Maekawa 1998. 81. Watanabe 2010; Furiya 2002, 24. 82. Koromogae literally means “changing wardrobes,” but it also refers to “changing furnishing materials” for the new season. Heian and Kamakura royals and aristocrats refurnished the interior décor of their households twice a year. 83. Kōkamon-in sponsored repentance rites, ceremonies of reciting the names of the buddhas, ceremonies of venerating Buddha relics, and memorial services for deceased family members, to name a few. For example, see Sankaiki 1:21 (Ninpei 2 (1152)/12/20); Hyōhanki 1:219 (Ninpei 3 (1153)/12/26), 2:105 (Hōgen 1 (1156)/5/15), 2:268 (Hōgen 2 (1157)/10/23); Gyokuyō, vol. 1: Kaō 2 (1170)/5/19, vol. 2: Shōan 2 (1172)/5/26, Shōan 3 (1172)/3/7, vol. 3: Angen 1 (1175)/2/19, vol. 5: Jishō 1 (1177)/2/27, vol. 6: Jishō 2 (1178)/10/6, and vol. 8: Yōwa 1 (1181)/10/27. 84. Despite his problematic characterization of female power as sacred, Gomi nonetheless challenges the older belief that retired sovereigns were dominant over nyoin. See Gomi 1990, 145, and 1991b. 85. Yamada 2005 and 2010. 86. Banse 1993, 49. Nyoin Power, Estates, and the Taira Influence 317

87. Banse 1993, 56–57. 88. KI 2:88–89 (# 687). 89. See Tamai 1954, 53–57; Gomi 1990, 115–116. 90. Hachijō-in later asked Minamoto no Yoritomo for help to solve the problem. See KI 2:88–89 (# 687). 91. KI 1:72–77 (# 108). See also Shinkō Kōya shunjū hennen shūroku (Revised version of the spring and autumn annals of Mount Kōya), 108 (entries of Ōhō 2 (1162)/4/15, 4/28, 5/24). 92. HI 6:2449–2451 (# 2979). 93. See Tanaka 1994, 147. Also see Sonpi bunmyaku 2001, 59:391. 94. HI 7:2582–2585 (#s 3235, 3236). 95. HI 7:2583 (#3235). 96. Hōkanshū 1952, 338–334 (#s 326, 327, and 328), contains Bifukumonin’s edicts showing her involvement in Arakawa Estate affairs. Also see Arakawa-no-shō in Nihon rekishi chimei taikei 31:243–244. 97. In other words, she made the Mount Kōya monastery handle all the matters related to the estate. See Hōkanshū 1952, 366 (# 352) and HI 7:2769 (# 3557). 98. Shinkō Kōya shunjū hennen shūroku 1982, 108 (Ōhō 2 (1162)/4/15, 4/28, and 5/24). See also KI 1:72–77 (# 108). 99. Satō 1993. 100. Gyokuyō, 1994–2002, 6:82 (Jishō 2 (1178)/11/12). To learn more about birth celebrations, see Kawai 2018. 101. An entry of Sekkan shō senge ruiju explains that just after the 1179 coup, Kiyomori asked Hachijō-in whether she could offer one of her houses to the newly appointed regent Motomichi. According to the precedent established by Fujiwara no Tadamichi, a new regent was to change his residence before making his first visit to a sovereign in order to show appreciation for the appointment (hatsuhaiga). Sekkan shō senge ruiju: This is a compilation of the conventions and ritual practices for the appointment of the regent position. The editor collected the information from courtier journals and arranged them according to individual regents. Included regents are Fujiwara no Tadazane, Fujiwara no Motozane, Fujiwara no Motomichi, Konoe Kanetsune, and Konoe Motohira. The date of compilation is unknown, but it seems not to have been long after the time in which the last regent, Konoe Motohira (1246–1268), is introduced. See DNS, vol. 3.26, 130–131. 102. Gyokuyō, 9:229 (Jishō 4 (1180)/12/18). 103. HI 8:3010 (# 3235). 104. Minamoto no Tamenaga acknowledged the order from Kiyomori’s administrative headquarters in 1164. HI 7:2609 (# 3285). 105. Murayama 1962, 197–202. 106. See Meigetsuki (Record of the full moon) 1:30 (Kenkyū 7 (1196)/4/15). 318 Power, Space, and Trade

107. See Kujōke monjo 1972, 2:45, doc. # 294; Ishida 1981. 108. To cite a later example, in 1231 Kitashirakawa-in (1173–1238, n. 1222) found that her chief of the year, Fujiwara no Ietoki, was incompetent in providing her with necessary resources when she was moving to her newly built mansion. Higuchi Kentarō (2008, 34) argues that the reason why Chikafusa failed to procure resources was because Kitashirakawa-in herself had failed to provide him with enough estates. Merely owning estates was not enough for a nyoin to fill her purse; she had to be able to move others to create wealth from her estates and bring it into her storehouse. See Minkeiki 1978, 2:233–234 (entry of Kanki 3 (1231)/2/17). 109. Gyokuyō 8:41, 63, 86 (Yōwa 1 (1181)/9/20, Yōwa 1 (1181)/12/5, and Juei 1 (1182)/2/9–11). 110. Gyokuyō 7:261–262 (Yōwa 1 (1181)/1/7). 111. Ibid., 261. 112. See Gyokuyō 5:55, 95, 106–107 (Jishō 1 (1177)/4/27 and Jishō 1 (1177)/6/5–6); 7:308–309, 317–318, 325 (Yōwa 1 (1181)/3/4–5, Yōwa 1 (1181)/3/21–22, and Yōwa 1 (1181)/4/26); 8:2, 53–54, 58 (Yōwa 1 (1181)/11/1–3 and Yōwa 1 (1181)/11/19). Also see Kikki 1:156, 167 (Jishō 5 (1181)3/3 and Jishō 5 (1181)3/21–22) (Zōho shiryō taisei vol. 29). 113. See Hyakurenshō, 106 (Jishō 5 (1181) 5/21). Also see Kikki 1:198, 237 (Jishō 5 (1181) 5/21 and Yōwa 1 (1181) 9/27) (Zōho shiryō taisei, vol. 29). Accord- ing to Kikki, Jōsaimon-in visited the neighborhood of her Hōkongō’in residence in 1183, two years after the fire. This indicates that she had not been living there. See Kikki 2:54 (Juei 2 (1183) 7/18) (Zōho shiryō taisei vol. 30). c h a p t e r 13

Networks of Wealth and Influence Spatial Power and the Estate Strategy of the Saionji Family in Early Medieval Japan

Rieko Kamei-Dyche

On the second day of the twelfth month in 1224, Saionji Kintsune (1171–1244) held a great banquet celebrating the completion of the massive villa he had constructed in Kitayama. The numerous luminaries in attendance were treated to an amazing sight, for Kintsune had transformed what had been just countryside into a site of luxurious splendor.1 High in the hills overlooking the capital, the villa consisted of a series of halls replete with beautiful images and pictures of Buddhist themes, surrounded by an elegant park with waterfalls and a giant arti- ficial lake. Unlike previous owners of the estate, Kintsune engaged in extensive landscaping and was the first to build a luxurious residence on the site. The villa sat on the location now occupied by the world-famous Golden Pavilion, the decorative pond in front of which was originally constructed by Kintsune. The luxurious villa and exquisitely landscaped surroundings, by virtue of their location from which one could peer over the capital as if at the refuge of an immortal, represented an expression in physical space of Saionji influence and prestige. Nor was this the family’s only villa: numerous residences and retreats dotted the landscape. To say that Kintsune lived a luxurious life at such residences would be a gross understatement. For instance, at his Suita villa, Kintsune, eager to enjoy the restorative waters from the hot spring in Arima, had the hot spring brought to him: during his time there, he had porters carry two hundred buckets of water from the spring to his villa, a distance of some twenty- five miles, on a daily basis.2

319 320 Power, Space, and Trade

Arima hot spring, in present-day Kobe, is known to be one of the three oldest in Japan, and due to its fame as a therapeutic spa for cur- ing disease, it had welcomed numerous guests since as early as the seventh century. Gotō Michiko points out that even during the Muromachi period few families had baths in their residences.3 The fact that more than a century earlier Kintsune not only enjoyed bathing at home but went to the extraordinary measure of having vast amounts of water regularly carried to him from his favorite hot spring speaks to how much wealth he had at his disposal. Even by the standards of the time this lifestyle was impressive, and it showcases the enormous economic resources commanded by leading courtier families like the Saionji. Such families reshaped the landscape to reflect and project their influence as they continued to exert power within the dual polity (Kyoto court and Kamakura shogunate) framework of the early medieval era (1192–c. 1333). The economic might of courtier families depended primarily on their estates and estate strategies, meaning the methods they deployed to acquire and manage estates to maximize their value, whether in terms of economic productivity from local resources or trade networks, or in terms of the cultural capital accrued from choice strategic or scenic locations.4 Shrewd families like the Saionji developed long-term strategies that developed their estates to provide multiple benefits. This chapter assesses the estates and estate strategies of the Saionji family by considering their varied economic and social roles, outlining how their estates facilitated the family’s wealthy lifestyle and prestige while serving as embodiments of their influence. I employ a macro- level approach to trace how the family acquired its estates, where these estates were, and what elements they had in common, in order to establish the key patterns involved in successful estate handling. The chapter emphasizes the importance of locating estates in space, revealing how estates functioned as parts of significant trade networks that reached both China and various parts of the Japanese archipelago. Trade played a crucial role in providing economic and cultural capital to families able to engage in it. In assessing this latter issue, I will examine not only what kinds of goods were traded but also how these goods were received in contemporary society and the particular roles courtier families played in the process. Finally, I also give considerable attention to estate management and the use of client families to maximize estates’ potential, which directly contributed to the wealth and influence of the Saionji family. Networks of Wealth and Influence 321

The Saionji Family The Saionji family originated as a subbranch of the Kan’in branch of the Fujiwara family in the late eleventh century and was originally known as the Ōmiya family.5 The family later took the name of their family temple, Saionji, built by the aforementioned Kintsune. It was under Kintsune’s stewardship that the family rose to prominence during the late twelfth century. This is an era once depicted as the twilight of the Kyoto court in the face of rising warrior authority, culminating in the formation of the Kamakura shogunate in 1192. More recent scholarship has revealed, however, the extent to which the court continued to exercise authority in the form of a power-sharing relationship with the shogunate, a system that persisted until the overthrow of the warrior government by Go-Daigo Tennō (1288–1339, r. 1318–1339) in 1333.6 Courtier families not only continued to exercise influence through the court, but in numerous cases also strove to take advantage of the new political order and the opportunities it offered. Men and women from courtier families possessed great economic, social, and cultural capital that appealed to warrior families looking to advance their status. Courtier families were recruited to provide tutelage in arts and scholarship, or practical skills like accountancy, to shogunal officials, further complicating the relations between the two power centers. In the case of the Saionji family, it gained a monopoly over the vital post of Kantō mōshitsugi—a liaison between the court and Kamakura—ensuring the Saionji a major role in the politics of the day. It also ensured them a place in Japanese historiography as one of the few courtier families to appear in the overwhelmingly warrior-focused studies of the early medieval era. Saionji influence was not, however, due only to their control over this choice position. The family proved immensely successful at wield- ing various strategies—“tools of authority” such as economic resources and social and cultural capital—to amass tremendous wealth and influence.7 Prominent members of the family—first and foremost Kintsune, renowned in contemporary accounts for his power and influence—were much more successful than other courtier families not because their strategies were unique but because of their proficiency in utilizing them. Naturally, one essential set of strategies pertained to the development of estates, upon which much of the family’s social and cultural capital endeavors depended and with which this study is concerned. 322 Power, Space, and Trade

Scholarship Estates and the estate system, as vital components of the medieval economic world, have been a perennial topic in the literature on medieval Japan. Only a relatively small part of this material is dedicated to the study of courtier estates, although such work has gradually increased, prompted, as Kanai Shizuka has observed, by the prominence given courtiers in Kuroda Toshio’s path-breaking kenmon theory and by the increasing availability of courtier-related documents.8 There have been studies of royal holdings, although mainly limited to the Hachijō-in and Chōkōdō estate clusters,9 and a range of studies of the estates held by the regental families (especially the Konoe and the Kujō).10 Scholar- ship on specific courtier families during the early medieval era includes Nakamura Naokatsu’s pioneering work on the Kajūji family,11 Naga- hara Keiji on Fujiwara no Teika (1162–1241),12 Usui Nobuyoshi and Sugawara Masako on the Yamashina,13 Okano Tomohiko on the Koga,14 Abe Takeshi on the Ōimikado,15 and most significantly for this project, Amino Yoshihiko on the Saionji.16 Amino, who reoriented the study of medieval Japan by his challenges to conventional approaches, did much to bring networks and heretofore marginal figures to the fore- front.17 His study of the Saionji estates, currently the only thorough treatment of the subject, represents one of the few studies of a courtier family’s holdings to consider the role of location and networks. In doing so, Amino points to a more effective way of collectively assessing a family’s holdings, and his work inspired this study. The current study seeks to build on Amino’s work with a more detailed assessment of the Saionji family’s estate strategies. To Amino’s argument that the Saionji’s holdings were clearly connected to trade routes, I add a consideration of the family’s development of spatial power based on strategies for acquiring estates and extracting maximum benefit from them. I also examine the vital role played by client families in estate management. In addition to the work of Amino, studies of the economy have shown increasing cognizance of medieval Japan as a networked society in which trade and distribution played key roles.18 Most of the scholarship on estates, however, focuses on transitions in the estate system or on analyses of particular estates. Little work has been done on how families developed estates as a set, rather than as separate entities. Few Networks of Wealth and Influence 323 scholars have focused on the locations of estates, even though courtier families clearly considered location to be of great importance when assessing the value of an estate. For early medieval courtier families, the ability to provide reliable access to, and project control over, a desirable location was a key concern. To understand the spatial power of estates, it is necessary to examine elements like trade, which played a key role in the economic base of medieval courtier society.

Spatial Power and Estates A single estate may have represented considerable economic or cultural value for its holder, but only when estates are understood collectively can their significance be fully grasped. This necessitates thinking about the location of estates in space: that is to say, how estates were situated in relation to each other and to adjacent trade and transportation routes. For example, viewed from the perspective of a maritime trade route, the true economic value of an otherwise undeveloped seacoast estate becomes readily apparent. In addition to the cultural capital generated from scenic locales, there was also the cultural and political capital that derived from control over nodes along domestic and foreign trade routes. Such networks of wealth and influence were essential concerns in early medieval courtier society. One way to handle the range of values associated with an estate is the concept I call “spatial power,” by which I mean control over geographic space, with all the attendant resources, wealth, and authority that such control brings. I favor the term “spatial power” over “geographic power” or “territorial power” because the latter two terms are too closely tied to the physical terrain and neglect the wealth and symbolic authority that are essential components of land and space. Fur- thermore, the ability to define space constitutes a significant form of power that can reproduce social hierarchies and assert status. This has been well established by sociologists concerned with the construction of social space as a stage for all social interaction, most famously associated with the concept of “dramaturgia” advanced by Erving Goffman in The Presentation of Self in Everyday Life.19 Goffman argued that everyday social interaction constituted a series of theatrical performances in which social actors play roles, complete with costumes, settings, and props. Social space is an essential component of performance. If we understand the social world as the sum of these interactions, then the 324 Power, Space, and Trade ability to convey authority or power through space is understandably significant. Historical treatments of space in Japan written in English also have much to offer on this theme, revealing connections between space and contemporaneous cultural and ideological frameworks. David Bialock, for example, shows how the delineation of social space was bound up with cultural and religious notions of space.20 My study, by employing the lens of spatial power as a means to examine the estate strategies of the Saionji family, seeks to encourage a more thorough understanding of space and its myriad implications within the world of early medieval Japan.

Estates and Wealth Accumulation Successful courtier families financed their lavish lifestyles and political advancement by making effective use of their estates. This necessitated acquiring a portfolio of estates through various means and managing them effectively. This section outlines the Saionji estates, the role of human resources in productively managing those estates, and, finally, one of the major investments of estate profits available to such powerful courtier families: trade with the continent.

Expanding a Family’s Landed Portfolio: Estates and Estate Acquisition As was the case with other powerful courtier families, the Saionji owned many estates across the realm. While a lack of documents pertaining to family holdings and inheritance frustrates attempts at a detailed study, it is still possible to reconstruct the ways the Saionji utilized their estates and the importance of the strategic location of those properties. Amino Yoshihiko’s research gives us a good starting point.21 Tracing the numbers of Saionji estates that are identifiable from surviving records—including estates they held temporarily—Amino noted a total of fifty-three, almost all of which were in western Japan.22 (For a map of the Saionji estates, see figure 13.1, which indicates all the estates the family is known to have owned, from Kintsune’s time up to and including the Muromachi period. For a listing of these estates, see table 13.1.)23 Networks of Wealth and Influence 325

Fig. 13.1. Saionji family estates. Template from Jeffrey P. Mass, ed., The Origins of Japan’s Medieval World, Stanford University Press, 1997. Originally drawn by Arnie Olds. Used with permission.

A significant number (thirty-three of the forty-seven in the west) were in the Kinai, Sanyō, Nankai, and Saikai regions.24 A smaller number were in the San’in and Hokuriku regions. Still other estates administered by the Saionji (as ryōke, estate proprietors) were landholdings that belonged to members of the royal family.25 This was not a pattern unique to the Saionji; many courtiers concentrated their estates in the west, as Nagahara Keiji observed when discussing the holdings of Fuji- wara no Teika.26 In particular, the Tobadono lands in Yamashiro Prov- ince, adjacent to the royal villa of the same name, contained numerous separate holdings.27 This pattern of land distribution, with many courtier families investing extensively in the region, was clearly connected to maritime transportation routes, as Amino has argued, a situation to which I will return shortly.28 Table 13.1 Landholdings of the Saionji family

Per Go-Daigo Per Saionji Province royal edict Sanekane Muromachi Name 1335/7/12 letter 1322/8/16 Period Other Region: Kinai Yamashiro Tobadono Kamishimo Katsura Shinmen Toyoda-no-shō Uji-Makishima Yodouoichi Mizuno-maki Koga-no-shō Ishihara- no-shō Shimomisu- no-shō Hatsukashi- no-shō Uji Makidono Katsura Shinmen Toba-no-shō Uda-no-shō Settsu Tomatsu-no-shō Tomatsu- Suitanishi-no-shō no-shō Kuradono Mizokui-no-shō Kawachi Shinkai-no-shō Kodakase-no- Shinkai-no-shō Ikeda-no-shō shō Ega-no-maki Shinkai-no- Fukuchi-no- shō * maki Ikeda-no-shō Takafuchi- no-gō Region: Tōkaidō Ise Hioki-no-shō Kurosaka Owari Kurota- no-shō Kitsuneana Table 13.1 (continued)

Per Go-Daigo Per Saionji Province royal edict Sanekane Muromachi Name 1335/7/12 letter 1322/8/16 Period Other Region: Tōkaidō Mikawa † Izu ‡ Suruga Oyakizu-no- Mikuriya Musashi Hangaya-no- Mikuriya Shimōsa Intō-no-shō Region: Tōsandō Ōmi Tanaka-no-shō Takashima- no-shō Ōkuni-jōge- no-shō Yamazaki-no- shō Mino Hidase-no-shō Hidase-no-shō Hachiyakita- Goudo-no-shō no-shō Shimokiri Shimotsuke Sano-no-shō Region: Hokurikudō Echizen Tomita-no-shō Wakasa§ Noto Hitoto-no-shō Ōizumi-minami- no-shō Region: San’indō Tanba Tokitsune-no-ho Tajima Hōki Inatsumi-no-shō Inaba Iwami Table 13.1 (continued)

Per Go-Daigo Per Saionji Province royal edict Sanekane Muromachi Name 1335/7/12 letter 1322/8/16 Period Other Region: Sanyōdō Harima Ōta-no-shō Goka-no-shō Ōta-no-shō Yokawashimo- Matsubara-nono-shō, shō Matsubara-no- shō Mimasaka Bizen Kayō Honjō Totori-no-shō Bitchū Oushi-no-shō Aki Nuta-no-shō Suō Yamashiro-no-shō Koga-no-shō Nagato Region: Nankaidō Kii Nakamura- Kishi-no-shō Kōnami-gō Hashimoto Kawabe Awa Tai-no-shō Ura-no-shō Sanuki Iyo ** Uwa-no-shō Uwa-no-shō Uma-no-shō Tosa Tamura-no-shō Region: Saikaidō Chikuzen Kusuhashi-no- Munakata shō Shrine Kanta-no-shō Bungo Anami-no-shō Uno-no- Hizen Mikuriya Higo Yamaga-no-shō Chart adapted from Amino 2008, 315. For information pertaining to individual estates, see Amino et al., 2005. * One gō (township) within Shinkai-no-shō † Mikawa was a proprietary province of Sanehira ‡ Izu was a proprietary province of Kinhira § Wakasa was a proprietary province of Saneuji ** Iyo was a hereditary proprietary province Networks of Wealth and Influence 329

Not all these estates were hereditary holdings. In addition to inheritance, an influential courtier family could at times acquire estates from other individuals or from families seeking to gain influence or curry favor. For example, the provincial governor of Bungo commended a holding called Anan to Kintsune. Gomi Fumihiko argues that this demonstrates that the provincial governor was subordinate to Kintsune.29 For Gomi, this case is reminiscent of how Ōta Estate in Bingo Province became a Taira holding due to the cooperation of a provincial gover- nor.30 As courtier families gained spatial power over an area, it was not uncommon for them to establish relationships like this that rewarded other officials with connections while further expanding their own spatial reach. Courtier families also actively sought, at times aggressively, to expand their landholdings. Saionji Kintsune’s near obsession with gaining Uwa Estate in Iyo Province is an example. Iyo had already been under Saionji control for generations as a hereditary proprietary province (chigyōkoku).31 Uwa Estate, however, had remained under the control of the Tachibana family, a situation Kintsune sought to rectify. The Azuma kagami (Mirror of the east), the official record of the Kamakura shogunate, shows that Kamakura saw no reason to take the estate from the Tachibana but eventually gave in to Kintsune’s relentless demands when the latter threatened in early 1236 to come to Kamakura and press the matter in person:

While [the shogunate] was not able to take either side, some time ago a letter from Kintsune concerning this matter arrived at the shogunate again. In the letter, he wrote that “if my request is not accepted, it will be like losing face in my old age. Now, I intend to come all the way down to Kamakura and make an appeal myself.” Since his coming down to Kamakura will cause much more trouble, the shogunate sent a messenger to [Kintsune’s] house manager, the novice Mutsu Dōbyū [Miyoshi Nagahira], saying that the land would become a holding of the Saionji. Katei 2 (1236/2/22)32

Uwa Estate was made a proprietary holding of Kintsune’s son Saneuji, then serving as minister of the right. It is not clearly stated why the shogunate was involved in this matter, but the Tachibana family had become retainers of Yoritomo33 and was under the administrative 330 Power, Space, and Trade system of the shogunate. Since Kintsune’s appeal drew on his relations with Kamakura, it would have been reasonable for the family head Tachibana no Kinnari, facing the loss of his stewardship, to bring the issue to the shogunate. As this case shows, courtier families needed to continually assert their rights to estates or risk losing them to other families aggressively expanding their estate portfolios. Even staunchly defending one’s ownership and going through proper legal channels, as the Tachibana did, was not enough to fend off powerful courtiers like Kintsune. A courtier family’s landholdings should thus be understood as a fluid entity, resulting not only from inheritance or rewards but also dependent on the strategies and will of the leading members of the family. Because estates represented an essential component of a family’s wealth and influence, families would go to considerable lengths to defend their interests. Kin tsune’s struggle for Uwa Estate is remarkable in this respect only for his stubborn attitude that resulted in a victory.

Human Resources in Estate Management and Wealth Accumulation In addition to acquiring estates through a variety of means, courtier families needed to cultivate human resources to manage those estates effectively. They frequently depended upon managers and staff, often drawn from other families with which they had forged a mutually beneficial (and hierarchical) relationship. In the case of the Saionji, the Miyoshi and the Tachibana (not the same family concerned with Uwa Estate above) families served this role. The Saionji family’s success at developing their economic foundation was no doubt partially due to the skills of these two client families. Most noteworthy among the Miyoshi was Nagahira (1168–1244), who served as Saionji Kintsune’s household manager and gained considerable prestige as a result. The Miyoshi family had long been known as a house of accounting, but it was not until Nagahira’s time that the family came to serve as house managers for the Saionji. The specifics of how this relationship came about are not known, but Ryō Susumu points out that a document dated Jōkyū 2 (1220)/11 lists Nagahira as director of Kintsune’s household,34 so it is certain that Nagahira served Kintsune before then. Considering his role in the Jōkyū incident, in which he was able to dispatch a warning from Kintsune to the shogunate just before the former’s arrest by royalist forces, it is obvious that Networks of Wealth and Influence 331 there was a strong and trusting relation between Nagahira and Kintsune by the time of the war in 1221, and it is thus likely that Nagahira had been serving Kintsune for at least several years before then. It was Nagahira who successfully conducted negotiations over the control of Uwa Estate.35 Even at the time, Nagahira’s role in supporting Kintsune’s economic power was well recognized. Taira no Tsune- taka’s diary Heikoki compares Nagahira with Fan Li (Tōshu), a respected Chinese statesman of the State of Yue during the Spring and Autumn period (c. 771–c. 475 BCE). Known for his business acumen, Fan Li retired as an immensely wealthy man and was later regarded as a folk deity associated with prosperity.

The monk Nagahira passed away yesterday, they say. He was very skilled at figures, and was beyond comparison with anyone [else] who served the priestly minister [Kintsune]. [Nagahira] was like Tōshu! Kangen 2 (1244)/3/2636

The Tachibana family that served the Saionji is far less known than the Miyoshi, but they also played a key role in facilitating Saionji wealth accumulation. Tachibana no Tomonobu (dates unknown), the first member of the family whose name appears as a household manager for the Saionji family, initially served Kintsune’s father Sanemune.37 He then became the Suō provincial resident deputy (mokudai) under Kintsune,38 and after the latter was appointed as chief administrator of the retired sovereign Go-Toba’s household office (Go-in no chō bettō), Tomonobu was entrusted with running the Toba villa (Tobadono), and he also came to be involved in the management of Saionji estates.39 In 1230 he was also in charge of repairing Hōjōji Temple, a luxurious complex established in 1020 by Fujiwara no Michinaga (966–1028). Tomonobu is said to have completed this task quickly.40 His son Tomo naka carried out various tasks for the Saionji, such as conducting a survey (kenchū) of Nuta Estate in Aki Province to determine the quality and productivity of its land.41 Like his father, Tomonaka also developed a reputation for reliable oversight of successful construction projects.42 Tomonaka’s son, Tachibana no Tomoshige (?–1263), also served as a house manager for Kintsune’s son Saneuji (1194–1269) or for the latter’s daughter, the influential Ōmiya-in (1225–1292, formerly Saionji Kitsushi).43 Tomoshige, like his father before him, became deputy pro- 332 Power, Space, and Trade vincial governor of Bizen Province (a Saionji proprietary province), but he also became the manager of the aforementioned Nuta Estate and of Munakata Shrine in Chikuzen Province, both of which were under direct Saionji control.44 Tomoshige’s skills generated considerable wealth for his own family as well as for his Saionji masters. In addition to these posts, Tomoshige was likely deeply involved in the maritime transport network that the Saionji operated in the Seto Inland Sea, and in the trade with Song China. Tomoshige also carried on the Tachibana tradition in construction management. Godai teiō monogatari (The tale of five sovereigns) relates that he built—and then, after a terrible fire, rebuilt—the Gojō Pavilion that served as Ōmiya-in’s residential palace. This source describes the pavilion as “so wonderful as to be incompa- rable to any other residence in Kyoto, aside from the Kan-in.”45 Subsequent generations of the Tachibana continued to flourish alongside their Saionji masters. Tachibana no Tomotsugu succeeded his father Tomoshige, and by the second month of 1269 was signing documents as the manager of the Munakata Shrine estates in Chikuzen.46 Furthermore, his name appears as the manager of Nuta Estate in 1281. Reflecting their rising status, Tachibana offspring were regularly promoted: for instance, in 1311, Tachibana no Tomonao (Tomotsugu’s son), became a member of the Council of State.47 His mother was a daughter of the chief priest of the Munakata Shrine, which the Tachibana had been managing as part of Nuta Estate.48 The elevation of the Tachibana family illustrates how economic success on behalf of a senior courtier family could translate into increased status for a lower-ranking family, a classic case of the interconvertibility of capital.49 Meanwhile, the important role played by client families in managing estates and supporting the wealth accumulation of more established courtier families demonstrates that networks of cliental relationships deserve more attention.

Investing in Influence: The China Trade While careful accumulation and management of estates was vital for a courtier family to secure its economic base, that was not the sum of the sources of wealth and influence available. Trade was another endeavor that could offer immense benefits to those families who had the means to engage in it. The most exclusive form was trade with the continent, which offered great opportunities for status and wealth but at high cost and risk.50 Official trade relations between Japan and the continent Networks of Wealth and Influence 333 had ceased with Sugawara no Michizane’s (845–903) proposal to stop sending official emissaries to Tang China in 894. This did not mean, however, that relations no longer existed. Rather, trade continued unof- ficially, and courtier families played a significant, if understudied, role. Trade with the continent, however, was a proposition fraught with immense risk: in addition to bad weather and the like, a large investment was needed just to recruit a ship and crew. Only a courtier family with reliable income from a large, established estate portfolio could safely absorb such a risk in its long-term financial strategies. Moreover, a varied and developed portfolio was necessary to ensure that the family had available goods with which to trade or the means to acquire them. As a result of these high costs and risks, even attempting trade with the continent testified to a family’s wealth and status. The real payoff, however, came in the form of rare and luxurious treasures from China, which served as visible indicators of the procurer’s elite status. A courtier family with access to such luxury goods could corner the market on an immense source of cultural capital, ensuring respect and easing the way to choice appointments. The courtier journal Yōkōki describes one case concerning the Saionji:

When [the retired sovereign Go-Saga] paid a visit to Hachiman Shrine, he was carrying a wooden baton. But the minister [Saionji Saneuji, a son of Kintsune] presented a baton made of ivory, so this ivory one was used [when Go-Saga visited Kamo Shrine]. Fashioning an ivory baton had long been ordered [by the monarch]; but, since it was hard to obtain ivory, it could not be made until now.51 Kangen 4 (1246)/4/29

In this instance, a member of the Saionji family was able to provide an item much desired by the sovereign, and at a perfectly opportune moment. The trade activities of the Saionji, in fact, had their roots in much older practices. One of the most famous examples of trade relations between Japan and the continent after 894 was the trade conducted by Taira no Kiyomori (1118–1181). Kiyomori, a distant relative of Saionji Kintsune, was typical in that he openly displayed the profits of his dealings with China, aware of the great cultural capital such goods represented. In the first chapter of The Tale of the Heike, in the section titled “Kiyomori’s Flowering Fortunes,” a description of precious goods from 334 Power, Space, and Trade

China attests to the great wealth of his Ise Taira family: “They lacked none of the Seven Treasures or myriad precious things—Yangzhou gold, Jingzhou pearls, Wujun damask, Shujiang brocade.”52 These luxury items were obtainable only through trade with Song China. How important such trade was to individual courtier families depended on their wealth, connections, and control over areas conducive to managing such trade. In the case of the Saionji, like the Ise Taira they invested heavily in trade with the continent and generated a great amount of wealth in doing so. Courtier families engaged in the China trade normally dealt with maritime merchants who served as middlemen, but on occasion the wealthiest families directly participated in the trade. Like Kiyomori, Saionji Kintsune sent a Chinese-style ship (karafune)53 to Song China. But while Kintsune acted as a private individual, his ship set off not to meet merchants but rather to visit the Song court itself. According to a contemporary record,

On the fourth day of the seventh month, someone said that the Chinese-style ship which the Ichijō Novice [Kintsune] had sent [to Song China] has returned. The ship has brought back one hundred thousand strings of cash.54 Moreover, various exotic treasures were included, they say. Among them were a bird that can speak human languages well and a water buffalo. The bird flies when music is played, and it speaks human languages without making any mis- takes. It is said that the water buffalo is twenty times stronger than a regular cow. The monk minister [Kintsune] sent cypress trees55 in order to build a pavilion as a gift for the emperor in the foreign land [the Southern Song]. The emperor was overjoyed at the gift, and gave various exotic treasures in exchange.56 Minkeiki, Ninji 3 (1242)/7/4

In other words, by choosing to participate directly in the Chinese tributary system, Kintsune received all manner of valuables from the continent. The animals mentioned in the account deserve particular attention. As Fujiwara no Akihira’s eleventh-century Shinsarugakuki (New monkey music) mentions, numerous goods were being imported to Japan from the continent almost two centuries earlier.57 Imported exotic animals, however, appear to have been quite new in Kintsune’s time, and they represented an especially lucrative source of cultural cap- Networks of Wealth and Influence 335 ital in courtier society. Several contemporary sources such as Tsurezure- gusa (Essays in idleness) or Meigetsuki (Record of the full moon) indicate that people were enthusiastic about these animals arriving in the capital, a phenomenon that invited cynical comments from the monk (and sometime social critic) Yoshida Kenkō (c. 1283–c. 1352); he reminds us that not everyone thought the China trade was worth the trouble.58 Yet the Saionji no doubt gained considerable influence socially by being able to display exotic animals from China right when such things were in vogue. Of course, the family may have also encouraged the fashion in exotic animals themselves for the purpose of increasing their status. The ability of Saionji family members to procure rare imported items like ivory batons and exotic animals can be easily understood in light of the trade relations that the family established with the continent. Whether it was rare treasures or exotic creatures, trade with the continent offered both profit and heightened status in courtier society, making this a lucrative if highly expensive form of investment of profits from the family’s estates.

Mapping Spatial Power: Locating Estates in Networks of Trade and Transportation As we have seen, careful accumulation of estates, human resource management, and trade with the continent were all key components of the Saionji family economy. The first two of these were important for any major courtier family seeking to enhance its wealth and status, while the prohibitive cost and risks of the third put it out of the reach of most. Seeking to benefit somehow from trade networks both domestic and foreign, however, was a staple in the estate strategy of numerous courtier families in the early medieval period. Courtier families generated wealth not only directly from individual estates, but also by strategically accumulating and developing estates to take advantage of transportation and trade networks to turn a profit. Assessing Saionji estate strategies with this in mind, while informed by the perspective of spatial power, sheds light on this heretofore little-understood process. Figure 13.1 (above) shows that many Saionji estates were on or near the seacoast. Of particular import are those along the Seto Inland Sea, including the aforementioned Uwa Estate in Iyo Province. The Seto Inland Sea, a major artery of transportation since ancient times, had developed into a hub of maritime transportation for the purposes of 336 Power, Space, and Trade conveying tribute from local areas in the west to the capital and for sending emissaries to China or Korea. Matsubara Hironobu argues, in fact, that maritime transportation on the Seto Inland Sea during the eighth and ninth centuries was not just limited to transport between the south and west of Japan and the capital regions, but was also closely connected to the overseas trade with Silla and Tang China59 (see figure 13.2). Taira no Kiyomori himself had further developed this Seto Inland Sea route to bring ships safely from Song China into the Kinai region. Chinese ships had been unable to come around northern Kyushu because routes were disorganized and provisioning was difficult.60 Kiyo- mori even built Itsukushima Shrine for the protection of this valuable maritime route. Kintsune was clearly Kiyomori’s successor in that he sought to control this maritime route and secure trade with China. Like Kiyomori, the Saionji took steps to ensure that their enterprise would benefit from spiritual protection. The best example of this was the family’s successful effort to gain control of Munakata Estate.

Fig. 13.2. Maritime trade routes. Template from Jeffrey P. Mass, ed., The Origins of Japan’s Medieval World, Stanford University Press, 1997. Originally drawn by Arnie Olds. Used with permission. Networks of Wealth and Influence 337

That estate was in Kyushu’s Chikuzen Province and was significant because it was the location of the Munakata Shrine, long regarded as a source of protection for maritime transportation.61 It was this traditional association, which in turn encouraged the growth of maritime facilities in the area, that made the estate valuable. At the time of Kintsune’s death in 1244, the estate was a holding of Shumeimon-in (1182–1264), Go-Toba’s consort. Its custodianship belonged to the warrior Miura Yasumura, but after his defeat at the Battle of Hōji in 1247, his rights were forfeited.62 The Yōkōki relates that in the wake of these events, Saionji Saneuji seized the opportunity to make the estate a holding of his patron Ōmiya-in, and Saneuji himself took custodianship of the property.63 The actual day-to-day management was taken care of by the Tachibana.64 This ensured that the Saionji would exercise control over the shrine and could count on its protective properties for their maritime activities. Developing estates in relation to maritime transportation routes could be an excellent way to generate wealth, not only from the route itself but also from domestic trade and distribution networks linked to the main routes. There was considerable wealth to be generated from both maritime and land transportation. Controlling land at points where networks connected gave a family like the Saionji flexibility and opportunities for profit from handling goods or supervising those who did. Family access to transportation routes could also be of benefit during unforeseen or difficult circumstances. The case of Kase Estate is particularly instructive. Yamauchi Shinji, who has studied Japan-Song commerce from the perspective of the sul- phur trade, analyzes part of The Tale of the Heike with an eye to trade and distribution issues.65 The tale relates how following the failed anti-Taira coup of 1177, three vassals of the retired sovereign Go-Shirakawa— Shunkan, Taira no Yasuyori, and Fujiwara no Naritsune—found themselves exiled to Kikai Island off Satsuma.66 After a while, a letter of pardon arrived for Yasuyori and Naritsune, who returned to the capital, leaving Shunkan struggling to survive with few basic necessities.67 The situation had not been so bad when the others had been present, for they had received “regular shipments of food and clothing to the island from Kase Estate, a Hizen property belonging to Lesser Captain Naritsune’s father-in-law, Norimori.”68 The Kase Estate referred to here was a property in the vicinity of what is now Saga City. Based on this description and on the fact that both Naritsune and Yasuyori stopped 338 Power, Space, and Trade at Kase Estate first on their way back to Kyoto, Yamauchi argues that Hizen Province and Kikai Island were connected via a maritime route.69 Furthermore, he argues that we can safely postulate that Yasuyori and Naritsune, upon leaving the estate, headed north along the west coast of Kyushu, whereupon they entered the Seto Inland Sea via the Kanmon Strait.70 The ability of the exiles to regularly acquire food and clothing on Kikai Island via ship, and smoothly travel up the coast when depart- ing, depended not only on the kindness of relatives but also on having family-held estates near maritime routes. Such estates could provide significant benefits for a family when it needed to move people and commodities. These examples clearly illustrate the significance of thinking about estates in terms not only of their own intrinsic economic value, but also of the spatial power they possessed based on how they were situated in relation to each other and to adjacent trade and transportation routes. Courtier families did not foolishly grab every parcel of land they could. Instead, they chose to neglect certain geographic areas to focus on others, fighting tooth and nail for control over certain estates but not others, even if the latter appeared to have greater economic value. In this regard the aforementioned Munakata Estate may seem to have had little economic value, but its significance for a family that already commanded adjacent maritime routes becomes readily apparent. The reasoning behind courtier family actions only becomes clear when their holdings are considered collectively, revealing a broader estate strategy. As Amino Yoshihiko rightly observes, “The proprietors of estates were not just [idly] waiting for land commendations in order to collect them; they [each] had their own frantic assertiveness and competed with each other in order to obtain estates or public land in various provinces based on their own strategies. The distribution of their lands must tell the outcome of their strategies.”71 In the case of the Saionji, their estate strategies in the Seto Inland Sea region take on new meaning considering the area’s significance for maritime trade and transportation. When considered together, seemingly disconnected actions reveal a careful attention to areas conducive to making maximum profits along this long-established trade route. The Saionji pursued estates in such a way that they collectively provided effective physical distribution along the coast. Within this context, Kin tsune’s obsession with Uwa Estate, discussed earlier, can be understood as part of a program for systematic spatial control. It is worthwhile to Networks of Wealth and Influence 339 consider Uwa and several other important estates in order to better understand this Saionji strategy. As shown earlier, Kintsune went to great lengths to take Uwa Estate— in Iyo Province, over which the family held hereditary proprietorship— from its previous holders. Iyo was the most significant of the several proprietary provinces that the Saionji family controlled. Kintsune himself became the provincial proprietor (chigyōkokushu) of Iyo in 1203, after which the honor was for the most part passed along the main line of the Saionji throughout the Kamakura period. This made Iyo unique among the Saionji proprietary provinces. It is not surprising that Kin tsune wished to couple this authority with direct proprietary power by gaining control over major estates like Uwa. According to Amino, the fact that Fujiwara no Sumitomo (d. 941)—a notorious pirate earlier in the region’s history—had chosen nearby Hiburi Island for his base strongly suggests that Kintsune was following suit and grasping a key locale on the maritime transport routes to the Kyushu region.72 Hiburi Island was already an important point along the Bungo maritime route. In other words, controlling Uwa Estate was tantamount to controlling those wishing to come and go along this stretch between the Seto Inland Sea and northern Kyushu. When one considers Saionji control over neighboring estates in Iyo, such as Uma Estate, and Kintsune’s involvement in trade, the value of Uwa Estate for the Saionji spatial strategy becomes readily understandable. As Amino argues, while Iyo Province was the central base for the Saionji estate strategy, they held numerous other landholdings distributed around northern Kyushu and the Seto Inland Sea, most of which were also involved in maritime transportation. The Saionji holding in northern Kyushu that seems to have been most strongly connected to the family’s proprietorship of Iyo Province is Anami Estate in Bungo Province. Just across from the Uwajima District in Iyo, it enabled the Saionji to oversee the main Bungo maritime route from both sides of the waterway. Amino points out that Kintsune gained Anami Estate in exchange for a commendation to Yusuhara Hachiman Shrine in the ninth month of 1230, suggesting that the estate was highly desired and surely related to Saionji strategies concerning Iyo.73 Of no less significance was Uno-no-Mikuriya.74 Like Uwa, this area had a long history of maritime activity. It was famous as a base of the Matsuura League, whose members came to be known as local sea lords in the region around Hizen.75 In other words, while many Saionji estates 340 Power, Space, and Trade were connected to maritime routes, two major estates—Uwa and Uno— were also historical bases for pirates or naval forces. This suggests that the Saionji were not especially innovative in their regional strategy, but rather that they were aware of the historical importance of these areas and pursued control over them as a result. The Saionji strategy in northern Kyushu left them in a position of considerable control over the maritime routes in the vicinity. As Amino observes, “There is no question that, through seizing Munakata Shrine and Uno-no-Mikuriya, the Saionji had [gained] great influence over the maritime routes of northern Kyushu.”76 The second area where the Saionji focused on acquiring estates was the Seto Inland Sea, which in addition to being a venue of major maritime networks and a gateway to the continent was also important as a domestic maritime route that linked northern Kyushu and the capital region. The Saionji controlled important estates linked to water transportation such as Matsubara Estate in Harima Province and Tai Estate in Awa Province. The best documentary evidence survives for Nuta Estate in Aki Province and for Kayō Honjō Estate in Bizen Province. Nuta Estate, mentioned earlier, was under Saionji control by at least the mid-thirteenth century, judging from an inspection record (kenchū mokuroku) for 1243 and 1252 signed by Tachibana Tomonobu, the aforementioned house manager for the Saionji.77 Moreover, it appears that Kobayakawa Shigehira, the estate steward who signed the document along with Tomonobu, also became a Saionji vassal and developed a deep relationship with the family.78 Amino holds that the Kobayakawa warrior family served as resident deputies of Iyo Province and that the Saionji employed a tactic of organizing cliental families—such as the Kobayakawa, who also became sea lords in later years—in local areas.79 A significant part of estate strategy was not just passively depending on pre-existing connections between areas, but on reinforcing them. Nuta Estate was already linked to Iyo Province, both by the Seto Inland Sea route and by a chain of locations that served as bases for locals engaged in fishing. Amino points out that by keeping the area under control while making the Kobayakawa family their clients, the Saionji neatly tied the estates together by means of maritime transportation networks.80 A similar argument can be applied to the case of the Kayō Honjō Estate in Bizen. This estate was yet another pivotal point along the main Seto Inland Sea maritime transportation network, and when this is considered in light of the fact that Saionji Saneuji also gained Networks of Wealth and Influence 341 proprietary rights over Bizen Province, there is no question that the Saionji succeeded in exercising great influence over Seto Inland Sea transportation. Neither was Saionji influence limited to the sea. Many Saionji estates occupied key positions on rivers that functioned as major waterways in the domestic transportation network. In particular, the locations of Kintsune’s villas are notable. The Suita villa, where he enjoyed bathing in water from the Arima hot spring, was on the banks of the Kanzaki River, estuary of the Yodo River, as shown on figure 13.3. The Yodo River, the headwaters of which meet Lake Biwa, joins the Uji, Katsura, and Kizu Rivers in Kyoto and then flows out into Osaka Bay, which is connected to the Seto Inland Sea, which in turn functions as the main artery connecting western and southern Japan to the central Kinai region. In the medieval era, the Suita area was a popular leisure spot for courtier visits. More importantly, it was situated along a main waterway that connected western Japan to Kyoto and Nara, where the households of powerful families were concentrated. As a result, many of the estates and pastures along the Yodo River and its estuaries had long played important roles in transportation and distribution networks. Since earlier times, estates there served as locations for storehouses or administrative offices that temporarily safeguarded tribute sent from estates in the western part of Japan.81 The estate on which Kintsune’s Suita villa was located was the Suita Nishi Estate. It may have been people from this estate who carried the buckets of hot water from Arima. Amino argues

Fig. 13.3. River Yōdo basin. Drawn by Philip Garrett. 342 Power, Space, and Trade that since ships must have been used to carry the hot water, the Saionji not only had a base here but also had the power to utilize local elites in the vicinity who were engaged in maritime transportation and distribution.82 In other words, Kintsune’s Suita villa was in a prime location: it was situated at a vital transportation point that connected the capital with western Japan, as well as with the peninsula and the continent. Another of Kintsune’s villas, Makishima in Uji,83 was due south of the Heian capital on the east coast of Ogura Pond, where the Katsura River joins the Kizu and Uji Rivers. Situated around the estuary where the Uji River flows into Ogura Pond, it too was a prime location. The area was a base for niebito, royal provisioners who, in addition to depending on transportation routes, also engaged in fishing in the vicinity. (For waterways around Heian-kyō, see figure 13.4.) In addition to their own villas, the Saionji managed some holdings for members of the royal family, which also offered opportunities for expanding their influence. Especially important in this regard was the

Fig. 13.4. Waterways around the capital. Drawn by Philip Garrett. Networks of Wealth and Influence 343

Toba villa of the retired monarchs, south of Kyoto. The Saionji family managed both the villa itself and the attached lands and stables. The Toba villa controlled bridges and the passage of ships on land adjacent to it, making it an important place for river transportation.84 Once the Saionji family had acquired the right to manage this area, this privilege became hereditary. Finally, although this discussion has focused primarily on maritime trade and transportation, land transportation was another area in which courtier families could invest. Although the Saionji did not have the same degree of economic interest in land transportation, they nevertheless cultivated influence in this arena as well. The family controlled numerous pastures near the capital, where horses and oxen presented to the royal family from local areas were cared for. Some of their estates were also recognized for their ability to produce oxen or horses of fine stock. For instance, Uno-no-Mikuriya, discussed above in relation to maritime transportation, was known for its fine oxen.85 The Shungyū ekotoba (Picture scroll of fine oxen)86 mentions how Saionji Saneuji presented an ox from this estate to the retired sovereign Go-Saga.87 This is significant from the perspective of land transportation because oxen and horses were vital to such networks and represented solid economic capital as a result.88 The Saionji family also had access to positions of influence over land transportation, gaining a near-monopoly over the post of director of the royal stables. Control over court positions concerned with stable management and pastures also gave the family influence over pastures and tribute lands in various provinces. Amino explains that as the Saionji controlled positions of managing royal estates for the retired sovereign (such as the Toba Estate with its villa and stables) as well as positions enabling them to recommend or appoint the head of the Bureau of Horses of the Left, they were acting as major figures in land transportation.89 The Saionji had a particular advantage when it came to generating profit from trade and transportation because they could link their investments in land transportation to their more established maritime ones. Major pastures were located near rivers and as a result were almost always linked with ports. This was because managing stables required organizing provisions via cart or boat as well as maintaining routes to move horses around quickly and reliably. Oxen and horses were carried to distant locations via ships. Human resources were also essential. Ship crews, shipping agents, and workers at stables and villas needed 344 Power, Space, and Trade to collaborate efficiently, and the Saionji were effective in this regard. Through their management of stables, the family had strong links with teamsters and shipping agents who ran carts to serve the transportation needs in the greater capital region. Ships the Saionji arranged to head upstream from Suita to the Minase detached royal palace90 were operated by crews mobilized from their pastures, the Toba villa, and members of the Bureau of the Horses of the Left, revealing how seamlessly the whole system came together. The Saionji case thus illustrates how early medieval courtier families could generate wealth through engaging in, and asserting a degree of control over, networks of trade and transportation and that the primary means of doing so was through projecting spatial power by means of estate holdings. Estates were accumulated and developed not merely for their own sake or for their individual economic worth, but rather as elements of long-term strategies that are apparent only when estates are assessed spatially. The Saionji family successfully secured pivotal points on routes between the capital and west and south Japan, thereby establishing a spatial foundation upon which to effectively manage both land and maritime routes domestically and abroad, with all the attendant wealth and prestige.

The spatial power of property holding in early medieval Japan could take many forms. On the one hand, spatial power could be symbolic, as seen in the reworking of the land around the Kitayama villa to produce a scenic landscape that manifested the high status of the Saionji family and brought recognition from other courtiers and the royal family. From this perspective, spatial power could encompass the conversion of economic capital into cultural capital, just as the wealth of estates could be invested in acquiring high-status luxury goods from the continent. On the other hand, spatial power also directly overlapped with economic power, as seen in the geographical location of Saionji estates or villas that offered effective control of both domestic and international trade routes. Through control of significant geographical points for both water and land transportation, the Saionji family presided over major distribution routes that served as the foundation of their wealth. In addition to the efforts of individual family members themselves to develop such spatial networks, a key role was also played by client families. In the case of the Saionji, the Miyoshi and Tachibana families provided managers for their households and estates. Through skilled Networks of Wealth and Influence 345 management of local workers, construction, and estate operation, these families contributed greatly to Saionji wealth accumulation and accordingly increased their own status within courtier society. Estates, whether owned or managed on behalf of others, were not passively held valuables. They were economic generators that were actively and strategically developed by courtier families and their associated families. It is in this broader economic context that the fabulous Kitayama villa with which I began my discussion also needs to be understood. Considering how carefully the Saionji worked to establish spatial networks based on estates, villas, and pastures, we can safely conclude that Kintsune’s choice of Kitayama was motivated by such concerns. The impressive location retained its significance in later years, when it was taken over in 1397 by the third Muromachi shogun, Ashikaga Yoshimitsu, who called himself the “King of Japan.”91 Yoshimitsu demanded that daimyō all over the country send a fixed number of workers for its renovation. Imatani Akira argues, “At the Kitayama pavilion, the top political decisions were made, and rituals at the national level also come to be conducted there. It is a huge mistake to call the Golden Pavilion (Rokuonji) the remains of Yoshimitsu’s villa; the site was rather the political center for the country, and deserves to be called a palace.”92 The reason Yoshimitsu chose this site in order to establish his own “palace” is not clear. Considered alongside Kintsune’s careful cultivation of spatial network strategy, however, there is no doubt that it was a prime location—situated in a lofty, scenic wonderland looking down on the capital—that demonstrated the owner’s authority. At the same time, the history of the location, bound up with the influential Saionji family and their numerous important visitors, had no doubt created considerable cultural capital at the site. The Saionji literal and figurative reworking of the spatial environment, cultivating strong links between space and power in order to provide themselves with an affluent lifestyle and enormous influence, reflected their legacy long after they lost that power.

Notes

1. Hyakurenshō, Gennin 1 (1224)/12/2, in DNS 5(2):445. The Hyakurenshō is a documentary record compiled in the late Kamakura period, covering the mid-tenth to mid-thirteenth centuries. The earliest volumes are no longer extant. Fujiwara no Teika described Kintsune’s villa following his visit on Gen- nin 2 (1225)/1/14 (Meigetsuki (Record of the full moon) 1911–1912, 2: 404), 346 Power, Space, and Trade and a description in some detail is given by Masukagami (The clear mirror), 298–299. (For English, see the translation by George W. Perkins, 1998, 72–73.) 2. Meigetsuki 1911–1912, 3: 319 (Kanki 3 (1231)/9/16). 3. Gotō 2009, 164. 4. In the conception of sociologist Pierre Bourdieu, cultural capital refers to both objectified cultural goods such as art and intangible cultural assets such as acquired skills or education. Along with economic and social capital it com- prises the three basic forms capital may take, and both into and from which it may be converted under the right circumstances (Bourdieu 1986). The status and authority associated with acquiring vast amounts of cultural capital made it a particular concern for courtier families in medieval Japan. 5. The Kan’in branch split into three houses headed by the sons of Fujiwara no Kinzane (1053–1109): the Sanjō (by Saneyuki, 1080–1162), the Ōmiya (by Michisue, 1090–1128), and the Tokudaiji (by Saneyoshi, 1096–1157) houses. 6. The work of Kuroda Toshio was especially significant in problematizing conventional narratives of the decline of courtier authority. See Kuroda 1994a. Hongō Kazuto (2000) provides an effective summary of Kuroda’s influence and the main debates. In English, see Adolphson 2000, and for an earlier but convincing formulation of the Kamakura-era system, see Hurst 1982. 7. For “tools of authority,” see Kamei-Dyche 2013. 8. Kanai 1999, 3. Some of the earliest studies of courtier landholdings were by Nakamura Naokatsu (1890–1976), especially Kajūji-keryō ni tsuite (1978b). On the economic world of medieval courtiers in general, see Sugawara 1998. See the introduction and chapter 1 in this volume for brief discussions of the kenmon theory. 9. See, for example, Nomura 1989b, 294–319. 10. See the pioneering study by Nakamura (1978b), and Takeuchi Rizō’s extensive fourteen-part series “Shōen to kizoku,” in Nihon rekishi, as well as Kamihara 1963 and Yoshie 1967. On the Konoe specifically, see for example Kawabata 1994, and on the Kujō, see Nishitani 1998a. 11. Nakamura 1978b. 12. Nagahara 1958. 13. Usui 1986 and Sugawara 1998. 14. Okano 1988. 15. Abe 1997. 16. Amino 2008b, originally published in Kokushigaku 146 (1992). 17. For example, Amino 1978b and 1982. 18. Pioneering works include Sasaki 1972 and Wakita 1969. More recent work includes Hongō 2004, Sakurai 1996, and Segal 2011. 19. Goffman 1959. Networks of Wealth and Influence 347

20. Bialock 2007, 217–271. Other important works on topics related to the spatial embodiment of sociopolitical hierarchies include Coaldrake 1996, and Fiévé and Waley 2002. 21. Amino 2008b. The earliest work on the Saionji estates began in the prewar era with Ono Kōji’s Nihon chūsei shōgyōshi no kenkyū (1989), which only touched on the estates briefly during the Muromachi era. Inui Naoko, meanwhile, discussed the Saionji estates to some extent in her work on the Sanjōnishi family (Inui 1980). The work of these two scholars, neither of whom dealt with the pre-Muromachi situation, was the sum of scholarship on the Saionji estates until Amino’s study in 1992 (see note 16). There continue to be occasional references, usually in studies of other families and later eras, such as Okano 2002. 22. Amino 2008b, 295. Amino defined western Japan as the area west of Owari, Mino, and Kaga. Among the fifty-three estates, only six were in the east. 23. For information pertaining to individual estates, see Amino et al. 2005. 24. Ibid., 296. 25. Ibid. 26. Nagahara 2004b, 112. 27. Amino 2008b, 296. He states that the Tobadono was later referred to as including the “thirteen estates.” 28. Ibid. 29. Gomi 1992, 117. 30. Ibid. Gomi observes that after the Jōkyū incident, the Saionji were as powerful as the Ise Taira had been. 31. Later, Iyo also became known as a location for warrior Saionji. See Ishino 1987 and Yamauchi 2012. 32. Azuma kagami 1988–1989 1:173. 33. Azuma kagami 1988–1989 1:59 (Jishō 4 [1180]/12/19). 34. Ryō 1957, 2:179. 35. Gomi 1992, 117. 36. Heikoki, in Zōho shiryō taisei, 32:280. 37. Tomonobu was the third son of a Prince Kaneyasu of the Hakke family but was adopted by Tachibana no Mochizane. It is not clear when and how Tomonobu came to serve the Saionji, but his name was listed as an outrider for Saionji Sanemune (Kintsune’s father) in 1205 (Gomi 1992, 120, and Sakurai 1987, 15–28). 38. Suō was Kintsune’s proprietary province from 1206 until 1215, when it returned to the control of Tōdaiji Temple. 39. Gomi 1992, 120. 40. Ibid. The Masukagami favorably compares Kintsune’s Kitayama villa to the Hōjōji in terms of its splendor (Masukagami, 299; trans. Perkins 1998, 72–73). 348 Power, Space, and Trade

41. The Saionji became the proprietors of Nuta Estate after the Jōkyū Inci- dent in 1221, whereupon Tomonobu conducted a survey as official inspector (Gomi 1992, 120). Thereafter, his descendants occupied the position of custodian there. The kenchū was a formal assessment of an estate’s productivity and was used to calculate appropriate rents. 42. For example, see Gomi 1992, 121. 43. The name “Ōmiya-in” was a reference to her family (see note 5). 44. On this, see Gomi 1992, 119–120. For more on Tomoshige and Tachibana management of Saionji estates, see also Amino 2008b. 45. Godai teiō monogatari, 439–440. 46. Gomi 1992, 121. 47. Ibid. 48. Ibid. 49. On the interconvertibility of capital, see Bourdieu 1986, 53–55. 50. On the trade with China, see von Verschuer 2006 and Batten 2007. On maritime commerce and piracy, see Shapinsky 2007, 2014. 51. Yōkōki 2004, 141. The Yōkōki is the journal of Hamuro Sadatsugu (1208– 1272), a courtier of the early thirteenth century. Except for the years 1246 to 1248, it survives only in fragmentary form. 52. Tale of the Heike 1988, 30. 53. Literally “Tang vessel,” but actually a general term for Chinese ships, other foreign ships, or ships built in such a style. 54. One string (kan) consisted of 1,000 coins, so 100,000 kan constituted an extremely large amount of money. Tanaka Kōji calculated the approximate value of one kan in current currency values in his essay (Tanaka 1996). Accord- ing to his article, one kan (kanmon) was equivalent to about ¥156,000 in 1990s currency. Even after allowing for some adjustments, 100,000 kan works out to more than ¥100 million. 55. In all likelihood, the lumber would have been chosen with care by Kin tsune’s Tachibana managers, who had a reputation as successful building contractors who could recruit and manage skilled craftsmen. Construction wood, including cypress, pine, and cryptomeria, appears to have been a common export good from Japan to China. See von Verschuer 2006, 69–70. 56. Minkeiki, in Dai Nihon kokiroku 8 (2001), 142. The Minkeiki is a journal written by the courtier Hirohashi Tsunemitsu (1213–1274), minister of civilian affairs (Minbukyō). The diary covers the years from 1226 to 1268. 57. The section titled “Hachiro Mahito” describes the goods imported by a Japanese merchant. These included woods, incense, ointments, pigments, furs, bowls, and many more items (von Verschuer 2006, 51). Networks of Wealth and Influence 349

58. “Even if we were deprived of Chinese goods, we should not miss them, except for medicines. Many Chinese books are available all over the country, and anyone who wishes can copy one. It is the height of foolishness that Chi- nese ships should make the dangerous journey over here, crammed with cargoes of useless things” (Essays in Idleness 1967, 101, episode 120. Also see 104, episode 121, and 126, episode 139). 59. See Matsubara 2004, especially chapter 2. 60. Charlotte von Verschuer (2006, 78) points out the important role that Kiyomori played in maintaining the sea route in the Inland Sea in order to invite a Chinese merchant to come as far as Ōwada on the coast, the port Kiyo- mori had reconstructed. 61. The Munakata Shrine actually consists of a set of three shrines. Worship there was originally concerned with land transportation but was soon super- seded by association with the sea and maritime travel. In ancient times when official relations were carried on between Japan and the Korean kingdoms and between Japan and China, envoys as well as traders prayed at the shrine for protection for their voyages before setting sail. 62. Miura Yasumura was the second son of Miura Yoshimura, a prominent vassal of the Minamoto. The Battle of Hōji resulted from an internal shogunate struggle between the Hōjō regents and the Miura. The defeat of the Miura meant the end of that family. 63. Yōkōki 2004, entries for Hōji 1 (1247)/8/18 and 8/27. 64. Ishii 1970, 453–492. 65. Yamauchi 2009. 66. Tale of the Heike 1988, 82. 67. Ibid., 98–100, 111–112. 68. Ibid., 89. 69. Yamauchi 2009, 56. 70. Ibid. 71. Amino 2008b, 297. 72. Ibid., 299. The ancestors of the previous owners of the estate had in fact received it as a reward for suppressing this same pirate. 73. Ibid., 300. 74. On Uno-no-Mikuriya in the early medieval era, see Fujimoto 2002. 75. The Matsuura League was a local naval force made up of families based in the Matsuura area of northwest Kyushu, a region long associated with fishing. They initially supported the Taira in the Genpei War but switched sides at the Battle of Dan-no-ura (1185) and became increasingly prominent during the thirteenth century. For more, see Sotoyama 2011. 76. Amino 2008b, 302. 350 Power, Space, and Trade

77. Ibid., 302–303. He signed the document as the official inspector. 78. Ibid., 303, drawing on the work of Ishii Susumu and Takahashi Masaaki. 79. Ibid. For more on the Kobayakawa family and their stewardship of Nuta Estate, see Takahashi 1981. 80. Amino 2008b, 303. 81. Amino et al. 2005, 7:301–302. 82. Amino 2008b, 304–305. 83. The Makishima villa, built by Kintsune in 1242, was in an area that was passed down in the Saionji family until the Sengoku period. 84. Amino 2008b, 306. 85. Ibid., 309–312, esp. 311–312. 86. This picture scroll was made around the end of the Kamakura period, but only the text portion is now extant. It records fine oxen since the end of the Heian period and indicates the value accorded good oxen. 87. Amino 2008b, 311–312. 88. One imagines that Kintsune got excellent use out of the water buffalo given to him by the Chinese emperor. 89. Amino 2008b, 309–310, 312. 90. The Minase Palace was a detached palace built as a summer retreat around 1199 where the Minase River joins the Yodo River. The palace was washed away in heavy rains in 1216 but was rebuilt by Minamoto no Michiteru (1187–1248) in an even more luxurious form than before. Go-Toba seems to have been particularly fond of this palace; the Meigetsuki alone records no fewer than thirty of his visits. 91. Imatani and Yamamura 1992. 92. Imatani 1990, 77. c h a p t e r 14

As Estates Faded Late Medieval Maritime Shipping in the Seto Inland Sea

Michelle Damian

Coastal settlements have long been recognized for their importance as transportation hubs. In ritsuryō times and during the age of the estate system, they were the conduits for submitting rents and dues to the central government and the point of departure for regular tributes sent to the monarch and the nobility. Coastal estates furnished their proprietors with bounty from the sea, and ports could be the sites of warehouses (kurashiki) for landlocked estates. The importance of these coastal settlements cannot be overstated in considering shipments to the court and central proprietors, and this has been explored in other studies. Less attention has been given to the role of ports in interactions between estates of different proprietors or estates that were not contiguous. Exploring those connections sheds light not only on the hierarchical relationships between estates and their proprietors but also on developments at the local level. This chapter examines late medieval economic and labor ties that linked coastal regions along the Seto Inland Sea, arguing that the maritime environment contributed to enhanced economic relations among estates as well as to the mobility of individual laborers. The chapter also will highlight increased commodification of goods as opposed to traditional payments of estate rents and dues (nengu). As the estate system broke down in the later medieval period, local economies grew stronger, which in turn facilitated more complex networks of individuals across wider areas. The relationship between estates and ports, particularly in the later medieval era as the estate system deteriorated, is sometimes hard to see. Here I will first trace a brief history of the development of two ports that

351 352 Power, Space, and Trade are representative of estate-related shipping: Yuge port in Yugeshima Estate (Iyo Province), and Onomichi, affiliated with Ōta Estate (Bingo Province) (for Onomichi see plates 25, 26, 27, and 28; for Yuge see plates 25 and 26). Then I will examine shipping trends throughout the Inland Sea region in the fifteenth century. By privileging the geography of ports in a study of Inland Sea shipping, networks of both people and goods become apparent, and what at first glance appeared to be minor ports are shown to have played a more significant role. Also notable is that estate ports that once shipped large quantities of rents to the court and nobility were not always the same ports that later shipped the greatest volumes of trade goods. Tracing the flow of trade goods in later medieval times reveals too that even though some ports shipped relatively few trade items, they actually played a key role in transshipment of items from more distant locales. Certain ports seem to have specialized in shipping particular items, although they likely did not actually produce those goods locally. Instead they procured them from other areas, which indicates direct trade between different regions. Compar- ing records of such trade goods with records of items earlier shipped as estate rent payments also reveals a difference in the types of goods that came to be produced for wider consumption in the later medieval age. Lack of documentation about certain estates introduced in the following analysis sometimes presents a challenge in drawing conclusions about specific estate development. (For a complete list of ports and their affiliate estates or proprietors, see appendix 1.) Nevertheless the overall shipping trends highlighted here clearly reveal that labor mobility and collaboration among different regions in the Inland Sea area increased during the fourteenth and fifteenth centuries as the estate system declined. The shift toward greater independence from estate proprietors in the late medieval period was one vector in the breakdown of the estate system. As we have seen throughout this volume, the estate system was based on powerful central proprietors and the shiki hierarchy, but the Nanbokuchō and Muromachi periods witnessed several phenomena that contributed to the flattening of that hierarchy.1 Particularly important was the tendency for central proprietors to take over administration of estates themselves as ichienryō, or fully held lands. Other developments were the shogunal half-tax (hanzei) system, by which military governors could appropriate 50 percent of rents from many estates for military purposes, and growing activism among cultivators, who with As Estates Faded 353 increasing frequency petitioned for greater autonomy in managing the administration of their own fields and rent payments.2 As we shall see in the following analysis, the resulting breakdown of the estate system had a twofold effect in regard to maritime shipping. Commodification of goods increased as cultivators protected their own interests and shipping agents sought their own profit, leading to thriving lateral trade. In addition smaller ports took on a critical role in transshipping goods that in turn strengthened lateral trade. It seems likely that under such conditions ties to central authorities, based on timely submission of rents and dues to estate proprietors, would have become less of a priority as rent collectors privileged local ties and personal profit.

Early Coastal Estates and Ports: Case Studies of Ports at Yuge and Onomichi Amino Yoshihiko was one of the first scholars to suggest the need to examine estate history through the lens of rivers and oceans.3 While the role of rice in paying rents and dues or other fees to an estate’s proprietor has been well studied, work such as Amino’s has revealed that goods other than rice were used to pay rents, and it also shows practices specific to a maritime culture. For example, in the fourteenth century oarsmen were sometimes assigned to ship rents and dues to the central court as part of their duties to their estate.4 These sailors and navigators were often given tax-exempt paddy lands in return for their services, which demonstrates the value placed on their specialized skills.5 Fur- thermore, marine goods such as salt became an accepted alternative payment means for rents and dues, particularly in the well-documented case of Yugeshima Estate. The earliest references to Yugeshima Estate and to Onomichi as the “warehouse” port for Ōta Estate are found in records from the late Heian period. For instance, documents from 1135 show that an Iyo provincial governor, Fujiwara no Tadataka, requested that the retired monarch Toba (1103–1156; r. 1107–1123) exempt Yugeshima from taxation on salt fields as well as paddy and dry fields.6 Onomichi is first mentioned in 1168 as “five chō of paddy and dry fields of Onomichi-mura,” when it was designated as a harbor and warehouse area for Ōta Estate.7 From the first, these two venues fulfilled quite different roles. Yugeshima Estate was a production point, mostly for salt, while Onomichi was a transshipment hub, serving as the loading point for rents from Ōta Estate to be 354 Power, Space, and Trade sent to its proprietor, initially the retired monarch Go-Shirakawa, who ceded it to the monastic complex at Mount Kōya in 1186. An examination of the roles of the ports of Yuge and Onomichi can shed light on the relationship among estates, ports, and the entire Inland Sea shipping system. Throughout the Kamakura period and into the Nanbokuchō era, Yugeshima remained an important producer and provider of salt. The estate was passed from Go-Shirakawa to his daughter Sen’yōmon-in, and finally in the mid-thirteenth to the mid-fourteenth century it was held by different groups within the Tōji temple community.8 Although rents were officially assessed on lands that were supposed to furnish agricultural products such as wheat and mulberry, in actuality the fees were paid in Yugeshima’s major product: salt. Indeed, in land surveys of the early fourteenth century, the dues were assessed directly as salt.9 Tōji placed a high value on the salt from Yugeshima, particularly for its use in purification rituals, and it claimed that shipments of salt should be exempted from checkpoint tolls due for that reason.10 Onomichi, for its part, occupied a key position for the shipping of goods to the Kinai region, primarily from Ōta Estate. Initially, however, Onomichi was also appreciated for its strategic position. The Tachibana family probably first opened Ōta Estate in the 1160s. One monk of Mount Kōya, mindful of earlier Tachibana-led uprisings, wanted the port to be physically separate from the estate to keep the estate dependent and better exert the temple’s control through the port.11 Rents would have been brought by horse from the landlocked estate to Onomichi and loaded there for transmission to Mount Kōya, possibly through the port at Sakai (Settsu Province).12 We know, for instance, that from 1278 to 1302 Enshin, assigned by Mount Kōya as Ōta Estate’s custodian (azukaridokoro), reportedly manipulated the assign- ment of oarsmen and sailors in the area to enhance his own power, which suggests the importance of controlling those skilled laborers.13 As time went on, collaboration between shippers from different port towns developed: there are suggestions in the mid-fifteenth century record that Onomichi shippers sometimes hired out boats from adjacent areas such as the port at Tomo to help with the transport of rents and dues.14 Other areas, including Yugeshima Estate, also borrowed ships from Aki and Bingo Provinces. Given that Onomichi was a major shipping center, it is highly likely that Onomichi itself was a source for those vessels.15 As Estates Faded 355

In the early Muromachi period (1336 onwards) when the Ashikaga shoguns dominated politics in the capital, Onomichi and Ōta Estate came under the jurisdiction of their military governor (shugo) from the Yamana family.16 (See appendix 2 for a list of military governors in the mid-fifteenth century.) Private trade had increased by that point, due in part to Onomichi’s naturally protected harbor and deep water that could accommodate larger ships, even the great vessels used for the China trade. Onomichi thus became a hub for official and private shipping for both domestic and overseas trade. An early fifteenth- century diary from a Korean envoy details his impressions of Onomichi: it recounts throngs of people standing at a market in front of Tenneiji temple while shouting out prices for their wares.17 In contrast, Yugeshi- ma’s economic development was largely internal, shifting from the use of local salt to cash as payment for rents and dues. While we are fortunate to have documentation about both Yugeshima and Onomichi, many of the ports and coastal estates along the Inland Sea do not appear as frequently in the written record. In many cases few if any documents provide detail about shipping practices or goods produced, particularly in some of the smaller locales more removed from the political center. The examples of Yugeshima and Onomichi nevertheless supply clues for positing the development of other ports along the Inland Sea and their local networks. These examples also suggest possible growth patterns for less-well-documented areas, particularly in the later medieval period when the traditional estate system organization was breaking apart. The following sections will demonstrate ways in which alternate forms of analysis, such as considering geographic ties and trade routes, can shed light on ports for which there is little or no historical record. In particular, lateral ties among cultivators and noncultivators become apparent in the transfer of trade goods in the coastal regions along the Inland Sea. Growing independence at the local level encouraged interactions with other coastal communities and individual seafarers, reflected in the increase of trade goods production and collab- orative shipping efforts outlined below. Not all the ports can be directly associated with a particular estate, as in Onomichi’s case (and other ports in Appendix 1), and as we have seen, Onomichi itself gradually grew beyond exclusive affiliation with Ōta Estate. Through examining the connections of people and products in the later medieval period, the continuous growth of lateral connections and, subsequently, local 356 Power, Space, and Trade economic development become clearer. It is yet another aspect of the transfer of power, both administrative and economic, from traditional estate proprietors and their hierarchy of operation to local residents who began to exercise more agency.

Studying Inland Sea Trade Practices: Sources and Methodology A key document for understanding the volume and types of maritime shipping during the late medieval period is the Hyōgo Kitaseki irifune nōchō (Register of incoming ships at the Hyōgo north gate; below, Regis- ter). Dated to 1445, this record has nearly two thousand entries on ships passing through the northern checkpoint (sekisho) at Hyōgo in Harima Province. The checkpoint system itself, which dated from ancient times, was established as a means for central powers to maintain their authority by monitoring travel along major roads and by stationing military troops. During the medieval period, however, checkpoints also took on an economic role by levying fees on those who passed through. For instance, the Hyōgo checkpoint was at the site of the port of Ōwada-no- tomari, established by Taira no Kiyomori in the late twelfth century as a safe mooring point both for ships coming from domestic ports and for trade vessels from China.18 Later, in 1308, the director (bettō) of Tōdaiji temple petitioned the court to establish a checkpoint there, with fees to be used for both rebuilding the port and maintaining the Hachiman Shrine at the Nara temple. The retired monarch Fushimi (1265–1317, r. 1287–1298) agreed to the request, and Tōdaiji was awarded the right to administer the northern checkpoint.19 In later years the fees collected were used to build a new pagoda and to repair walls within Tōdaiji’s precincts.20 Tōdaiji assigned a manager (zasshō) and several magistrates (bugyō) to the checkpoint to oversee inspections and collect the fees, ensuring a constant Tōdaiji presence there.21 Meanwhile there was also a southern checkpoint at Hyōgo that was administered by Kōfukuji, but since no similar port records remain, it is impossible to compare the operations of the two checkpoints in detail. The records from the northern checkpoint nevertheless provide a wealth of information about the types of goods brought from the far reaches of the Inland Sea via Hyōgo in 1445. Specifically, each entry in the Register noted the port of registration for the ship (not necessarily the ship’s port of origin), the quantities and types of cargo, fees levied on the cargo, the name of the captain, As Estates Faded 357 and the name of the warehouse manager (toimaru) who received the goods.22 The ports of registration ranged from near the checkpoint at Hyōgo itself to as far east as Sakai, as far south as Maehama (Tosa Prov- ince), and as far west as Shimonoseki (Nagato Province), thus encompassing the length and breadth of the Inland Sea. Many, though not all, of the ports had the same names as a corresponding estate. The items shipped were varied, including marine goods such as fish and seaweed, salt, rice, beans, indigo, grasses, and iron. A superficial analysis reveals only the flow of goods from the different ports to Hyōgo, gateway to the Kinai region. A more in-depth geographical analysis, however, provides more information about the flow of goods from place to place along the Inland Sea, particularly in entries of specific types of salts or rice that were often named after their points of origin (e.g., “Aga salt,” “Sanuki rice”). Furthermore, looking in still greater detail at production points for each of the items provides further clues to interactions among communities. Analysis of the Register reveals much information about the types and volumes of goods that were shipped from place to place, the people involved, and the practices common to mariners throughout the Inland Sea. While estates are only infrequently directly referenced in this record of 1445, some of the goods that passed through Hyōgo would have come from the few remaining estates still operating in western Japan. Using a geographic information system (GIS) to analyze the data from the Register is particularly useful for revealing patterns in the movement of people and objects.23 While most studies of estates rely on documentary or archaeological evidence, incorporating geospatial analysis provides new information on interactions among various places. The maps below will show fewer details about specific estates. Rather they demonstrate shipping trends throughout the Inland Sea, such as the high level of trade interactions independent of an overarch- ing authority—the court or a proprietor—by the mid-fifteenth century.

Trade across the Inland Sea: Transshipment Ports and Patterns The quantity of goods being shipped along the Inland Sea that is reflected in the 1445 Register is truly staggering to consider. In the late fourteenth and early fifteenth centuries, the ability of individuals to make their own profits increased with the growing productivity of the land, the development of a cash economy, and the availability of a 358 Power, Space, and Trade

growing surplus.24 New developments in agriculture included double- and triple-cropping of paddies, better fertilizing and irrigation techniques, and a wider use of livestock, all resulting in greater growth that was reflected in the quantity of trade goods shipped to Hyōgo.25 Over the span of thirteen months, from 1445 to the beginning of 1446, nearly two thousand ships from more than a hundred different ports carried quantities of goods measuring from less than one hundred koku to more than one thousand koku to Hyōgo.26 Literally thousands of pounds of salt, rice, marine goods, and beans were brought to the Kinai region. Since the port of registry for the ships was not necessarily the port of origin of that ship’s voyage, it cannot be taken for granted that all the cargo was loaded at the port associated with the ship. Items produced at one place could have been transshipped through a port associated with an entirely different place. By looking at traceable cargoes, however, the origins and paths of dissemination of certain goods become clear and help determine major ports for transshipment. A note of caution, however: While it seems a safe assumption that ports shipping the largest amount of goods to Hyōgo would also be the most important transshipment hubs, that was not always the case. As we trace these trade routes, we can see the growing lateral connections, regional affiliations, cross- provincial interactions, and transshipment practices that characterized the region as the estate system weakened. The product brought to Hyōgo most frequently from within the Inland Sea region was salt (see plate 25). The propensity to record the production point of the salt in the Register (“Aga salt,” “Mihara salt,” “Bingo salt,” and the like) greatly helps in identifying the different routes that salt took to Hyōgo. Even though the map shown here reflects a large amount of data, the overall color distribution at first glance suggests that salt by and large was going directly to Hyōgo from nearby its point of production. Suō salt came from the ports of Tomita, Yanai, and Kaminoseki, all in Suō Province. Bingo salt in general was shipped from Bingo and Aki ports, Aga salt from the Harima ports near Aga, Kojima salt from the ports surrounding Kojima on the Bizen peninsula, and so forth. Even the naming trends for salt provide some insight into regional specialization and the prominence of certain locales and estates. Most names given to salt denote a wide swath of land, often at the provincial level (Bingo salt, Suō salt) or regional level (Aga salt, Mihara salt). Notably very few are estate-specific. Even knowing, for example, the prominence of salt production in Yugeshima Estate as As Estates Faded 359 outlined above, “Yuge salt” is not recorded specifically in the Register. Instead, ships from Yuge port are recorded as having brought Bingo salt to Hyōgo. Based on this we can say that by 1445 the importance of the estate had paled in comparison to regional affiliation, another indicator of the breakdown of the estate system. Removing from the analysis ports that shipped less than five hundred koku of salt to Hyōgo over the year reveals more detail about the flow of salt throughout the Inland Sea (see plate 26). It makes clear that the western Inland Sea area was a central salt production and shipment area. Onomichi, Yuge, Setoda (Aki Province, today part of Onomichi City), and Tajima (Bingo Province) ports each shipped more than 3,500 koku of Bingo salt alone. Though it is unclear if Onomichi, Tajima, and other Bingo Province ports produced the salt near the port area or purchased it from other local providers, the quantity of Bingo salt shipments from those ports certainly implies that salt was produced locally. In Yugeshima’s case, as described above, it is extremely likely that much of this salt originated as surplus initially produced to pay rents. As society shifted to a more cash-based economy, that surplus would have been sold elsewhere for cash, leading to the production of more salt for trade purposes.27 It is telling, however, that although Yuge port was in Iyo Province, its ships were registered as carrying only Bingo salt. The large volumes of Bingo salt from Yuge suggests either that the “brand name” of Bingo extended to salt produced in Yugeshima of Iyo Province, or alternatively that Yuge port effectively imported salt from Bingo Prov- ince to ship as its own. Regardless, either case demonstrates significant cross-province interactions or influence. Bingo and Aki ports shipped other products in addition to Bingo salt. One such example is Takasaki in Aki Province. Two centuries earlier it was listed in a survey as being part of the New Nuta Estate in 1243, where the Kobayakawa warrior family were assigned as military stewards sometime after 1221. It was still part of that warrior family’s holdings in 1491, when another inventory was made.28 Nevertheless in comparison to neighboring ports at Onomichi and Setoda, Takasaki was not a major trade port according to the 1445 Register. It shipped only approximately 5,200 koku of goods to Hyōgo throughout that year, paying about 25,500 coins in toll fees levied at the checkpoint.29 In contrast, Setoda sent nearly 18,300 koku of goods and paid fees of 122,770 coins, while Onomichi shipped approximately 12,700 koku and paid fees of 78,275 coins. Yet even though Setoda and Onomichi were busier ports, 360 Power, Space, and Trade

Takasaki played an important role in Inland Sea travel. For instance on his pilgrimage to Itsukushima in 1389, Shogun Ashikaga Yoshimitsu (1358–1408) spent the night in Takasaki, as did a vessel with Korean envoys en route to the capital in 1420. In 1451, a China trade ship employed sailors based at Takasaki harbor.30 So despite Takasaki’s overall lack of domestic trade destined for Hyōgo, it was still a key stopping point for ships traveling long distances. This role helps explain one standout aspect of Takasaki trade goods seen in the Register. Unlike even its neighboring ports, Setoda and Onomichi, which handled great quantities of trade goods, Takasaki was a transshipment point for Suō salt. Though all other Suō salt in the Register was shipped from ports in Suō Province (Kaminoseki, Yanai, and Tomita), Takasaki ships brought six hundred koku of Suō salt to Hyōgo, more than was shipped even from Suō ports. Takasaki lacked official ties with the Suō ports that could have resulted in deeper trade relationships, yet the shipment trends demonstrate that unofficial ties existed. Moreover, no ports east of Takasaki shipped Suō salt; Takasaki first became an initial destination for Suō salt and then sent it on to Hyōgo. I will show why below, when I discuss Takasaki’s eastern neighbor port of Kamagari. Transshipment trends are seen elsewhere as well. Hiketa port, in Sanuki Province on Shikoku, was once a township in the royal Ōuchi Estate in the thirteenth century but was under the jurisdiction of the Hosokawa shogunal deputy (kanrei) in Muromachi times. More salt from its neighbor Katamoto was shipped out of Hiketa—810 koku— than Hiketa’s own salt (only 566 koku).31 On the eastern edges of the Inland Sea, Amagasaki also sent diverse types of salt to Hyōgo. With 469 koku of Mihara salt, 160 of Bingo salt, 100 of Shima salt, 55 of Aga salt, and 45 of Higashiyama salt (referring to salt mostly distributed from points scattered in Harima and Settsu), most of the salt being shipped on Amagasaki ships to Hyōgo did not come from areas on the route between the two ports. Either the ships left Amagasaki and visited other areas as part of their regular trade routes, picking up those cargoes of salt and then proceeding to Hyōgo before returning to their home port; or other places sent their salt to Amagasaki and Amagasaki-registered ships then brought it to Hyōgo. The latter possibility may have been an attempt to avoid middleman fees, as I will discuss below. Either choice indicates a high level of interaction regardless of geographic location. Similar to the cases of Takasaki and the Suō ports above, the lands affili- As Estates Faded 361 ated with those likely production points shared no common provinces or estate proprietors, implying direct trade between locales without any involvement of central authority. Perhaps the most unusual transshipment trend was embodied in Tsurajima (Bizen Province, possibly developed as part of Tōji’s Tsura- jima Estate). Its salts came from the Kojima area in Bizen Province (935 koku), on the same peninsula where it was located, from Bingo (140 koku), and from Shiwaku (380 koku) across the strait in Sanuki. But by far the greatest volume of the salts shipped from Tsurajima to Hyōgo (2,280 koku) was from Shima (Sanuki Province), a port on the island now known as Shōdoshima, between Tsurajima and Hyōgo. The province was under Hosokawa military governors in Muromachi times. For Tsurajima ships to carry Shima salt, they would have needed to stop in Shima en route to Hyōgo or have Shima salts shipped to Tsurajima, in the opposite direction from Hyōgo. Regardless of which route was taken, this suggests a high volume of direct interregional trade and contact. In this case, however, it is possible that since both Shima and Tsura- jima were connected through the Hosokawa family governance, lateral trade may have been facilitated through official channels. In comparing rice shipments with salt, particularly with the case studies of Onomichi and Yuge, certain specialization trends become evident. Rice was sometimes recorded in the Register with a name that reflected a certain style of baling rather than a production point.32 Unlike the brand name of “Bingo salt,” which was shipped from both ports, the majority of rice shipped from Onomichi was unspecified, with no indication of its production point (see plate 27). It is likely, however, that the rice came from Ōta Estate and other nearby areas that used the port at Onomichi not only to ship their rents and dues but also to send their surplus rice for commercial transactions. Yuge, on the other hand, shipped no rice at all to Hyōgo. Yuge’s profits and access to cash came from Yugeshima’s salt production. Similar to the trends noted in the salt shipments, the greatest revelations from plate 28 come from examining the major ports that handled multiple baling styles of rice. Places that sent a wide variety of rice to Hyōgo are more likely to have been transshipment ports. Traveling from west to east, the first identifiable port handling more than two styles of rice is Kamagari (Aki Province, part of Hidaka Estate). Throughout the year it sent 700 koku of Sanuki rice, 410 koku of Bungo rice, 20 koku of Buzen rice, and 70 koku of unspecified rice to Hyōgo. It is the only port 362 Power, Space, and Trade that shipped rice to Hyōgo from both Buzen and Bungo in Kyushu. Even the ports in the Register that were closer to Kyushu, such as Shi- monoseki and Moji, did not send Kyushu rice. This suggests that similar to Takasaki’s role in salt shipping, Kamagari became a transshipment port, receiving rice from Bungo and Buzen and then in turn shipping it on their own vessels to Hyōgo. Notable too, Kamagari shared no official ties with estates or ports in Suō that may have helped facilitate that trade, which suggests the presence of direct unofficial ties between cargo owners. One likely reason the relatively small ports of Takasaki and Kamagari took on major roles in transshipment was their geographic locations. Takasaki and Kamagari may have been relatively safe stopping-off points for western captains since further east there were many smaller islands that posed a challenge to navigation and in later years became pirate enclaves.33 Captains coming from western Honshu and Kyushu may have offloaded their goods at Takasaki and Kamagari, the captains of which would have been familiar with the more treacherous eastern waters. The comparison with Takasaki, a port that shipped an annual amount of cargo to Hyōgo similar to that from Kamagari, also suggests the possibility of specialization. Even though Kamagari was a more westerly port than Takasaki and would have therefore likely been a convenient transshipment point for Suō salt as well, there are no records of Suō salt coming from Kamagari. Instead Suō salt bypassed Kamagari for the slightly more eastern port of Takasaki and progressed from there to Hyōgo. Conversely, if Takasaki had been the region’s major transshipment hub for goods from westernmost Japan, the rice from Bungo and Buzen would likely have skipped Kamagari in favor of Takasaki. Since this was not the case, it seems likely that the different ports specialized in transshipping particular goods. Additional research is required to understand more about the mechanics of specialization and possible ties to estates. Boats labeled “jige” (“local”) in the Register were a special case, and they highlight the authority of the ships’ captains. This notation refers to boats that originated from nearby Hyōgo and claimed it as their home port. The majority of ships passing through the checkpoint throughout the year were registered as local, and as a result, the volume and value of the goods from local ships were the highest of those from all ports in the Register. It is possible that local ship captains voyaged to other areas and picked up trade goods to bring back through the checkpoint. But As Estates Faded 363 it is also likely that other captains brought their wares to trade directly with local captains, who would then bring those goods headed to the inland markets through the Hyōgo checkpoint. Examining the timing of the ships manned by Captain Awaya/Sae- mon Gorō34 and the goods he transported through the checkpoint as one example, it becomes apparent that he could not have personally gathered each type of good at its production point. He passed through Hyōgo on the fifth day of the fourth month with salt from Mihara, and then again four days later with salt from Bingo. The distance from Hyōgo to the Bingo area is approximately two hundred kilometers; with the sailing technology available at that time, such a round-trip journey would have been close to impossible. It is therefore far more likely that Saemon Gorō brought in another person’s shipment of Bingo salt, either from a transshipment port or from a Bingo Province captain who brought him his own product for trade. The reasons for such a large amount of local trading nearby Hyōgo are not totally clear, but one hypothesis is that traders wanted to elimi- nate the role of warehouse manager. In the Register, few of the entries for ships from local ports (most notably those labeled local, Amaga- saki, or Kuise) include the name of a warehouse manager. So it is possible that the local captain acted as his own agent,35 enabling him to bypass a middleman and avoid an additional service fee. Similarly, captains from more distant ports might have utilized local connections and traded their goods to local captains at a fair price, offloading their wares without needing to pass through the checkpoint. Ship captains had a widespread maritime network that encouraged collaboration among many parties, and unofficial local trade may have been a manifestation of those practices.36 This is additional evidence that suggests how, as the estate system weakened, individuals were able to work for their own profit rather than answering to the demands of an estate proprietor. Trends visible through geographical tracking of trade goods recorded in the Register suggest that transshipment and interregional trade for goods was a relatively common practice in 1445. Even an overview analysis of salt and rice shipping patterns demonstrates that the products made their way from points of origin to areas throughout the Inland Sea, and it suggests direct trade between the various coastal communities. Estate affiliations are not noted in the record, suggesting that they were less important by the mid-fifteenth century.37 364 Power, Space, and Trade

Commodities vs. Rents and Dues: A Return to Yugeshima and Onomichi The above analysis shows the importance of lateral trade routes within the Inland Sea region by the mid-fifteenth century, and it emphasizes the connections among various local entities. It is also worth noting some of the differences between shipping of commodities for trade and shipping rents and dues from estates to the central proprietors, even if these become harder to evaluate, in the later medieval period. As discussed above, by the early fourteenth century Yugeshima’s rents and dues had come to be assessed directly in salt, signifying a recognition of the value of salt in its own right rather than as a substitute for other products. In another case, in examining the rents and dues paid from Ōta Estate to its proprietor Mount Kōya, however, the gap between production of goods as commodities and for rent payment becomes more apparent. These were additional aspects of the breakdown of the estate system. Specifically, Tōji, Yugeshima Estate’s proprietor, had long been opposed to the commodification of salt: as noted above, from 1305 the temple repeatedly issued orders that salt should be exempt from taxes or tolls due to its ritual value, based on its use in religious ceremonies. Nevertheless Yuge ship captains had begun to recognize the potential profits in selling salt and made use of warehouse connections, presumably to increase their own individual profit. In at least one early case from 1291, a warehouse manager took some Yugeshima salt and sold it to one merchant for two hundred coins, only to find that the merchant later sold it at double the price to a Kyoto merchant for a great increase in profit.38 The increasing commodification of goods and profits that could result provided great incentive for ship captains to claim encounters with “pirates” or “storms” that prevented them from paying their rents to Tōji. As they came to understand the situation, the ruling body of monks in turn issued a directive that even if a ship encountered such disasters, Yugeshima Estate still needed to submit its dues.39 The tension between proprietor and local estate personnel is visible in this scenario, in which local individuals sought to prioritize the potential for their own profits at the expense of the estate proprietor. Fifteenth- century documents regarding Yugeshima’s connections with Tōji are scant, although it is known that the temple was trying to rearrange some payment of rents under the auspices of the Murakami sea lords even in As Estates Faded 365 the 1450s and 1460s.40 What is clear, however, is that the amount of salt (3,513 koku) shipped from Yuge port in 1445 confirms that Yugeshima remained an important source of salt production, much of it for commercial distribution, in 1445. Circumstances were different at Ōta Estate. An undated document compiling a list of eight years of unsubmitted rents and dues between 1439 and 1447 provides insight into the expected contributions from Ōta to Mount Kōya. Given that this series is roughly contemporary with the 1445 Hyōgo Register, it provides an excellent comparison of the types of goods required for submission as dues and those shipped as commodities. The documents, usually denoted as being submitted from Sakai, gateway port to Mount Kōya, listed the amounts and types of cargo shipped as well as the names of the ships’ captains who finally transported the rents and dues. In most cases the captains are also affiliated with a particular location. Eighty percent of the captains came from Onomichi, long known as the warehouse port for Ōta Estate, and many of the others came from the nearby ports of Mihara, Innoshima, or Tajima. Furthermore, by 1445 all those locations were under the control of the Yamana military governor, which indicates the potential for some coordination under the auspices of a common authority for shipping the dues to Mount Kōya. While the organization of the documents is not totally clear, the process of rent payment to the estate proprietor was nevertheless tortu- ous. The submission of rents and dues appears to have been delayed from three months to as long as one year in some cases, and in several instances there appears to have been a gradual repayment of a shipment that had been lost at sea. With no supplemental documentation, it is difficult to determine if this was truly a loss or if, as in the case of the Yuge warehouse manager cited above, it was a form of embezzle- ment for the captain’s or manager’s own benefit. Regardless, the overall tendency to delay the shipment of dues indicates a relatively weak hold at this point over Ōta Estate by Mount Kōya. While dues were eventually paid, doing so in a timely manner was no longer a given. Notable too is the major difference in terms of the variety of goods being shipped as rent, according to the Kōyasan document, and those shipped as commodities and recorded in the Register. Rents for Mount Kōya consisted of rice, soybeans, and small amounts of cloth and egoma (perilla) berries, which were used to make lamp oil. On average, Ōta Estate shipped approximately three hundred koku of rice, seventy-five 366 Power, Space, and Trade koku of soybeans, two hundred koku of mixed rice and soybeans, and only 0.1 koku of egoma and 1.75 bolts of cloth per year to Kōya, totaling slightly under six hundred koku of all goods submitted as payment of rents and dues. On the other hand, as we see in table 14.1, commodities shipped from Onomichi to Hyōgo included a significant variety of goods. Onomichi’s longtime role as Ōta Estate’s warehouse suggests that Ōta Estate was likely the source of many of these products, though it is possible that some came from cultivators in neighboring lands as well. In particular, the large quantity of rice and beans shipped as commodities suggests that unless 1445 was a bumper crop year, the late submission of rents and dues to Mount Kōya was not necessarily due to crop failure or similar problems. Instead, it seems likely that indifference to timely submittals may have been the culprit, which again suggests a weakening authority of Mount Kōya’s proprietorial power over Ōta Estate as its holding by 1445. In the cases of Yugeshima and Ōta Estates, then, it appears that during the fourteenth and fifteenth centuries the tendency to find ways to circumvent timely payment of rents and dues increased. It is possible that the lateral ties between ports and estates helped to contribute to that tendency. That is to say, as the surplus of goods increased, so did the potential to trade that surplus for other goods or for cash. Ships’ captains and cargo owners would thus have sought out other markets in neighboring regions, leading to more connections with people and more demand for locally produced goods. Ultimately the realization that more profit could be made by selling the goods, combined with decreasing local presence of a proprietor’s representative, would have further contributed to the breakdown of ties between estates and proprietors.

The relative autonomy of seafaring and maritime trade in the late medieval period is apparent. In the mid-fifteenth century it is no longer necessarily clear which port belonged to which estate, which itself suggests a tendency toward autonomous local activity. Furthermore, the flow of trade goods shows little reliance on administrative bonds such as those of provincial governance or estate proprietorship. The late medieval trend of increased autonomy and collaboration that grew in response to the breakdown of the estate system is reflected in the voyages of mariners: ships’ captains and their trade routes. As Estates Faded 367

Table 14.1 All goods shipped on Onomichi boats to Hyōgo, 1445

Small sardines 50 horseloads Red sardines 140 horseloads Lumber 10 koku Pine 20 koku Magnolia 35 koku Rice (all types) 1680 koku Bingo salt 9410 koku Egoma 18 koku Woven mats 200 mats Iron 100 horseloads Soybeans 40 koku Azuki beans 10 koku Beans (all other types) 1030 koku

Tracing the flow of commodities also indicates interactions among different places. If the ships’ captains rendezvoused at certain ports, it is likely that goods would have been traded there before ever arriving in Hyōgo. Evidence from the salt and rice records in Hyōgo, particularly when compared with the records of submission of rents and dues from Yugeshima and Ōta Estates, certainly implies a robust trade. As the multilayered hierarchies of the estate system broke down, lateral ties between communities and individuals gained in importance. In the mid-fifteenth century there does not seem to have been a strong correlation between estate proprietors or overseers and the flow of commodities. While a common proprietor might in theory have found it easier to compel collaboration between local communities and visiting ships, or to have captains from neighboring estates voyage together, most of the examples above had no shared authorities. And as the number of fully held lands increased, it appears that local communities and their residents carried on personal trade, such as in the case of the Yuge warehouse manager appropriating salt for his own profit. There are few official indications that the local proprietors of fully held lands obtained a profit from maritime-based trade. As in the case of Ōta Estate and its late submission of its dues, however, it is possible that the relaxed attitudes toward dues submission may have been the result of 368 Power, Space, and Trade local representatives acting for their own benefit rather than for that of the absentee proprietor. In addition, some entries in the Register record exemptions for the checkpoint fees based on the status of the local lord or an affiliate temple, most notably in ships registered to certain ports affiliated with the Yamana or Hosokawa warrior families. Those are far more of an exception than the rule, however, and overall trends suggest that maritime trade was conducted relatively independently from warrior lords. The old tripartite oligarchic world dominated by courtiers, warriors, and clerics had crumbled. To conclude, by the mid-fifteenth century maritime connections among different locales drove the flow of commodities without relying on central governance and hierarchies like those of the shiki system. Mar- itime shipping was obviously tremendously important in transporting quantities of goods from island to island, and sea-based travel and trade transcended political or administrative boundaries. The strengthening of local connections likely benefited from the ambiguous boundaries of the time, and, in turn, contributed to their dynamic deterioration. As local connections increased and the potential for profits became apparent, so did the tendency to evade submitting rents and dues to central authorities. Continued research in these matters will shed more light on the relationship between local economic activity and the decline of the estate system.

Appendix 1: Selected Ports and their Administrative Histories through Time Hiketa

Sanuki Province. In the thirteenth century Hiketa was a township within Ōuchi Estate. In the later medieval period it was under the jurisdiction of the Hosokawa shogunal deputy, and as of 1369, the Hosokawa retainer Yasutomi oversaw half of the estate (Amino 2005, 36).

Kaifu

Awa Province. It is unclear exactly which estate Kaifu may have been affiliated with, though it was certainly within the confines of Kaifu Dis- trict, which included the royal Shishikui Estate. The latter first appears in the record in 1191 and later passed to the proprietorship of Mount As Estates Faded 369

Kōya, which held it into the fifteenth century. From the Nanbokuchō through Sengoku periods, Kaifu was overseen by the Kaifu family, working in conjunction with the Hosokawa military governor (Nihon shōen shiryō 1998, 778; Nihon rekishi chimei taikei 2000, 37:687).

Kamagari Aki Province. Part of Kōfukuji’s Hidaka Estate that appears in the record in the later twelfth century, Kamagari was overseen by the Tagadani family during the Nanbokuchō era, when they were appointed its overseers in 1351 by a Prince Hitachi (Nihon shōen shiryō 1998, 718; Amino 1999, 229).

Kaminoseki Suō Province. Originally known as Kamadoseki in Heian times, it was part of an estate for which the Kamobetsuden Shrine was proprietor. It was under the power of the Murakami Genji “sea lords” in Muromachi times. Chinese travel diaries from the fifteenth century note the presence of the Ōuchi family there. That family held the military governorship of Suō from Kamakura times onward, and they became a dominant force in both domestic and foreign trade in the Muromachi period, until they were destroyed by the Mōri warlords in Sengoku times (Nihon rekishi chimei taikei 1980, 36:161; Arnesen 1979; Shapinsky 2014).

Kannoura Tosa Province. Known from early Heian times. In 1207 the holy man Hōnen landed there on his way to his place of exile, so it was a port for the province at that time. No documentary evidence remains specifically about Kannoura in the medieval period. In later medieval times the area seems to have been called “Asama Estate,” and it appears in a Chōsokabe record of the mid-sixteenth century under that name (Nihon rekishi chimei taikei 1983, 40:48).

Shima Sanuki Province. The island of Shōdoshima encompassed several estates, including Shōdoshima, Ikeda, Kusakabe, and Hido. It is unclear where 370 Power, Space, and Trade the port designated “Shima” was located, so it is difficult to ascertain which estate it may have been affiliated with; but in the Nanbokuchō era most of the island was administered by the Hosokawa military governor of Sanuki. In the Muromachi era, dues from Kusakabe and Hido Estates were divided between the retainer appointed by the Hosokawa family, Yano Tokuhisa, and Chōkōji Temple (Amino 1999, 32–33).

Shishikui Awa Province. Shishikui port developed almost certainly within the boundaries of Shishikui Estate, a Mount Kōya holding in the thirteenth and fourteenth centuries. In the fifteenth century it came under control of the Hosokawa military governor (Amino 2005, 31–32).

Tachibana Awa Province. Tachibana may have been part of Kuwano Township, which has few records associated with it. In the late Kamakura period its steward was a member of the Yamada family, thought to have been associated with the Yamada of Owari. Records from 1334 and beyond refer to a “Kuwano mikuriya,” an estate held by Ise Shrine (Nihon shōen shiryō 1998, 777; Amino 2005, 30).

Takasaki Aki Province. Part of New Nuta Estate in the thirteenth century. Since the Jōkyū Disturbance of 1221, the New Nuta Estate had had military stewards from the Kobayakawa warrior family. After the Ōnin War (1467– 1477), they were still there (Nihon rekishi chimei taikei 1982, 35:462).

Tomita (also read Tonda) Suō Province. Tonda was almost certainly affiliated with Tonda Town- ship and was originally overseen by a military land steward, perhaps from the Ōuchi family. Tonda-no-ho was ceded to Tōdaiji in 1334—the temple held the province as its chigyōkoku, which meant that it received taxes from it—and the Ōuchi were military governors there. In the fifteenth century they obtained greater control over some of the public lands in the Tonda area (Amino 1999, 306–307). As Estates Faded 371

Tsurajima

Bizen Province. Little documentation is available for Tsurajima, but a record from 1418 indicates it may have been part of Tsurajima Estate in Kojima District. By Muromachi times, the Hosokawa military governor gained control of what may have been a Tōji estate (Nihon rekishi chimei taikei 1988, 34:718).

Ushimado

Bizen Province. While there are few detailed early records about Ushi- mado that provide information about its direct oversight, it was referred to as Ushimado Township in late medieval documents related to Hon- renji Temple. It was widely referred to in travel diaries as a stopping-off point for those on trade and pilgrimage voyages, suggesting a relatively public nature (Amino 1999, 16–17).

Yanai

Suō Province. Yanai port was almost certainly within the royal Yanai Estate, which had Go-Shirakawa’s vow temple Rengeōin as its proprietor in the thirteenth century. In the Kamakura era a military land steward was appointed by the shogunate at Yanai Estate, but in the Muromachi period, Takayama- dera took over its administration. Given Tōdaiji’s long authority over the province during medieval times, however, its affiliation came into dispute in the mid-fifteenth century. After the Ōnin War, the Ōuchi family of military governors took control (Nihon shōen shiryō 1998, 733; Amino 1999, 302).

Yura

Awaji Province. Though records are sporadic, Yura port almost certainly developed as part of Yura Estate, which was established with Taira Yori mori as its original holder in late Heian times. Later it was held by the New Kumano Shrine, Zenrinji, and the royal house. In 1394 the estate and its port were overseen by an administrator appointed by the shogunate, and in 1476 a percentage of its rents and dues was given to Kibune Shrine in Kyoto. In general Yura was still affiliated more with central powers than with local administrators (Nihon shōen shiryō 1998, 772; Nihon rekishi chimei taikei 2001, 29.1:611). 372 Power, Space, and Trade

Appendix 2: Military (Provincial) Governors in 1445 Aki Province: Yamana Mochitoyo (1433–1454) Awa Province: Hosokawa Mochitsune (1430–1449) Awaji Province: Exact appointee unknown; under Hosokawa governance Bingo Province: Yamana Mochitoyo (1433–1454) Bizen Province: Yamana Noriyuki (1441–1446) Iyo Province: Exact appointee unknown; under Kōno governance Sanuki Province: Hosokawa Katsumoto (1442–1473) Suō Province: Ōuchi Norihiro (1441–1465) Tosa Province: Hosokawa Katsumoto (1442–1473)

Notes

1. See the chapters by Joan Piggott (chapter 1) and Noda Taizō in this volume. 2. Nishitani 2011, 27–28; Shimizu 2013, 154–158; Nagahara 1998, 239–241. 3. Amino and Inaba 2007, 37. 4. Shinjō 1994, 40–41. 5. Ibid., 71. 6. Yamauchi 1985, 12–14. In English see Shapinsky 2014, 71–104. 7. KM 1968, 1 (# 37). 8. Yamauchi 1985, 17–20 and 94–96. 9. Shapinsky 2014, 74. Other products regularly delivered to proprietors over the thirteenth and fourteenth centuries included seaweed, whitefish, hemp, and timber. 10. Ibid., 81. 11. Matsushita 1994, 188. 12. Aoki 1971, 1:477. Records of a shipment of rents show that the goods were shipped to Sakai, where they were exchanged for coins that were subsequently shipped to Kōyasan (Kongōbuji monjo 1936, 2:380–411 (# 165). This undated document describes the accumulation of heretofore unsubmitted rents from 1439 to 1447, and will be discussed in more detail below. 13. Shinjō 1994, 41. 14. Nagahara 1998, 173, from records of Onomichi’s shipments of rents and dues from 1439 to 1447. 15. KI 37:212–213 (# 28836). 16. Amino 1999, 186–187. 17. Rōshōdō Nihon kōroku 1968, 71. As Estates Faded 373

18. Kamiki 1996c, 38–41. 19. Kamiki 1996b, 108. 20. Hyōgo Kitaseki irifune nōchō 1981, 224–226. 21. Ōshima 1964, 47. Also see Hatakeyama 2003 for a detailed discussion of the complex relationship among Tōdaiji, the retired monarch Fushimi and his ties with Tōdaiji affiliates, and the various offices within the checkpoint complex. 22. For a reproduction of the Register, see Hyōgo Kitaseki irifune nōchō 1981, 3–218. 23. Kemp 2010, 31. GIS is an information science that helps collect, model, manage, display, and interpret geographic information. 24. Tonomura 1992, 86. 25. Farris 2006, 129–136. 26. While most maritime-related uses of the term koku refer to weight, as in the total displacement of the vessel (i.e., a thirty-koku ship, similar to the use of tonnage measurements today), it is unclear whether its use in the Register nota- tions are for volume or weight of bulk goods such as rice and salt. It does not appear to denote the value of the item, as that is indicated through calculations of the fees levied on the cargo and valued in coins. Further adding to the confu- sion, the volume of a koku fluctuates over time. Medieval-era koku may have only been the equivalent of about 116 liters of rice, while Tokugawa-period koku are commonly thought to have been 180 liters (Bassino et al. 2010, 4). If the medieval koku did not drastically fluctuate in volume, in 1445 when the Register was recorded one koku of rice would have weighed approximately 213 pounds and one koku of salt would have weighed approximately 211 pounds. 27. See the chapter by Sakurai Eiji in this volume for more information on the development of a medieval cash-based economy. 28. Nihon rekishi chimei taikei, 35:496. 29. Various scholars have examined the intricacies of payment at the Hyōgo checkpoint. See Shinjō 1994, 521; Takebayashi 1994, 2; and Yamatoya 2006, 5–10. It is unclear, however, whether coins actually exchanged hands or whether a value was simply placed on the cargo and then a portion of the cargo was turned over to pay the fees. The Register simply notes that the fees were assessed as numbers of mon coins, usually copper, but does not specify the method of payment. 30. Nihon rekishi chimei taikei, 35:462. 31. The first reference to Hiketa in the historical record dates from 1247 (KI 10:81–82 (# 7060). Also see KI 23:92–93 (# 17578) and 29:363–376 (# 22661), and Nihon rekishi chimei taikei, 38:65–66). Hiketa was in today’s Aioi City. An early reference to Ōuchi Estate is dated 1258. For clues to its history see Nihon shōen shiryō 1998, 784; Amino 2005, 36. 374 Power, Space, and Trade

32. Though “Sanuki rice,” for example, is being shipped to Hyōgo from as far away as Shimonoseki and Moji, the rice itself was not necessarily originally from Sanuki (Kamiki 1993, 45). Hanshō rice most often came from areas close to the Kinai, including ports such as Aboshi, Uozaki, Nakasho, Matsue, and Aga. The term hanshō possibly denoted a small receptacle originally used for hand washing that was used as a measure of volume in homes throughout Yamato, Yamashiro, and Harima Provinces (Takefuji 1981, 255). Meanwhile “Awaji” rice was centered in the ports on Awaji Island such as Mihara, Iwaya, Murotsu, Sumoto, and Takenoguchi. 33. The term “pirates” is problematic, since Peter Shapinsky demonstrates in his monograph that many of the seafarers on the Inland Sea were acting within the accepted social hierarchy to exert local power and should more accurately be called “sea lords” (Shapinsky 2014, 13–14). There are numerous instances of travelers, be they traders, travelers submitting official rents and dues, or pilgrims, referring to these seafarers in fearful or argumentative contexts. In such cases I use the term “pirates” to denote the trouble they caused for shipping. Their enclaves were built on the smaller islands offshore near Onomichi, including Innoshima, Nōshima, and Kurushima. 34. These records specify both names in each entry, possibly to differenti- ate Saemon Gorō affiliated with Awaya, likely representing a particular warehouse, from another captain with the same name. 35. Kobayashi 1981, 291. 36. Damian 2015, 222–238. Applying GIS to the captains of the ships mentioned in the Register as well as the cargo shows a strong correlation between captains from similar regions. This suggests the presence of a well-developed network among the captains in Inland Sea waters. 37. According to Shimizu Ryō (2013, 172), the assassination of Shogun Ashikaga Yoshinori in 1441 essentially marked an important terminal moment for central support of the estate system. See Joan Piggott, chapter 1 of this volume. 38. Shapinsky 2014, 80. Also see Nihon engyō taikei 1977, 1:178 (# 117). 39. Shapinsky 2014, 81. Also see Nihon engyō taikei 1977, 1:226 (# 150). 40. Amino 1978a, 324. pa rt v i

Power Challenges and Conflicts

c h a p t e r 15

Bad Neighbors and Monastic Influence Border Disputes in Medieval Kii

Philip Garrett

In 1241—the middle of the Kamakura period—a dispute broke out between Nate Estate and Niunoya Village in Kii Province (modern Kinokawa City, Wakayama Prefecture). On the face of it, this was a simple dispute between neighboring communities over the right to draw water from the stream that separated them and to harvest timber from the hill at the head of the stream. The dispute, however, grew into a conflict that drew in local, regional, and national authority; provoked judgments by the Kyoto and Kamakura governments; demanded military intervention by the military governor (shugo) both in the estate and at the proprietary temple; and resulted in destructive territorial warfare by local residents, warrior estate managers, and armies of monks. The fundamental cause of the dispute—control of the stream and hill— remained unsolved for more than two hundred years (see plate 29). Disputes such as this were a common cause of conflict in the thirteenth and fourteenth centuries, with numerous similar incidents occurring across Kii Province. Border disputes and water rights disputes arose from a combination of factors, both territorial and class-based, within and between estates. The Nate-Niunoya dispute cut through all levels of society, exposing class and regional tensions, as well as the basic competition for resources. Water rights—a fundamental necessity for an agricultural community—remained central to the conflict, but the fuel that sustained the enduring dispute was the social and political tension underpinning the balance of power in northern Kii. While local in origin, such disputes were linked to broader provincial and national politics, with warrior bands, proprietors, and patrons drawn into village conflicts for economic, political, or social gain.

377 378 Power

Examining the record of cross-border crime and border disputes (sōron) therefore allows us to gain an understanding of the local adjustments of real power and political influence in the provincial communities of medieval Japan, and the contemporary struggle for control of land, revenue, and people. These conflicts demonstrate the limitations of social control and prevention of crime within the estate system, as well as the difficulty of regulating interaction between different “private” organizations in a state dependent on indirect administration. More importantly, such disputes also demonstrate the process of the formalization of changing provincial power. Incidents such as the Nate-Niunoya dispute show that violence and direct military action were becoming part of the interaction between rival groups in the early medieval period, but that this violence was still constrained by a dependence on external legitimization. Low-level local violence was frequent, but it was within the established framework of authority and the legitimization of the law courts. This system linked village disputes to regional and national level arbitration, thus moderating the scope of violent disorder, a demonstration of the continued involvement of central government in provincial affairs. The limited ability of central legal authority to resolve disputes satisfactorily, however, and the reliance of estate proprietors on forceful occupation and violence to achieve their aims indicate the increasing regionalization of power in the early medieval period. Disputes are thus a window into the articulation of central and provincial power in medieval Japan and of the relationship between local, regional, and national politics in the period. The proprietors of Nate Estate and Niunoya Village were both nearby Buddhist temples. Temples and shrines were among the greatest landholders of late classical and medieval Japan, easily rivaling the extent of church and monastic land in medieval Europe. Together with the court nobility in Kyoto and the military government headquartered at Kama kura, religious institutions were one of the three key power blocs in the medieval Japanese state. Each group controlled extensive networks of land, patronage, subordinates, and legal jurisdictions, in what Kuroda Toshio termed the kenmon or “influence/authority” system.1 There was a complex relationship between the belligerent estates and their proprietors, in which the estate’s status was guaranteed by the legal influence of the elite proprietor in exchange for a share of its revenue via a series of intermediaries. Bad Neighbors and Monastic Influence 379

Border Disputes in Provincial Society The Nate-Niunoya dispute was one of many fought on the margins of the growing bloc of estates under the administration of the Shingon Bud- dhist temple complex at Mount Kōya in the early medieval period. Mount Kōya was a heterogeneous community of subtemples and halls among which the largest establishment was Kongōbuji, proprietor of Nate Estate. Kongōbuji vigorously expanded its estate holdings in the immediate area surrounding the temple in the early medieval period, frequently bringing it into conflict with other temples and aristocratic proprietors. Kongōbuji was spiritually and politically important as both a sacred site of pilgrimage and worship and as an organization with close links to the aristocracy and military elite. The temple used this religious and social influence to acquire and expand its proprietorship of estates in Kii, seeking pious land donations and using its legal clout to expand its rights to that land. At the regional level, Kongōbuji competed for control of estate proprietorships against rivals such as aristocratic families and other religious institutions such as Iwashimizu Hachimangū, from which it took Suda Minami, Tomobuchi, and Ogawa-Shibame estates in the early medieval era. At the local level, it was also in competition with other estate proprietors to secure control of resources and land for its own estates, leading to legal disputes and border conflicts with neighboring villages.2 A high and growing percentage of estates in northern Kii were held by temples based within the province, close to their estates, and the margins of their holdings were flashpoints for border disputes. Kongōbuji estates such as Nate and Arakawa were frequently in conflict with the estates of neighboring Kokawadera, and there were even violent conflicts among different organizations within Mount Kōya, both among monks on the mountain and among the estates they controlled. Kongōbuji seized on local disputes as a means of strengthening its local sphere of influence, politicizing village conflicts, and providing military as well as legal support to its estate residents. The fact that disputes on the borders of Mount Kōya landholdings were political conflicts between regional powers as well as local grievances between neighboring communities is demonstrated by the record of where they occurred in Kii Province. On estate borders between Kongōbuji estates there were no major disputes, but on the borders between Kongōbuji and non-Kongōbuji estates, conflicts were common, 380 Power violent, and long-running. The Nate-Niunoya dispute is perhaps the most famous and of longest duration, but Mount Kōya forces directly participated in disputes in several estates. In Kongōbuji’s Makuni Estate, mountain ascetics (yamabushi ) from Mount Kōya’s Amano Shrine fought a border war against Iwashimizu Hachimangū’s Tomobuchi Estate to the east from 1258 to 1263, following a similar conflict with Iwashimizu shrine functionaries (jinin) in Nogami Estate to the west.3 In the 1220s, the inhabitants of Kongōbuji’s Shizukawa Estate fought against Jingoji’s Kaseda Estate and a local league of warriors, the Yuasa-tō, for control of water rights.4 In each of these disputes, Kongōbuji supported the claims of its estate residents not only by lawsuit and petition to central authority but also by military intervention by forces of Kongōbuji monks and allied local warriors. Kongōbuji’s increasing direct participation in disputes from the thirteenth century onward was part of the transformation of the temple from estate proprietor to local lord, supporting conflicts with soldiers levied from its other estates and Mount Kōya itself. In this respect, the temple was behaving in a manner similar to that of provincial warrior leaders, the commanders of kinship-based war bands and province-wide alliances, who constituted the elite of provincial society and the local representatives of the Kamakura military government. In common with these local lords, Kongōbuji developed a territorial fief grounded in the sworn service of estate officials and influential local families, in which the temple’s vassals supported it financially and militarily in exchange for influence within the temple and its protection. Thus when an estate became embroiled in a territorial feud, Kongōbuji was able to supplement its national-level lawsuits with military forces marshaled from its surrounding territory. The evolution of the nature of provincial disputes during this transitional period is of particular interest, involving changing estate structures, the nature of authority and local control, and the transformation of provincial political geography. Border disputes were complex and diverse, but there is clear evidence in the conflicts in Kii of the expansion and consolidation of local power. Such disputes were part of the process of restructuring the fragmented control and indirect relationships of earlier estates into more unified territorial domains, presaging the localized focus of later medieval Japanese landholding. The classical estate model of a legitimacy-based hierarchy of indirect landholding spread across the realm was evolving into a local spatial-based relation- Bad Neighbors and Monastic Influence 381 ship. In this emergent system, the estate was guaranteed by a combination of local connections and alliances, legal and religious authority, and the physical ability to communicate with and organize local administration. The victors of Kii’s disputes were those who could not only prove their claim in the courts but also physically take possession of and administer the land.

Case Study: The Nate-Niunoya Border Dispute Kongōbuji’s involvement in a long-running border dispute in Nate Estate is a good example of how the temple became directly involved in such disputes and in estate affairs in general in the Kamakura period. In doing so it came to dominate both estate management and the local area, building multigenerational relationships between estate society and the temples on the mountain. Nate was a narrow estate running north from the bank of the Ki River in Naga District of Kii Province. The estate was administered by Kongōbuji, which held the proprietorship and controlled the supervisory positions of local manager (geshi) and reeve (kumon), while Mount Kōya’s head temple in Kyoto, Tōji, controlled the post of custodian (azukaridokoro).5 Nate sat on the border between the local spheres of influence of Mount Kōya and Kokawadera, a branch temple of Shōgoin in the capital.6 The border demarcating the land of the two temples was the River Minase (the modern R. Nate), which flowed south from the foothills of the Izumi Mountains to join the River Ki. This river separated Kongōbuji’s Nate Estate from Kokawa Estate, and in the mid-Kamakura period it was the question of the ownership of this river and the hill from which it flowed that provoked violent conflict between the residents of Nate and their western neighbors, Niunoya Village in Kokawa.7 Conflict first broke out in 1241, following years of tensions within Nate Estate and between the two communities that quickly developed into an endemic conflict that continued into the mid-fifteenth century.8 The dispute initially concerned tensions between the neighboring communities and across the class divide between the yeomanry (hyakushō) and local warriors, but it soon took on a broader political context, and as the conflict progressed into the Nanbokuchō and Muromachi periods, it became a private territorial war between the monks and managers of Kongōbuji estates and their counterparts at Kokawadera. The water rights quarrel that underlay the conflict was 382 Power typical of disputes in Kii and was paralleled by similar disputes in neighboring estates. The significance of the Nate-Niunoya conflict lay not only in its duration but also in the manner in which it linked monastic power, estate managers, warrior leagues, and national authority during a transitional era. Both temples sought to assert their power as local landowners, directly involving themselves in the dispute through interaction with both local residents and central authority. Kongōbuji’s involvement in the dispute secured its interests in Nate by protecting the estate’s resources but also by exercising the temple’s protection over the estate and its residents: the temple was a necessary medium through which the cultivators and local elites could obtain protection and representation at the national level as well as the muscle to oppose a local rival. From Mount Kōya’s perspective, the temple’s assiduous representation of Nate’s interests strengthened the position of the estate as a Kongōbuji possession far beyond the temple’s original limited claim to ownership.

The Origins of the Dispute The Nate-Niunoya conflict was, at heart, a vendetta between two communities in competition for resources, but it reflects the wider structure of medieval provincial society as well. The dispute resulted initially from actions of officials of neighboring estates. A local warrior, Minamoto no Yoshiharu, was originally the reeve of Nate Estate in the early thirteenth century, but he had committed various crimes, including seizing the annual rents and causing suffering to the residents. Furthermore, he had declared an area of mountainside under his control as being Niunoya land despite its traditional use by Mount Kōya, and as a result he was expelled from the estate.9 Following his expulsion from Nate, Yoshiharu took up residence in Niunoya and, with the collusion of the military steward’s deputy (jitōdai) at Niunoya, Taira no Nagayasu, he was able to divert water from Nate to irrigate his fields in Niunoya, destroying Nate’s irrigation sluice to do so.10 Yoshiharu and Nagayasu were men at the top level of estate society, members of the warrior families who held hereditary managerial positions and had control over an extensive share of land within the estate. Yoshiharu’s family, the Uno Minamoto, had settled in the Kokawa-Nate area as minor proprietary lords and reeves of Nate Estate, holding fields Bad Neighbors and Monastic Influence 383 in both Nate and Niunoya.11 The locally headquartered reeve was often the most powerful individual on an estate. Mount Kōya depended on the services of local strongmen such as Yoshiharu to administer estates, but the same strongmen were also attempting to increase their own wealth and control over the estate, putting them in potential conflict with the lower-ranking yeomanry and the temple.12 Yoshiharu’s actions can be interpreted as an attempt to expand his personal control and influence in the Nate-Niunoya area, especially by strengthening the family power base that centered around the claim to control the forestry of Mount Shiio to the north of the two estates.13 In withholding revenues from Mount Kōya, he was doing as many local warrior managers attempted to do: exploiting the indirect relationship with the external proprietor to retain a larger proportion of the estate’s products and revenues for himself. This process was occurring across the countryside as proprietors struggled to force managers to forward taxes, but Mount Kōya’s proximity and growing involvement in the estate made it possible for the temple as proprietor to remove Yoshiharu from his post and expel him from Nate. The temple frequently expelled uncooperative managers, either confiscating their posts or exiling them from the estate. Nate’s local manager and reeve, for example, was replaced by the temple in 1262, and at Arakawa Estate too the reeve was dispossessed and exiled in 1193 and again in 1285, and the local manager was arrested, dispossessed, and placed under the supervision of resident monks in 1290.14 Kongōbuji was thus able to enforce the dispossession of local estate officials like Yoshiharu by virtue of its proximity and hence its ability to send armed monks and temple attendants to intervene in estate matters, as well as by its connections to local warrior leagues, which provided both potential military assistance and a source of new managers. After the expulsion of Minamoto no Yoshi- haru, Kongōbuji appointed Kishi no Yukimasa of the Yuasa-affiliated Kishi war band as local manager of Nate. As a warrior from outside the estate, he was more dependent on the temple for his authority, a tactic that the temple also applied in estates such as Makuni and Ara- kawa.15 Kongōbuji also benefited from the appointment, given that the exercise of governmental authority in the province was fundamentally dependent on war bands such as the Kishi and Yuasa. Thus when, in 1243, the Kamakura government ordered the arrest of Yuki- masa pursuant to investigations into the dispute, the shogunal agent 384 Power

(ryōshi) sent to apprehend him was Yukimasa’s in-law Yuasa no Mun- enari, who therefore failed to do so.16 Kongōbuji was therefore able to remove Yoshiharu from Nate, but his exile did not mean destitution for the ex-manager. Yoshiharu simply moved from Nate to the area under his control in Niunoya, reinforcing his position there by claiming the right to control Mount Shiio as a part of Niunoya and diverting water to his fields. Nor did exclusion of Yoshiharu end the relationship between the Uno Minamoto and Mount Kōya, with members of the family continuing to be important as estate managers and monks at the temple in subsequent centuries, with some even reaching the abbotship.17 The Niunoya branch of the family also developed a close relationship with Kokawadera.18 The mobility of the local elite indicates a difference between their relationship with external authorities and that of the yeomanry below them. The Uno Minamoto were originally from Yamato Province but had settled in the Kokawa area of Kii, intermarrying with local warrior families to establish a landed power base in Niunoya and Nate that was unrestricted by the boundaries of individual estates and secured by possession of managerial titles such as the reeve’s position.19 Con- necting the temple proprietors to the land-based cultivators, managers were under pressure from the demands of both sides, but they also had the potential to develop local power through resident lordship or local military strength.20 The yeomanry, by contrast, was made up of minor landholding cultivators and was dependent on the infrastructure and community of the individual estate, which resulted in community- based action in which the whole estate community banded together in response to threats.21 Following the expulsion of Yoshiharu from Nate, Niunoya Vil- lage claimed ownership of Mount Shiio, the hill from which the River Minase flowed. Along with the destruction of Nate’s irrigation sluice, the annexation of Shiio to Niunoya provoked angry exchanges between the proprietors of the two estates. According to the petition made by Kokawadera to the shogunate, which is no longer extant but is preserved in part as a quotation in Mount Kōya’s rebuttal of 1241:

Regarding Mount Shiio, the issue of the [land] west of R. Minase is clear. Since long ago, it has been off-limits to [the residents of] Nate Estate. Accordingly, the mountain’s owner [Rinshū], controlled its use for forestry.22 Bad Neighbors and Monastic Influence 385

Niunoya’s claim to the land rested on the original claim of Rinshū, as he was a resident of Kokawadera Estate.23 Rinshū had probably lived in Niunoya, and as the mountain’s “owner” (jishu), he had blocked access to the mountain for the people of Nate when they attempted to gather timber there in the twelfth century.24 Furthermore, Yoshiharu—now in Niunoya—held a claim to the hill on the grounds that his father had married Rinshū’s daughter, and this claim would then pass to Nagayasu, the Niunoya deputy military steward, after Yoshiharu’s death.25 Mount Kōya rebutted Niunoya’s position by denying Rinshū’s original control of the mountain.

This [area] is Kaori-no-misono. What’s more, when the elders were asked about the mountain owner’s control and the use of hatchets [forestry], they said it was a mistake . . . the idea that the mountain’s owner Rinshū controlled it or harvested timber there is an egre- gious lie!26

Kaori-no-misono (literally, “the honorable fragrant garden”) on Mount Shiio had been periodically used by both communities for timber and slash-and-burn agriculture. In 1214, Nate’s officials and residents asserted their right to cultivate it to make incense in perpetuity; they claimed they had felled timber there for the monks’ chancellery (mandokoro) of Mount Kōya.27 This declaration is indicative of the focus of yeoman activity in Kii. Traditional Japanese historiography has tended to equate the yeomanry with farmers on the basis of the foundation of the estate system on rice land, and to assess the activity and output of estates in terms of the rice they produced. Amino Yoshihiko challenged this assumption, calling it Japan’s “agrarian fundamentalist ideology,” and argued that medieval yeomen engaged in a variety of occupations.28 This reassessment is important in the context of mountainous land such as that of Nate Estate. The villages of Naka and Nishi at the southern end of the estate were on the flat alluvial plan of the River Ki, but much of the estate, particularly Nogami and Egawa Villages, lay on higher ground, with wooded hills stretching up to hundreds of meters above sea level. On Kongōbuji estates like this, there was therefore both a shortage of high-grade agricultural land and a demand for temple-related goods such as incense and lumber, and for mulberry cultivation to produce silk and paper, so Shiio’s “gardening” may have been quite important to the revenue of the estate. 386 Power

While the ownership of Shiio was financially important to Mount Kōya and Nate Estate, it gave rise to a much more serious cause of conflict between Nate and Niunoya. The two estates’ claims to Shiio rested not solely on the inheritance of Rinshū’s original ownership, but on the legal delineations of the estate borders. Such boundaries were primarily defined by four border posts marking the limits of the estate in the car- dinal directions as recorded in the estate’s founding document.29 This extremely simple arrangement was clearly limited in its ability to describe complex parcels of land with boundaries with contiguous neighboring estates. This simple definition of boundaries left them open to dispute, and while their limited delineation was often rectified by detailed maps, these were frequently not drawn up until after border disputes had broken out.30 When Kokawadera claimed ownership of Mount Shiio on the basis of the course of the River Minase, which separated the estates, the monks of Kongōbuji were quick to seize the opportunity.

According to Kongōbuji’s defense statement, “The section of [Kokawadera’s] letter about the place called Shiio being west of the River Minase is an unparalleled falsehood! There is a river to the west of the aforementioned Shiio, and it is the source of the Minase.” According to Kokawadera’s letter bearing a provincial seal: “The eastern border [of Kokawa Estate] is R. Minase, and the stream flows continuously from the peak.” So it is the river west of Shiio that reaches the peak, whereas the eastern valley of Shiio does not reach the peak. Therefore Shiio is within Nate Estate.31

At the base of Mount Shiio, the Minase splits into two small tributaries, and it was this that provided the basis for the conflicting interpretations of the border. Kokawadera took the eastern stream to be the main branch and claimed that as the Minase formed the eastern boundary of Niunoya and Kokawadera Estates, Shiio was part of Niunoya. Mount Kōya countered this, pointing out that the branch of the river that ran into the hills west of Shiio “reaches the peak,” while the eastern branch did not, and so the western branch must be considered the proper course of the River Minase. If the Minase flowed to the west, then by Kokawadera’s own logic it could not be part of Niunoya and must be part of neighboring Nate. Shiio’s position as a foothill of higher peaks means that the two tributaries of R. Minase do indeed rise up on either side of the hill, with the larger branch to the east but the smaller west- Bad Neighbors and Monastic Influence 387 ern branch originating some fifty meters higher.32 The implications of the inclusion of this mountainous parcel of land in one estate or the other, however, were considerably more complex than a line on a map. The issue of the ownership of Shiio brought into question the ownership of the Minase and hence the right to draw its water for irrigation. While incense and timber were important to the estate, the river’s water was essential for irrigating Nate and Niunoya’s lower and more fertile fields. As the irrigation ditches that brought water from the Minase to Nate’s fields ran through the estate, they were a focus for the community, drawing the yeomanry together to defend a common concern, thus involving all the estate’s villages. Any threat to the irrigation system was a threat to the entire estate community (see figure 15.1). The redirection of water from Nate Estate to fields in Niunoya therefore triggered the confrontation between the cultivators of Nate and Niunoya and the involvement of Mount Kōya and Kokawadera. After the destruction of Nate’s irrigation dam by Yoshiharu and Nagayasu, the officials of the two estates met to try to agree on the division of water between them. This meeting failed spectacularly.

Kokawadera along with the Niunoya deputy military steward Nagayasu harvested the crops in the middle of the night without permission—a dastardly deed! As a result, the people of Nate harvested the little that remained within the land belonging to the estate.33

Kokawadera then seems to have accused Nate of having sent an armed force of hundreds to carry out a retaliatory harvest of the “little that remained,” although Mount Kōya strenuously denied this in its defense statement. Furthermore the failure of negotiations set the tone for the dispute as, unable to agree on the division of irrigation water between the neighboring communities, residents and monks attacked the irrigation systems of both estates. Subsequently, after Nate’s cultivators tried to redirect water back to their estate from the area damaged by Yoshiharu, Kokawadera’s monks retaliated.

On Ninji 2 (1241)/6/27, the monks of Kokawadera along with the cultivators of Niunoya headed to Nate Estate’s two weirs, and on the same day several dozen Kokawadera monks armed themselves and, in command of an army of cultivators, destroyed both weirs. They Fig. 15.1. Northern Naga District, including Nate Estate. Drawn by the author. Bad Neighbors and Monastic Influence 389

also broke the pipe and dug up the ditch so that none of the irrigation water would flow. Moreover, while protecting the weirs on the following day, the 28th, the people of Nate . . . went to the upper weir and stopped it up, and here violence occurred between them—the Kokawadera monks and Niunoya cultivators had plotted the fight in advance.34

With Nate’s two weirs on the Minase destroyed, the water supply to the estate’s fields was cut off, and armed and armored Nate yeomen then destroyed Niunoya’s weir in retaliation, along with committing other violent acts including cutting people down.35 These weirs, referred to as “dams” or “wells,” were low dams built across the river, with channels cut into them to allow the siphoning off of a set volume of water that was essential to the valley’s irrigation.36 With both monks and estate residents engaged in violence, Kokawadera petitioned the shogunate for recognition of Niunoya’s rights to Shiio and Minase, at which point Mount Kōya responded in the same fashion. The shogunate initially investigated the dispute through its system for resolving legal and armed disputes, collecting submissions from both parties and holding a hearing at its Rokuhara office in the capital in 1244.37 Hearings, or monchū, consisted of a series of representations by both sides, supported by any documentary evidence provided by the plaintiff and defendant. The Nate-Niunoya monchū consisted of seven days of hearings with Nate represented by monks from Mount Kōya, envoys from Tōji, and the proxies of the estate’s local manager and reeve, and Niunoya was represented by Kokawadera monks, an envoy from the head temple Shōgoin, and the proxy of Niunoya’s military steward.38 The record of the hearings is interesting inasmuch as the attendees reflect the involvement of different groups in the dispute as well as various levels of estate management. The Tōji emissary was the highest authority the Nate delegation could muster, representing the patron at the head of the estate hierarchy, and the two estate residents represented the opposite end of the social and political spectrum. The summoning of the abbots of Tōji and Shōgoin—the head temples for Kongōbuji and Kokawadera—was unusual and indicates the all-out nature of the conflict.39 The largest contingent of the Nate delegation, however, was from Mount Kōya, consisting of six monks involved in running the estate. The attendance at the Rokuhara hearings shows Kongōbuji’s active support of local estates within its local domain, and 390 Power the considerable effort made by the temple reflects the direct relationship developing between the Nate villages and Kongōbuji as their protector. The hearings at Rokuhara illustrate Kamakura justice. Investigators were dispatched to Kii to survey the estates, and evidence was presented by both sides. The lawsuit failed, however, to resolve the dispute: in 1247 the shogunate refused to rule on the matter, declining all responsibility for resolving the dispute on the grounds that it would not interfere in “border disputes in the western provinces.” It then forwarded the record of the hearings to the Kyoto court, leaving the matter as the latter’s responsibility.40 This sudden refusal to offer a verdict is interesting. In the early years of the Kamakura administration, it initially accepted a wide range of cases to hear from across the country, but it was quickly overwhelmed by the claims relating to the disorder caused by the Genpei War. The shogunate thereafter limited its purview to cases involving its warriors, limiting the number of lawsuits it had to process and making affiliation with the military administration appealing to those who would benefit from access to its judicial system.41 Despite this, the Nate-Niunoya dispute ought to have fallen under Kamakura’s jurisdiction, as Niunoya Village had a Kamakura-appointed military steward and his proxy who were already involved in the dispute. The reason for the shogunate’s refusal to rule on the matter despite initially accepting the suit is unknown. The court, on the other hand, offered its verdict in 1250. The court considered the evidence already presented by Kongōbuji and Kokawadera, the maps made during the shogunal investigation, and the statements of local residents before offering a compromise. The court’s edict recognized Niunoya’s claim to Mount Shiio on the grounds that there was no evidence that Shiio had ever actually been controlled by Nate, and correspondingly declared the eastern valley to be the official course of the river. Nevertheless the court rejected both parties’ claim to control the waters of the Minase, declaring it to be a public resource.42

Estate Violence and Governmental Intervention Needless to say, this resolution failed to satisfy the Nate residents, who embarked on a series of raids on Niunoya in the 1250s, killing, stealing from villages, and destroying irrigation works. The disregard of Bad Neighbors and Monastic Influence 391 the court’s edict by Mount Kōya and the Nate residents reveals the limited reach of the law in the Kamakura period. Without a governmental mechanism to enforce edicts in the provinces, the court’s decision depended on the disputants’ recognition of the ruling and their subsequent adherence to its terms. The fact that within a year of the ruling Nate’s inhabitants had invaded Niunoya and dug new irrigation works in the allegedly “public” river suggests that neither the estate residents nor the proprietor Mount Kōya greatly feared any consequences from central authority. As Cornelius Kiley notes, “This new lack of power to curb the larger shrine and temple honjo, whose priestly soldiers had long been a problem, created new opportunities for the extension of Bakufu influence. Only the Bakufu was now in a position to mediate between such forces and ultimately subordinate them to itself.”43 And, despite previously having withdrawn from the legal wrangling over the estates’ border, the shogunate (bakufu) did indeed attempt to mediate between the opposing forces. In the seventh month of 1251, the Nate yeomen redressed the unfavorable verdict by attacking their neighbor: “Carrying bows and arrows, wearing armor and helmets, they forcefully invaded Niunoya, dug new irrigation works, and committed violent acts such as beat- ings and slashings.”44 And they did so yet again in 1252, going so far as to attack Kokawadera itself. It was then that the military government of Kii, under the shogunate’s constable, attempted to physically impose a compromise on Nate and Niunoya.45 Specifically, in 1253, the deputy military governor (shugodai) of Kii mobilized shogunal retainers (gokenin) in the province and constructed two new weirs on the River Minase.46 Each weir had two sluices, one siphoning off water to irrigation channels and one returning excess water to the river. The two dams were of equal width, with the small size of the gaps—some two and a half feet wide and six inches high—demonstrating that the river lived up to its name as one of scant water supply. The upper dam siphoned water to Nate and the lower to Niunoya.47 Nate residents, however, proved equally scornful of the deputy military gover- nor’s attempt to resolve the dispute: the estate’s officials and yeomen promptly destroyed the dams, doing the same after the deputy military governor and shogunal retainers returned and rebuilt them in 1254. This direct opposition to Kamakura’s chief representative in the province and the assembled retainers was an act of direct opposition to the military government. 392 Power

This defiance was only made possible by the Kongōbuji clergy and their allies, the Kishi and Yuasa, who rose up in support of Nate’s continued attacks on Niunoya during the 1250s and 1260s as they escalated in scale and violence.48 Repeated governmental mobilizations of the province’s warriors were unable to deter Nate from further aggression, and Kongōbuji appears to have had no intention of curbing this violence. Indeed, by resisting the deputy military governor’s intervention, Kongōbuji was safeguarding its control of the estate and home territory. The military governor was a rival source of local power, and if Kongōbuji had allowed his influence in Nate to grow, it ran the risk of seeing its own influence curtailed.49 Kongōbuji’s behavior is thus indicative of the balance of power and complex relationships between local proprietors, warrior officials, and government authority. It was not just protecting its interests by defending the estate’s income: by involving itself directly in the conflict it was providing assistance that the yeomanry might otherwise have sought from other power groups with a stake in the estate, allied to Kongōbuji or not. Nate’s cultivators and officials continued sporadic attacks on Niunoya throughout the 1260s, apparently ignoring a summons from Rokuhara, which further underlined Kamakura’s inability to resolve the dispute.50 Against this backdrop of continuing instability and rising violence, Kokawadera withdrew from direct litigative involvement in the case, with the military steward of Niunoya taking over representation of the estate in Kamakura. Kongōbuji by contrast intensified its involvement in the dispute, gradually transforming its role as protector of Nate’s interests from that of bridge between cultivators and government to that of primary protector. The military governor of Kii and his deputy made repeated attempts to broker compromise and impose water-sharing earthworks on the Nate and Niunoya cultivators for the remainder of the Kama kura period, but these continued to be violently resisted by both Nate and Kongōbuji. Such attempts to impose a resolution can be viewed as not only an attempt by central authority to curb provincial disorder through the agency of the provincial military governor, but also as an effort by the military governor to assert his position as the paramount authority in the province and to expand his control over local warriors. The rejection of these measures by Nate’s residents was encouraged by Kongōbuji’s support, positioning the temple in opposition to the military governor and the assertion of external control inside the Bad Neighbors and Monastic Influence 393

Mount Kōya domain. Kongōbuji offered estates like Nate a means of access to justice through representation of their cause at the national level to the shogunate and the Kyoto court. Moreover, in tandem with legal representation, Kongōbuji’s extensive network of local estates, affiliated warriors, and armed clergy provided it with a suitable military force to defend Nate’s interests outside the courts. Kongōbuji thus provided support for the combined forces of the temple complex and its surrounding domain, and access to a legitimizing authority that, by virtue of its basis in local religious power, was less dependent on the continued authority of central government than was the military governor.

The Dispute in the Late Kamakura to Muromachi Periods Lawsuits and compromises failed to resolve the fighting between Nate and Niunoya between the 1240s and the 1260s, and surviving records show that violence periodically broke out again on an even larger scale in the fourteenth and fifteenth centuries.51 In contrast to the temple’s primarily litigative involvement in the mid-thirteenth century, by the 1340s it was mobilizing large numbers of monks and estate residents from across the Mount Kōya domain to fight in battles over Nate’s border. The later development of the water rights dispute is thus marked by transition from the provision of legalistic and legitimizing support to military assistance, with Kongōbuji estates exchanging tithes and military service for defense and support, a “local lordship” sharing more in common with warrior leaders than with religious estate proprietors. This local lordship is characteristic of Mount Kōya’s adaptation to the erosion of the early medieval “influence” system and the gradual collapse of indirect administration guaranteed by the central government. By offering military assistance as well as legal defense, the temple was putting itself in a position similar to that of leading provincial warriors such as the Yuasa; but, more importantly, its proximity to the estates meant that it could provide services that distant patrons such as its own head temple Tōji in Kyoto could not. Tōji’s influence over Nate and other Kongōbuji estates waned over the Kamakura period as Mount Kōya developed stronger local relationships with local warrior families and yeomen. Such relationships were not dependent on the chain of legitimization running to the capital that had characterized earlier estate holdings. 394 Power

Whereas the initial confrontations in the dispute offer an insight into the chains of legitimacy and authority that underpinned the late classical and early medieval estate system, the later development of the dispute shines a light on the direct personal relationships between Nate Estate and temples within Mount Kōya, and the intertwining of Kongōbuji’s interests with those of the local elite. In 1345, as Nate and Niunoya’s inhabitants once again engaged in battle, Kii military governor Hatakeyama Kunikiyo proclaimed that “the Niunoya-Nate water rights matter, which has resulted in fighting, is absolutely unacceptable. . . . The decision from on high is that if they go into battle they should be stopped. And if they do not acquiesce, then both sides’ irrigation works should be destroyed.”52 Nate’s force was aided by an army of monks from Kongōbuji, who mustered the officials of nearby Kongōbuji estates such as the local manager of Tomobuchi Estate to the south.53 This force then rampaged through Niunoya as well as Kokawadera’s nearby Arami Estate, killing people and burning down villages.54 The significance of this battle is not simply that an army of Mount Kōya monks directly participated, but that the temple mobilized the officials of neighboring Kongōbuji estates to support Nate. Kongōbuji was integrating the individual communities of the neighboring estates surrounding the temple into a broader unity, that of the “Mount Kōya domain.” Estate officials, originally independent contractors who managed the estate on behalf of the temple and forwarded a share of the revenue in exchange for legitimization of their role, were now behaving like vassals, riding to their lord’s banner in times of war. The growing connections between these families of the local warrior elite and Kongōbuji are shown clearly in this incident. A fragmentary document attached to the military governor’s order names the leader of the assault on Niunoya and Arami as Raisen of Renjōin.55 Renjōin was one of a number of subtemples within Kongōbuji controlled by local warrior families that became increasingly politically powerful from the fourteenth century onward. Renjōin performed memorials for the ancestor of the Uno Minamoto–Niunoya family, and many of its monks were drawn from it.56 It was part of the group of subtemples that controlled Kongōbuji’s land administration and military organization, with the result that the monks tasked with collecting Nate’s rents from the Minamoto estate officials were also Uno Mina- moto family members.57 Bad Neighbors and Monastic Influence 395

Kōya shunjū hennen shūroku states that Raisen was born in Nate Estate into the Yuasa family, close relatives of the Uno Minamoto whose relationship with Renjōin lasted for several hundred years, and that he was a landholder in Naka Village, Nate Estate, in his own right.58 Kenshun nikki confirms that Raisen’s conspirator in 1345 was the Uno Minamoto reeve of Nate, so Raisen’s leadership of the attack supported his own kin and income, as well as the broader benefit of the temple as institutional proprietor. The battle of 1345 was thus not simply a mobilization of the temple and surrounding estates in defense of Kongōbuji’s interests, but also in defense of the personal and familial interests of monks at the temple itself.59 Moreover, the value of the broader Kongōbuji domain to its inhabitants is also demonstrated: continued membership in this network of estates afforded mutual protection from external force. The expansion of local temples “at the bottom of the mountain” expanded Mount Kōya itself from the association of halls and temples atop the Mount Kōya plateau to a domain-wide presence, allowing close involvement with estate affairs and the ability to provide military support in times of confrontation.60 Further strengthening this connection was the large population of local monks and temple affiliates living within the estates, linking monastic and estate society. The temple complex’s continued involvement with, and intervention in, Nate’s boundary dispute helped to develop the idea of Nate as a part of the wider domain even when the temple’s central control was weak.61 The character of the dispute had thus changed since its beginning in the 1240s, with the original focus on the needs and personal feuds of the local community overshadowed by the vested interests of temple proprietors. The dispute became an endemic border conflict that— while still triggered by the water rights quarrel between two entrenched communities—was fought between rival local powers for control of the Kinokawa Valley. Nate Estate stood at the boundary of the Mount Kōya sphere of influence, and the River Minase came to mark the de facto border dividing Naga District in Kii between Mount Kōya and the domains of Kokawa and Kongōbuji’s great rival, Negoroji.62 The changed political landscape of the later medieval age also affected the litigative options still open to Kongōbuji. While it remained officially neutral during both the fall of the Hōjō government and the protracted civil war between Northern and Southern Courts in the fourteenth century, both sides in each conflict were 396 Power anxious to win the temple’s favor for fear that if not suitably flattered, Mount Kōya would abandon its neutrality in favor of the other side. With the strategic importance of Kii to the conflict between Kyoto and Yoshino, Kongōbuji was able to petition the rival factions for recognition of Nate’s claims in the dispute. In 1363, Kongōbuji petitioned the southern court monarch Go-Murakami for a resolution to the dispute, and the temple then received a terse but highly favorable edict recognizing Nate’s claims.63 Go-Murakami’s edict was the first time in the course of the dispute, which had by this time been fought for more than a hundred years, that the temple had achieved a legal victory. Its favorability was, no doubt, due to Go-Murakami’s hope that Mount Kōya would lend some of its military force to the Yoshino cause.64 Nate’s legal victory was short-lived however, since the defeat of the Southern Court saw the decision reversed by the military governor Ōuchi Yoshihiro, who ordered a return to the division of the 1250s. Even the end of the civil war in 1392 and Mount Kōya’s reassertion of control over its war-damaged estates was not sufficient to bring the Nate-Niunoya dispute to a close. Large-scale conflict continued across the River Minase border in the fifteenth century, with the deputy military governor of Kii mobilizing an army to prevent Mount Kōya’s army from marching on Kokawadera during a Nate-Niunoya battle in 1433. This led to a confrontation between the temple’s scholars, who backed down, and the workers, who refused and went on to destroy more than half of Mount Kōya in the ensuing fighting. The largest, and last, recorded battle in the dispute occurred in 1467, more than two hundred years after Minamoto no Yoshiharu first broke Nate’s irrigation weir, when a tit-for-tat burning of Niunoya and a village in Nate Estate acted as a trigger for open conflict between their proprietors. An army composed of the resident monks of Nate, Shizukawa, Ōzu, and Kaseda Estates attacked the opposing bloc of Kokawadera and its eleven villages, causing Kongōbuji to send an army from the temple in support. Negoroji, Kongōbuji’s long-time adversary, then sent its own army to support Kokawadera, with control of Naga District divided between the opposing forces. Following the warfare, a ceasefire was brokered by the deputy military governor in which both sides agreed not to obstruct the river.65 Bad Neighbors and Monastic Influence 397

Conclusion: The Balance of Legal Legitimacy and Military Authority Disputes—often violent—were an inevitable feature of the medieval land system in Japan. Estates were in competition for land and resources, with proprietors and warriors competing for control of those estates. Weak central control was exacerbated by the nature of many estate boundaries; the poorly defined border that caused much of the Nate- Niunoya dispute was far from unusual. Disputes reflect the tensions in provincial society during the period. The yeomanry was in conflict with neighboring communities in communal conflicts rooted in access to resources at the estate or village level. Estate officials participated in these disputes, often joining in raiding parties, but they were also active beyond the estate level: they either had personal connections to neighboring estates or were summoned by Kongōbuji to do military service. Proprietors competed at a higher level, and the nature of that competition changed most significantly over the period. The lawsuits designed to redefine estate boundaries or confirm usage rights, and to readjust the balance of power between neighboring proprietors, slowly gave way in the face of declining utility, paralleled by a rise in direct application of force by local proprietors such as Kongōbuji or provincial warlords in defense of territorial blocs. The long-running dispute between Nate and Niunoya is typical of Mount Kōya’s involvement in provincial disputes. Examining the Kinokawa Valley in northern Kii, the extent and scope of disputes is clear. Following the northern bank of the river downstream across Ito and Naga Districts, every single estate along the bank was involved in disputes over boundaries, water rights, or overall ownership rights with neighbors. Running downriver from the provincial border, Suda Estate, an Iwashimizu Hachimangū possession, invaded neighboring Ōga Estate, a Daidenbōin possession.66 Ōga was embroiled in an ongoing dispute with its westerly neighbor Kanshōfu from 1134 onward, as a doctrinal schism between proprietors Daidenbōin and Kongōbuji led to violence both within Mount Kōya and in their estates67 (see figure 15.2). Kanshōfu too was in dispute along its other border, with Kaseda Estate to the west.68 Kaseda, a Jingoji possession, engaged in disputes with its own westerly neighbor Shizukawa Estate over water rights and with Daidenbōin’s Shibuta Estate to the south.69 Shizukawa’s western neighbor was of course Nate Estate, the focus of the long-running 398 Power

Fig. 15.2. Upper Kinokawa valley. Drawn by the author.

conflict with Niunoya. The fact that there were six major disputes in one fifteen-mile stretch of valley speaks volumes as to the historical significance of inter-estate disputes in Kii.70 The two hundred years of intermittent conflict between the neighboring communities of Niunoya Village and Nate Estate, and the wider neighboring domains of Kokawadera and Kongōbuji, demonstrate the close interrelation between different power groups in provincial Japan in the medieval era. The unilateral actions of the Nate yeomanry may have been the backbone of cross-border violence from the beginning of the dispute, but the military, political, and familial connections among estate managers, warrior bands, and temple proprietors were key to the balance of power on the ground in this long-running conflict. These disputes are significant events in the medieval history of Kii. They are well recorded: each incident was surrounded by lawsuits, petitions from dispossessed cultivators, and lobbying letters from Kongōbuji to the political elite. Records of disputes are in turn a key source for our understanding of how medieval landholding operated. Long-running conflicts like the Nate-Niunoya dispute also show how this system changed over time, and they relate the wider downward flow of government and authority to the local level. The story of the clashes between individual estates and communities is part of the evolution of localized Bad Neighbors and Monastic Influence 399 direct temple administration and is evidence of the relationship among Mount Kōya, local warrior families, and estate inhabitants. Attempts to impose external resolutions to disputes were protracted and frequently unsuccessful, and they exposed the weakness of patrons and absentee proprietors. Local proprietors—the provincial temples, major warrior families, and the subtemples and shrines of Mount Kōya—capitalized on their connections to both external authority and local land by involving themselves in local disputes. By entering a dispute, proprietors such as Kongōbuji were fulfilling an obligation to the residents of an estate by responding to their petitions, and by doing so they were able to strengthen their connection to their proprietary holding. The changing nature of the assistance offered by Kongōbuji in disputes reflects the changing background of medieval society and the nature of local conflict as Kongōbuji developed territorial lordship founded on a contiguous local domain. The temple’s continuing ability to offer support and its continuing incentive to do so were predicated on its proximity to Nate and on Nate’s position on the frontier of the bloc of Mount Kōya estates. In doing so, it was providing military assistance the distant capital elite could not provide, supported by an unimpeachable religious might to which other local military providers—local lords and warrior leagues—could not aspire.

Notes

1. Taira 1996, 431. Also see Kuroda Toshio 1994a and 1994b; and in English see Adolphson 2000, especially the introduction. 2. Suda Minami, 1334; Tomobuchi, 1333; Ogawa-Shibame annexed in late thirteenth century, legally recognized in 1333. Koyama 1990, 3. 3. Tomobuchi dispute: KM 7:186–187 (# 1588, 1263); KM 1:503–509 (# 447, 1271); Yamakage 2002, 180–182. Nogami dispute: KI 6:422–437 (# 4430, 1232). 4. Koyama 1987b, 264. 5. Kinokawa Ryūiki Shōen Shōsai Bunpu Iinkai 2004, 4, hereafter Kino- kawa Iinkai. 6. Shōgoin, situated in modern Sakyō Ward, Kyoto, was in the medieval period a branch temple of the Miidera/Onjōji school of Tendai Buddhism. 7. Modern Kami- and Shimo-Nyūya, Kinokawa-shi, Wakayama Prefecture. 8. Koyama et al., 2004, 112. 9. Koyama 1987b, 200; KM 4:250–254 (# 114, 1241). 10. Kinokawa Iinkai 2004, 8. 400 Power

11. Kinokawa Iinkai 2004, 5. 12. Amino 2012, 228. 13. KM 1: 427–435 (# 396, 1250). 14. KM 3:652–660 (# 819, 1193); Kōya shunjū hennen shūroku 1912, 181 (Kōan 9 (1286) 12/3) and 182 (Shōō 3 (1290) 8/8). 15. KM 1:425 (# 393, 1243); 425–426 (# 394, 1244). Kōyasan appointed Kishi no Seishinbō as reeve of Iwabashiri Village in Makuni in order to obtain control over this disputed area during a border war against the shrine functionaries and cultivators of Iwashimizu Hachimangū’s Tomobuchi Estate, and he appointed Mike Shinjō as Kongōbuji’s agent in Arakawa during conflict in the thirteenth century. KM 7:186–187 (# 1588, 1263); KM 6:442 (# 1394‑1, 1307); Yamakage 2002, 188. 16. KM 1:425 (# 393, 1243). 17. E.g., the tenure of Minamoto no Sadaharu and Tameharu in the 1290s as warrant officer (sōtsuibushi) and community leader (tone) of Nate, and of Raisen as 110th administrator (kengyō) of Kongōbuji. KM 4:270–271 (# 125, 1291); Kōya shunjū 1912, 219; Yamakage 1988, 312. 18. Kinokawa Iinkai 2004, 5. 19. Koyama 1987b, 201. 20. Amino 2012, 228 note 12. 21. Koyama 1987b, 208. 22. KM 4:246–250 (# 113, 1241). 23. KM 1:427–435 (# 396, 1250). 24. Kinokawa Iinkai 2004, 5; for the position of jishu, see Asakawa 1929b, 89. 25. Koyama 1987b, 189; Kinokawa Iinkai 2004, 5. 26. KM 4:246–250 (# 113, 1241). 27. KM 4:246 (# 112, undated), Kinokawa Iinkai 2004, 8; KM 1:427–435 (# 396, 1250). 28. Amino 2012, 25. 29. Keirstead 1992, 51 note 14. 30. In the case of Nate, Rokuhara ordered maps to be drawn up following the lawsuit hearing in 1244. 31. KM 4:246–250 (# 113, 1241). 32. Calculated from data provided by Kokushi Chiri’in, Geospatial Infor- mation Authority of Japan, map reference 34º19’50”N 135º25’45”E. 33. KM 4:246–250 (# 113, 1241). 34. KM 4:246–250 (# 113, 1241). 35. Koyama 1987b, 190. 36. 堰、井水、井口. For a diagram of one of the Minase sluices, see Koyama 1987b, 197. Bad Neighbors and Monastic Influence 401

37. KM 1:425–426 (# 394, 1244). 38. KM 1:425–426 (# 394, 1244). 39. Koyama 1987b, 193. 40. Adolphson 2000, 230; Koyama et al., 2004, 111. 41. Mass 1995, 34. 42. KM 1:427–435 (# 396, 1250). 43. Kiley 1982, 36. 44. “Kenchō sannen shichigatsu nichi Nate-no-shō rōzekijin kyōmyō chūbun’an” (Kōyasan Miedō monjo) in Koyama 1987b, 195. 45. KM 4:260–261 (# 117, 1252); 4:264 (#119, 1252). 46. Adolphson 2000, 230. 47. Koyama 1987b, 197; Koyama et al. 2004, 111. 48. Koyama 1987b, 200; KM 7:396–464 (# 1661, undated). 49. A situation that can be seen in Kanshōfu and, dramatically, in Minabe Estate on Kii’s coast, where the estate was eventually partitioned between the military governor and Mount Kōya, leading to the temple’s losing all control and most of its income. Varley 1966, 42. 50. Koyama et al. 2004, 112. 51. Koyama 1987b, 204. 52. KM 4:654–655 (# 389, 1345). 53. Koyama 1987b, 204. 54. Koyama 1987b, 205. 55. KM 4:654–655 (# 389, 1345). 56. Kii zoku fudoki 1962, 4:241. 57. Yamakage 1988, 314 and 335. Raisen was responsible for tax collection in Nate in the 1350s. For an examination of the role of Kongōbuji’s “Shōshūe” subtemples and local warrior families in the internal politics of Kongōbuji and Mount Kōya, see Hongō 1986, 502–527. 58. Kōya shunjū 1912, 219; Yamakage 1988, 312. Kōya shunjū is a collection of documents and histories of various halls on Mount Kōya, compiled by the Kōyasan administrator (kengyō) Kaiei in 1694–1719. 59. KM # 389 (above) lists “Renjōin Sōsenbō” as the leading Kōyasan monk. The identity of Sōsenbō as Raisen is confirmed by KM 7 #s 1661 and 1662 (7:396–464 and 7:465–521); the reeve was named by the military steward of Niunoya, Shinagawa Mitsukiyo, in Kenshun nikki (by Daigoji abbot (zasu) Ken shun, 1299–1357), cited in Koyama 1987b, 204. 60. In Nate, the Nishimura Myōhōji and Nakamura Kannonji. Kinokawa Iinkai 2004, 6. 61. At the beginning of the medieval period, Kongōbuji’s control over estates such as Nate was limited due to the weakness of the external hierarchies 402 Power of the estate system, other estate rights-holders such as Tōji, and independent local families. The monastic complex’s direct control gradually increased, however, with a domain-wide reassertion of control in the early fifteenth century. 62. Koyama et al., 2004, 113. 63. KM 4:672–673 (# 409, 1363). 64. Koyama 1987b, 204. 65. Kinokawa Iinkai 2004, 5; Koyama 1987b, 204. 66. Ōga was invaded by Iwashimizu Hachimangū in 1132 and 1137 and was engaged in a dispute with Kanshōfu from 1134 onward—Negoro yōsho 1994, 10–12 (leaves 15–19, 1139); Noritake 2002, 2. 67. In the early twelfth century, the establishment of new, well-funded cloisters at Mount Kōya and a dispute between these and the older halls over matters of doctrine, ritual, and precedence led to a schism on the mountain. This resulted in the division of Mount Kōya Shingon Buddhism into two schools, the orthodox (kogi) and reformist (shingi), represented by Kongōbuji and Daidenbōin respectively, leading to 150 years of disputes and violence on the mountain. In 1288, following numerous violent incidents, temple burnings, and killings, Daidenbōin left Mount Kōya and relocated to Negoroji, at the heart of the bloc of its estates in northwest Kii. Daidenbōin estates that were closer to Mount Kōya—Ōga and Shibuta—were then annexed by Kongōbuji in 1333. 68. Negoro yōsho 1994, 1139/7/28, cited in “Kanshōfu-shō,” Nihon rekishi chimei taikei, 31:57. 69. Kobayashi 2002, 41–44; Maeda 1999. 70. Cartographical data for the locations and borders of estates in Kii were drawn from the following sources: Kokushi Chiri’in; Takeuchi 1975–1976, 2:253–264; Kinokawa Iinkai, 2004; Toyoda 1977, 367, 383, 423, and 500; Koyama 1990; Nihon shōen dētabēsu; Nihon rekishi chimei taikei 31 (1983); Yamakage 2002, 1:27 and 45; 2:3, 6, 28, 67, 89, 97, and 104; Matsuda 1984. c h a p t e r 16

The Akutō on Ōbe Estate Lawsuits, Evidence, and Participation in the Late Kamakura Legal System

Dan Sherer

[Question:] In provinces all over the country the situation is the same, but I have heard that in this province [of Harima], the activities of akutō [“evil bands”] are especially troublesome. Since when has it been like that?

[Answer:] From some time in the Shōan era (1299–1302) and the Kengen era (1302–1303), reports of such things came from here and there. Estates were invaded and plundered, pirates appeared in bays and inlets, and people were constantly harassed by thieves, rob- bers, highwaymen and brigands who roamed the mountains. —From the Mineaiki1

In the mid-thirteenth century, Japan began to grapple with its akutō problem. The shogunate, both at its headquarters at Kamakura and at its compound in the Rokuhara area of Kyoto, issued missives demanding that akutō be quelled as far afield as Kyushu in the west and Mutsu in the northeast, while estate proprietors complained that akutō were raiding their lands.2 The problem seemed to get only worse as the years went on, and eventually the shogun’s deputies at Rokuhara went on akutō-hunting expeditions, killing and capturing dozens and razing their fortifications. These forays would disperse the akutō for a brief time, but they would quickly return in force. For the last half century of the Kamakura period, the akutō seemed only to increase. On Ōbe Estate in Harima Province, three cases related to akutō made their way through the shogunal courts. This chapter will explore

403 404 Power the first two, which are better-documented than the third. In the first case, a dismissed estate official named Ōmura Shigemasa “assembled numerous akutō ”3 and raided the estate, plundering residents’ homes. In the second, a group of warriors attempted to seize land within Ōbe’s borders. Through these two cases we will examine how the charge of being an akutō was used in the Kamakura legal system. The focus on the legal system is particularly important in understanding the structure of estates, which were first and foremost entities defined by law. Specifically, the concept of the estate was derived from tax law, in that certain lands were exempted from certain taxes. This legal concept was extremely important in the Kamakura period: of the roughly thirty-five thousand documents in the Kamakura ibun collection, more than ten thousand mention estates. Many of these (including all the documents from the two cases we will examine) are extant because they were part of the legal process, either being depositions in legal cases or having been preserved as part of the evidence for a case. One of the specific problems the Kamakura legal system had to deal with was the invasion of estate territory by akutō. Many of the relevant documents were produced by estate residents, officials, or proprietors.4 In fact, of the surviving akutō documents, more than four-fifths deal with a specific estate, and low-level officials figure prominently as both vic- tims and perpetrators.5 The link between akutō and estates has not been ignored by previous scholarship: much of it has suggested that akutō were directly opposed specifically to estate proprietors. Previous scholarship on both akutō and the Kamakura legal system has focused on estate proprietors as actors in legal disputes, but through the lens of the akutō problem, this chapter will show that estate residents were active participants in the legal process as well. It will also clarify the process itself, and in doing so will shed light on the complex interactions among estate proprietors, officials, residents, and akutō.

What Is an Akutō? To begin, we need to clarify the definition of akutō. Originally a Chinese term that we would translate as “evil bands,” it is one of several synony- mous Chinese terms that made their way into the Japanese language at some point before the Heian period, such as akuto and kyōto. All these terms essentially indicated “evil doers.”6 Until the thirteenth century, Akutō on Ōbe Estate 405 these terms were used interchangeably, but after that, there is a spike in the use of just akutō, especially in documents issued by or directed toward the Kamakura shogunate. There is not a single use of any of the old synonyms in a shogunate document for thirty-five years after 1239, and the streak was broken only by a description of the invading Mongol fleet, which the shogunate saw fit to dub kyōto.7 So why did the shogunate focus on the term akutō? Most scholars in postwar Japan have described akutō as a social phenomenon, though there has been a major and protracted disagreement as to exactly what sort of social phenomenon they were.8 Kuroda Toshio, for example, argued that they were landholders who responded to increasing monetization by gambling themselves into debt and, having lost their lands, turned to crime.9 Amino Yoshihiko followed a different line, linking the akutō to social outcasts and seeing them as part of the distinctly nonagricultural stream of Japanese society upon which his scholarship focused.10 Nagahara Keiji thought that akutō were trying to make themselves into feudal lords, and he argued that they represented an important point in the transition between stages of feudalism.11 And yet despite these varying theories, these scholars all interpreted “akutō ” as a distinct category of people, who greatly increased in numbers in the late Kamakura period. The main argument was about the nature of this distinction. This argument still dominates Japanese historiography on akutō, in particular the work of the Akutō Kenkyūkai (Akutō Research Group), which has published several volumes of essays on the subject beginning in the 1980s.12 The leader of the group for several decades was the late Satō Kazuhiko, one of the most important scholars in the social- phenomenon school, whose work on akutō has been described as one of the “three famous essays that reordered the field of akutō studies.”13 Along with the essays the group has published, its books have also included bibliographies of both akutō primary sources and secondary scholarship. In English, the most thorough treatment of akutō as a social phenomenon is Lorraine Harrington’s 1982 essay, which argued that akutō cases were a reaction to both changes in the nature of landholding and the Hōjō regency’s increasing tendency toward autocracy.14 Pierre Souy- ri’s The World Turned Upside Down likewise deals with the issue from this perspective, primarily following Amino and Satō.15 There is, however, a line of scholarship that takes a totally different tack, pioneered by Yamakage Kazuo in the late 1970s.16 Yamakage 406 Power argued that akutō were not a social phenomenon, that is, that there was no group of people who could be identified as akutō. Rather, Yamakage made the case that the sudden explosion in the use of the term akutō in the historical record is due to its use in the shogunal laws in the 1230s to describe people who committed certain serious crimes, crimes that the shogunate considered within its own jurisdiction.17 The use of the word akutō, then, does not indicate a social class or occupation but is rather a legal category. By invoking this category, plaintiffs might compel the shogunate to act on their behalf, since anyone in that category was not only the enemy of the plaintiff but of the shogunate itself.18 This line of inquiry has been pursued and complicated by later scholars such as Shimozawa Atsushi, Kondō Shigekazu, and Morten Oxenboell.19 I believe that in general Yamakage’s reading has merit, and in the following I will explore two akutō cases on Ōbe Estate as examples of legal maneuvering rather than as accounts of banditry.

Harima and the Mineaiki During the thirteenth and fourteenth centuries, Harima (the southwest section of modern Hyōgo Prefecture) was the site of numerous akutō cases. Indeed, several scholars go so far as to say that Harima in particular had an akutō problem.20 One important source for this claim, and indeed for akutō studies as a whole, is a text called the Mineaiki,21 a Harima gazetteer presented as a conversation between a young monk and an old monk at Keisokuji, a temple in what is now Himeji. The Mineaiki was likely written at the end of the 1340s and the author is assumed to have been a monk at Keisokuji, though no one knows for certain.22 The most well-studied section of the work relates to Harima’s akutō, and it has often been used to examine the akutō phenomenon as a whole.23 In my brief summary of the account, I will follow Morten Oxenboell’s general division of the Harima akutō story into three sections.24 The first section relates the appearance and actions of akutō from 1299 to 1303, during the Shōan (1299–1302) and Kengen (1302–1303) eras. Of particular note to scholars is the description of their dress:

These bandits disguised themselves in a most peculiar way, completely unlike normal people. They wore persimmon coloured katabira (thin kimono) and roppongasa (straw hats) instead of eboshi (‘normal’ headwear) and hakama (divided skirts). They never con- Akutō on Ōbe Estate 407

fronted people directly, but sneaked about with irregular bamboo quivers on their backs. At the waist they carried long swords with unadorned hilts and scabbards, and in their hands they held nothing but long shafted weapons made of sharpened sticks of bamboo or hard wood. They did not even wear any kind of armor.25

Satō Kazuhiko, for example, used this passage to argue that akutō were connected to outcasts, noting especially the persimmon-colored robes that were common wear among some outcast groups,26 while Arai Takashige has argued that the robes show a connection to the itinerant ascetics known as yamabushi.27 Further description points out that the akutō fought with what can be called guerrilla tactics, and that they were fond of gambling.28 This passage was important evidence for the arguments of both Amino and Kuroda, as noted above. The second section in the Mineaiki recounts Kamakura’s response. The shogunate sent warriors, including one of the Rokuhara deputies (tandai), Ōsaragi Koresada,29 to Harima to fight against the akutō in 1319. The warriors killed scores of the akutō, burned their fortifications, and even set up fortresses at Akashi and Nagaishi to keep them pacified. These fortresses were on the coastline and near river mouths, suggesting that the targets of the raid here were connected to trade over water. The old monk in the Mineaiki laments, however, that once the deputy returned to the capital, the akutō returned in force. In the third section, which begins in the late 1320s, those called akutō show marked differences from those described earlier.

Now mounted on fine horses and with spare ones in tow, groups of fifty or a hundred warriors appeared. They carried with them heavy chests, quivers and all other kinds of martial equipment, and they were wearing shining armor with inlays of gold and silver. They usurped otherwise undisputed land, claiming to be retainers of the original holders. They exchanged pledges with like- minded fellows and formed gangs, promising to assist each other when necessary.30

These akutō are well-equipped, riding horses and clad in orna- mented armor. Further, the Mineaiki states that they also constructed more sophisticated fortifications and were so ruthless and militarily powerful that even the military governor of the province was afraid of 408 Power them. Finally, we see an exchange of pledges, a practice that eventually came to be associated with the idea of ikki (leagues or solidarities).31 Oxenboell has argued, and I agree, that the akutō in phase one, the poorly equipped gamblers, and the akutō in phase three, the well- organized and well-equipped land encroachers, are in fact different groups of people.32 Both groups were called akutō only because both were accused of serious crimes. The former group seems like a gang of thieves. The latter group seems to be what we would otherwise call “warriors”; their dress and appearance were likely quite like that of the Rokuhara party that was sent to quell them in 1319.33 I do not believe that the Mineaiki account should be read as an accurate description of all akutō; indeed, it is likely heavily embellished even when describing the Harima akutō. But because it is a local history of sorts, it provides context in which to consider the akutō troubles on Ōbe Estate in that province.

A Guide to the Kamakura Court System Another important question is: How might an akutō case have proceeded through the Kamakura judicial system?34 While in general criminal matters were the domain of land stewards (jitō) or military governors (shugo), criminal cases could be brought before the shogunal court either because one of the litigants was a shogunal houseman (gokenin) or with the recommendation of the local shogunate officials. Procedurally, there was not much difference between civil and criminal cases. First, one of two shogunal organs received the case and appointed a judge. In cases in western Japan, such as both of the Ōbe cases here, an organ at the Kyoto shogunal headquarters at Rokuhara called the Rokuhara Office of Criminal Matters (Rokuhara Kendan Kata) heard the case.35 The agency that received the case then assigned it to one of the five panels of the Rokuhara Board of Coadjudicators (Hikitsuke Kata).36 The Board of Coadjudicators would have reviewed the first of the Ōbe cases. However, after 1300, the coadjudicators lost their jurisdiction over criminal cases, which then became the sole responsibility of the Rokuhara Office of Criminal Matters. Further, in cases wherein temples were litigants (both the cases at issue here), a full panel of coadjudicators was not involved; only the head of the panel heard the case. After the case was assigned, the litigants were then asked to send their accounts of the facts and matters of law at issue. It is notable that Akutō on Ōbe Estate 409 the shogunate usually performed no investigation into the facts or the law involved; it relied on the litigants for both. In all, there would be three rounds of documents in which the litigants put forward their version of the facts and the applicable law.37 In the cases at Ōbe below, it is possible that the defendants never sent messages in response, as none of them are extant. If that was so, it seems likely that the court held the defendants in contempt (ihai) and simply found for the plaintiff. If the defendants in our cases did respond, those documents were likely either never included in, or deliberately excised from, the archives of Ōbe’s proprietor, Tōdaiji. Assuming that a full trial was held—hardly a bad assumption, as in both cases the defendants included housemen with some manner of military post—the process continued as follows. After the three rounds of documents, the litigants would be summoned to appear before the court in person. The shogunate had a great deal of trouble actually compelling people to appear, even though the penalty for failing to do so was to lose the case. Once the litigants or their proxies (daikan) were assembled, a preliminary hearing would be held before the head judge for the case.38 The next step would be a full hearing before all the judges. The exact procedure for these hearings is unknown, but mondō, the term used in the fourteenth-century legal primer Sata mirensho, can be translated as “debate” or “argument,” implying that there was an exchange between the two parties.39 After this hearing, both parties would be asked if there was anything they wished to add to the case, and if so, another hearing would be held. Fol- lowing this, the judges themselves would question the parties and then dismiss them to deliberate. The head judge would draft a document with the findings of the court, which the other members would then need to sign. If there were only a single judge, presumably he would draft this document on his own. What was the basis for the decision? The judges were, for the most part, not trained in law or legal procedure. The only information about matters of law was provided by the litigants. The closest thing to a consis- tent and basic legal principle was dōri, which can be roughly translated as “reason.” Steenstrup has defined dōri as the effective consensus of the warrior community, but he notes that it could also be defined as the ethics or morality of the court. He further argues that this concept allowed a lot of leeway for court decisions and even permitted them to ignore precedent or law (though not entirely, or even particularly often).40 410 Power

This written judgment would then be sent up to the highest-level court, the Board of Councilors (Hyōjōshū).41 There were three such boards, one in Kamakura, one at Rokuhara, and one at the shogunate’s Kyushu headquarters in Hakata. The board was to read out the case document submitted to it, and then, in an order chosen by lot, each judge would give his opinion as to whether the document was correct or incorrect. If the Board of Councilors agreed with the original decision, they marked the document correct and gave it back to the presiding judge from the Board of Coadjudicators, who would have an official opinion drawn up by a scribe and signed by the deputies (in Kamakura, by the two highest-ranking officials there), and then would present the judgment to the winning litigant. If the Board of Councilors disagreed with the document, the coadjudicators had to reconsider the case. There were avenues for appeal, but barring those the case was at this point concluded. Of course, the ruling did not guarantee that anyone would actually enforce the decision. Unfortunately, precisely how the Ōbe cases discussed below made their way through the judicial system is unclear, and the later case shows signs of having been presented to other legal authorities, such as the retired sovereign. Based on the above process and assuming a full trial in the shogunal courts, however, here is how the cases likely advanced through the system.

1. Tōdaiji submitted its suit to the Rokuhara Office of Criminal Matters. 2. The cases were assigned to judges by the Office of Criminal Mat- ters. In the first case (1295), because the plaintiff (Tōdaiji) was a temple, only the head of one of the panels of the Board of Coadju- dicators would have been assigned the case instead of a full panel. In the second case (1322), criminal cases under Rokuhara’s jurisdiction had become the exclusive domain of the Office of Crimi- nal Matters, so the judge was likely the head of that office. 3. The judges having been decided, both sides sent the court their versions of the facts and applicable law in three exchanges. 4. The litigants (or proxies) were summoned to appear before the court. 5. A preliminary hearing was held before the head judge, followed by a full hearing before the full court. Because in both these cases the panel was likely composed of only one judge, the preliminary Akutō on Ōbe Estate 411

hearing may have been omitted. After this hearing, both sides got one final chance to present new facts to the court. 6. The judge or judges questioned the litigants and then retired to deliberate. The judges’ findings were drafted into a document, which was sent to the Board of Councilors. 7. The Board of Councilors voted on whether they agreed with the lower court’s findings. If they disagreed, then they likely returned to step 5 and went through the remaining steps until the councilors agreed with the decision. If they agreed, they sent the decision back to the presiding judge on the Board of Coadjudicators or the Office of Criminal Matters, which drafted a formal opinion and presented it to the winner (in both cases, Tōdaiji).

We should keep this framework in mind when considering the cases below.

The First Case, 1295 In 1294, citing his lawlessness and violence, the great Nara temple Tōdaiji, Ōbe Estate’s proprietor, dismissed its centrally appointed administrator (zasshō), one “Lord Sanuki.”42 He was, in fact, one of three such officials who were fired for the same reason, violence and nonpayment of rents, in a fairly short stretch of time.43 Sanuki was replaced by Ōmura Shigemasa,44 but Tōdaiji soon accused Shigemasa too of not forwarding rents, and he was removed from his post that same year. According to documents45 written by the Ōbe Estate yeomen, or hyakushō,46 Shigemasa did not take his dismissal with grace. He “raised numerous akutō ” and together with them raided the homes of the yeomen and stole “wealth and goods” including agricultural items, horses, and cattle. Shigemasa’s underlings then kidnapped the yeomen’s wives and children. Even worse for the yeomen, soon afterwards a number of shrine associates (jinin)47 from Tōdaiji’s Hachiman Shrine arrived and demanded compensation for food, horses, and travel, which the yeomen decried as onerous. In the first month of 1295 (Einin 3), the yeomen responded formally by sending Tōdaiji a petition demanding that Ōmura be punished and that the temple cease sending down shrine associates. Tōdaiji’s three deans (the sangō), the monastic authorities responsible for temple and monks, forwarded this letter to Rokuhara after omitting the section 412 Power about the shrine associates and attaching their own complaint.48 There was further investigation by both the temple and the shogunal office at Rokuhara, which uncovered another alleged conspirator on the estate, and later documents from Ōbe Estate suggest that Shigemasa was indeed arrested.49 The most interesting thing about this case is the evidence Tōdaiji submitted: the yeomen’s petition, which complicates the question of how the term “akutō ” was used at the time. As noted earlier, scholars following Yamakage view it as a shogunal legal term that was inten- tionally used by others as a means of encouraging the shogunate to intervene in a case. Most of the studies that have followed this line, however, have focused on the shogunate and estate proprietors and do not give as much attention to documents produced on site. To help illuminate the way the term akutō was used and understood, I will examine two extant versions of the same document: the petition the yeomen sent to Tōdaiji and the copy of that petition Tōdaiji forwarded to Rokuhara. To begin with, the versions are in different collections. The one that was forwarded to Rokuhara is in the Tōdaiji monjo collection.50 It is dated Einin 3 (1295), first month. It is just short of 780 characters and is solely devoted to the issue of Shigemasa’s—and his akutō band’s—depravities. The other document is in the private collection of the Tsutsui family, members of which even today serve as senior monks at Tōdaiji.51 Also dated Einin 3, first month, this latter version is almost 1,300 characters long, since it includes the complaints about the shrine workers who were dispatched to the estate. Except for places where the document is damaged and characters are missing, comparison shows that the Tōdaiji monjo document is a truncated version of the Tsutsui family document, almost to the individual character. This tells us a few things. First, the Tōdaiji monastic deans edited the documents they forwarded to Rokuhara. The charitable reading of this particular edit is that the section on the shrine associates was not considered germane to the complaint. The critical reading is that the excision was done to protect Tōdaiji’s image as a benevolent religious institution defending its own interests and staff. Recall that the judges in the shogunal courts were for the most part not trained legal experts and were expected to come to opinions based on a murky and moralistic sense of “reason.” The demands of the shrine associates might well have struck them as unreasonable and thus weakened Tōdaiji’s case against Ōmura Akutō on Ōbe Estate 413 and his band. In any case, we can be sure that this is not the only example of Tōdaiji, or any proprietor, editing the documents it submitted to authorities, although the issue requires further study. All this tells us that the yeomen documents and the suit that followed cannot be read as simple extensions of the will of the temple. This case is not merely between the proprietor and the accused akutō; in the yeomen we have a third voice, albeit one that is being mediated by the temple. So what can we make of this voice? Unfortunately, details are scarce. The documents are merely signed “Ōbe no shō hyakushōra” (the yeomen-cultivators of Ōbe Estate). This suggests that the petition was written, or at least commissioned, by the Ōbe yeomen as a group. Nothing therein suggests how the decision to petition Tōdaiji for relief was reached or whether there was negotiation with the temple beforehand. The section that complains about the shrine workers refers to “hyakushō satanin,” which could be rendered “yeomen agents.” It is clear that the author of the original complaint was well-educated and possessed legal knowledge. Even if we do not accept the premise that the very use of “akutō ” is a deliberate legal maneuver used to draw the shogunate into the case, the petition shows a working knowledge of legal precedent in both court and shogunate law. For example:

As a result [of the importance of this holding] a royal order was given, exempting [this estate] permanently from the Usa [Hachi- man] duty, and there was an order from the shogunate forbidding illegal intrusions by the military governor in perpetuity. What is more, at the time of the Jōkyū military disturbance [1221], as a result of the temple’s [Tōdaiji’s] suit, there was an order from the shogunate stating that it would strictly forbid illegalities on this estate.52

While I have been unable to find documentation for the claims of exemptions from an “Usa duty,” there is a copy of an order from Minamoto no Yoriie’s chancellery (Shōgunke Mandokoro) dated 1203 which specifically forbids illegal entry by the military governor and his representatives into Ōbe Estate.53 In other words, the yeomen’s petition refers to legal precedents that were ninety years old. Moreover, the Jōkyū disturbance had occurred in 1221, more than seventy years in the past. At the very least, the author had knowledge of the precedents and court and shogunal orders he could use in this claim. 414 Power

The author also had at least a passing knowledge of Buddhist doctrine. The introduction to the petition reads “The noble temple [Tōdaiji] venerates the sacred vow of a holy age; it is a good land where the wise practice. It is a monastery without equal in Jambu,54 the undu- plicated protector of this valley of the sun [ Japan].”55 It should be noted that the document submitted by actual monks (the dean’s letter) has no real religious language in it, outside of listing all the rites funded by dues from Ōbe Estate.56 The author of the yeomen’s petition was not necessarily a scholar-monk, but he knew how to add Buddhist flourishes to his writing. Perhaps it was seen as good form to use Buddhist terms when petitioning temple authorities. Likewise, the author certainly had a working knowledge of Chinese sources. For example, the proverb “Never has a straight body cast a bent shadow, nor has order above yielded chaos below”57 is in every document we have from the yeomen in this case. The source for this saying is the Zhenguan zhengyao (Jp: Jōgan seiyō), a record of the sayings of the second Tang emperor, Taizong, on the subject of good governance. It was widely read in Japan from the Heian period onwards, and Minamoto no Yoritomo’s wife Hōjō Masako (1157–1225) is known to have commissioned a version written in the kana syllabary in the early thirteenth century. The work was clearly well read and frequently cited: there are also references to the work in numerous warrior-related narratives, such as the Shōmonki, the Heike monogatari, and the Taiheiki.58 In addition, the introduction to the yeomen’s petition includes the following: “The peace of the realm is reflected in the temple’s [Tōdaiji] rise and fall, the governance of the Hundred Kings [of previous ages] is manifested in the protection of the gods and Buddhas.” This is a paraphrase of a poem by the Tang poet Bai Juyi.59 By including it, the petition’s author was holding up Tōdaiji as the protector of good governance and the realm. This not only demonstrates the erudition of our unnamed author but also his ability and willingness to advocate on his own behalf (or, at least, on behalf of those who commissioned the petition). And in terms of his reference to akutō specifically, it indicates that by this point the use of “akutō ” as a legal strategy was not merely the domain of proprietors and the shogunate but was also a weapon in the arsenal of Ōbe yeomen in the late Kamakura period. While there is no specific evidence as to the identity of the author of these documents, there are several likely candidates. One possibility Akutō on Ōbe Estate 415 is that one or several of the yeomen wrote it themselves, perhaps one of the “hyakushō satanin” noted above. The satanin not only suffered the depredations of Shigemasa’s raids, but also the demands of the shrine associates.60 While Ōbe Estate’s internal structure is unclear, we do know that the yeomen sometimes met in council to make decisions for the whole estate.61 On other estates, satanin were among the highest ranking of the yeomen, drawn from the attendant (samurai- shū) or named-field holder (myōshu)62 class of the already prosperous yeomen.63 Another option is that an on-site estate official wrote the documents. The chief local administrator on the estate was the reeve (kumon),64 and since the original function of the reeve was the writing and organizing of official documents, that official could have been the author. That said, the reeve does not appear in the documents anywhere, so his authorship seems unlikely. A third option is that a local monk was the author. This would make sense given all the religious imagery. The temple Jōdoji was on Ōbe Estate, so it could well have served as a meeting hall for the yeomen, and one of its resident monks could have composed the petition. While I consider this possibility more likely than that of the reeve as author, the lack of mention of Jōdoji in any of the documents leads me to believe that the most likely author was one or more of the hyakushō satanin, though perhaps with outside input. But even if we knew for certain who wrote the petition, it is unclear to which subdivision of Tōdaiji the yeomen would appeal. The estate had at some point not too long before shifted from the portfolio of the Tōnan’in, a cloister within Tōdaiji, to that of the corporate Tōdaiji community (on this, see chapter 4, by Endō Motoo).65 It is likely that the yeomen sent the letter to the office of the three deans, as representatives of the Tōdaiji clergy66 (shuto), since it was they who forwarded the letter to Rokuhara. Notable too is that the akutō in this case are not, as one would expect if we took the Mineaiki at face value, described as poor, ill-equipped, or strangely dressed, as were the akutō in the Mineaiki’s first phase. Indeed, the only detail of their accoutrement in the record is that they are armed and armored,67 and Oxenboell cites Ōmura Shigemasa as being almost a perfect example of the sort of akutō that the Mineaiki suggests were common in the 1320s (roughly thirty years later).68 I will note, however, that the complaint doesn’t actually include Shigemasa among the akutō, 416 Power but it accuses Shigemasa of organizing them. I think it is significant that in this case, the akutō are nameless. In the case I will discuss next, they have names.

The Second Case, 1322 The second case took place approximately twenty-five years later. In 1322, the assembled monks of Tōdaiji sent a petition to the shogunate claiming that a group of “famed akutō ”69 had entered the grounds of the estate and harvested crops from fields that belonged to the reeve. These “famed akutō ” were listed in the petition and in an attached list of names (see table 16.1; for the locations of akutō residences, see figure 16.1). That the akutō are named here already represents a major difference from the first case, in which the akutō were unnamed followers of Ōmura Shigemasa. And we not only know the names of these men, we know where they lived and in some cases for whom they worked. If we can trust this document, two of these men were in the employ of one Shirai Hachirō, who was the land steward of Kuboki, a village on

Fig. 16.1. Harima Province. Drawn by Philip Garrett. Akutō on Ōbe Estate 417 the border of Ōbe Estate. In fact, the shogunate, when it finally sent a response on the issue, named Shirai as the ringleader.70 There are a number of monkish names on the list, but these men were probably not monks. More likely they were older men of means who had nominally retired and taken the tonsure but continued to live a lay life, as was common practice at the time. The residences listed suggest something very interesting as well: If these men did constitute a network of thieves or even of warriors, they covered more than half of the province, with the farthest two listed residences being nearly forty miles apart. Their residences are also all close to either rivers or the coast, and indeed one of the accused akutō is also called a pirate. There is perhaps also a connection to the two fortifications mentioned in the Mineaiki (at Akashi and Nagaishi), both near river mouths on the Inland Sea coast. What stands out the most to me about this case is how Tōdaiji presented it to the shogunate. As we have seen, the evidence in the first case is contained in a petition from the yeomen. The evidence in this

Table 16.1 List of Akutō in the second Ōbe Estate case, 1322

Name Residence Notes Anji Norifusa Anji Estate Hayashida, Noto Dharma Hayashida Estate Bridge Tsutsumi Gorō Tarui Village Dharma Bridge Honchi Vassal of Shirai Hachirō The Monk Enryō Houseman of Shirai (Sukezane) Hachirō. Later called Taiyūbō Nagao Bizenbō Hirano Estate Representative (daikan) of Nakanojō Keibu no Shō Tarōsaemon (Initiate Shimohata Estate Enshin)

Source: OS 4:629–630 (# 332) 418 Power second case, however, consists of the list of names and two documents from a completely different case, which revolves around depredations by several of the actors on the list from nearby Hirano Estate, a holding of the Tōin courtier family. Hirano was on the coast of Harima, about ten miles southeast of Ōbe. The first of the two documents, referred to by Tōdaiji as the “retired monarch’s edict pertaining to treason,” is an inzen, or order issued by the retired monarch Go-Fushimi (1288–1336, r. 1298–1301).71 Dated 1316, it complains that although the shogunate had already ordered the arrest of the perpetrators, their violent activities had actually worsened. Another document presented as evidence, dated later in the same year, is an order from the Rokuhara deputies to a warrior named Kasuya Yajiro72 to collect all the housemen in the province and, along with one Nagahama Rokurō Nyūdō,73 to lead an attack on the criminals, capture them, and to act against any later akutō activities in the region without first seeking Rokuhara’s approval.74 The Mineaiki notes an expedition to Harima in the first year of the Gen’ō era (1319). It was led by someone named Koresada, who was almost certainly the southern Rokuhara deputy at the time, and it included as well a warrior named Kasuya Jirō. I believe that the raid being ordered by the Rokuhara deputy is the one described in retrospect by the author of the Mineaiki. These documents are the only evidence we have for the Hirano Estate case. Presumably they were used in the Ōbe case because the same group that raided Hirano also raided Ōbe. One wonders how Tōdaiji even got these documents in the first place, as none of them is addressed to Tōdaiji or an affiliate.75 But what strikes me as more significant is the total lack of evidence from Ōbe Estate itself. What became of the yeomen? If someone were stealing crops on estate land, wouldn’t someone on the ground be losing income?76 Where is this someone and why is his report not part of the case? We must ask, moreover, why there is no sign that the yeomen showed an interest in the case. Yeomen documents do not suddenly disappear in the 1320s; in fact, several such documents from Ōbe Estate are extant.77 Perhaps the yeomen as a collective had no interest in what was at issue, and therefore no motivation to involve themselves in the case. It is also possible that for whatever reason they were opposed to the claim Tōdaiji was making, although there is no evidence to suggest this. Also, since this plot of land was called the kumon myō, the absence of any documents from or about the reeve (kumon) is also curious.78 Akutō on Ōbe Estate 419

There was likely more to the case than the simple theft of crops, in my view. Several scholars have linked the crime of harvesting and stealing crops (karita rōzeki) to another emerging legal concept: tōchigyō. Speaking broadly, this is the legal concept that if one is using the rights to a piece of land as though one owns it, then one may make a claim to own it legally. Like akutō, the term tōchigyō appears in the Goseibai shikimoku, which states that one could claim rights to a piece of land in the absence of documents if one had occupied it for twenty years. As time went on, the importance of the twenty-year holding period declined, and simply making use of the land became the only requirement, especially under the Muromachi shogunate (1338–1573). As several scholars have noted, complaints about illegally cutting crops can be read as a first step in making a tōchigyō claim, that is, a way of denying someone else’s claim.79 Perhaps Shirai and his followers were trying to stake a claim to parts of Ōbe Estate permanently. Or perhaps Tōdaiji was trying to claim land that had not previously been part of the estate. Morton Oxenboell has argued too that the increase in complaints about akutō in this period (late Kamakura) should be read not as proof of increased lawlessness but of an increase in the resort to law. In other words, cases like this one indicate the increasing use of laws proclaimed by central authority in local disputes, especially in areas like Harima, which was close enough to the capital and to Tōdaiji for the shogunate and the temple to maintain control, even as more remote estates became less and less reliable.80 This case and the evidence presented therein seem to support this argument. Recall also the description of akutō in the last part of the Mineaiki account, which describes a period beginning in 1324: “They usurped otherwise undisputed land, claiming to be retainers of the original holders.” The Hirano documents are dated 1316, and the Ōbe Estate complaint is dated 1322. The times are tantalizingly close, especially when we consider that the Mineaiki is meant to be a recollection written decades later. Further, like the akutō in the account of 1324 in the Mineaiki, the akutō in this case seem to resemble ordinary warriors. The alleged ringleader, Shirai Hachirō, is a land steward. Nagao Bizenbō is listed as being the local representative for a high-ranking official at Rokuhara, indeed, the head of one of the panels of the Board of Coadjudicators.81 Two others are listed as being Shirai’s vassals. There is a high probability that most, if not all, of the men Tōdaiji complained about here were housemen of Kamakura, which might partly explain 420 Power why the shogunate did not execute them after the incident at Hirano and during the later raid on the province.82

So what are we to make of these two cases? One conclusion is that Tōdaiji succeeded in using the charges of akutō activities, as temple authorities were able to move the shogunate in both cases to demand that local housemen investigate. In the first case, they succeeded in having the perpetrator arrested. In the second case, the result is less clear. The evidence from the first case demonstrates clearly that not just proprietors, which have been the focus of much of the scholarship on akutō cases, but also local elites could and did make use of the term “akutō ” to get what they wanted: the shogunate’s attention. I have also argued here that the second case is most likely an argument over the legal status of a piece of land and that Tōdaiji seems to have won, at least in the legal arena. In terms of the estate system itself, that residents and low-level estate officials were involved in the legal system suggests how and why the system held together. It has long been noted that the proprietors and the estate officials on site were dependent on each other, and the participation in estate operations by anyone, whether legally compelled or not, was on some level contractual. In the thirteenth century one of the reasons the yeomen of Ōbe (or perhaps on other Tōdaiji estates) were willing to put up with the demands placed on them by the temple was that the relationship gave them a means of calling on the power of the shogunate against their enemies by initiating suits. Given that the two cases are almost thirty years apart, it is tempting to see them as evidence of changes over time in shogunate policy or proprietor tactics concerning akutō cases. I think, however, that it is pre- mature to conclude this based on only two cases on one estate. Instead, we should look at these two cases as an example of the variety of forms that these akutō cases could take. In the first case, the complaint of both the yeomen and Tōdaiji was about theft of moveable property, violence, and kidnapping. In the second, the temple complained about theft of produce from the land, and potentially the land itself. In the first case of 1295, the yeomen on site played an active role in pursuing the issue, and their petition formed the bulk of the evidence for the initial complaint. In the second case of 1322, there were no on-site documents used by Tōdaiji to make its case. Nonetheless in both cases the shogunate ruled for Tōdaiji and made some attempt to deal with the accused. So while the first case demonstrates that the legal tactic of calling one’s oppo- Akutō on Ōbe Estate 421 nent an akutō could be used by actors other than estate proprietors, both cases demonstrate that the proprietor made the ultimate decision about what evidence to use or which cases to bring before the authorities. While the yeomen were more involved than some previous scholarship suggests, the proprietor did not need the yeomen in order to lodge an akutō complaint with the shogunate. I would also like to stress here too that in the first case we have seen that Tōdaiji edited documents it forwarded to Rokuhara. This suggests a policy of deleting elements that were either not germane to the case or that reflected poorly on the temple. Whichever was the case here, we can be almost sure that among the other evidentiary documents we use as sources are documents edited in the same way, and we must be mindful of this possibility as we utilize sources in the future. In any case, because fortunately we have two different versions of a document from different archives, the first Ōbe Estate akutō case provides an opportunity to see an aspect of the Kamakura legal system that has for the most part remained well hidden.

Notes

I would like to thank Morten Oxenboell for his help with this project, both in sharing his research and offering advice on changes to this chapter. I would also like to thank Joan Piggott and Janet Goodwin, who introduced me to Ōbe Estate and likewise provided helpful guidance. Finally, I would like to thank Endō Motoo and Takahashi Toshiko for help with finding additional sources and reading documents. Any errors are, of course, my own. 1. Oxenboell 2006, 6–7. 2. A list of all known akutō documents, including the shogunate missives and the estate proprietor’s complaints, can be found in Ōtake 2005. 3. OS 4:575 (# 272). 4. Ōtake 2005. 5. Ōtake’s list of akutō records includes 425 documents, of which only 79 do not deal with a specific estate. This becomes more pronounced as time progresses. For documents after 1250, fewer than 17 percent are not specific to a single estate. After 1300, fewer than 5 percent are not estate-specific. See Ōtake 2005. 6. Negayama 2011, 72–73. 7. Ibid., 76. The document that so refers to the Mongols is KI 15:332 (# 11741). 422 Power

8. My account of the historiography of the field largely follows Oxenboell 2005 and 2009, with my own additions. 9. Kuroda 1974. 10. Amino 1995b. 11. Nagahara 1955. 12. Akutō Kenkyūkai 1998, 2005, and 2013. 13. Watanabe 2013, 210–211. The essay in question is Satō 1969. The other two essays are Amino 1963 and Amino 1970. 14. Harrington 1982, 221–250. 15. Souyri 2001, 100–110. 16. Yamakage 1977, 1–41. 17. Perhaps the most important law code put out by the Kamakura shogunate was the Goseibai shikimoku (also called the Jōei shikimoku) of 1232, which included akutō in its third article, concerning traitors and arsonists. Most of the laws relating to akutō are part of what are now called “Tsuikahō” (roughly “supplemental laws”), which were added in the years after 1232. See Satō 1996, 76–80, for a large selection of the laws related to akutō. 18. Over time the specific crimes required to make one an akutō became less well defined, to the point where eventually one could justify an akutō claim with nonviolent offenses, such as gambling. See Shimozawa 2003. 19. Shimozawa 2003; Kondō 1993; Oxenboell 2005, 2009. 20. OS 2002, 1:411. 21. It is also, in keeping with the numerous ways of reading the characters, sometimes called the Hōsōki or the Bushōki. 22. For an excellent study of the Mineaiki, its composition, and its use as a historical source, see Ōyama 2002. 23. Oxenboell 2006, 6; 2009, 59–61. 24. Oxenboell 2006, 6–10. 25. Translation from Oxenboell 2006, 7, with one alteration. It should be noted that Oxenboell is breaking with some scholars in saying that the akutō did not wear hakama and eboshi, but I believe that his reading makes the most sense given that the opening says that they dressed abnormally. My own change here is to the phrase yoroi haramaki, which Oxenboell translates as “armor and breast- plates,” but which I believe should be rendered as “armor of all/any kind.” The entirety of the original text of the Mineaiki’s akutō account can be found in Zoku gunsho ruijū 28.1: 223–224. 26. Satō 1996, 50–51. 27. Arai 1990, 157–158. 28. See the continuation of the above quoted passage from Oxenboell 2006. Akutō on Ōbe Estate 423

29. The Mineaiki refers to him as “Ōshū Koresada.” Ōsaragi was given the title of governor of Mutsu (Ōshū). Kokushi daijiten s.v. Ōsaragi Koresada. There were two deputies, the “Northern” and “Southern” deputies (so termed because of their residences), in the shogunal office at Rokuhara in Kyoto. 30. Oxenboell 2006, 8–9, with some revisions. 31. Ibid., 9. The idea that akutō are the predecessors of ikki has been put forth several times before. See Harrington 1982, 240. For an introduction to ikki, see Souyri 2001, 128–131, 161–166, and 183–202. 32. Oxenboell 2006, 9–14. 33. Some scholars have argued that the descriptions in this phase indicate that abnormally showy outfits were worn, indicative of a new and subversive style called basara. For a brief English discussion of the basara argument, see Souyri 2001, 108–109, and for a rebuttal of the argument see Oxenboell 2006, 13. 34. This follows Steenstrup 1987 and 1980. 35. If the case was from eastern Japan, it was the domain of the Board of Retainers (Samuraidokoro) in Kamakura. 36. There were three such panels, one at Kamakura (for the Kantō), one at Rokuhara (for the area around the capital), and one in Hakata (for Kyushu). 37. Members of the East Asian Law Research Group of the Project for Premodern Japan Studies (PPJS) at the University of Southern California have translated and annotated many chapters of the textbook Shiryō de yomu Nihon hōshi (Murakami and Nishimura 2016). Therein, chapters that focus on issues of Kamakura law (esp. chaps. 3, 4, 19, 20, 24) supplement this overview of the Kamakura court system. For these translations see the archive at uscppjs. org/Research. In addition, translations of various shogunal formularies and supplemental laws can be found in the notes of the USC Kambun Workshops (esp. 2013, 2017), at uscppjs.org/Kambun Workshop. 38. In cases in which a full panel wasn’t used, the preliminary hearing might have been eliminated. 39. The Sata mirenshō can be found in Chūsei hōsei shiryōshū, vol. 2, 355–376. An English translation is Steenstrup 1980, 405–435. Another helpful guide to issues in Kamakura law is Mizubayashi et al. 2001, 105–230. 40. Steenstrup 1987, 97. 41. Steenstrup 1987, 81, refers to this body as the “Supreme Court cum Council of State.” I have used Jeffrey Mass’ translation (1995, 289). 42. Sanuki was one of the provinces of Japan, and “lord” (kō) was an appellation indicating high rank. Presumably this Sanuki had served in one of the governorship postings in Sanuki at some point, but his identity is unclear. 424 Power

43. OS 4:571–572 (# 268). These included Kusunoki Kawachi Nyūdō, who several scholars have suggested could be the father of the later Nanbokuchō- period (1333–1392) hero, Kusunoki Masashige. See OS 2002, 1:413–414. 44. Many of the documents list him as Tarumi Shigemasa, and it is unclear which, if either, is correct. I use Ōmura, following the compilers of Dainihon komonjo and Ono-shi shi. Further discussion of Ōmura and his family can be found in OS 2002, 1:416. 45. My focus in this section will be on one of these documents, OS 4:571– 572 (# 268), but I will at times reference other documents in this case presented by the yeomen. 46. I am indebted to Philip Garrett for this translation. See his chapter in this volume. I should also note that hyakushō is not used here as in modern Japanese to mean “farmer” or “peasant.” The yeomen represented the highest echelon of the estate’s society (excluding the estate officials such as the reeve (kumon) and the administrators). These people were almost certainly prosperous and probably rented land and equipment to poorer farmers. 47. Also read jinnin. 48. OS 4:575 (# 272). 49. For co-conspirators, see OS 4:581 (# 280). For Shigemasa’s arrest, see OS 4:583–587 (# 286). 50. OS 4:571–572 (# 268), KI 24:223–224 (# 18734). See Endō Motoo’s chapter in this volume for a discussion of Tōdaiji document collections. 51. OS 4:573–574 (# 269), KI 24:222 (# 18733). 52. OS 4:571–572 (# 268), KI 24:223–224 (# 18734). 53. OS 4:536–537 (# 226). Yoshiie (1182–1204) was shogun from 1202–1203. 54. Jambu (Jp. Enbu) is the Sanskrit term for the human world. Digital Dic- tionary of Buddhism, s.v. Jambu. 55. OS 4:536–537 (# 268). 56. OS 4:575 (# 272), KI 24:231 (# 18748). 57. Original Chinese: 未有身正而影曲、上理而下乱者. 58. Jōgan seiyō 1978, 12. 59. 四海安危表寺家興衰、百王理乱、顕鎮守擁護 from OS 4: 571–572 (# 268) is a paraphrase of 四海安危照掌内。百王理乱懸心中, “[When the Emperor uses wise ministers,] the peace of the realm is reflected [as in a mirror] in the palm of his hand, and the governance of the Hundred Kings [of previous ages] is reflected in their hearts.” From Bai Juyi’s poem “Bai lian jing” (Japanese Hyakurenkyō), a poem in the Wakan roeishū compilation. My translation is based on Sugano’s notes in Wakan rōeishū 1994, 342. 60. OS 4:571–572 (# 268), KI 24:223–224 (# 18734). Akutō on Ōbe Estate 425

61. OS 4:583–585 (# 286). 62. For more on myōshu, see Keirstead 1985, 311–330. 63. Kokushi daijiten s.v. satanin. 64. See OS 2002, 1:434–452 for a summary. 65. Indeed, the second yeomen’s petition in this case [OS 4:575 (# 272), KI 24:231 (# 18748)] is the first source that shows the changeover. See Konishi 2003, 45, and Endō Motoo’s chapter in this volume. 66. For example, see OS 4:585–586 (# 288), KI 26:186 (# 19792). 67. OS 4:585–586 (# 273) thus describes the akutō’s dress: “鎧甲冑、帯弓箭” (armored, carrying bows and arrows). 68. Oxenboell 2009, 74. 69. OS 4:630–631 (# 334). 70. OS 4:650 (# 344). 71. OS 4:624 (# 323), KI 34:27 (# 25875). 72. Probably Kasuya Shigeyuki (1270–1333), a vassal of one of the Roku- hara deputies. Shigeyuki killed himself along with a number of other Roku- hara vassals at Rengeji in Ōmi after Rokuhara fell in the battles that ended the Kamakura shogunate in 1333. Takahashi 1989, 42; “Ōmi banbashuku Rengeji kako chō,” in Gunsho ruijū 29:245. 73. The Taiheiki (1994, 1:533 and 508 note 27) notes a man with the same name in the service of Nitta Yoshisada. 74. OS 4:624 (# 324). 75. Most likely someone on Ōbe Estate had sent Tōdaiji information about what had occurred at Hirano, and Tōdaiji requested copies from the Tōin family. The latter was no doubt anxious to be rid of the troublemakers as well. 76. This may seem like argument from absence, but it is not. There are not several disparate documents that serve as our basis for understanding this case, there is only one, consisting of copies of several documents (including the Hirano documents) in Tōdaiji’s collection. This document was likely made in case the temple needed to relitigate the case or the precedent was useful in some other matter. I therefore find it unlikely that there was once a document from Ōbe itself in this bundle of documents that was somehow lost. If there was an on-site document that was used in this case, then Tōdaiji’s archivist choose not to include it, which suggests that any hypothetical on-site document was seen as less useful to future litigation than documents from the Hirano case. Ono-shi shi treats all the documents for this case as separate documents but notes that they were in one batch, but Tōdaiji monjo treats them as one document (see OS 4:624 (# 323) and Tōdaiji monjo 19:333). 77. For example, OS 4 #s 349 and 350 are both from the Karyaku era (1326–1329). 426 Power

78. Examples of documents by the reeve include OS 4 #s 330, 333, and 335, all written during the same year as the complaint. It should perhaps be noted that a long-lasting dispute over the reeve’s position had begun by this point; the first legal filings began in 1322. I see no reason, however, why this should prevent either (or both) of those litigants from complaining about raids on land they reportedly owned. It is also possible (though unlikely) that while the parcel of land was called “the kumon myō ” (the reeve’s name-fields), that one of the previous reeves had sold it and that kumon myō was just a place name at this point. For more on the dispute over the kumon position, see OS 2002, 1:434–452. 79. Kobayashi Kazutake, quoted in Ōtake 2005, 303. 80. Oxenboell 2009, 79–82, 227. Further, while Harrington and others have noted that the vast majority of akutō cases are from the Kinai region, Oxen- boell has pointed out that in the shogunate’s own laws and edicts, akutō are for the most part identified outside of the Kinai. This disconnect suggests that the akutō threat that the shogunate was concerned with was not precisely the same threat that the estate proprietors complained about. See Harrington 1982, 237, and Oxenboell 2009, 78. 81. Nakanojō Keibu no jō (Kaitō Hiromasa) was head of one of the panels of the Board of Coadjudicators at Rokuhara at the time of this lawsuit. See Mori 2005, 148. 82. Although to be fair, the record generally shows that the shogunate rarely executed captured akutō. c h a p t e r 17

Warriors and Estates in Muromachi-Period Harima

Noda Taizō Translated by David Eason

The Tōdaiji landholding Ōbe Estate is one of several examples of the ties between warriors and estates in Harima Province. Beginning in the latter half of the Kamakura period, Ōbe Estate was beset with warrior- related problems that ranged from rampaging “evil bands” (akutō), to various Ō family claimants arguing over appointment to the post of reeve (kumon), to the diversion of irrigation water to neighboring holdings by land stewards (jitō) and other warriors. Yet these were certainly not the only instances of interaction between estates and warriors, whether those within or outside the borders of any given estate. Rather, throughout both the Nanbokuchō and Muromachi periods, residents of estates such as Ōbe, Yano, and Ikaruga (see figure 16.1) in Harima maintained regular ties with the warrior establishment as represented by the provincial military governor (shugo), a relationship that raises fundamental questions about residents’ dealings with such figures as well as the more general impact warrior actions exerted over the lives of local inhabitants. Both of these points are concerns here.

Estates and Provincial Administration during the Nanbokuchō and Muromachi Periods Let us start by confirming connections between estates and military governors in Harima Province throughout the Nanbokuchō and Muroma- chi periods.1 Particularly during the former era of pronounced unrest, as a province just beyond the outer perimeter of the capital region,

427 428 Power

Harima was frequently transformed into a battlefront and was the site of repeated altercations. The Akamatsu family, entrusted with the post of military governor, accordingly established a province-wide system of conscription for the purpose of waging war.2 Specifically, they first proceeded to impose a diverse assortment of taxes and duties on both estates and public lands in the province. They collected lumber, fodder, commissariat rice, and other resources as “requisite fortress provisions.” For instance, extant tax invoices from the Kyoto temple Tōji indicate that such imposts were levied on Yano Estate, one of the temple’s holdings in southwestern Harima. Thus, when in 1354 the shogun Ashikaga Yoshiakira (1330–1367) encamped in the province at a place called Hiroyama, Yano Estate was ordered to furnish lumber to construct the shogun’s field headquarters as well as laborers to plaster its walls. More- over, the same estate was also required to contribute men and commissariat rice to aid in the building of the Akamatsu family’s main fortress at Kinoyama (see figure 16.1), and they were further burdened with paying for expenses incurred by the military governor’s agents.3 The next obligation, that of military service, was similarly borne by the deputies (daikan) and lower-ranking officials (satanin) who resided on the estates of noble and religious proprietors. The holding of these positions by warriors was not new, though they had rarely been subjected to summons from either shoguns or provincial military governors in the past. From the start of the Nanbokuchō period, however, hard- pressed governors no longer hesitated to contact the warrior residents of courtier and ecclesiastical estates. In the case of Ōbe Estate, during the period from 1338 until 1342, both the custodian (azukaridokoro) and the reeve joined with the forces of the military governor to fight against par- tisans of the Southern Court in the mountainous border area between Harima and Settsu Provinces.4 Those who did not respond affirmatively to mobilization orders were viewed as enemies and deemed legitimate targets for attack. In addition, the practical issue of preserving authority over one’s landholdings was also now linked to demonstrations of allegiance to the military governor. Moreover, what had initially been temporary and limited imposts by military governors became increasingly routine as instability intensified, until they at last emerged as a standard feature of provincial administration under the rubric of “duties owed the military governor” (shugoyaku). The precise mechanisms through which military governors arranged to collect these imposts are well illustrated through the following exam- Warriors and Estates in Muromachi-Period Harima 429 ple. During the Nanbokuchō period the Ogawa family, a lineage of local notables who resided in a place called Sakamoto at the foot of Mount Shosha (the location of Enkyōji; see figure 16.1) on the northern outskirts of Harima’s provincial seat (present-day Himeji City), held the court-appointed post of deputy provincial supervisor (kokuga gandai). In Harima the functions of the provincial seat persisted even into the Nanbokuchō and Muromachi periods, while sources reveal the continued presence of a resident deputy to the provincial governor (komokudai) together with other closely affiliated lesser officials entrusted with the supervision of criminal investigations. As a post equivalent to that of resident deputy in earlier times, the deputy provincial supervisor initially would have been expected to handle the tasks of tax collection associated with both mundane land management and extraordinary province-wide levies. Beginning in the latter half of the 1370s, however, the Akamatsu military governor came to exercise direct control over such taxes, exemptions to them, and, after 1392, even appointments to the supervisor post itself. And as a means to strengthen their rule over the entire province, it is little surprise that Akamatsu leaders sought to incorporate well-connected local bureaucrats with extensive knowledge of taxation like the Ogawa family into their administrative apparatus, especially since members of the Ogawa lineage presumably maintained an official land ledger that contained details concerning the ownership history and area of all estates and public lands throughout the province, a powerful tool indispensable for the assessment and collection of taxes. As military governors of Harima, Akamatsu leaders also confirmed the prerogatives of another key provincial family, the Ōi, in their management of Iwa, the primary government-sponsored shrine (ichinomiya) in the province, thus enhancing Akamatsu control over the shrine as well as the shrine complex within the precincts of the provincial headquarters. (For Iwa Shrine, see figure 16.1.) Furthermore they regularly ordered Enkyōji Temple on Mount Shosha and Ichijōji Temple on Mount Hokke (see figure 16.1), two of the six leading Tendai establishments near the provincial headquarters, to offer prayers for the blessings of peace and prosperous harvests in all parts of Harima. In this way, by organizing religious observances on an extensive scale, the Aka- matsu family was able to promote an image of the military governor as one who ensured regional order and tranquility, thereby encouraging a heightened degree of acquiescence to their rule. 430 Power

Such efforts to absorb the functions and prestige associated with the duties and customs of provincial governors from ages past were not limited to the Akamatsu military governors. Indeed, during the Yamana family’s tenure in that office in the second half of the fifteenth century, end-of-season dues were forwarded to the Yamana governor during the first, eighth, and final months of the year.5 In addition a “young herbs offering” (wakanafu) and other offerings once made at the beginning of the year from local estates to proprietors went instead to the Yamana governor. In this way military governors came to be deeply involved in the affairs of shrine and temple holdings with ever more frequency during the Nanbokuchō and Muromachi periods. At this time the government of the Ashikaga shoguns wielded the greatest authority to guarantee the management of estates, with either the shogun or his subordinates issuing official directives (migyōsho) in response to requests put forth by parties with competing claims. Yet whether or not such documents proved effective depended entirely upon the reception they garnered from those who presided over local affairs. In practical terms, then, it was none other than military governors who ensured the continued management of estates. For this reason it remained essential for estate managers to forge amicable ties with military governors and their followers, a process that depended on offering both scheduled and extraordinary “tribute” in the form of money, goods, and labor, as well as a regular supply of products such as charcoal, timber, and tea.

Shrine and Temple Holdings in Harima In the spring of 1429 a local league (ikki) rose up in the neighboring province of Tanba, prompting a disturbance that the military governor there, Hosokawa Mochimoto (1399–1429), quickly moved to quash. The movement was mostly concentrated in “those lands controlled by temples and shrines,” and a policy of strict reprisals was decided upon. Although the noble and religious overlords of these estates made numerous entreaties, their pleas went unanswered. Rather, the method chosen to resolve the conflict corresponded to one that had already been instituted when dealing with recalcitrant estates in Harima Prov- ince the year before.6 Specifically, that previous winter a movement demanding debt relief (tokusei ikki) had mobilized throughout Harima, at which time the military Warriors and Estates in Muromachi-Period Harima 431 governor there dispatched messengers and military forces to forcibly pacify and occupy the affected areas. Requisitioning of commissariat rice also appears to have taken place.7 A review of estate-related sources confirms that Tōji’s Yano Estate and Hōryūji-held Ikaruga Estate, as well as Iwa Shrine lands in Shisō District, all experienced disturbances, lending credence to the claim that the estates of noble and religious overlords served as epicenters of unrest. Indeed, there was a trend among noble and temple-run estates to shun the interference of warriors. Even a letter of compliance offered by Ōbe Estate’s on-site deputy included the assertion that “no agreements will be entered into with the military governor or other higher authorities,” evidence of the disdain with which the military governor’s intervention was viewed.8 Administered with relative moderation in comparison to lands controlled by warriors, the estates of noble and religious overlords presented an attractive stag- ing ground for the formation of leagues for precisely this reason.

Military Governorships after 1441 and the Impact on Provincial Society Following his assumption of the post of shogun in 1428, Ashikaga Yoshi- nori (1394–1441) began to intervene aggressively in the succession patterns of leading warrior families to an unprecedented degree. Among his targets were the military governor and family headship posts held by members of the Yamana, Shiba, Kyōgoku, Isshiki, and other powerful warrior lineages. In the first month of 1441, for instance, he compelled Hatakeyama Mochikuni (1398–1455) to yield his family’s headship to a younger brother and in the sixth month of the same year, he forced the military governor of Kaga Province, Togashi Noriie, to abdicate his position to a sibling as well. Both of these instances inspired little overt resistance. Yet just a few months later, when Yoshinori attempted a similar maneuver to replace the established heir to the Akamatsu family with an alternate of his own choosing, the outcome was quite different. Informed of this impending reassignment, Akamatsu Noriyasu (1423–1441) acted pre-emptively by inviting the shogun to a banquet and arranging for his unceremonious assassination during the height of the festivities. Noriyasu and his father Akamatsu Mitsusuke (1381–1441) thereafter returned to their home province of Harima bearing Yoshinori’s sev- ered head. Attacked in response by other leading families within the 432 Power

shogunate, both men fell in battle while attempting to defend Kinoyama Castle during the ninth month of 1441, and the post of military governor of Harima was subsequently bestowed upon Yamana Mochitoyo (1404–1473). Governorships in the provinces of Bizen and Mimasaka were given to other members of the Yamana family. Three of the sixteen districts in Harima had been entrusted to Akamatsu Mitsumasa (d. 1445), a close retainer of the Ashikaga, but even these few holdings were reassigned to the Yamana family in the first month of 1444. As a result, Mitsumasa banded together with his kinsman Akamatsu Norihisa (1425–1455) to travel to Harima, raise troops, and launch an abortive rebellion that was ultimately crushed the following year. And then again more than ten years later, in the winter of 1454, Norihisa once again journeyed to Harima to regain control over these three districts, only to be pursued by the shogun’s forces and driven to suicide within a few short months. So political instability continued to wrack Harima after 1441, visible not only in the transition to a new military governor, but also in the turnover among local warriors. For while warriors who had served the Akamatsu found their positions either diminished or discontinued, an entirely new cadre of Yamana family followers gained access to posts and land. For example, during the tenth month of 1441, when the magistrate dispatched to Shisō District by the Yamana family entered the region, he also seized the residence of the Akamatsu-appointed deputy military governor Uno Mitsutaka. As a result Mitsutaka had no choice but to relocate to a nearby temple, though it was said that his eleven- year-old brother opted to side with the Yamana in order to retain some of the family’s prior influence in the region.9 At this juncture new figures also began to emerge in and around Ōbe Estate. This included Tarui Nakatsukasa-no-jō, a warrior completely missing from earlier sources and whose actions now appear in written records for the first time. Likely hailing from a place called Tarui, just to the south, he appears to have been among those warriors selected by the Yamana to become intimately involved with matters at nearby Ōbe Estate. In replacing the Akamatsu as military governors of Harima, Yamana leaders systematically dispatched deputy military governors (shugodai) and district magistrates (gundai) while also stationing deputies and local officials (kyūnin) throughout the province. Their rule was exceed- ingly unpopular with the owners of estates. Writing in his diary Ken- Warriors and Estates in Muromachi-Period Harima 433 naiki (The journal of Kennai), the court noble Madenokōji Tokifusa (1394–1457) lamented, “The estates and annual revenues that belong to both nobles and religious establishments, as well as those due warriors, are plundered with a severity terrible to behold. Should the Yamana family gain appointment as military governors, then the province is certainly doomed.” Furthermore, Tokifusa noted that in contrast to the Hosokawa and Hatakeyama families who, among military governors, “knew propriety,” the Yamana were “destroyers of order.”10 That the Madenokōji family held properties in the provinces of Harima and Mimasaka then under Yamana control helps to explain the apprehen- sion expressed concerning the new military governors’ rule. Ōbe Estate lay within the boundaries of Katō District, which, along with Akashi and Miki, was one of the three districts briefly entrusted to Akamatsu Mitsumasa following the events of 1441. Because this territory had formed the core of the Akamatsu’s holdings prior to its reassignment to Yamana Mochitoyo in 1444, Yamana rule there was harsh in the extreme. This, too, was related in Kennaiki, where it is recorded that Mochitoyo ordered district-wide cadastral surveys to be conducted for all noble and religious estates in the region.11 The surveys consisted of multiple components. First, each estate was subject to a review of field dimensions, the types and amounts of taxes and levies, and whether or not those laborers conscripted for long-term corvée obligations had yet to return. Next, the surveys recorded whether or not an estate was administered directly by an agent with ties to a noble or religious establishment and, moreover, whether or not there had been any change in the division of an estate’s revenues between the portion forwarded to the owners and that remitted to the military governor since the time of Akamatsu Mitsumasa’s tenure or earlier. In addition to seeking answers to these questions, investigations also centered on whether or not the managers of individual parcels of estate land (myōshu) once in service to the Akamatsu were now recognized followers of the Yamana, and orders were issued directing that in cases to the contrary, their holdings were to be confiscated forthwith. Next, Yamana leaders ordered inquiries into the holdings of their own retainers and whether or not they claimed lands previously acquired from Akamatsu Mitsumasa as part of their original benefice. Lastly, the Yamana even demanded the submission—together with accompanying oaths (kishōmon) witnessed by the local populace—of detailed reports specifying the full expanse of fields and total labor obligations expected for all recently confiscated 434 Power holdings. Under the authoritarian rule of the Yamana, the confiscation of estates, reassignment of local officials, forceful appropriation of revenues, refusal to forward rice taxes and additional dues, and unlawful seizure of lands all increased. Faced with this situation, estate proprietors petitioned the Ashikaga to issue official directives that would draw upon the authority of the shogun and his subordinates to bring the Yamana family’s “outrages” to an end. Those members of the Hosokawa and Hatakeyama families who alternated in occupying the post of deputy shogun (kanrei) were comparatively faithful in executing the orders of the shogun. For this reason contemporary estate proprietors praised them as those who “knew propriety,” in contradistinction to the “order-destroying” Yamana. In this way, by the time of the reign of Ashikaga Yoshimasa (1436–1490), the leading warrior families had gradually coalesced into factions. Ten- sions between the Hosokawa and Hatakeyama, a rivalry that pitted the Yamana against the Hosokawa—who backed a revival of the Aka- matsu—and opposition among members of the Hatakeyama and Shiba produced struggles for positions of leadership within each of these families. Numerous political rivalries of this sort overlapped and intertwined until, ultimately, these warrior families split into two competing camps responsible for initiating the great disorder of the Ōnin era, beginning in 1467. At the risk of overstating the case, it might be said that the shogun’s administration consisted of two opposing factions. The first group was headed by the Hosokawa and pursued a policy of preserving the holdings that belonged to nobles and religious establishments, while the rival camp was represented by the Yamana. These were the seeds that would sprout into the Ōnin War thirty years hence.

Killing Gorōzaemon at Ōbe Estate The impact on local affairs of the 1441 transfer of the military governorship from the Akamatsu to the Yamana is particularly evident when viewing the details surrounding the murder of a prominent cultivator named Gorōzaemon at Ōbe Estate in 1454. Gorōzaemon was one of a group of influential cultivators on Ōbe Estate, and both he and his son Gorō came to serve as agents of the Yamana family. Specifically he took over the post of a record keeper (zushi) formerly held by an individual named Genzaemon-no-jō and thus came to wield substantial power within the boundaries of the estate. Warriors and Estates in Muromachi-Period Harima 435

In the winter of 1454, however, beset by fellow cultivators who called for debt relief, Gorōzaemon retreated from his holdings just as the inhabitants from one part of the estate, Fukuda Township (ho), put his home to the torch. (For Fukuda Township, see figure 16.1.) Thereafter, after seeking the protection of the Yamana family, Gorōzaemon returned to the estate only to suffer yet another assault in the form of a nighttime raid that claimed his life.12 The people who killed Gorōzaemon initially attacked him under the pretext of demanding debt relief, but it was said too that he had been secretly conspiring with the Akamatsu family. Shortly after the death of Gorōzaemon, Akamatsu Norihisa was appointed military governor of Harima by the shogun Ashikaga Yoshi- masa. In Kyoto the opposition to this move expressed by Yamana Mochitoyo earned Mochitoyo both Yoshimasa’s ire and an early forced retirement. Meanwhile upon entering Harima Province, Norihisa attempted to advance his rule in various ways, for example by ordering the stockpiling of tax rice at the fortress of Shirahata (see figure 16.1) and requisitioning ten koku of rice from Ōbe Estate to provide support to his impoverished followers. Nevertheless, by the fourth month of the next year Ashikaga policy had undergone a reversal and Norihisa became a target of officially sanctioned pursuit. Norihisa’s last days were spent battling a Yamana-led invasion of Harima before, in the fifth month, Norihisa was finally driven into neighboring Bizen Province, where he and a number of family retainers perished by their own hands. It seems that Gorōzaemon’s surreptitious correspondence with Aka- matsu Norihisa introduced into Ōbe Estate the specter of unwelcome chaos that spurred fearful residents to carry out their murderous assault on him. With Norihisa’s arrival in Harima, Gorōzaemon’s son Gorō was able to return to Ōbe Estate before he too was quickly prompted to flee.13 As influential inhabitants of Ōbe Estate, both father and son had served not only as agents of the Yamana family but also as secret infor- mants for the Akamatsu, revealing the extensive ties of service they had with military governors. This kind of close relationship between estate residents and warrior authorities gave rise to a host of upheavals and other events of consequence on local estates. Thus, during the period after Gorōzaemon was killed and his son was driven headlong in flight, the former’s posts were promptly reassigned to officials selected by the Yamana family.14 Warrior intervention was common when confronted with disputes over competing interests or political upheavals within individual estates. 436 Power

For instance, a well-known conflict concerning the post of reeve at Ōbe Estate raged among members of the Ō family throughout the first half of the fourteenth century, until the position was taken over by the Toyofuku family in 1359 and then in 1392 was transferred to Tōdaiji as part of a donation made by Ashikaga Yoshimitsu (1358–1408).15 Like- wise, the post of record keeper seized by Gorōzaemon and confiscated by the Yamana after his death had become, by the opening years of the Ōnin War, a subject of fierce contention between Kōzuki Genjirō and Mishima Ukyō-no-suke, two individuals both in service to the Aka- matsu.16 In addition, at the time of Akamatsui Norihisa’s demise in 1455, the Yamana had indicated their intent to appoint a list of followers to positions even at Tōji-managed Yano Estate.17 In this way warrior forces, from the military governor on down, intruded into estates in instances of dispute or political unrest. And through the steady accretion of appointed subordinates and seized posts, they developed warrior domains carved out from the holdings of noble and religious institutions, a part of the economic underpinnings for daimyō rule during the subsequent era of Warring States.

Notes

1. During the Nanbokuchō period (the period of Northern and Southern Courts), there were two rival royal courts, one in Kyoto and the other in the mountains of Yoshino in Yamato Province. This was a period of civil war, which ended in 1392 when the southern court acquiesced and the two courts were reunited. The military authority of the time was the Muromachi shogunate, established in 1336 and located in Kyoto. The Nanbokuchō period is sometimes considered the initial phase of the Muromachi period. 2. Itō 1993. 3. Aioi-shi shi 8.1:190–193 (# 207). 4. Hyōgo-ken shi 5:227–232 (#s 159–161, 164, 166–170). 5. Aioi-shi shi 8.2:831–838 (# 1034) and 840–848 (# 1038). 6. Mansai Jugō nikki,17 (Shōchō 2 (1429)/2/5). 7. Aioi-shi shi 8:178–184 (# 780), 202–209 (# 793). 8. OS 4:753–754 (# 477), 767–768 (# 486). These sources are also available in Hyōgo-ken shi 5:268–269 (# 227), 270–271 (# 233). 9. Kennaiki, Dainihon kokiroku 14:4, 190 (Kakitsu 1 (1441)/10/28). 10. Kennaiki, Dainihon kokiroku 14:4, 130–132 (Kakitsu 1 (1441)/intercalary 9/20). Warriors and Estates in Muromachi-Period Harima 437

11. Kennaiki, Dainihon kokiroku 14:7, 29–30 (Bun’an 1 (1444)/4/14). 12. Hyōgo-ken shi 5:439–441 (#s 272, 274, 275). 13. Hyōgo-ken shi 5:441 (# 276). 14. Hyōgo-ken shi 5:442 (#s 277, 278). 15. Hyōgo-ken shi 5:253 (# 202), 262–263 (# 217). 16. OS 4:1020 (#s 570, 571). 17. Aioi-shi shi 8.2:777–778 (# 1018). pa rt v i i

Getting the Word Out

c h a p t e r 18

Teaching Japanese Estates Old Challenges and New Opportunities

Ethan Segal

Shōen, most commonly referred to in English as “private estates,” are among the most important institutions for understanding pre-1600 Japanese history. Any study of landholding, early peasant communities, or center-periphery relations must take estates into account. Although much attention is given to premodern elites, most of whom were liter- ate and resided in urban settings, the majority of the people lived in the countryside, many of them on provincial estates. Rent records, judicial pronouncements concerning disputes over estate lands, and correspondence between rural estates and their proprietors constitute a significant percentage of the extant primary source corpus. These documents provide crucial information on a wide range of topics including economics (taxation rates), agriculture (double cropping and field use), religious institutions (Buddhist temple land administration), and juris- prudence (early medieval legal systems). In addition, because estates took on various forms over a period of eight hundred years, they offer valuable insights into the wider political and economic concerns of their times. The earliest estates, which first appeared in the eighth century, reflected the evolution of the Taika Reform principles that informed the design of classical government and land management. In the eleventh century, estates became “political footballs” in the contest between the retired emperors and the aristocratic Fujiwara family as they vied for control of the government. During Japan’s medieval era, appointment to an estate became the prize plum for many thirteenth-century warriors who served the military government at Kamakura, and later, the ability to claim half of an estate’s income was crucial to the economic strength of fifteenth-

441 442 Getting the Word Out century provincial warlords (shugo). Given the useful ways in which they illustrate key political and economic issues in each age, it seems hard to imagine that students might learn about premodern Japan without developing an appreciation of the landholding system. The reality, however, is that many American high school and college students do just that: they take classes that deal in whole or part with premodern Japan but do not address estates. As instructors across the United States, Japan, and elsewhere know firsthand, there are significant challenges to teaching Japan’s premodern estate system. The organization of the estate system is difficult to explain, and there was tremendous variation among estates across regions and over time, so that lessons learned from studying a particular estate in one period do not necessarily provide much insight into a different estate in a later period.1 For those teaching outside of Japan, textbook coverage is limited, there are few good materials available in translation, and instructors feel uncertain about how to effectively teach estates. These are just a few of the issues that stand in the way of successful instruction about estates in American classrooms. Given these obstacles, it should come as no surprise that economic historian Kozo Yamamura wrote of estates as “a difficult subject to teach or research” and that “survey histories and college textbooks slight this subject.”2 Distinct from the other chapters in this volume (which present new research findings on and historical analyses of estates), this chapter is focused on issues in education about estates. First, it explains in greater detail the challenges to teaching estates in American classrooms. Next, it reviews the coverage of medieval estates in the textbooks used in high school and college classrooms. The chapter then turns to the roles scholars have played in shaping the ways we think about estates. In the final section, it offers some alternatives and suggests some approaches that instructors might find useful in teaching the premodern Japanese landholding system. Of course, these are merely suggestions; each instructor will want to decide for him or herself on whether or not the ideas proposed are appropriate for his or her classroom.

The Challenges of Teaching Estates Although estates hold a special place in premodern studies as a “fundamental economic, social, and political institution in Japan for nearly a millennium,”3 there are many challenges to helping students understand Teaching Japanese Estates 443 their significance. First, as even specialists know, estates were quite complex. As detailed in the introduction and first chapter of this volume, medieval Japanese estates were managed and run by means of a hierarchical, multilayered administration that included resident supervisors of cultivation, estate managers, absentee proprietors, and elite patrons. Under this system, different parties held distinct rights to income in the same estate. Further complicating things is that warriors sometimes held competing rights on these same estates following the creation of the Kamakura shogunate in the late twelfth century. The many layers of rights to income make it difficult to speak about estate “ownership” in the same way that one might discuss it for a medieval European manor or a plantation in the American South. Second, the multitude of terms used to refer to these individuals can be confusing, especially when working with documents in translation. On the estate itself, one might encounter an azukaridokoro or a shōkan, just two of the titles often rendered into English as “manager” (used to refer to those who supervised the estate on site). Absentee landlords almost certainly included a proprietor (ryōshu or ryōke) and very possibly a patron (honke), each of whom were entitled to a different share of the estate’s income, but the distinctions among these officeholders is unclear without careful explanation. Attempts to use broad descriptive terms in English—such as, for example, “powerful cultivators”—require caution as well, since the Japanese terms that they are intended to substitute for change over time. Powerful, well-to-do cultivators were known as tato in the eleventh-century primary source Shinsarugakuki (New monkey music), but distinctly different powerful cultivators known as myōshu were key figures in managing estate fields during the twelfth century.4 Furthermore, not only officeholders’ titles but also patterns of estate creation changed over time. The earliest estates (shoki shōen) were born from exceptions granted to the effort at nationalizing all land under the Taika Reforms, whereas later estates were dependent on certain government documents (kanshōfu shōen) or on a process of “commendation” to powerful individuals who provided protection (kishin shōen).5 Third, we must acknowledge that estates do not hold students’ interest as well as some other premodern Japanese topics do. Estate documents do not have the romantic story lines or psychological depth of a work like The Tale of Genji, the philosophical and religious conundrums that are wrestled with in An Account of My Hut, or the valiant heroes and stirring battle scenes of The Tale of the Heike. Instead, estate documents 444 Getting the Word Out often contain much legalistic, formulaic language, are filled with personal and place names that are unfamiliar to non-Japanese and usually left untranslated, and, for some students, are as dry as a modern mortgage agreement or judicial court document. Even the images an instructor might choose to use in a lecture on estates are not very excit- ing: a photograph of rice fields on land that, centuries ago, was part of an estate, or a reproduction of an estate document written in Japanese calligraphy (and therefore unintelligible to American students). Faced with challenges like these, instructors need to make an extra effort if they hope to instill in students an appreciation for the significance of estates. And many who teach courses on premodern Japan struggle with the topic as a result. In an admittedly unscientific survey, I solicited information about teaching estates from colleagues at a variety of American colleges and universities who were Japan scholars but not specialists in Heian or medieval history. One respondent wrote, “I remember a book with what I thought was a very provocative title about shōen that thoroughly killed the subject for me. There was nothing in the book of interest to someone who cares deeply about medieval Japan.”6 Another respondent wrote, “I’ll be honest, I thought the literature on shōen was incredibly dry when I was in grad school. I understood why shōen mattered from the perspective of institutional history, but . . . I couldn’t imagine assigning anything to undergrads at that time.”7 Several commented that they felt there was a lack of good materials available to use in teaching estates. Many relied on the class textbooks, and few had any specific materials designed to promote education about estates. How effective are the textbooks available in English at explaining estates? And are there any other materials that might be used to supplement them?

Estates in English-Language Educational Books A review of the materials available to educators in English reveals that there are few world history books that deal with Japanese estates in a meaningful way. I looked first at world history textbooks used in high schools. Of the six books that I examined, only one, World History: Conti- nuity & Change, attempted a substantive explanation of estates.

Samurai were financed primarily through a land distribution system based on the shōen, or “estate.” Unlike the medieval European Teaching Japanese Estates 445

manor, the shōen was not a single piece of land with a manor house surrounded by grazing land. Instead, a shōen was made up of scattered farmland that was governed as a single unit. Those who held a shōen did not necessarily live on it; they simply had a right to its income or its harvest, usually in rice. Often many people would have a share in a single shōen, and most would also have shares in other shōen.8

The passage notes correctly that (a) many estates consisted of scattered holdings, (b) proprietors usually did not live on their estates, and (c) many people had “shares” in the estate’s harvest. Although specialists might find fault with the simplicity of this account, it is much better than what can be found in the other five textbooks. One contends that the nobles gained “great wealth and power” by owning “estates, or large tracts of land” outside of the capital.9 It tells readers that peasants worked on these estates, which provided income that allowed the nobles to defy the emperor. Another book gives control of estates to the samurai, observing, “Peasants, who made up 75 percent of the population, formed the backbone of feudal society in Japan. Peasant families cultivated rice and other crops on the estates of samurai.”10 The remaining three books make no mention of estates at all.11 World history textbooks aimed at the university classroom are slightly better, although it depends on the text. Of the six that I reviewed, two make no mention of land, landholding practices, or estates in their sections on pre-1600 Japan.12 Two others provide only cursory treatment of estates, just two or three lines scattered throughout these very large volumes.13 A fifth title, Agriculture in World History, provides more substantial coverage: it describes the breakdown of the Taika system of state land ownership, the acquisition of estates by landlords and Bud- dhist monasteries, the payment of dues in kind, and the introduction of double cropping, but there are some problems with other parts of the explanation provided.14 The best coverage (of the six volumes that I reviewed) appears in Duiker and Spielvogel’s Essential World History, which provides a brief narrative account of estates and explains the commendation process.

The central government’s attempts to impose taxes directly on the rice lands failed, and rural areas came under the control of powerful families whose wealth was based on the ownership of tax-exempt 446 Getting the Word Out

farmland called shōen. To avoid paying taxes, peasants would often surrender their lands to a local aristocrat, who would then allow the peasants to cultivate the lands in return for the payment of rent. To obtain protection from government officials, these local aristocrats might in turn grant title of their lands to a more powerful aristocrat with influence at court. In return, these individuals would receive inheritable rights to a portion of the income from the estate.15

To be fair, we must acknowledge that world history textbooks have an almost impossible task: to try to tell the history of the world in a limited number of pages and with some type of coherent narrative. Dif- ficult choices with regard to what is and is not included are inherent in the project, and experts in any national or regional history can easily find fault with the results. Given these constraints, it is heartening to find that at least some of these books—and especially a volume on agriculture around the world—include premodern Japanese estates. But we also must recognize that students will not be able to develop a decent understanding of estates from these books alone. One way instructors might enhance their teaching about estates is by using supplementary materials such as videos, educational websites, and targeted curricula. Unfortunately, though, there are very few such materials available to aid in instruction on estates. Columbia University’s Asia for Educators website, for example, contains outstanding online educational resources, but there is no unit that addresses estates. Simi- larly, there are no articles on this topic in the journal Education About Asia. Professional educators at the Stanford Program for Intercultural Education (SPICE) and the University of Colorado Program for Teach- ing East Asia, both of which design and publish supplementary curriculum units, are not aware of any materials that focus on estates.16 My own online search yielded very few leads. One of the most promising was a video created by Oregon Public Broadcasting on “Early Economies.” The website is teacher friendly in that it includes links to readings, activities, and some good discussion questions (though no answers are provided). The video itself devotes about five minutes to Japan and estates. It includes discussion of Emperor Go-Sanjō’s failed effort to stop the proliferation of estates in the eleventh century, the wide variety of goods (not just rice) that were paid as rents, and the ways in which cultivators sometimes protested against unfair treatment.17 It is among the better such supplementary materials available, though caution is Teaching Japanese Estates 447 required as it uses images that are clearly not from the period under discussion. Another promising lead was From Buke to Bushi: Medieval Japan, a set of supplementary lesson plans and readings created by the History Teachers’ Association of Victoria (Australia). The work contains some good content, but there are problems too. Despite the title’s implication that the materials focus on Japan’s medieval age, the forty-four-page booklet covers all of Japanese history, from geography of the islands (i.e., prehistory) to the Meiji period (which it delightfully titles “Post- Feudal”). Most of the medieval section addresses the Tokugawa period (which scholars today generally consider “early modern” rather than “medieval”) and conflates things among the Kamakura, Muromachi, and Tokugawa periods. As for estates, it mentions them as contributing to (a) weakening of central government control and (b) the rise of powerful families such as the Fujiwara and Tokugawa.18 Although there is a need for materials like this one, and the author has done a fine job of designing it in a way that teachers and students are sure to appreciate, some issues with content make me reluctant to recommend this title to those looking for supplementary units on premodern Japan. The last set of educational materials under review consisted of college-level textbooks focused specifically on Japan. As one would expect, these books provide the most detailed coverage, and recent titles seem to be the most effective. For example, A Brief History of Japa- nese Civilization does a good job of highlighting the differences between estates in different periods, explaining the hierarchy of rights and the commendation process in the Heian period, and addressing the roles of warriors and the shogunate in the Kamakura period.19 East Asia: A Cul- tural, Social and Political History also explains the conditions on estates that were unique to each era and includes excerpts from relevant early medieval primary sources in a special boxed section titled “The Estate Steward in Legal Documents.”20 Japan to 1600: A Social and Economic His- tory is the most colorful, offering rich descriptions of agricultural practices and life in the countryside for estate residents.21 It also includes some detailed discussion of the ups and downs of specific estates such as Ōyama and Kuroda. Japan Emerging: Introductory Essays on Premodern History devotes an entire chapter to estates (though I leave it to others to evaluate that chapter’s merits).22 These books hold the most promise for effectively introducing estates to American students. But why should that be the case? How has the field come to be shaped in ways that 448 Getting the Word Out

discourage scholars in other areas of study from developing interests in or teaching materials about Japanese estates?

Shaping the Field of Shōen Studies Early research on estates was dominated by Japanese scholars with particular interests that were, in some regards, unique to Japan in the first half of the twentieth century. Actually, they provide excellent case studies of one of the important lessons that any first-year history student must learn: that the past and the present are in continuous dialogue with each other. Just as our knowledge (or ignorance) of the past influences our understanding of the present, so too our contemporary concerns affect the ways in which we think about the past. Early twentieth-century historians of premodern Japan illustrate this principle very well, since their present-day concerns shaped the ways in which they conceived of the distant past, including estates. Let me explain with two examples. First, during the Meiji period, a major concern of many educated Japanese (including academics) was their country’s efforts to become a “first-rate” nation. The great states of that time were the European colo- nial powers, so Japanese historians contributed to their nation’s effort by seeking out parallels between the histories of the western European countries and Japan. This led them to reconceptualize Japanese history and create a medieval age for Japan when none had existed before. As Thomas Keirstead has argued, “medieval” (chūsei) in its current form was first applied to Japanese history only as recently as the 1890s; the Ministry of Education did not adopt it until 1906. One of the keys to the appeal of medieval studies was that they “supplied parallels between Japan and the West that assured Japan’s parity in the international scene.”23 Of course, no one living during Japan’s medieval age (generally conceived of as the late twelfth to late sixteenth centuries) cared about parity with the West; such comparisons were born from the contemporary concerns of Japanese historians living in the early twentieth century. Estates were interpreted by historians as the Japanese version of medieval European manors and therefore played an important part in the parallels drawn between Europe and Japan. Formative scholars of medieval Japanese landholding such as Nakada Kaoru and Asakawa Kan’ichi drew upon their knowledge of Germanic law and French history to help them understand Japanese estates.24 As Asakawa, in a famous article on land tenure, wrote: Teaching Japanese Estates 449

Western Europe and Japan, with no mutual relation between them, independently evolved in the course of a long time those peculiarly definite and exhaustive social adjustments that are known as feudal, under conditions and upon principles which were, if significant in their minor differences, also striking in their extraordinary similarity in the main features.25

This approach continues to influence the ways in which estates are taught today, close to a century later and an ocean away. Although most Japan scholars today are wary of using the label “feudal,” and few if any would recommend studying medieval European manors in order to better understand Japanese estates, a quick survey of schoolteacher PowerPoint presentations and students’ reports on the Internet reveals that many place Japanese estates in a broader framework that compares them with manors. The second example concerns some of these same academic researchers but reflects a different concern: freedom from the intel- lectual oppression of the wartime Japanese state. As is well known, in the decades leading up to World War II, there was increasing censor- ship and condemnation of academics who tried to publish anything that might be seen as an affront to the imperial institution.26 Kimura Shigemitsu argues convincingly that historians of classical and medieval Japan responded to this pressure by electing to publish on the topic of estates precisely because estates were thought to have contributed to the collapse of the imperial land-management system.27 In other words, estates were appealing as a research topic in part because they were “subversive”: they allowed these scholars to highlight alternatives to control by the imperial system. As with the first example, historians’ contemporary issues influenced their decisions about how to interpret the remote past. Such thinking continues to influence the ways in which estates are taught to this day, even though the wartime state and its intel- lectual controls were eliminated more than seventy years ago. We see it in the assertion that estates were private, in the link between the spread of estates and the weakening of imperial authority, and in the emphasis on certain fundamental qualities of estates: the right to not pay taxes and the right to deny entry to government officials. Both of these examples (highlighting parallels with medieval Europe and seeking institutions that undermined imperial authority) reflect concerns that were unique to Japan. They also demonstrate the impor- 450 Getting the Word Out tance of Japanese scholars in the early development of the field. But what of scholarship published in English? How has it influenced the ways estates are taught outside of Japan? During the first half of the twentieth century, Asakawa Kan’ichi (who, although a Japanese citizen, spent most of his career as a professor at Yale University) was the most influential scholar of premodern Japanese landholding outside of Japan. His work in institutional history, especially landholding and the Iriki Estate, made him the leading authority of his time publishing in English.28 As already noted above, Asakawa called attention to similarities between European manors and Japanese estates. He also established the essential features that defined estates for decades to follow, noting that “The shō was largely an illegal growth . . . seen to possess three characteristics: it contained, as its chief original element, a recently cultivated area of rice-field; was under private possession and private management; and enjoyed or claimed a degree of fiscal immunity.”29 He was not blind to the significant differences between European and Japanese landholding, but he was eager to engage in comparative research and recognized the importance of using terminology and points of reference that would allow for such dialogue.30 Yet Asakawa never became known as a popu- larizer of early Japanese history; instead, it would fall to postwar historians such as John Whitney Hall to stimulate broader interest in medieval estates and premodern studies.31 Hall was more successful as a teacher and mentor (training many scholars) and as a popular author (publishing a textbook and several works for general audiences in addition to his scholarship). His work built on that of Asakawa, but he also differed from his predecessor at Yale in some important ways. One was that Hall did not devote as much attention to Western institutions for comparative purposes. He believed that (1) Japanese history had to stand on its own, (2) even extensive Chinese influence did not alter fundamental native patterns of development, and (3) it invited misunderstanding to substitute English-language terms for seemingly similar Japanese ones. Another difference was that Hall did not describe estates as “illegal” institutions that contravened the imperial land-management system. Instead, he noted that estates were created with the approval of the highest government authorities and that by the twelfth and thirteenth centuries, they differed little in nature from public lands under the administration of provincial governors.32 Of course, not all of Hall’s observations have stood the test of time, even though they continue to appear in teaching materials about estates. For Teaching Japanese Estates 451 example, Hall wrote of warriors accumulating estates over the duration of the Heian period, which gave them the economic base to challenge courtiers. This is not accurate—warriors held different rights to income than courtiers, and the two groups remained distinct in their levels of estate control throughout the Heian period—but one can still find such notions in textbooks published within the past ten years.33 Jeffrey P. Mass, a student of Hall, expanded his mentor’s work through careful examination of primary source documents, many of which he published in translation and are available as potentially rich materials for use by instructors. These documents offer a different perspective on classical and medieval Japan than the better-known literary sources. Yet while many of the documents that Mass worked with concerned estates, they were not his primary interest (which concerned people, especially warriors, and the legal system). He tended to view estates themselves as passive sites of conflict, writing, “Most Kamakura documents deal instead with the mundane matters of shōen administration—local appointments, tax collection, and the improvement of agriculture.”34 Kozo Yamamura, trained as an economist, offered a very different take on estates. Seeking an alternative to the Marxist interpretations coming out of Japan in the postwar period, he instead tried to explain the estate system by analyzing its participants as if they were rational economic thinkers. For example, in his study of Tara Estate, he drew upon land-to-man ratios, labor inputs, and income security to explain the changing size of myōshu holdings.35 It is an interesting approach, but one that has not found many followers beyond those with an interest in economics. Thomas Keirstead made estates and the estate system a major focus of his research in the 1980s and 1990s. His book The Geography of Power in Medieval Japan remains one of the most stimulating and provocative examinations of estates and cultivators available in English.36 It offers new insights not only into the actions of medieval peasants and estate authorities but also into the discursive field of estate studies, which was influenced by Western scholarship on medieval Europe and only conceived of as a “system” in the early twentieth century. Keirstead’s work is heavily influenced by the French philosopher-historian Michel Fou- cault but draws on a variety of primary sources, many of them concerning the aforementioned Tara Estate. His analysis of cultivators’ protests, petitions, and, in some cases, decisions to abscond—all part of what he refers to as the “theatre of protest”—offers a different window into 452 Getting the Word Out estate studies, one that is unique among shōen scholars. Although scholars and instructors are sure to find the work interesting, some might find the theoretical approach and sophisticated reasoning more appropriate for a graduate seminar than an undergraduate lecture hall.37 The fine scholarship of these and other historians has been crucial in raising our understanding of how estates worked and what they contributed to classical and medieval Japanese society. But they have not led to an abundance of articles or other materials that instructors feel comfortable assigning to undergraduates. Are there other resources that educators might turn to for educating students about estates?

New Approaches, New Possibilities The scholars and books referenced above are still relevant to an informed understanding of estates, but there are also alternatives that teachers might consider for their classrooms, ranging from recently published textbooks and monographs to active learning techniques and new approaches to content. Along with the three textbooks already discussed earlier in this chapter (Friday’s Japan Emerging, Ebrey and Walthall’s East Asia, and Schirokauer, Lurie, and Gay’s A Brief History of Japanese Civilization), there are several recent monographs that provide excellent coverage of land, agriculture, and commoners. Rethinking Japanese History is Alan Christy’s translation of Amino Yoshihiko’s work, and it provides an excellent introduction to the writings of that famous and controversial historian of medieval Japan.38 Topics include commercial exchange, land taxes, commoners engaged in nonagricultural occupations, women’s status, and much more, including estates. Wil- liam Wayne Farris’ Japan’s Medieval Population is another recent title that offers valuable insights into commoners’ living conditions, agricultural advances in the medieval period, and land management (among other topics).39 Although it is not focused on estates per se, Farris’ book provides wonderfully detailed information on what life must have been like for cultivators living on estates. And, of course, there are the chapters in this volume, which represent groundbreaking research into estates. Even if an instructor decides that assigning one of these titles in its entirety would be too much for students, that instructor is sure to gain much from reading these books him- or herself. Another possibility is to incorporate primary sources in translation into a lesson. This will require planning and careful explanation, since Teaching Japanese Estates 453 as noted at the outset of this chapter, primary sources in translation can be confusing for students unless they are given adequate context. But the potential rewards are significant, as primary sources allow students to gain the most by engaging directly with the same types of evidence (mediated through translation, of course) that professional historians use in their efforts to understand the past.40 Complete translations of estate documents are available in David Lu’s Japan: A Documentary His- tory, Kan’ichi Asakawa’s The Documents of Iriki (also available online), and in several books by Jeffrey P. Mass, including The Kamakura Bakufu and Lordship and Inheritance.41 Mass and Asakawa provide many more documents than Lu, although the volume of materials and the different ways they are organized will require some additional planning and selection by an instructor. Lu’s volume offers fewer documents but provides context and might be sufficient for those seeking to deal with estates in a limited time frame. Keirstead also includes some excerpts from documents in translation and expert analysis of the parts of those documents, though his book does not include any translated documents in their entirety.42 Hopefully, future scholarship (perhaps by some of the authors contributing to this volume) will provide additional primary sources in translation that can be assigned to students. Finally, there are a number of instructional approaches and active learning techniques that teachers might utilize to help students better engage with estates. For example, one could have students engage in debate over whether or not estates were “legal,” or role play by assigning various parts including cultivators, myōshu, manager, proprietor, patron, provincial governor, and local warriors. Alternatively, an instructor could present estates (perhaps through carefully chosen primary source excerpts) as a puzzle that the class must work together to solve. Why would an estate manager choose to commend his land to a central aristocrat? How did non-samurai landholders deal with the challenges posed by warriors appointed to their estates in the Kamakura period? Rather than simply tell students the answers to these questions, a selection of relevant materials could instead be provided that would allow students to figure out the answers on their own through group or pair work. Another option is to pose problems in class or on review sheets that force students to think about the “big picture” issues of estates, land management, and how both evolve over time. Questions I have given to my own students include: (1) How did inheritance and landholding practices change (or not change) over the course of the medieval 454 Getting the Word Out period? Why did they change (or not change)? (2) Identify and analyze major changes in the way people “owned” and managed land over the period from the eighth to the twelfth centuries. (3) What changes can we see in landholding from the twelfth to the sixteenth centuries? How were estates the “battleground” between warriors and civilian proprietors? Regardless of which (if any) of these techniques an instructor might opt to employ, careful explanation is always essential. Estates can be a difficult topic for students, especially if they fail to realize that not all shōen are the same and that the differences in how and why estates were organized over the centuries is part of what makes them so interesting and worthwhile as objects of study!

Notes

1. As Thomas Keirstead has written, “Shōen have been revealed to encompass just about every kind of holding imaginable, and to incorporate all manner of economic activity, not just agriculture.” Keirstead 1992, 5. 2. Yamamura 1981, 349. 3. Ibid., 349. 4. For primary sources on tato and myōshu (including the description of a tato in Shinsarugakuki, for which also see Piggott, chap. 7), see Lu 1997, 99–101. For more on the rise and decline of myōden (named fields) and myōshu (holders of named fields), see Keirstead 1985. 5. For more on these various titles and the different means by which estates were created, see Segal 2012. 6. Personal communication, May 12, 2012. 7. Personal communication, May 23, 2012. 8. Hanes et al. 1999, 314. 9. Jacobs et al. 2005, 377. 10. Ellis and Esler 2007, 392. 11. Krieger, Niell, and Jantzen 1992; Farah and Karls 1999; Beck et al. 2007. 12. Overfield 2012; Tignor et al. 2014. 13. Hansen and Curtis 2014 notes, “The government no longer redistrib- uted land every six years; instead powerful families amassed large, permanent estates” (p. 224). Morillo 2014 explains, “The power of these clans rested on private estates that were exempt from the land tax” in the Heian period (p. 308), and “Shares of income rights from rural estates went to both courtiers and warriors” in the Kamakura period (p. 381). Teaching Japanese Estates 455

14. Tauger 2011, 62. Problematic passages include, “The shoen were masters of a sho, the Japanese counterpart to the western medieval manor,” which incorrectly suggests that shōen were landowners and shō were their properties (no macrons in the original). Later in the same paragraph, “peasants living on the sho disposed of the lands as though they owned them” inaccurately gives all authority for control of estate lands to the cultivators. 15. Duiker and Spielvogel 2014, 282–283. More recent theories of estate formation, emphasizing a top-down process in which elites instituted estates, are just beginning to make their way into the English-language scholarly literature. 16. The Asia for Educators website of Columbia University is available at http://afe.easia.columbia.edu/ (accessed April 30, 2014). Education about Asia, a publication of the Association for Asian Studies, is available at http://www .asian-studies.org/eaa/ (accessed April 30, 2014). Personal communications with Dr. Rie Kijima of SPICE (May 28, 2014) and University of Colorado Pro- gram for Teaching East Asia director Lynn Parisi (May 27, 2014). 17. “Early Economies” is Unit 8 of the Bridging World History curriculum, available at http://www.learner.org/courses/worldhistory/unit_main_8.html (accessed May 21, 2014). 18. Bramley 2007. References to estates and/or land ownership appear on p. 4 and p. 18. 19. Schirokauer, Lurie, and Gay 2013. 20. Ebrey and Walthall 2013; the boxed section is in chapter 11. 21. Farris 2009. 22. Friday 2012; the chapter on estates can be found on pp. 167–177. 23. Keirstead 2004; the quoted passage is from p. 10. The term chūsei was sometimes used by Japanese historians prior to the late nineteenth century, but Keirstead contends that it was simply a term to mark centuries that were neither recent nor distant but something in the middle; in other words, chūsei for those earlier authors did not convey the parallels with Europe that the label has carried since the late nineteenth century. Keirstead 1998, 54–55. 24. Nakada 1948 [1906]. 25. Asakawa 1914, 1. 26. One of the most famous examples was Tsuda Sōkichi, professor at Waseda University, whose research on the ancient chronicles Kojiki and Nihon shoki led him to question the divine origins of the imperial house. He was accused of desecrating the dignity of the imperial family, forced to resign his position, and sentenced to prison. 27. Kimura 1996, 19–21. 28. Asakawa published many works in English; the most famous is The Docu- ments of Iriki (1955 [1929a]). Many of his translated documents are now avail- 456 Getting the Word Out able online courtesy of the University of Tokyo Historiographical Institute at http://www.hi.u-tokyo.ac.jp/IRIKI/eng_index.html. For a brief introduction to Asakawa and the Iriki-in document collection, visit http://www.hi.u-tokyo .ac.jp/IRIKI/eng_note.html (URLs accessed on June 17, 2014). 29. Asakawa 1914, 5. 30. This is a point he made on several occasions, writing, “These differences, however, important as they clearly are, should not blind us to the similarity that is obvious in many respects between the domainal systems of East and West.” Asakawa 1965b, 59. 31. Hall wrote of Asakawa, “What expectations Asakawa had as a teacher we do not know. His approach to history from a fairly early period had turned to institutional analysis and feudalism in particular, and this was not the most conducive to undergraduate appeal. . . . He wrote no textbook or popular history, and so it was Murdoch and Sansom who ‘fed the student mind on Japanese history.’ ” Hall 1965, 9–10. 32. Mass 1992, 7, 15. 33. On Hall’s misconception, see Mass 1992, 17. An example of this thinking in a recent textbook is Beck et al. 2007, 341–343. 34. Mass 1976, 10. 35. Yamamura 1981. 36. Keirstead 1992. 37. One instructor who responded to my informal survey specifically noted that although she thinks very highly of Keirstead’s interpretations of estate dis- course, her undergraduate students struggled with some of his writing and that therefore she has been hesitant to assign it again. 38. Amino 2012. 39. Farris 2006. 40. For more on the benefits of using primary sources in the history classroom, see Kobrin 1996 and Veccia 2004. 41. Lu 1997, Asakawa 1955 [1929a], Mass 1976, 1989. 42. Keirstead 1992. The Bridging World History “Early Economies” unit (described above) cites Keirstead and appears to draw on some of his translated materials and analysis in the unit video. Some Afterthoughts

Janet R. Goodwin

Studies of the estate system reveal much about Japan’s medieval period in which estates matured, flourished, and faded. Estates are, in fact, one defining characteristic of the medieval age in the theories of many Japanese scholars. As Sakurai Eiji points out, medieval society has often been regarded as fragmented and plural, in distinction from the more unified society of the early classical (ritsuryō) age and from the Edo-period society that followed. On the other hand, approaches such as Kuroda Toshio’s kenmon theory of an elite triumvirate—the court, religious institutions, and the military—do lend an overlay of unity to this fragmented view.1 The chapters in this volume show how estates fit into this disparate picture. As I pointed out in the introduction, not all estates looked alike. Depending in part on location, they varied in degree of proprietarial control or local autonomy; they yielded different products; they related to local and realm-wide government in different ways. This volume has suggested such variety but could not encompass every possibility. Yet estates—or at least most of them about which we know—were governed by similar principles, the chief of which was the layered system of rights and responsibilities called shiki. But that too changed over time. The collapse of shiki and the advent of fully held lands was certainly a major transformation of the estate system, as was the increasing incursion of warriors at all levels of the system. As the structure of medieval Japan changed, so too did the structure of its most important landholding system. At its apex the estate system produced both cooperation and conflict: in theory everyone involved had a stake, contributing to ties between capital and countryside or between elites and ordinary people. In fact,

457 458 Afterthoughts as many of the chapters in this volume have demonstrated, people often fought over the take rather than cooperating in producing and distrib- uting it. While “benevolence” was expected of proprietors, especially if they were religious institutions, those likely to suffer from bad harvests were far more likely to have been local cultivators rather than monks or aristocrats. Thus local people resisted submitting rents, and proprietors sent agents to demand payment, often in violent ways. Local people, too, formed cooperative ties among themselves, constructing irrigation facilities, trading goods, sharing resources, and supporting local shrines and temples, in ways that contributed to the emergence of self- governing villages and organizations in the later medieval age. One important point made by several chapters in this volume is that medieval Japan cannot be seen as separate from the rest of East Asia, as emphasized by Sakurai Eiji in both his chapter in this volume and in his introduction to the first volume of medieval history in the most recent Iwanami Kōza Nihon no rekishi series.2 Contact with the continent, whether it resulted in innovative architectural techniques, exotic products, or new understandings of Buddhism, was reflected in many ways within the experience of the estate system. Another important point is the involvement of religious teachings in the operation of the system. Estates with religious institutions as proprietors are perhaps the most studied of all such landholdings, probably because temples and shrines were particularly adept at maintaining archives; but estates were not just revenue generators for many of those involved in the management of such holdings. Rather, they supplied the means by which religious teachings could be propagated, often in seemingly secular ways. Aristo- cratic proprietors, too, connected religious beliefs with the operation of their estates, since many courtiers’ holdings were dedicated to temples that specialized in prayers and offerings for dead parents and other forebears. Thus estates can provide one illustration of the ways that religion and the secular were integrated in almost all aspects of daily life on all levels of society. Perhaps the most important factor in understanding estates is that they were where people lived and worked. It’s hard to envision ordinary people in medieval Japan. For the most part, they didn’t write documents, compose tales or poetry, or carve and paint exquisite images, and thus did not produce most of the memorable artifacts that remain as traces of the medieval past. We see these people mostly in the aggre- gate, or as they were seen—often negatively—by people above them on Afterthoughts 459 the social ladder. But if we listen carefully, we can still hear their voices: as harvesters of rice and other crops, transporters of goods, makers of pottery, payers of rents or resisters to them, devotees of the gods and buddhas, petitioners for succor from violence or relief from unfair rule. It is our hope that this volume enables you, as readers, to understand what they had to say.

Notes

1. See Sakurai 2013, 3–9. 2. Ibid., 21–24.

Glossary

akutō 悪党: literally, “evil bands.” They were accused of stealing crops, murdering residents, torching houses, and wreaking havoc in general. The term can be seen too as a legal designation for those opposed by the shogunate and other elites. arata 荒田: wasted fields azukaridokoro 預所: custodian; a high-level estate supervisor who represented the proprietor’s interests. The custodian was sometimes an aristocrat who carried out his or her duties in the capital, and sometimes a warrior, monk, or local elite who lived on the estate. bantō 番頭: local elites, headmen; a low-ranking member of the estate management staff be 部: service groups, including cultivators, in the late sixth and seventh centuries bessho 別所: autonomous temple. The term particularly designates several small temples used by Chōgen as bases for his fundraising activities on behalf of Tōdaiji. betsuin 別院: detached cloister betsumyō 別名: a partially tax-exempt hereditary unit overseen by a local official bettō 別当: a temple director or administrator. Bettō could be appointed to a particular temple by the court. bōji 牓示: estate border markers bugyō 奉行: to serve as an administrator; act as a magistrate (bugyōnin) 奉行人 buninjō 補任状: letter of appointment buppō 仏法: the Buddhist law. Often used in opposition to ōbō, or the king’s (secular) law. Both were supposed to work together for a well-governed state.

461 462 Glossary chigyōkoku 知行国: a province whose management and revenues were turned over to a member of the royal family, an elite courtier, or a religious institution that was known as the chigyōkokushu 知行国主 chinjō 陳状: a statement of defense in a lawsuit chō 町: a unit of area measurement, about 2.94 acres, although the size varied in medieval times chokushiden 勅旨田: royal grant fields chōrō 長老: an elder monk daibutsuyō 大仏様: an architectural style represented today by the Jōdodō at Jōdoji in Hyōgo Prefecture daikan 代官: a representative, proxy, agent, or contractor. Daikan had many different functions. daikan ukeoi 代官請負: a system in which a contractor managed estate land in exchange for a portion of the rent daikanjin 大勧進: chief alms collector for a given project Daikanjinshiki 大勧進職: the Chief Alms Collector’s Office at Tōdaiji, which was responsible for soliciting donations for the rebuilding and repair of Tōdaiji structures and images dainō 代納: payment by substitution (usually of goods other than rice) daisennō 代銭納: payment by cash commutation danna 旦那, 檀那: patrons of religious institutions dezukuri 出作り: fields outside estate borders (see kanōden) dokō 土壙: earthen graves dōri 道理: principle or reason; a sense of justice and ethics dōshū 堂衆: lower-ranking service monks fuko 封戸: benefice residence units, whose taxes went to designated high- ranking courtiers or religious institutions funyū 不入: immunity from entry into an estate by provincial officials, one of the rights, along with fuyu (below), that helped to define medieval estates fuyu 不輸: exemption from provincial taxation for an estate gechijō 下知状: a directive, often used by the warrior government to announce its judgment in a lawsuit gekokujō 下克上: “inferiors overthrowing superiors,” a term often used as a complaint by elites in medieval times Glossary 463 genin 下人: low-level cultivators or laborers, often dependents of higher- status cultivators or small landholders genzen 現前: the appearance of Amida Buddha “before one’s eyes” gesu, geshi 下司: an estate manager; often the highest ranking local official on an estate gō 郷: township, originally a unit in the codal provincial administrative system, province (kuni) → district (gun) → township (gō) → subtownship (ho). In later classical and medieval times townships and subtownships functioned much like villages. goganji 御願寺: a temple established by royal vow gokenin 御家人: a retainer (houseman) of the Kamakura shogunate gorintō 五輪塔: five-tiered stupa goshi 五師: “the five masters,” elders of the monastic community at Tōdaiji gōshi 郷司: township heads gōso 強訴: forcible petitioning by shrine affiliates or temple monks, a religious protest. Typically protesters carried a palanquin said to house their deity to a location occupied by their foes, intending that the deity would frighten their foes into compliance. gun 郡: district, a subdivision of a province gundai 郡代: in the Muromachi and Warring States eras, a warlord’s representative in a district gunji 郡司: district chieftain hanzei 半済: “half tax.” Half of an estate proprietor’s rents were turned over to the military governor (shugo), in a system dating from the mid-fourteenth century. hatake 畠: dry fields (as opposed to wet rice paddies) hijiri 聖: a religious practitioner who was not permanently affiliated with a particular temple. Sometimes translated as “holy man.” Hijiri included recluse ascetics, solicitors of donations to temples, and managers of temple estates. hitosato 人里: village ho 保: originally a subunit of the gō; both are often translated as township. hōgen 法眼: “eye of the law.” An honorary Buddhist title. hōkei shūkōbo 方形周溝墓: square, ditched grave honjo 本所, honke 本家: supreme proprietor of an estate, the highest- ranking party in the estate hierarchy, and usually a member of the 464 Glossary

royal family or a powerful courtier. Often translated as patron or guarantor. hyakushō 百姓: yeomen, prosperous cultivators hyakushōuke 百姓請け: cultivators’ contracted rights to collect and pay their own rents ichienryō 一円領: “fully held lands,” an estate fully administered by the proprietor, common in the Nanbokuchō and Muromachi periods ichinomiya 一宮: the leading shrine in a province ikki 一揆: a solidarity or league, sometimes established for a protest or uprising ikkoku heikinyaku 一国平均役: tax assessed equally on all fields in a province in 院: retired sovereign; in late Heian and Kamakura times, he often led the court. inju 院主: head monk at a temple in-no-chō 院庁: office of the retired sovereign in-no-chō kudashibumi 院庁下文: an order from the office of a retired sovereign inzen 院宣: order of a retired sovereign iryōden 井料田: fields whose revenues were used to support irrigation jingūji 神宮寺: a temple affiliated with a shrine and located on its grounds jinin (jinnin) 神人: purveyors or functionaries who performed miscellaneous duties for important shrines; their duties ranged from money lending to rent collection to coercive enforcement. See kugonin, yoriudo. jinu 寺奴: “temple slaves”; the term was sometimes used to designate residents of temple-held estates. jishi 地子: levies or rents jitō 地頭: military steward. The Kamakura shogunate originally placed them on estates confiscated from the defeated Taira, but eventually they were placed on many more estates. They were responsible for keeping order and making sure the proprietor received rents, but their primary loyalty was to the shogunate. jitōdai 地頭代: a military steward’s proxy. The jitōdai acted on behalf of the jitō when the latter did not actually reside on the estate to which he was appointed. Glossary 465 jōin 定印: a hand position (mudra) of Buddhist images signifying meditation. jōroku 丈六: a measurement for Buddhist images; standing jōroku images were about sixteen feet, while seated images were about eight feet. junbo 准母: the “equivalent mother” of a sovereign; an ally of the reigning monarch, usually his sister or aunt. jūnin 住人: residents jūsō 住僧: a temple’s resident monks kairō 回廊: corridor kaison 海村: fishing villages kajishi 加地子: “added rent.” Lesser cultivators sublet parcels from local holders and myōshu, paying them rents in addition to the rents owed proprietors. kandaka 貫高: a system to calculate taxes in cash kanjin hijiri 勧進聖: religious practitioners who collected donations for temple projects kanmotsu 官物: comprehensive tax assessed on fields annually kannō 勧農: literally, encouragement of cultivation, a prerogative of the proprietor kannōken 勧農権: authority to manage cultivation kanōden 加納田: “added fields” that protruded outside an estate’s permitted boundaries. See dezukuri. kanpaku 関白: viceroy, chancellor. The most powerful figure at the mid- Heian court, and generally a member of the northern branch of the Fujiwara family. The kanpaku frequently spoke for the sovereign (tennō). See sesshō, Sekkanke. kanrei 管領: in the Muromachi shogunate, the shogunal prime minister or deputy shogun, and often the most powerful figure in the Muromachi government. kanshōfu 官省符: special court orders authorizing the establishment of an estate Kantō mōshitsugi 関東申次: liaison between the court and the Kamakura shogunate karita rōzeki 苅田狼藉: illegitimately harvesting another party’s crops kata-arashi 片あらし: fields left fallow in alternate years kawanari 河成: flooded land; also land that was repurposed for irrigation channels or irrigation ponds. 466 Glossary kebiishi 検非違使: royal police, capital police. They performed police and judicial functions in Kyoto on the monarch’s order. kebutsu 化仏: a miniature Buddhist image on the crown of a larger image kechien 結縁: religious connections. By donating alms or performing good deeds, one formed connections with the Buddha, a particular temple, a religious leader, or other believers. kenchū 検注: land survey; records of the survey were known as kenchūchō 検注帳 or kenchū mokuroku 検注目録. kendanken 検断権: authority to survey the land, generally the prerogative of the proprietor. kenmon 権門: “gates of power.” In the theory of historian Kuroda Toshio, the court, the military, and powerful religious institutions formed an elite triumvirate that controlled the medieval government, society, and economy. kidan 基壇: packed earth foundation kishin 寄進: commendation of land or other resources to a higher-level individual or institution, to gain favor and protection kishōmon 起請文: oath. When sworn to a deity, it bound its declarer on pain of gruesome punishment. ko 戸: benefice residence units kobyakushō 小百姓: lower-level cultivators kogosho 小御所: a small structure from which the Byōdōin Phoenix Hall can be viewed kokudaka 石高: a system to calculate taxes in rice kokuga gandai 国衙眼代: deputy provincial governor (目代 mokudai) kokugaryō 国衙領: land under the control of provincial officials, as opposed to that of estates. Also known as kōryō. kokujin 国人: powerful local warlords, especially in Muromachi and Sengoku times komokudai 小目代: resident deputy to a provincial governor kondō 金堂: a golden hall, the name often given to the main hall of a temple korō 古老: village elders kōryō 公領: lands taxed by, and under the control of, the provincial government; provincial domain. See kokugaryō. koshiro 子代: a service group, including cultivators, that served the Yamato monarch in the late sixth and seventh centuries Glossary 467 kudashibumi 下文: an edict of permanent duration issued by powerful institutions such as the court, the retired monarch’s office, or the shogunate. Often used for appointments. kugonin 供御人: purveyors or functionaries who performed miscellaneous duties for the court or religious institutions. See jinin, yoriudo. kumon 公文: an estate official whose duties were similar to those of the reeve in medieval England. While technically a rent collector and registrar, the reeve was often the highest resident official on an estate, and as such sometimes had peace-keeping duties. kurōdo 蔵人: secretary and provisioner, especially for the monarch or retired monarch Kurōdodokoro 蔵人所: the Royal Secretariat, an extracodal office created in the early ninth century kuyō 供養: a Buddhist service to honor or memorialize a person, sacred image, or sacred building kyūnin 給人: local officials kyūshu 給主: an estate proprietor’s deputy mandokoro 政所: a chancellery, frequently the leading administrative office in the household of a courtier, retired monarch, or the shogunate. Often involved in estate management. matsuji 末寺: a branch temple, frequently one that paid dues to its main temple, which also had authority over the temple’s monastic staff menden 免田: fields exempt from rents and other levies, granted to estate officials, military stewards, and skilled artisans, among others migyōsho 御教書: communiqué or directive of instruction by a courtier of the third rank or higher, or by a shogun or vice-shogun in medieval times mikuriya 御厨: tribute lands that provided for the throne and important shrines misshū 密衆, 密宗: monks who undertook esoteric practices; also a category of Tōdaiji monks miyake 屯倉: a residence with storehouses and fields that provisioned Yamato great kings and other high elites in the sixth and seventh centuries mokkan 木簡: inscribed strips of wood used for communiqués or shipping tags mokudai 目代: representative of the provincial governor, governor’s deputy 468 Glossary monchū 門注: judicial hearing; the Monchūjo was an investigative and judicial body of the Kamakura shogunate monzeki 門跡: a cloister headed by a member of the royal family mōshijō 申し状: petition or report submitted to higher authority, especially the court mukaekō 迎講: ceremony replicating Amida’s welcoming of the dead, also called raigō-e mura 村, murazato 村里: village myō 名: a unit of rent collection or taxation; a “named field unit.” Also called myōden (名田). myōshu 名主: holder of a myō. An upper-level cultivator or overseer in charge of paying rents [to a proprietor] or taxes to provincial authorities or a proprietor. naikenchō 内検帳: record of surveys of agricultural land nashiro 名代: service group, including cultivators, in the late sixth and seventh centuries. See koshiro. natsuhata 夏畠: dry fields harvested in the early summer nayosechō 名寄帳: list of those responsible for paying rents nenbutsu 念仏: reciting the name of Amida Buddha nengu 年貢: annual rents or taxes nenyo goshi 年預五師: the monk among the “five masters” (goshi) in a temple monastic community who was in charge for a particular year, at Tōdaiji and elsewhere niebito 贄人: those who provided nie—royal provisions including fish, poultry, and other goods such as hunted game—for the royal table nuki 貫: in architecture, a horizontal rail that connects pillars by running through them, a piercing tie rail nyoin 女院: a titled royal woman, and the female equivalent of a retired sovereign (in) nyūdō 入道: a lay monk, someone who takes abbreviated Buddhist vows but then remains at home ōban toneri 大番舎人: royal attendants from the provinces of Settsu, Izumi, and Ōmi who served the Northern Fujiwara Regents’ House in rotation ōbō 王法: royal law, rule by the tennō and court. See buppō. ōjō 往生: rebirth in paradise, in particular Amida Buddha’s Gokuraku Paradise Glossary 469

ōtabumi 大田文: record of a survey of agricultural fields, usually by shogunal officials in Kamakura times raigō 来迎: Amida’s welcome of the dying to his paradise raigō-e 来迎会: a ceremony that performs the above, also called the mukaekō rinji 臨時: special, extraordinary, irregular; for land holdings, it frequently indicated levies assessed for a specific purpose rather than a regular levy. risshō 立荘: establishment of an estate by authorities at court ritsu 律: a general term for penal laws, also used for the Buddhist precepts (Vinaya). The Ritsu school (Risshū) was one of the early Buddhist schools imported into Japan. There were ritsu temples and ritsu monks. ritsuryō 律令: penal (ritsu) and administrative (ryō) legal codes, patterned after those of China and first adopted by the Japanese government at the turn of the eighth century. The codes were fleshed out and amended by supplementary laws (kyaku 格) and protocols (shiki 式) from later Nara times onward. Many historians refer to the eighth and ninth centuries as “the ritsuryō age,” because the codes remained a focus of court concern and scholarship at the time. rōmyō 籠名: “imprisoning a name.” An act of cursing whereby monks would write the name of an opponent on a piece of paper and lock it up inside their temple. Punishment by a deity was expected to ensue. ryōke 領家: an estate proprietor, which might be a major temple, shrine, or noble house. The ryōke was often actively involved in managing the estate. ryōshu 領主: proprietor; landholder. A more general term than ryōke, above. saifu 割符: bills of exchange, similar to modern promissory notes or drafts sakute 作手: right to cultivate a unit of land, sakuteshiki 作手職 samurai-shū 侍衆: warrior attendants sangō 三綱: “the three deans,” monastic administrators, at Tōdaiji and other temples satanin 沙汰人: local estate officials and agents sekisen 関銭: fees assessed at checkpoints or toll barriers sekisho 関所: checkpoint, toll barrier Sekkanke 摂関家: the Regents’ Line. These powerful courtiers from the Northern Fujiwara family monopolized the positions of regent 470 Glossary

(sesshō) and kanpaku (viceroy, chancellor) in the Heian and early medieval periods. senji 宣旨: an order transmitting a royal command sesshō 摂政: a regent for an underage sovereign. See kanpaku, Sekkanke. shami 沙弥: a novice monk shigyō 執行: a temple administrator. Also pronounced shugyō or shikkō. shihai monjo 紙背文書: “reverse-side documents”: documents preserved because their reverse sides were reused for other written material shiki 職: shared rights to income from land, with associated responsibilities shitaji chūbun 下地中文: an agreement to divide the land of an estate between the proprietor and the military land steward shitomi-do 蔀戸: in Buddhist temple architecture, overhung latticed doors that can be raised shōen 荘園, 庄園: a landed estate, characterized by shared rights (shiki) to income from the land. Supreme authority over a shōen was exercised by a powerful individual, house, or religious institution, rather than by the court government. shōen seiri rei 庄園整理令: court regulations allowing provincial authorities to curb the expansion of estates shōen-kōryō (kokugaryō) sei 荘園—公領 (国衙領)制: the medieval landholding system, comprising both estates and the provincial domain Shōji 荘司: Shōen Office, a unit within Tōdaiji to manage and defend the temple’s estates shōkan 荘官: estate managers shōkanjin 小勧進: assistant alms collector for a given project shoki shōen 初期荘園: early estates, in the eighth and early ninth centuries shoryō 所領: a private holding shōshu 荘主: on-site estate managers shugo 守護: provincial constables, initially those close to Minamoto no Yoritomo, who were appointed to head up the housemen of a province on behalf of the Kamakura shogunate. Later they became increasingly powerful military governors. shugodai 守護代: a deputy constable or military governor Shūnōsho 収納所: provincial tax payment office shuto 衆徒: the monastic community of a temple Glossary 471 sō 惣: corporate village; self-governing village community sō 僧: monk sojō 訴状: document of complaint sōron 争論: dispute, often about estate borders or water rights sōtsuibushi 総追捕仕: warrant officer tadokoro 田所: an office within an estate or a provincial headquarters charged with various tasks of land management tan 段, 反: a unit of area measurement, 1/10 of a chō, about 1/3 acre tansen 段銭: a special levy, often assessed for repairs of the royal palace or infrastructure repairs ordered by the shogun tato 田堵: field contractors who specialized in organizing the work of land opening; agricultural specialists tōchigyō 当知行: rights deriving from long-term management of land for at least twenty years, a medieval Japanese version of “squatters’ rights” toimaru 問丸: warehouse manager and transport specialist toiya 問屋: transport specialists and trade brokers tokuden 得田: fields on which a proprietor could assess rents tokusei ikki 徳政一揆: a protest demanding debt relief tokuseirei 徳政令: edicts canceling debts tone 刀袮: community leader tonseisha 遁世者, tonseisō 遁世僧: world-renouncing ascetics, as opposed to regular monks affiliated with a particular temple; they were sometimes but not always ordained. tōryō 頭領: headman tōsaku 当作: currently cultivated fields tsuikahō 追加法: supplementary laws, statutes added to Kamakura shogunal law after the promulgation of the Jōei Formulary (Goseibai shikimoku) in 1232 ujidera 氏寺: the temple of a family or lineage ukesho 請書: a contract, often to deliver rents to estate proprietors in exchange for a percentage of the receipts yashiki 屋敷: residence yoriudo, yoryūdo 寄人: dependent or low-ranking functionary who performed miscellaneous duties, including supplying goods, for royal or courtier households, or large religious institutions. 472 Glossary

Yoriudo sometimes lived on estates whose proprietors were not their own patrons. See jinin, kugonin. zaichi ryōshu 在地領主: local landholders; resident landlords zaichōkanjin 在庁官人: hereditary provincial officials who hailed from and resided in a province zaike 在家: residences, homesteads zasshō 雑掌: an agent, usually of a proprietor or other elite, who performed miscellaneous duties zasu 座主: the head monk or abbot at a temple zōyaku 雑役: a tax on individuals zushi 図師: record keeper Bibliography

Primary Sources and Source Collections Editors’ note: Primary sources and source collections often appear in several editions; authors use the edition available to them. Because readers will want to do the same, all editions used for this volume are listed below.

Aioi-shi shi. 1984–1995. 8 volumes. Ed. Aioi-shi Kyōiku Iinkai. Aioi-shi. Azuma kagami. 1976–1979. Ed. Nagahara Keiji and Kishi Shōzō. 6 vols. Shin Jinbutsu Ōraisha. Azuma kagami, Kikkawa bon. 1976. Ed. Sōsho Kankōkai. 3 vols. Meicho Kankōkai. Azuma kagami. 1988–1989. In Shintei zōho Kokushi taikei, ed. Kuroita Katsumi, vols. 32–33. Yoshikawa Kōbunkan. Chōshūki. 1975. The journal of Minamoto no Morotoki. Zōho shiryō taisei, vols. 16–17. Chūsei hōsei shiryōshū vol. 1, Kamakura bakufu hō, 1955; vol. 2, Muromachi bakufu hō, 1957. Comp. Satō Shin’ichi and Ikeuchi Yoshisuke. Iwa nami Shoten. Confessions of Lady Nijō. 1973. The memoirs of Go-Fukakusa-in no Nijō. Trans. Karen Brazell. Stanford University Press. Dai Nihon komonjo (DNK). 1901–. Comp. Tōkyō (Teikoku) Daigaku Shiryō Hensanjo. Tōkyō (Teikoku) Daigaku Shuppankai. Dai Nihon shiryō (DNS). 1901–. Comp. Tōkyō (Teikoku) Daigaku Shiryō Hensanjo. Tōkyō (Teikoku) Daigaku Shuppankai. Engishiki. 1931. Comp. Koten Dokkyūsho, Zenkoku Shinshokukai. 3 vols. Rinsen Shoten. Reprint 1992, comp. Torao Toshiya. Engishiki. 2000–. 2 vols. to date. Shūeisha. Engishiki. 1970. Translated by Felicia Bock. Procedures of the Engi Era: Books 1–10. 2 vols. Sophia University. (For a translation of Books XVI and XX, see Bock 1985).

473 474 Bibliography

Essays in Idleness: The Tsurezuregusa of Kenkō. 1967. By Yoshida Kenkō. Trans- lated by Donald Keene. Columbia University Press. Godai teiō monogatari. 1960. In Gunsho ruijū, vol. 3 (Teiōbu). Gunsho Ruijū Kanseikai. Gyokuyō. 1908. The journal of Kujō Kanezane. 2 vols. Ed. Takatō Chūzō. Kokusho Kankōkai. Gyokuyō. 1994–2002. Kujōke-bon Gyokuyō. In Toshoryō sōkan, ed. Kunaichō Shoryōbu. 8 vols. Meiji Shoin. Heian ibun, katei ban (HI). 1963–1976. Comp. Takeuchi Rizō. 13 vols; Kin- sekibun hen (1965). Tōkyōdō Shuppan. A digital version is available. Heikoki. 1983. The journal of Taira Tsunetaka. In Zōho shiryō taisei, vol. 32. Rinsen Shoten. Hiroshima-ken shi: Kodai-chūsei shiryō. 1974. 5 vols. Hiroshima Prefecture. Hōkanshū. 1952. In Tokyo Daigaku Shiryō Hensanjo, comp. DNK Iewake, vol. 1, Kōyasan monjo, vols. 1–2. Tōkyō Daigaku Shuppankai. Hyakurenshō. 1901. In Kokushi taikei, vol. 14. Keizai Zasshisha. Hyakurenshō. 1922. From entries in Tōkyō Daigaku Shiryō Hensanjo, comp. DNS, vol. 5, pt. 2. Hyakurenshō. 2000. Ed. Kuroita Katsumi. Vol. 11, Shintei zōho kokushi taikei. Yoshikawa Kōbunkan. Hyōgo Kitaseki irifune nōchō. 1981. Ed. Hayashiya Tatsusaburō. Chūo Kōron Bijutsu Shuppan. Hyōgo-ken shi: Shiryō hen chūsei. Ed. Imai Rintarō. 1983–1997. 9 vols. Hyōgo- ken shi Henshū Senmon Iinkai. Hyōhanki. 1965. The journal of Taira no Nobunori. Zōho shiryō taisei, vols. 18–20. Ed. Zōho Shiryō Taisei Kankōkai. Rinsen Shoten. Iga no kuni Kuroda no shō shiryō (KS). 1975–1979. Comp. Takeuchi Rizō. 2 vols. Yoshikawa Kōbunkan. Jōdoji monjo. 1905. From entry in Tōkyō Daigaku Shiryō Hensanjo, comp. DNS, vol. 4, pt. 4. Tōkyō Teikoku Daigaku. Jōgan seiyō. 1978. Ed. Harada Taneshige. 2 vols. Meiji Shoin. Kamakura ibun, komonjo hen (KI). 1971–1995. Comp. Takeuchi Rizō. 46 vols. Tōkyōdō Shuppan. Kennaiki. The journal of Madenokōji Tokifusa. 1963–1986. 10 vols. In Dai Nihon kokiroku, ed. Tōkyō Daigaku Shiryō Hensanjo. Iwanami Shoten. Kii zoku fudoki: Kōyasan no bu. 1962. Kōya University Zoku Shingon-shū Zensho Kankōkai. Kikki. 1965. The journal of Fujiwara no Tsunefusa. In Zōho shiryō taisei, vols. 29–30, ed. Zōho Shiryō Taisei Kankōkai. Rinsen Shoten. Kongōbuji monjo. 1936. In Kōyasan monjo, vol. 2, ed. Kōyasan Shi Hensanjo. Kōyasan Monjo Hakkōkai. Bibliography 475

Konjaku monogatarishū. 1971. 4 vols. Comp. Mabuchi Kazuo et al. In Nihon koten bungaku zenshū, Shōgakukan. English translations: Marian Ury, 1979, Tales of Times Now Past, University of California Press; Yoshiko Dykstra, 1998–2003, The Konjaku Tales, 3 vols., Intercultural Research Institute, Kansai Gaidai University; Royall Tyler, 1987, Japanese Tales, Pantheon; Robert Brower, 1952, “The Konjaku Monogatarishū,” 3 vols., PhD dissertation, Stanford University. Kōya shunjū hennen shūroku. 1912. In Dai Nihon Bukkyō zensho, vol. 131. Bussho Kankōkai. Kōyasan monjo (KM). 1904–1907. In DNK, Iewake, vol. 1. 8 vols. Tōkyō Daigaku Shuppankai. Republished 1968. Kugyō bunin. 1986. 5 vols. In Shintei zōhō Kokushi taikei. Yoshikawa Kōbunkan. Kujō Masamoto. 1961. Masumoto-kō tabi hikitsuke. In Toshoryō sōkan, ed. Kunaichō Shoryōbu, Yōtokusha; and 2001, in Shinshū Izumisano-shi shi, vol. 5. Seibundō. Kujōke monjo. 1971–1977. In Toshoryō sōkan Kujōke monjo, ed. Kunaichō Shoryōbu. 7 vols. Meiji Shoin. Kujōke rekisei kiroku. 1989–1999. In Toshoryō sōkan Kujōke rekisei kiroku, ed. and pub. Kunaichō Shoryōbu. 4 vols. Kyōōgokokuji monjo. 1960–1970. Ed. Akamatsu Toshihide. 9 vols. Heirakuji Shoten. Mansai Jugō nikki. 1918–1920. The journal of Mansai Jugō. 3 vols. Kyoto Teikoku Daigaku Bunka Daigaku. Mansai Jugō nikki. 1928–1929. In Zoku gunsho ruijū (Hoi), ed. and pub. Zoku Gunsho Ruijū Kanseikai. vols. 1, 2. Reprint 1974–1975. Masukagami. 1965. In Nihon koten bungaku taikei, vol. 87, comp. Iwasa Masashi et al. Iwanami Shoten. English translation by George W. Per- kins, 1998: The Clear Mirror: A Chronicle of the Japanese Court During the Kamakura Period (1185–1333). Stanford University Press. Masamoto-kō tabi hikitsuke. 1961. In Toshoryō sōkan, ed. Kunaichō Shoryōbu, Yōtokusha; and 2001, in Shinshū Izumisano-shi shi, vol. 5, comp. Izumi- sano Shishi Hensan Iinkai. Seibundō. Masamoto-kō tabi hikitsuke: Honbun hen, kenkyū shōroku hen, sakuin hen. 1996. Volumes containing the text, abstracts of research findings, and index. Izumi Shoin. Also see Shinshū Izumisano-shi shi, vol. 5. Meigetsuki. 1911–1912. The journal of Fujiwara no Teika. 3 vols. Kokusho Kankōkai. Michiyukiburi. 1994. The journal of Imagawa Ryōshun. In Chūsei nikki kikōshū. Shinpen Nihon koten bungaku zenshū, vol. 48, ed. Nagasaki Ken, 389– 426. Rendered in modern Japanese by Inada Toshinori. Shōgakukan. Mineaiki. 1976. In Zoku gunsho ruijū, vol. 28.1 (Shakkebu), ed. Zoku Gunsho Ruijū Kanseikai, 217–253. 476 Bibliography

Minkeiki. 1978. The journal of Hirohashi Tsunemitsu. In Dai Nihon kokiroku, ed. Tōkyō Daigaku Shiryō Hensanjo. Iwanami Shoten. Reprint 2001. Namuamidabutsu sazenshū. 1955. By Namuamidabutsu Chōgen. Comp. Nara Kokuritsu Bunkaizai Kenkyūjo. Shinyōsha. Also in Ono-shi shi (OS) 1997, vol. 4, 537–545 (# 227) and Bijutsu Kenkyūjo, ed., Bijutsu kenkyū 30 ( June 1934), 39–51. Negoro yōsho—Kakuban kiso shiryō shūsei. 1994. Sōhonzan Daigoji. Nigatsudō-e engi. 1926. In Zoku gunsho ruijū 27.2:128–134, comp. Zoku Gun- sho Ruijū Kanseikai. Nihon engyō taikei, Shiryō hen kodai, chūsei. 1977. Ed. Nihon Engyō Taikei Henshū Iinkai. Bunkasha. Nihon shōen ezu, Shakumon hen. 2007–. Comp. Tōkyō Daigaku Shiryō Hensanjo. 2 vols. to date. Tōkyō Daigaku Shuppankai. Nihon shōen ezu shūei. 1988–2003. Comp. Tōkyō Daigaku Shiryō Hensanjo. 7 vols. Tōkyō Daigaku Shuppankai. Nyoinki. 1977. In Gunsho ruijū, vol. 29, ed. Hanawa Hokiichi. Gunsho Ruijū Kankōkai. Ōmi banbashuku Rengeji kakochō. 1934. In Gunsho ruijū 29 (Zatsubu). Zoku Gunsho Ruijū Kanseikai. Ono-shi shi (OS). 1997. Comp. Ono-shi Hensan Senmon Iinkai. Vol. 4 (Shiryō hen I) and Bekkan: Bunkazai hen. Ono City. Ritsuryō. 1976. Comp. Inoue Mitsusada et al. In Nihon shisō taikei. Iwa- nami Shoten. For translations of selections in English see Piggott, forthcoming, Shōen Documents and Materials in English Translation, http://www.uscppjs.org/resources/#/estate-translation-project/. Rōshōdō Nihon kōroku. 1968. By Sō Kikei. Zoku Gunsho Ruijū Kanseikai. Ruijū sandai kyaku. 1980. In Shintei zōhō Kokushi taikei. 2 vols. Yoshikawa Kōbunkan. French translation: Francine Herail, 2008–2011, Recueil de décrets de trois ères, méthodiquement classés, 2 vols. Librairie Droz. Sankaiki. 1965. The journal of Nakayama Tadachika. In Zōho shiryō taisei, vols. 26–28, ed. Zōho Shiryō Taisei Kankōkai. Rinsen Shoten. Sata mirenshō. In Chūsei hōsei shiryōshū, vol. 2, Muromachi bakufu hō, 1957, comp. Satō Shin’ichi and Ikeuchi Yoshisuke, 355–376. Iwanami Sho- ten. English translation: see Steenstrup 1980. Sazenshū. See Namuamidabutsu sazenshū. Shinkō Kōya shunjū hennen shūroku. 1982. Ed. Hinonishi Shinjō. Meicho Shuppan. Shinsarugakuki. 1979. Annotated by Ōsone Shōsuke. In Kodai seiji shakai shisō, vol. 8 of Nihon shisō taikei, ed. Yamagishi Tokuhei et al., 133–152. Iwanami Shoten Kankōkai. For a selection of English translations by Joan Piggott, see Shirane 2007, 491–497. Bibliography 477

Shinshū Izumisano-shi shi. 2001–2009. Comp. Izumisano-shi Shi Hensan Iinkai. Shiryō hen, 5 vols. Seibundō. Shoku nihongi. 1989–2000. 6 vols. In Shin Nihon koten bungaku taikei, vols. 12–16 & Bekkan, comp. Aoki Kazuo et al. Iwanami Shoten. Shōsōin hennen monjo (SHM). 1901–1940. 25 vols. In Tōkyō Daigaku Shiryō Hensanjo, comp., DNK. Tōkyō Daigaku Shuppankai. Sonpi bunmyaku. 2001. By Tōin Kinsada. 5 vols. In Shintei zōho Kokushi taikei, vols. 58–62, ed. Kuroita Katsumi and Kokushi Taikei Henshūkai. Yoshikawa Kōbunkan. Originally published in 1964. Taiheiki. 1994. Ed. Hasegawa Tadashi. 4 vols. Shinpen Nihon koten bungaku zenshū, vols. 54–57. Shōgakukan. Reprint 2007. Tale of the Heike. 1988. Translated by Helen Craig McCullough. Stanford University Press. Tamakiwaru. 1994. The journal of Kengozen. In Towazugatari, Tamakiwaru. Shin Nihon koten bungaku taikei, vol. 50, ed. Misumi Yōichi and Satake Akihiro. Iwanami Shoten. Tōdaiji bettō shidai. 1932. In Gunsho ruijū, vol. 4 (Buninbu), 569–597. Zoku Gunsho Ruijū Kanseikai. Tōdaiji bettō shidai. 1971. In Tōdaiji yōroku, 160–204. Tōdaiji monjo. 1944–. 23 vols. to date. In Tōkyō Daigaku Shiryō Hensanjo, comp., DNK, Iewake, vol. 18. Tōkyō Daigaku Shuppankai. Tōdaiji yōroku. 1944. Comp. Tsutsui Eishun. Zenkokushobō. Republished 1971, Kokusho Kankōkai. Tōdaiji zoku yōroku. 1907. In Zokuzoku gunsho ruijū, vol. 1, comp. Kokusho Kankōkai. Tōdaiji zoku yōroku. 2013. Comp. Tsutsui Kanshū. Kokusho Kankōkai. Tōdaiji zōryū kuyōki. 1929. In Dai Nihon Bukkyō zensho, vol. 121 (Tōdaiji sōsho, vol. 1), comp. and pub. Bussho Kankōkai , 47–57. Tōji hyakugō monjo. 2004–. 12 vols. now published. Comp. Kyoto Furitsu Sōgō Shiryōkan. Shibunkaku. Tōji monjo 1925–. 17 vols. to date. In Iewake, vol. 10, Tōkyō Daigaku Shiryō Hensanjo, comp. DNK. Tōkyō Daigaku Shuppankai. Wakan rōeishū. 1994. Comp. Fujiwara Kintō. In Shinpen Nihon koten bungaku zenshū, vol. 19, ed. Sugano Hiroyuki. Shōgakukan. Reprint 2008. Eng- lish translation: J. Thomas Rimer and Jonathan Chaves, ed., Japanese and Chinese Poems to Sing: The Wakan Rōei shū. Columbia University Press, 1997. Wamyō ruijushō. 1992. In Nagoya-shi Hakubutsukan shiryō sōsho, vol. 2. Nagoya- shi Hakubutsukan. Yōkōki. 2004. The journal of Hamurō Sadatsugu. 2004. Yagi Shoten. Zen’yaku Azuma kagami. See Azuma kagami 1976–1979. 478 Bibliography

Zoku gunsho ruijū. 1872–1929. Comp. Hanawa Hokiichi. 71 vols. Zoku Gun- sho Ruijū Kanseikai.

Reference Works Digital Dictionary of Buddhism. www.buddhism-dict.net. Kokushi daijiten. 1979–1997. 15 vols. Comp. Kokushi Daijiten Henshū Iinkai. Yoshikawa Kōbunkan. Nihon kokugo daijiten. 1972–1976. 10 vols. Shōgakukan. Nihon rekishi chimei taikei. 1979–2005. 50 vols. Ed. Heibonsha Chihō Shiryō Sentā. Heibonsha. Nihon shōen shiryō. 1998. 2 vols. Ed. Kokuritsu Rekishi Minzoku Hakubut- sukan. Yoshikawa Kōbunkan. Online database: http://www.rekihaku .ac.jp. Accessed September 25, 2017. Nihon shōenshi daijiten. 2003. Ed. Seno Seiichirō. Yoshikawa Kōbunkan.

Secondary Sources Abe Takeshi. 1972. Nihon shōenshi. Shinseisha. ———. 1997. “Shiryō shōkai: Ōimikado-keryō ni tsuite.” Teikyō shigaku 12: 341–355. ———. 2011. Kenkyū nyūmon Nihon no shōen. Tōkyōdō Shuppan. Abe Takeshi and Satō Kazuhiko. 1990. Jinbutsu de tadoru Nihon shōen. Tōkyōdō Shuppan. Adolphson, Mikael S. 2000. The Gates of Power: Monks, Courtiers, and Warriors in Premodern Japan. University of Hawai‘i Press. ———. 2007. The Teeth and Claws of the Buddha: Monastic Warriors and Sōhei in Japanese History. University of Hawai‘i Press. Aida Hanji. 1964. Chūkai Yōrōryō. Yūshindō. Akamatsu Toshihide. 1972. “Somaudo to shōen, Iga no kuni Tamadaki to Kuroda no shō.” In Kodai chūsei shakai keizaishi kenkyū, 153–348. Hei- rakuji Shoten. Akutō Kenkyūkai, ed. 1998. Akutō no chūsei. Iwata Shoin. ———. 2005. Akutō to nairan. Iwata Shoin. ———. 2013. Chūsei shōen no kiso. Iwata Shoin. Amino Yoshihiko. 1963. “Taikai ni furete—‘akutō ’ no hyōka no hensen wo chūshin ni.” Rekishigaku kenkyū 283: 68–71. ———. 1966. Chūsei shōen no yōsō. Hanawa Shobō. ———. 1970. “Akutō no hyōka wo megutte—Nihon chūsei kenkyūshi no ichi danmen.” Rekishigaku kenkyū 362: 29–40. ———. 1978a. Chūsei Tōji to Tōjiryō shōen. Tōkyō Daigaku Shuppankai. Bibliography 479

———. 1978b. Muen/kugai/raku: Nihon chūsei no jiyū to heiwa. Heibonsha. ———. 1980a. Nihon chūsei no minshūzō. Iwanami Shoten. Republished 2009, Amino Yoshihiko chosakushū vol. 8: Chūsei no minshūzō. Iwanami Shoten. ———. 1980b. “Heian jidai makki—Kamakura jidai ni okeru shio no seisan.” Nihon Senbai Kōsha, ed. and pub., Nihon engyō taikei: Genshi, kodai, chūsei. Republished 2008, Amino Yoshihiko chosakushū, vol. 9: Chūsei no seisan to ryūtsū, 3–143. Iwanami Shoten. ———. 1982. Higashi to nishi no kataru Nihon no rekishi. Soshiete. ———. 1985. “Chūsei no seien to shio no ryūtsū.” In Kōza Nihon gijutsu no shakaishi, vol. 2 (Engyō, gyogyō), 43–91. Nihon Hyōronsha. Repub- lished 2008, Amino Yoshihiko chosakushū, vol. 9: Chūsei no seisan to ryūtsū. Iwanami Shoten. ———. 1989. “Shōenshi no shikaku.” In Kōza Nihon shōenshi, vol. 1, ed. Amino Yoshihiko, Ishii Susumu, Inagaki Yasuhiko, and Nagahara Keiji, 1–88. Iwanami Shoten. ———. 1991a. “Shōen wo shūshu taikei to ryōshu keizai.” In Kōza Nihon shōenshi, vol. 2, ed. Amino Yoshihiko, Ishii Susumu, Inagaki Yasuhiko, and Nagahara Keiji, 196–206. Iwanami Shoten. ———. 1991b. Nihon chūsei tochi seido shi no kenkyū. Hanawa Shobō. ———. 1995a. “Shōen ni ikiru hitobito.” In Chūsei no mura, ed. Ishii Susumu, 19–31. Tōkyō Daigaku Shuppankai. ———. 1995b. Akutō to kaizoku: Nihon chūsei no shakai to seiji. Hōsei Daigaku Shuppankyoku. ———. 1998. Nihon chūsei no hyakushō to shokunōmin. Heibonsha. Repub- lished 2009, Amino Yoshihiko chosakushū, vol. 8: Chūsei no minshūzō. Iwa- nami Shoten. ———. 1999. Chūgoku chihō no shōen. Yoshikawa Kōbunkan. ———. 2005. Shikoku, Kyūshū chihō no shōen. Yoshikawa Kōbunkan. ———. 2008a. Chūsei no nariwai to ryūtsū. Amino Yoshihiko chosakushū, vol. 9. Iwanami Shoten. ———. 2008b. “Saionji-ke to sono shoryō.” In Amino Yoshihiko chosakushū, vol. 3, 293–320. Iwanami Shoten. ———. 2012. Rethinking Japanese History. Translated by Alan S. Christy. Cen- ter for Japanese Studies, University of Michigan. Amino Yoshihiko and Inaba Nobumichi. 2007. Amino Yoshihiko chosakushū, vol. 12: Kaimin no shakai. Iwanami Shoten. Amino Yoshihiko, Ishii Susumu, Inagaki Yasuhiko, and Nagahara Keiji, eds. 1989a. Shōen nyūmon, Kōza Nihon shōenshi, vol. 1. Yoshikawa Kōbunkan. ———. 1989b. Shōen no kaitai, Kōza Nihon shōenshi, vol. 4, ed. Amino Yoshi- hiko, Ishii Susumu, Inagaki Yasuhiko, and Nagahara Keiji. Yoshikawa Kōbunkan. 480 Bibliography

———. 2005. Shikoku •Kyūshū chihō no shōen, Kōza Nihon shōenshi, vol. 10, ed. Amino Yoshihiko, Ishii Susumu, Inagaki Yasuhiko, and Nagahara Keiji. Yoshikawa Kōbunkan. Andō Seiichi. 1978. Wakayama no kenkyū, vol. 2: Kodai, chūsei, kinsei. Seibundō. ———, ed. 1990. Kishūshi kenkyū, vol. 5: Kōyasan sōtokushū. Kokusho Kankōkai. Aoki Shigeru, ed. 1971. Onomichi shishi (shinshū), vol. 1. Onomichi Shiyakusho. Arai Takashige. 1990. Chūsei akutō no kenkyū. Yoshikawa Kōbunkan. ———. 2001. Tōdaijiryō Kuroda no shō no kenkyū. Azekura Shobō. ———. 2005. Kuroda akutōtachi no chūseishi. NHK Books. Arnesen, Peter. 1979. The Medieval Japanese Daimyo. Yale University Press. ———. 1982. “Suō in the Age of Kamakura.” In Court and Bakufu in Japan: Essays in Kamakura History, ed. Jeffrey P. Mass, 92–120. Yale University Press. ———. 1984. “The Struggle for Lordship in Late Heian Japan: The Case of Aki.” Journal of Japanese Studies 10, no. 1: 101–141. Asahi Shinbun. “Mohara haiji konryū no yurai.” http://www.ne.jp/asahi /yuhi/net/haiji/yurai.htm. Accessed June 12, 2012. Asakawa, Kan’ichi. 1914. “The Origin of the Feudal Land Tenure in Japan.” American Historical Review 20, no. 1: 1–23. ———. 1929a. The Documents of Iriki. Yale University Press. Also published in 1929 by Greenwood Press. Republished 1955, Japan Society for the Promotion of Science. ———. 1929b. “Agriculture in Japanese History.” The Economic History Review 2, no. 1: 81–92. ———. 1965a. “The Early Shō and the Early Manor.” In Land and Society in Medieval Japan, 262–268. Japan Society for the Promotion of Science. ———. 1965b. “Ushi-ga-hara Shō, Yechizen.” In Land and Society in Medieval Japan, 37–68. Japan Society for the Promotion of Science. Asano Keisuke. 2014. “Nihon kodai ni okeru mura no seikaku.” Shigaku zasshi 123, no. 6: 37–68. Asao Naohiro. 1961. “Jūnanaseiki ni okeru sangyō kōzō no tokushitsu.” Nihonshi kenkyū 56, 59–70. Republished 2004, Asao Naohiro chosakushū, vol. 2, Kinai kara mita bakuhansei shakai. Iwanami Shoten. Baba Hajime. 2010. “Nifuda ga kataru kodai no zeisei.” In Mokkan kara kodai ga mieru, ed. Mokkan Gakkai, 30–37. Iwanami Shinsho. Banse Akemi. 1993. “Inseiki~Kamakuraki ni okeru nyoinryō ni tsuite: Chūsei zenki no ōke no arikata to sono henka.” Nihonshi kenkyū 374: 46–75. Bassino, Jean-Pascal, Kyoji Fukao, and Masanori Takashima. 2010. “Rice Price, Grain Wages of Carpenters, and Skill Premium in Kyoto ca. Bibliography 481

1260–1600: A Comparison with London, Florence, Istanbul, and Cairo.” Economic History Society Annual Conference, University of Durham, Collingwood College. http://www.ehs.org.uk/dotAsset /85e81444‑8b65‑4f60-b950-cc6dd0aaceb2.pdf. Batten, Bruce. 1986. “Foreign Threat and Domestic Reform: The Emer- gence of the Ritsuryō State.” Monumenta Nipponica 41, no. 2: 199–219. ———. 1993. “Provincial Administration in Early Japan: From Ritsuryō Kokka to Ōchō Kokka.” Harvard Journal of Asiatic Studies 53, no. 1: 103–134. ———. 2003. To the Ends of Japan: Premodern Frontiers, Boundaries, and Interac- tions. University of Hawai‘i Press. ———. 2007. “Cross-border Traffic on the Kyushu Coast, 794–1086.” In Heian Japan, Centers and Peripheries, ed. Mikael Adolphson, Edward Kamens, and Stacie Matsumoto, 357–383. University of Hawai‘i Press. Beck, Roger B., Linda Black, Larry S. Krieger, Phillip C. Naylor, and Dahia Ibo Shabaka. 2007. World History: Patterns of Interaction. McDougal Littell. Bialock, David. 2007. Eccentric Spaces, Hidden Histories: Narrative, Ritual, and Royal Authority from The Chronicles of Japan to The Tale of the Heike. Stanford University Press. Bock, Felicia. 1985. Classical Learning and Taoist Practices in Early Japan, with a Translation of Books XVI and XX of the Engishiki. Arizona State Univer- sity Press. Bourdieu, Pierre. 1986. “The Forms of Capital.” In Handbook of Theory and Research for the Sociology of Education, ed. J. Richardson, 241–258. Greenwood Press. Bramley, Rohan. 2007. From Buke to Bushi: Medieval Japan. History Teachers’ Association of Victoria. Bray, Francesca. 1986. The Rice Economies: Technology and Development in Asian Societies. University of California Press. Brown, Delmer M. 1949. “The Japanese Tokuseirei of 1297.” Harvard Jour- nal of Asiatic Studies 12: 188–207. ———, ed. 1993. The Cambridge History of Japan, vol. 1 (Ancient Japan). Cam- bridge University Press. Brown, Delmer M., and Ichirō Ishida. 1979. The Future and the Past: A Trans- lation and Study of the Gukanshō, an Interpretive History of Japan Written in 1219. University of California Press. Coaldrake, William H. 1996. Architecture and Authority in Japan. Routledge. Collcutt, Martin. 1981. Five Mountains: The Rinzai Zen Monastic Institution in Medieval Japan. Harvard University Press. Conlan, Thomas Donald. 2003. State of War: The Violent Order of Fourteenth- Century Japan. Center for Japanese Studies, University of Michigan. 482 Bibliography

Damian, Michelle. 2015. “Late Medieval Japan’s Seto Inland Seascape: Shipping, Sailors, and Seafaring.” PhD dissertation, University of Southern California. Dobbins, James C. 1996. “Kuroda Toshio and His Scholarship.” Japanese Journal of Religious Studies 23, no. 3–4: 217–233. Duiker, William J., and Jackson J. Spielvogel. 2014. The Essential World His- tory, vol. 1 (To 1800). 7th ed. Wadsworth Cengage Learning. Ebisawa Tadashi. 2011. “Shōen keikan no hozon to katsuyō.” Nihon rekishi 752: 63–67. ———. 2013. “Chūsei ni okeru suiden kaihatsu to tetsu seisan—Bitchū no kuni Niimi no shō no baai.” In Kōkogaku to chūseishi kenkyū, vol. 10 (Mizu no chūsei—Chisui, kankyō, shihai), ed. Ono Masatoshi, 39–70. Koshi Shoin. ———. 2014. Chūsei no shōen kūkan to gendai. Bensei Shuppan. Ebisawa Tadashi, Sakai Kimi, and Shimizu Katsuyuki, eds. 2014. Ajia yūgaku: Chūsei no shōen kūkan to gendai. Bensei Shuppan. Ebisawa Tadashi and Toshiko Takahashi, eds. 2014. Chūsei shōen no kankyō, kōzō to chiiki shakai. Bensei Shuppan. Ebrey, Patricia, and Anne Walthall. 2013. East Asia: A Cultural, Social, and Political History. Wadsworth Cengage Learning. Ellis, Elizabeth Gaynor, and Anthony Esler. 2007. Prentice Hall World History: Michigan. Pearson Education, Inc. Endō Motō. 1933. “Shokunin no soshiki toshite no ‘za’ no ichi kōsatsu.” Shakai keizai shigaku 3, no. 2: 1–29. Endō Motoo. 2011. “Chūsei Tōdaiji monjo wo fukan suru.” Mita chūseishi kenkyū 18: 1–39. ———. 2018. “Hensan to kenkyū no nōto.” https://www.hi.u-tokyo.ac.jp /personal/endo/index.html. Accessed September 27, 2017. Endō Yuriko. 2007. “Jūniseiki matsu no sensō wo tōshite miru kishin no ichi kōsatsu.” In Saikō chūsei shōensei, ed. Endō Yuriko, Kuramochi Shige- hiro, and Tamura Noriyoshi, 185–223. Iwata Shoin. Farah, Mounir A., and Andrea Berens Karls. 1999. World History: The Human Experience. Glencoe/McGraw-Hill. Farris, William Wayne. 1995. Heavenly Warriors: The Evolution of Japan’s Mili- tary, 500–1300. Harvard University Press. ———. 2006. Japan’s Medieval Population: Famine, Fertility, and Warfare in a Transformative Age. University of Hawai‘i Press. ———. 2009. Japan to 1600: A Social and Economic History. University of Hawai‘i Press. Fiévé, Nicolas, and Paul Waley, ed. 2002. Japanese Capitals in Historical Perspective: Place, Power and Memory in Kyoto, Edo and Tokyo. Routledge. Bibliography 483

Foard, James. 1977. “Ippen Shōnin and Popular Buddhism in Kamakura Japan.” PhD dissertation, Stanford University. Friday, Karl. 1992. Hired Swords. Stanford University Press. ———, ed. 2012. Japan Emerging: Essays in Premodern History. Westview Press. Frölich, Judith. 2003. “Land Administration in Medieval Japan, Itō no shō in Chikuzen Province, 1131–1336.” History 88, no. 229: 3–16. ———. 2007. Rulers, Peasants and the Use of the Written Word in Medieval Japan. Peter Lang. Fujimoto Yorihito. 2002. “Chūsei shoki ni okeru Uno no mikuriya no kōzō to henshitsu.” Aoyama shigaku 20: 7–28. Fujisawa Takako. 2013. “Kamakura jidai ni okeru Taima mandarazu no juyō—Juyōsō kaen ryūzō Amida raigōzu nado.” Bukkyō geijutsu 328: 25–78. Fujita Hirotsugu. 2009. Shōen ezu ga kataru kodai, chūsei. Yamakawa Shuppansha. Fukui-ken, ed. 1994. Fukui-ken shi: Tsūshihen, vol. 2. Fukutō Sanae, ed. 2002. Rekishi no naka no kōjotachi. Shōgakukan. Furiya Tetsuo. 2002. “Hiraizumi shutsudo no bōeki tōji: Yanagi no gosho ato shutsudo no Kanhantō san tōjiki kara mieru ryūtsū keiro.” Hirai- zumi bunka kenkyū nenpō 2: 15–26. Gay, Suzanne. 2001. The Moneylenders of Late Medieval Kyoto. University of Hawai‘i Press. ———. 2009. “The Lamp-oil Merchants of Iwashimizu Shrine: Transre- gional Commerce in Medieval Japan.” Monumenta Nipponica 64, no. 1: 1–51. Gekkan bunkazai hakkutsu shutsudo jōhō, reports on Natsumi haiji: 1991.8, 59; 1991.12, 221; 1992.4, 135; on Yokoe Estate: 1991.1, 62. Gernet, Jacques. 1972. A History of Chinese Civilization. 2nd ed. Translated by J. R. Foster and Charles Hartman. Cambridge University Press. Goble, Andrew. 1996. Kenmu, Go-Daigo’s Revolution. Harvard Council on East Asian Studies. ———. 2011. Confluences of Medicine in Medieval Japan: Buddhist Healing, Chinese Knowledge, Islamic Formulas, and Wounds of War. University of Hawai‘i Press. Goffman, Erving. 1959. The Presentation of Self in Everyday Life. Anchor Books. Gomi Fumihiko. 1984. Inseiki shakai no kenkyū. Yamakawa Shuppansha. ———. 1990. “Sei・bai・en: Onna no chikara.” In Nihon josei seikatsushi, chūsei 2, comp. Joseishi Sōgō Kenkyūkai, 109–146. Tōkyō Daigaku Shuppankai. ———. 1991a. “Buke seiken to shōensei.” In Shōen no seiritsu to ryōyū, Kōza Nihon shōenshi vol. 2, ed. Amino Yoshihiko, Ishii Susumu, Inagaki Yasu hiko, and Nagahara Keiji, 118–163. Iwanami Shoten. 484 Bibliography

———. 1991b. Fujiwara no Teika no jidai. Iwanami Shoten. ———. 1991c. “Hachijō-in wo meguru shokenmon. ” In Nihon chūsei seiji shakai no kenkyū, ed. Zoku Gunsho Ruijū Kanseikai, 29–66. ———. 1992. Bushi to bunshi no chūseishi. Tōkyō Daigaku Shuppankai. Goodwin, Janet R. 1990. “The Buddhist Monarch: Go-Shirakawa and the Rebuilding of Tōdai-ji.” Japanese Journal of Religious Studies 17, nos. 2–3: 219–242. ———. 1994. Alms and Vagabonds: Buddhist Temples and Popular Patronage in Medieval Japan. University of Hawai‘i Press. Goodwin, Janet R., and Kevin Wilson. 2015. “Memories and Strategic Silence in Jōdoji engi.” Japanese Journal of Religious Studies 42, no. 1: 109–131. Gotō Michiko. 2009. Sengoku wo ikita kuge no tsumatachi. Yoshikawa Kōbunkan. Groner, Paul. 2002. Ryōgen and Mount Hiei: Japanese Tendai in the Tenth Cen- tury. University of Hawai‘i Press. Grossberg, Kenneth. 1981a. Japan’s Renaissance. Harvard East Asia Center. ———. 1981b. The Laws of the Muromachi Bakufu. Monumenta Nipponica, Sophia University. Hall, John Carey, trans. 1906. “Institutes of Judicature (Goseibai shikimoku).” In Japanese Feudal Law, 19–44. University Publications of America. Hall, John Whitney. 1961. “Foundations of the Modern Japanese Daimyo.” Journal of Asian Studies 20, no. 3: 317–329. ———. 1965. “Kan’ichi Asakawa: Comparative Historian.” In Land and Soci- ety in Medieval Japan, ed. Asakawa Kan’ichi, 3–21. Japan Society for the Promotion of Science. ———. 1983. “Terms and Concepts in Japanese Medieval History.” Journal of Japanese Studies 9, no. 1: 1–32. Hall, John Whitney, and Jeffrey P. Mass, ed. 1974. Medieval Japan: Essays in Institutional History. Yale University Press. Republished 1988, Stanford University Press. Hamada Takashi. 1979. “Ryūzō Amida raigōzu seiritsushi kō.” Bukkyō geijutsu 126: 21–42. Hanes, William Travis, Toyin Falola, and Theodore K. Rabb. 1999. World History: Continuity & Change. Holt, Rinehart and Winston. Hansen, Valerie, and Kenneth R. Curtis. 2014. Voyages in World History, vol. 1 (To 1600). 2nd ed. Wadsworth Cengage Learning. Harrington, Lorraine. 1982. “Social Control and the Significance of Akutō.” In Court and Bakufu in Japan: Essays in Kamakura History, ed. Jeffrey P. Mass, 221–250. Yale University Press. Haruta Naoki. 2010. “Chūsei kaison no seigyōreki.” Kokuritsu rekishi minzoku hakubutsukan kenkyū hōkoku 157: 51–82. Bibliography 485

Hashimoto Michinori. 1995. “Harima-no-kuni Ōbe no shō.” In Chūsei no mura, ed. Ishii Susumu, 97–108. Tōkyō Daigaku Shuppankai. Hashimoto Yoshihiko. 1986. “Nyoin no igi to enkaku.” In Heian kizoku, 144– 171. Heibonsha. Hatakeyama, Akira. 2003. “Chūsei Tōdaiji ni yoru Hyōgo-seki no keiei to sono soshiki.” Nihon rekishi 494: 47–65. Hatano Junko. 2005. “Heishi to Aki no kuni ōkeryō shōen seiritsu katei, chigyōkokusei no suii to tomo ni.” Shigaku kenkyū 250: 32–50. Hattori Hideo. 1989. “Chōsa kenkyū no hōhō.” In Shōen nyūmon, Kōza Nihon shōenshi, vol. 1, ed. Amino Yoshihiko, Ishii Susumu, Inagaki Yasuhiko, and Nagahara Keiji, 187–253. Yoshikawa Kōbunkan. ———. 1995. Keikan ni saguru chūsei. Shinjinbutsu Ōraisha. ———. 2004. Bushi to shōen shihai. Yamakawa Shuppan. Hayashi Rokurō and Suzuki Yasutami. 1985. Fukugen Tenpyō shokoku shōzeichō. Gendai Shichōsha. Hidaka Katsuhiro. 2015. “Shōensei wo ika ni oshieru ka?” Rekishi to chiri 250: 15–22. Higuchi Kentarō. 2008. “Nyoinsei no tenkai to shitsuji.” In Kamakura jidai no kenryoku to seido, ed. Uwayokote Masataka, 31–56. Shibunkaku Shuppan. Hikoyoshi Mieko. 1973. “Jishō Juei sōranki ni okeru Hachijō-in kurōdo Ashikaga Yoshikane no yakuwari.” Seiji keizai shigaku 84: 17–20. Hiraoka Jōkai. 1980. Tōdaiji jiten. Tōkyōdō Shuppan. Republished 1995. ———. 1988. Nihon jiinshi kenkyū—chūsei, kinsei hen. Yoshikawa Kōbunkan. Hirata Tenshū. 2002. “Suzuyaki no rekishi.” Tōsestu 597: 11–24. Hirose Kazuo. 1986. “Chūsei e no taidō.” In Iwanami kōza Nihon kōkogaku 6: Henka to kakki, ed. Kondō Yoshirō, Amakasu Ken, Sahara Makoto, Tozawa Mitsunori, Yokoyama Kōichi, Katō Shinpei, and Tanaka Migaku, 295–356. Iwanami Shoten. Hirota Kōji. 2001. “Hineno・Iriyamada mura bantō shikiji ichiran.” In Shinshū Izumisano-shi shi, vol. 5, ed. Izumisano-shi Shi Hensan Iinkai, 788. Seibundō. ———. 2006a. “Chū・ka i dankyūmen no suiri kaihatsu shi.” In Shinshū Izumisano-shi shi, vol. 12 (Kangai suiri hen), ed. Izumisano-shi Shi Hensan Iinkai, 38–53. Seibundō. ———. 2006b. “Shōen iseki no rinen to kōsei,” Hisutoria 202: 41–60. ———. 2008a. “Masamoto-kō tabi hikitsuke ni miru Hine no shō no seikatsu tan’i, shihai kubun, bantō, densū.” In “Masamoto-kō tabi hikitsuke to Hine no shō,” Shinshū Izumisano-shi shi, vol. 1, ed. Izumisano-shi Shi Hensan Iinkai, 432–441. Seibundō. ———. 2008b. “Sengokuki no mura no jianshū to shinkō shūdan.” In “Chū, kinsei ikōki no zaichi shakai,” Shinshū Izumisano-shi shi, ed. Izumisano- shi Shi Hensan Iinkai, vol. 1, 571–609. Seibundō. 486 Bibliography

———. 2011. “Masamoto-kō tabi hikitsuke (Kujō Masamoto): kuge no zaishō jikimu to sengoku shakai.” In Nikki de yomu Nihon chūseishi, comp. Motoki Yasuo and Matsuzono Hitoshi, 243–257. Minerubā Shobō. Hisano Shūgi. 1978. “Chūsei jiin seiritsu ni kansuru ichi kōsatsu, jū-jūniseiki Tōdaiji wo megutte.” Shirin 61, no. 4: 38–75. Hōgetsu Keigo. 1963. “Chūsei sangyō to gijutsu.” In Iwanami kōza Nihon rekishi 8 (chūsei 4), ed. Ienaga Saburō, 79–108. Iwanami Shoten. ———. 1983 [1943]. Chūsei kangaishi no kenkyū. Yoshikawa Kōbunkan. Hongō Kazuto. 1986. “Chūsei jiin no shakaiteki kinō ni tsuite no ichi kōsatsu: Kōyasan wo rei toshite.” Shigaku zasshi 95, no. 4: 502–527. ———. 2000. “Chūsei chōtei soshō ni tsuite no hōhōron: arata na jisshōteki seika ni moto tsuite.” Harukanaru chūsei 18: 60–68. Hongō Keiko. 2004. Chūseijin no keizai kankaku: okaimono kara sagaru. Nihon Hōsō Shuppan Kyōkai. Horiike Shunpō. 1955. Chōgen Shōnin no kenkyū. Nanto Bukkyō Kenkyūkai. ———. 1980. “Tōdaiji yōroku hensan ni tsuite.” In Nanto bukkyōshi no kenkyū, 315–345. Hōsōkan. Hosokawa Ryōichi. 1987. Chūsei no Risshū jiin to minshū. Yoshikawa Kōbunkan. Hotate Michihisa. “Toshiken to kizoku hanchū,” hotatelog.cocolog-nifty .com. Viewed November 15, 2015. ———. 1984. “Chūsei minshū keizai no tenkai.” In Kōza Nihon rekishi vol. 3, chūsei vol. 1, ed. Rekishigaku Kenkyūkai and Nihonshi Kenkyūkai, 167–206. Tōkyō Daigaku Shuppankai. Huey, Robert N. 2002. The Making of the Shinkokinshū. Harvard University Press. Hurst, G. Cameron. 1976. Insei: Abdicated Sovereigns in the Politics of Late Heian Japan, 1086–1185. East Asian Institute, Columbia University. ———. 1982. “The Kōbu Polity: Court-Bakufu Relations in Kamakura Japan.” In Court and Bakufu in Japan: Essays in Kamakura History, ed. Jeffrey P. Mass, 3–28. Stanford University Press. ———. 1999. “Insei.” In Cambridge History of Japan, vol. 2 (Heian Japan), ed. Donald Shively and William McCullough, 576–643. Cambridge Uni- versity Press. Hyōgo-ken Shi Henshū Senmon Iinkai, ed. 1997. Hyōgo-ken shi shiryō hen: Chūsei 9. Hyōgo Prefecture. Ichikawa Hideyuki. 1999. “Hakkutsu seika kara mita kaku jidai no Sayama ike.” In Sayama ike, comp. and pub. Maizō Bunkazai, 495–504. Ichimura Takao. 2011. “Chūsei Setouchi no minatomachi to senshu/toi no nettowa-ku.” In Saigoku ni okeru seisan to ryūtsū, ed. Kawaoka Tsutomu and Koga Nobuyuki, 171–212. Seibundō. Ida Toshikuni. 2006. “Tsuchimaru ・ Ōgi no suiri kaihatsu shi.” In Shinshū Izumisano-shi shi, vol. 12 (Kangai suiri hen), ed. Izumisano-shi Shi Hen- san Iinkai, 29–38. Seibundō. Bibliography 487

Ihara Kesao. 1993. “Honjo saibanken no ichi kōsatsu, Denka sata ni tsuite.” Nihon rekishi 547: 19–38. Iinuma Kenji. 2008. “Zeni wa dōzairyō to naru no ka—kodai, chūsei no dōseisan, ryūtsū, shinkō.” In Beppu Daigaku Bunkazai Kenkyūjo kikaku shiriizu 1: Hito to mono to kankyō ga kataru, kyōzutsu ga kataru chūsei no sekai, ed. Oda Fujio, Hirao Yoshimitsu, and Iinuma Kenji, 3–20. Shi- bunkaku Shuppan. Ikebe Wataru. 1981 [1916]. Wamyō ruijūshō gungoriekimei kōshō. Yoshikawa Kōbunkan. Imatani Akira. 1981. “Setonaikai seikaiken no suii to irifune nōchō.” In Hyōgo Kitaseki irifune nōchō, ed. Hayashiya Tatsusaburō, 272–288. Chūō Kōron Bijutsu Shuppan. ———. 1986. “Biwakō to Yodo no suikei.” Shūkan Asahi hyakka Nihon no rekishi 20, no. 5: 270–278. ———. 1990. Muromachi no ōken: Ashikaga Yoshimitsu no ōken sandatsu kei- kaku. Chūō Kōronsha. Imatani Akira and Kozo Yamamura. 1992. “Not for Lack of Will or Wile: Yoshimitsu’s Failure to Supplant the Imperial Lineage.” Journal of Jap- anese Studies 18, no. 1: 45–78. Inaba Nobumichi. 1975. “Chūsei Tōdaiji jiin kōzō kenkyū josetsu.” Nenpō chūseishi kenkyū 1: 1–38. ———. 1997. Chūsei jiin no kenryoku kōzō. Iwanami Shoten. Inaba Tsuguharu. 2012. “Chūsei no shakai taisei to kokka.” Nihonshi kenkyū 600: 54–81. Inui Naoko. 1980. “Muromachi kōki kuge keizai no ichi kōsatsu: Sanjōnishi-ke no shōryō shihai wo chūshin ni.” Nenpō chūseishi kenkyū 5, 87–102. Irumada Nobuo. 1965. “Kuroda no shō dezukuri chitai ni okeru sakute no seiritsu to shokaisō.” Bunka 29, no. 3: 397–424. Ishida Hisatoyo. 1988. Nihon bijutsushi ronshū. Chūo Kōron Bijutsu Shuppan. Ishida Yūichi. 1981. “Shodaibu to sekkanke.” Nihon rekishi 392: 22–39. Ishii Susumu. 1970. Nihon chūsei kokkashi no kenkyū. Iwanami Shoten. ———. 1985. “Formation of Bushi Bands.” Acta Asiatica 49: 1–14. ———. 1986. “Kuroda no shō to Tōdaiji: Chūseiteki sekai no keisei.” Shūkan Asahi hyakka Nihon no rekishi 2 (chūsei 1, vol. 530): 9–43. Asahi Shinbunsha. ———. 1988. “Genpei sōranki no Hachijō-in shūhen: Hachijō-in monjo wo tegakari ni.” In Chūsei no hito to seiji, ed. Ishii Susumu, 1–130. Yoshikawa Kōbunkan. ———. 1991. “ ‘Chūseiteki sekai’ to Ishimoda shigaku no keisei.” In Chūseishi wo kangaeru: shakairon, shiryōron, toshiron, 131–146. Azekura Shobō. 488 Bibliography

———. 1995a. “Kunisaki no shōen no miryoku.” In Nihon no mura, ed. Ishii Susumu, 3–17. Risōsha. ———. 1995b. Chūsei no mura: keikan wa katarikakeru. Tōkyō Daigaku Shuppankai. ———. 2000. Chūsei no mura wo aruku. Asahi Sensho. ———. 2005. Ishii Susumu chosakushū, vol. 7. Iwanami Shoten. Ishikawa Kenritsu Maizō Bunka Sentā, ed. and pub. 1994. Ōyahiyama yōseki・Jikekurobatake yōseki: kokuei Suzu daini chiku kaitaku kensetsu jigyō ni kakaru bunkazai shikutsu chōsa hōkokusho 2. Ishikawa Kenritsu Rekishi Hakubutsukan, ed. and pub. 2000. Wakayama no shō wo aruku: Noto saidai no chūsei shōen. Ishikawa-ken Matto-shi Kyōiku Iinkai, ed. and pub. 1979. Tōdaijiryō Yokoe shōke ato. Ishimoda Shō. 1946. Chūseiteki sekai no keisei. Tōkyō Daigaku Shuppankai. ———. 2006. “Japan’s Medieval World.” Translated by Joan R. Piggott. In Capital and Countryside in Japan 300–1180, ed. Joan R. Piggott, 326– 361. Cornell East Asia Series. Ishino Yae. 1987. “Nanbokuchō Muromachichō no Iyo Saionji: kuge daimyō seiritsu no zentei.” Kokugakuin zasshi 88, no. 10: 40–56. Itō Keisuke. 2010. “Tōdaiji fuko yunō to jishi.” Kodai bunka 61 no. 4: 126–135. Itō Masayoshi. 1989. “Tōdaijiryō Echigo no kuni Ishii no shō sōron to Toyota no shō sōden jikken shōkō.” In Chūsei Nihon no shosō, vol. 1, ed. Yasuda Motohisa Sensei Tainin Kinen Ronshū Kankō Iinkai, 205–293. Yoshikawa Kōbunkan. Itō Nobuo. 1983. “Tenjikuyō to Chōgen.” In Chōgen ・ Eison ・ Ninshō, ed. Nakao Takashi and Imai Masaharu, 137–149. Yoshikawa Kōbunkan. Itō Toshikazu. 1993. “Chūsei kōki ni okeru ‘shōke’ to chiiki kenryoku.” Nihonshi kenkyū 368: 29–55. Iwata Shigeki. 2004. “Mokuzō busshu–Iga Shin-Daibutsuji honzon no te de aru kanōsei wo megutte.” Museum 588: 3–15. Izumisano Kyōiku Iinkai, comp. and pub. 2001. Hine no shō chūsei sekizōbutsu chōsa hōkokusho. Izumisano-shi. 2008–2009. Shinshū Izumisano-shi shi, Tsūshi hen. 3 vols. Seibundō. Jacobs, Heidi Hayes, Michal L. LeVasseur, Kate Kinsella, and Kevin Feld- man. 2005. History of Our World. Pearson/Prentice Hall. Kagawa Prefecture. 1989. Kagawa-ken shi, vol. 2 (chūsei ). Shikoku Shinbunsha. Kainuma, Yoshiko. 1994. “Kaikei and Early Kamakura Buddhism: A Study of the Anamiyō Amida Form.” PhD dissertation, University of California, Los Angeles. ———. 2014. “Chōgen’s Jōdoji Amida Triad and Its Environment: A Theat- rical Effect of the Raigō Form.” Artibus Asiae 74, no. 1: 97–127. Bibliography 489

Kaizu Ichirō. 2011. Kii-no-kuni Kaseda no shō. Dōseisha. Kamakura Saho. 2008. “Shōensei no seiritsu to bumon no tōgō.” In Kama kura Saho and Nagai Eiji, “Chūsei no shakairon to funsō kaiketsu.” Rekishigaku kenkyū 846: 47–55. ———. 2009. Nihon chūsei shōensei seiritsushi ron. Hanawa Shobō. ———. 2013. “Shōensei to chūsei nengu no seiritsu.” In Iwanami kōza Nihon rekishi, vol. 5 (chūsei 1),” ed. Ōtsu Tōru, Sakurai Eiji, Fuji Jōji, Yoshida Yutaka, and Yi Song-si, 129–162. Kamakura Saho and Ono Takashi. 2013. “Sekkanki no shōen.” In Shōenshi kenkyū handobukku, ed. Shōenshi Kenkyūkai, 33–63. Tōkyōdō Shuppan. Kamei-Dyche, Rieko. 2013. “Tools of Authority: The Saionji Family and Courtier Society in Early Medieval Japan.” PhD dissertation, Univer- sity of Southern California. Kamihara Eiko. 1963. “Kamakura jidai Sekkanke no keizaiteki kiso.” Rekishi kyōiku 11, no. 6: 45–51, 59. Kamikawa Michio. 2010. “Heian makki no sanrin jiin to chiiki shakai– Mikawa no kuni no himerareta moji shiryō tankyū.”Aichi Kenritsu Daigaku moji bunkazai kenkyūjō nenpō 3: 1–15. Kamiki Tetsuo. 1993. “Jūgoseiki nakaba Setonaikai ni okeru kome yusō: Bun’an ninen ‘Hyōgo Kitaseki irifune nōchō’ no kisō bunseki,” part 2. Kokumin keizai zasshi 168, no. 3: 45–62. ———. 1996a. “Jūgoseiki nakaba Setonaikai ni okeru kaisen to shōhin unsō: Bun’an ninen ‘Hyōgo Kitaseki irifune nōchō’ no kisō bunseki,” part 3. Kokumin keizai zasshi 173, no. 4: 1–14. ———. 1996b. “Chūsei no Setonaikai to Hyōgo-tsu: ‘Hyōgo Kitaseki irifune nōchō’ ga kataru mono.” In Kaidō rekishi no taaminaru: Hyōgo no tsu no monogatari, ed. Kamiki Tetsuo and Sakiyama Masahiro, 103–150. Kōbe Shinbun Sōgō Shuppatsu Sentaa. ———. 1996c. “Taira no Kiyomori no bei eria kaihatsu: Ōwada-no-tomari kara Hyōgo-tsu e.” In Kaidō rekishi no taaminaru: Hyōgo no tsu no monogatari, 27–56. Kōbe Shinbun Sōgō Shuppatsu Sentaa. Kanai Shizuka. 1999. Chūsei kugeryō no kenkyū. Shibunkaku Shuppan. Kaneko Hiroyuki. 1989. Kodai no miyako to mura. Kōdansha. Karikome Hitoshi. 1998. “Shunjōbō Chōgen ni yoru daikanjin jigyō no minshūteki kiban.” Nihon shigaku shūroku 21: 1–26. ———. 2004. Shōen shakai ni okeru shūkyō kōzō. Azekura Shobō. Katō Tomoyasu. 1995. “Shoki shōen.” In Iwanami kōza Nihon tsūshi, vol. 5 (kodai 4), ed. Asao Naohiro, 315–332. Iwanami Shoten. Katsumata Shizuo. 1996. “Sengoku jidai no sonraku—Izumi no kuni Iri- yamada mura, Hineno mura.” In Sengoku jidai ron, 93–128. Iwanami Shoten. Originally published in Shakaishi kenkyū 6, 1985. 490 Bibliography

Katsuyama Seiji. 1995. Chūsei nengusei seiritsushi no kenkyū. Hanawa Shobō. Kaufman, Laura. 1980. “Ippen Hijiri-e: Artistic and Literary Sources in a Buddhist Handscroll Painting of Thirteenth-Century Japan.” PhD dissertation, New York University. Kawabata Shin. 1994. “Sekkankeryō shōengun no keisei to denryō: Konoeke ryō no seiritsu.” In Kodai chūsei no seiji to bunka, ed. Inoue Mitsuo, 121–148. Shibunkaku Shuppan. ———. 1996. “Harima-no-kuni Ōbe no shō no kaihatsu to suiri.” In Shōen wo yomu, aruku, ed. Kyōto Daigaku Hakubutsukan, 90–94. Shibunkaku Shuppan. ———. 1999. “Ōbe no shō no kaihatsu to suiri.” In Genzai ni ikiru chūsei no keikan: Harima no kuni Tōdaijiryō Ōbe no shō, comp. and pub. Hyōgo- ken Ono-shi Kyōiku Iinkai, 15–20. ———. 2000. Shōensei seiritsushi no kenkyū. Shibunkaku Shuppan. Kawai Masaharu. 1977. “Provincial Aspects of Muromachi Politics.” In Japan in the Muromachi Age, ed. John W. Hall and Toyoda Takeshi, 65–88. University of California Press. Kawai, Sachiko. 2007. “The Lady of the Eighth Ward: Political, Economic, and Military Power of Nyoin during the Twelfth Century.” MA thesis, University of Southern California. ———. 2015. “Power of the Purse: Estates and the Religio-political Influ- ence of Japanese Royal Women, 1100–1300.” PhD dissertation, Uni- versity of Southern California. ———. 2018. “Life-Cycle Rituals for Newborns in Heian Japan.” In Birth and Death in the Royal House: Selections from Fujiwara no Munetada’s Journal Chūyūki, ed. Christina Laffin et al. Cornell East Asia Series. Kawane Yoshiyasu. 1984. Chūsei hōken shakai no shuto to nōson. Tōkyō Daigaku Shuppankai. ———. 2010. Hōkensei riron no shomondai, Kawane Yoshiyasu chosakushū, vol. 3. Bunrikaku. Kawane Yoshiyasu, commentator, and Watanabe Setsuo, translator. 1987. “Maruku Burokku no Nihon hōkenseiron—Asakawa Kan’ichi no shochosaku e no shohyō.” Atarashii rekishigaku no tame ni 188: 24–30. Kawashima Shigehiro. 1981. “Kantoku shōen seiriryō to Tengi jikken.” Nihonshi kenkyū 227: 61–81. Keirstead, Thomas E. 1985. “Fragmented Estates. The Breakup of the Myō and the Decline of the Shōen System.” Monumenta Nipponica 40, no. 3: 311–330. ———. 1990. “The Theater of Protest.” Journal of Japanese Studies 16, no. 2: 357–388. ———. 1992. The Geography of Power in Medieval Japan. Princeton University Press. Bibliography 491

———. 1998. “Inventing Medieval Japan: The History and Politics of National Identity.” The Medieval History Journal 1: 47–71. ———. 2004. “Medieval Japan: Taking the Middle Ages Outside Europe.” History Compass 2: 1–14. Kemp, Karen. 2010. “Geographic Information Science for Spatial Analysis and the Humanities.” In The Spatial Humanities: GIS and the Future of Humanities Scholarship, ed. David J. Bodenhamer, John Corrigan, and Trevor M. Harris, 31–57. Indiana University Press. Kentani Shōtarō. 1981. “Tsugaru kaikyō wo watatta Suzuyaki to umi agari no Suzuyaki.” Tōsetsu 344, 25–30. Kikutake Jun’ichi and Yoshida Hiroshi, eds. 1981. Kōrai butsuga. Asahi Shinbunsha. Kiley, Cornelius J. 1974. “Estate and Property in the Late Heian Period.” In Medieval Japan: Essays in Institutional History, ed. John W. Hall and Jef- frey P. Mass, 109–124. Yale University Press. ———. 1982. “The Imperial Court as a Legal Authority in the Kamakura Age.” In Court and Bakufu in Japan: Essays in Kamakura History, ed. Jeffrey P. Mass, 29–44. Yale University Press. Reprint 1988, Stanford University Press. ———. 1999. “Administration and Land Tenure in Early Heian.” In Cam- bridge History of Japan, vol. 2 (Heian Japan), ed. Donald Shively and William McCullough, 236–340. Cambridge University Press. Kimura Eiichi. 2008. “Kamakura jidai no jisha funsō to Rokuhara tandai.” Shigaku zasshi 117, no. 7: 1276–1300. Kimura Mamiko. 2005. “Minamoto Toshiaki, In to Sekkanke to wo tsunaida ‘sōnaki’ Shirakawa In bettō.” In Ōchō no henyō to musha, ed. Motoki Yasuo, 265–293. Seibundō. Kimura Shigemitsu. 1982. “Daikaikon jidai no kaihatsu—sono gijutsu to seikaku. ” In Gijutsu no shakaishi 1: kodai, chūsei no gijutsu to shakai, ed. Miura Keiichi, 150–205. Yūhikaku. ———. 1985. “Chūsei zenki no nōgyō seisanryoku to hatasaku.” Nihonshi kenkyū 280: 52–82. ———. 1992. Nihon kodai, chūsei hatasakushi no kenkyū. Azekura Shobō. ———. 1996. “Kodai kara chūsei e.” In Zenkindaishi no atarashii manabikata, ed. Rekishi Kyōikusha Kyōgikai, 19–26. Aoki Shoten. ———. 2000. Chūsei no minshū seikatsushi. Aoki Shoten. ———. 2010. “Chūseiteki nōgyō seisan no keisei.” In Nihon nōgyōshi, 82–96. Yoshikawa Kōbunkan. ———. 2012. Sengo Nihon chūseishi kenkyū to mukiau. Aoki Shoten. ———. 2013. “Chūsei shōen no kōzō.” In Shōenshi kenkyū handobukku, ed. Shōenshi Kenkyūkai, 127–152. Tōkyōdō Shuppan. Kinokawa Ryūiki Shōen Shōsai Bunpu Iinkai, ed. 2004. Kinokuni Nate no shō—Shizukawa no shō chiiki chōsa. Chūō Insatsu. 492 Bibliography

Kishida Hiroshi. 1999. “Muromachi bakufu shugo to shōen.” In Shōen no kaitai, Kōza Nihon shōenshi, vol. 4, ed. Amino Yoshihiko, Ishii Susumu, Inagaki Yasuhiko, and Nagahara Keiji, 1–64. Yoshikawa Kōbunkan. Kitō Hiroshi. 2002. Nihon no rekishi 19: Bunmei toshite no Edo shisutemu. Kōdansha. Kitō Kiyoaki. 1985. Kodai no mura. Iwanami Shoten. ———. 1991. Review of Ishimoda Shō chosakushū. Shigaku zasshi 100, no. 3: 62–71. Kobayashi Kazutake. 2002. “Shibuta no shō.” In Kinokuni shōen no sekai, vol. 2, ed. Yamakage Kazuo, 33–64. Seibundō. Kobayashi Shōji. 2000. Nihon kodai no sonraku to nōmin shihai. Hanawa Shobō. Kobayashi Takeshi. 1965. Shunjōbō Chōgen shiryō shūsei. Nara Kokuritsu Bunka- zai Kenkyūjo shiryō, vol. 4. Nara Kokuritsu Bunkazai Kenkyūjo. ———. 1971. Shunjōbō Chōgen no kenkyū. Yūrindō. Kobayashi Yasuo. 1981. “Irifune nōchō ni miru kokuryō to kasho.” In Hyōgo Kitaseki irifune nōchō, ed. Hayashiya Tatsusaburo, 289–302. Chūō Kōron Bijutsu Shuppan. ———. 2008. “ ‘Hyōgo Kitaseki irifune nōchō’ ni miru edabune.” Ritsumei- kan bungaku 605: 795–807. Kōbe Shinbunsha, ed. 1959. Sosen no ashiato 3. Kōbe Shinbunsha. Kobrin, David. 1996. Beyond the Textbook: Teaching History Using Documents and Primary Sources. Heinemann. Kodama Kōta and Hiroyuki Kihara, ed. 1997. Kagawa-ken no rekishi. Yama kawa Shuppansha. Kohara Yoshiki. 2009. “(Chōgen isan) sono ato: shoki Kanjinjo to Tōdaiji,” Nihonshi kenkyū 566: 1–28. Koizumi Nobusuke. 1963. “Tōdaiji no jiryō shakai ni kansuru ichi kōsatsu.” Nihon bukkyō 16: 8–19. Kojima Shōsaku. 1932. “Sekkanke ōbanryō toshite no Ina no shō.” Reki- shi chiri 60, no. 5: 435–445. Kokuritsu Rekishi Minzoku Hakubutsukan, comp. 1998. Nihon shōen shiryō. 2 vols. Yoshikawa Kōbunkan. ———, comp. 2008. Seigyō kara miru Nihonshi. Yoshikawa Kōbunkan. Kokushi Chiri’in. Denshi kokushi. Geospatial Information Authority of Japan Digital Map Service. Komeie Taisaku. 2002. Chū, kinsei sanson no keikan to kōzō. Azekura Shobō. Kon Masahide. 1994. “Hōgen shōen seiri rei no rekishiteki igi: Heian kōki shōen seiri seisaku wo meguru seifu to kenmon.” Nihonshi kenkyū 378: 1–28. Kondō Shigekazu. 1992. “Kamakura bakufu no seiritsu to tennō.” In Kōza: zenkindai no tennō: Tennō kenryoku no kōzō to tenkai, vol. 1, ed. Naga- hara Keiji, 175–206. Aoki Shoten. Bibliography 493

———. 1993. “Akutō meshitori no kōzō.” In Chūsei no hakken, ed. Nagahara Keiji, 21–39. Yoshikawa Kōbunkan. ———. 2008. “Japanese Medieval Documents from the Beinecke Rare Book and Manuscript Library.” http://todai-yale.jp/resources/docs /Kondo.pdf. Accessed September 10, 2015. Konishi Mizue. 2003. “Todaijiryō Harima no kuni Ōbe no shō ni tsuite ichi kōsatsu.” Ōsaka Shoin Joshi Daigaku ronshū 40: 41–53. Kōtaishi Naruhito Shinnō. “Kyōto to chihō wo musubu mizu no michi: kodai chūsei no Biwa-ko Yodogawa suiun wo chūshin toshite.” http://www .kunaicho.go.jp/okotoba/02/koen/koen-h15az-mizuforum3th.html. Accessed August 11, 2014. Koyama Yasunori. 1973. “Shōensei keiseiki no ryōshu to nōmin, Iga Kuroda no shō.” In Shōen no sekai, ed. Inagaki Yasuhiko, 107–138. UP Sensho. ———. 1981. “Kodai shōen kara chūsei shōen e.” Rekishi chiri kyōiku 329. ———. 1987a. “Shōen sonraku no keikan.” In Shūkan Asahi hyakka Nihon no rekishi 588: 154–159. ———. 1987b. Chūsei sonraku to shōen ezu. Tōkyō Daigaku Shuppankai. ———. 1990. “Kōyasan Goshuin engi to shōensei.” In Kishūshi kenkyū, vol. 5 (Kōyasan sōtokushū), ed. Andō Seiichi, 1–24. Kokusho Kankōkai. ———. 1998. Chūsei jisha to shōensei. Hanawa Shobō. Koyama Yasunori, Takeuchi Masato, Sakaehara Towao, Yumikura Hiroto- shi, Kasahara Masao, and Takashima Masaaki, eds. 2004. Wakayama- ken no rekishi. Yamakawa Shuppansha. Krieger, Larry S., Kenneth Neill, and Steven L. Jantzen. 1992. World History: Perspectives on the Past. D. C. Heath and Company. Kubota Hiroshi. 1989. Kaisen taihōkō: Sumida Shōichi Hakushi kaisen shikimoku no kenkyū shūi. Ōsaka Keizai Hōka Daigaku Shuppanbu. Kubota Kazuhiko. 1989. “Nabari gunji Hasetsukabe-shi ichizoku to Yanase ho.” In Chūsei Nihon no shosō, vol. 1, ed. Yasuda Motohisa Sensei Tainin Kinen Ronshū Kankō Iinkai, 173–203. Yoshikawa Kōbunkan. Kudō Keiichi. 1969. Kyūshū shōen no kenkyū. Hanawa Shobō. ———. 1978. “Nankyō ‘ichi no akusō’ Kakunin, Iga no kuni Tōdaijiryō no kanseisha.” In Shōen no hitobito, 69–85. Kyōikusha. ———. 1983. “Shōen.” Acta Asiatica 44: 1–27. ———. 2002. Shōensei shakai no kihon kōzō. Azekura Shobō. Kurata Bunsaku. 1971. “Harima Jōdoji no Mida sanzonzō.” Bukkyō geijutsu 80: 1–11. Kuriyama Keiko. 2001. “Junbo rikkōsei ni miru chūsei zenki no ōke.” Nihon- shi kenkyū 465: 1–24. Kuroda Akinobu. 1994. Chūka teikoku no kōzō to sekai keizai. Nagoya Daigaku Shuppankai. 494 Bibliography

———. 2003. Kahei shisutemu no sekaishi—‘Hitaishōsei’ wo yomu. Iwanami Shoten. Kuroda Hideo. 1978. “Jū, jūisseiki no shishi ni tsuite.” In Minshūshi no kadai to hōkō, ed. Minshūshi Kenkyūkai, 62–86. San’ichi Shobō. ———. 1979. “Chūsei seiritsuki ni okeru hatasakushi no hatten.” Minshūshi kenkyū 17: 29–60. ———. 1984. Nihon chūsei kaihatsushi no kenkyū. Azekura Shobō. ———. 1989. “Shōen ezu no shiryōgaku.” In Kōza Nihon shōenshi, vol. 1, ed. Amino Yoshihiko, Ishii Susumu, Inagaki Yasuhiko, and Nagahara Keiji, 148–186. Yoshikawa Kōbunkan. Kuroda Hiroko. 1985 [1982]. “Kamakura kōki ni okeru ike-chikuzō to sōson no seiritsu.” In Chūsei sōsonshi no kōzō, 111–151. Yoshikawa Kōbunkan. Kuroda Toshio. 1967. Shōensei shakai (Taikei Nihon rekishi 2). Nihon Hyōronsha. ———. 1974. Nihon chūsei hōkenseiron. Tōkyō Daigaku Shuppankai. ———. 1975. “Chūseishi josetsu.” In Iwanami kōza Nihon rekishi, chūsei 1, 1–34. Iwanami Shoten. ———. 1976. Nihon chūsei no kokka to shūkyō. Iwanami Shoten. ———. 1980. Jisha seiryoku. Iwanami Shinsho. ———. 1994a. Kenmon taiseiron, vol. 1 of Kuroda Toshio chosakushū. Hōzōkan. ———. 1994b. “Shōensei shakai to bukkyō.” In Kuroda Toshio chosakushū, vol. 2, 3–44. Hōzōkan. Kurushima Noriko. 1994. “Tōji no sōkura—nengu shunō no ba.” In Shūkan Asahi hyakka Nihon no rekishi, Bessatsu 13, 38–42. Asahi Shinbunsha. ———. 2000. A review of Nagahara Keiji’s Shōen. Nihon rekishi 622: 106–109. Kyōto Daigaku Bungakubu Hakubutsukan, ed. 1996. Shōen wo yomu ・ aruku. Shibunkaku Shuppan. Kyoto National Museum, ed. 1975. Jōdokyōga. Heibonsha. Laffin, Christina, Joan Piggott, and Yoshida Sanae, eds. 2018. Birth and Death in the Royal House: Selections from Fujiwara no Munetada’s Journal Chūyūki. Cornell East Asia Series. Lopez, Robert S. 1971. The Commercial Revolution of the Middle Ages, 950– 1350. Prentice-Hall. Lu, David, trans./comp. 1997. Japan: A Documentary History. M. E. Sharpe. Maeda Masaaki. 1999. “ ‘Kinokuni Kaseda no shō ezu’ ni egakareta ‘shima’ no ryōyū mondai—chūsei kara kinsei ni kakete no Shibuta- jima chiiki no kaihatsushi.” Wakayama Kenritsu Hakubutsukan kenkyū kiyō 4: 1–22. Maekawa Kaname. 1998. “Chūsei Noto Suzugama no seisan keitai to ryūtsū kikō ni tsuite no ichi kōsatsu.” In Narasaki Shōichi Sensei koki kinen ronbunshū, ed. Narasaki Shōichi Sensei Koki Kinen Ronbunshū Kankōkai, 247–258. Shinyōsha. Bibliography 495

Maki Kenji. 1932. “Sekkanke no ōbanryō to Izumi no kuni Ōtori no shō.” Rekishi to chiri 30, no. 1: 1–19; 30, no. 2: 23–44; 30, no. 3: 28–47. Maki Michio. 1993. Insei jidaishi ronshū. Zoku Gunsho Ruijū Kanseikai. Maruyama Yukihiko. 1973. “Jūseiki ni okeru shōen no keisei to tenkai, Tōdaijiryō Itabae Soma wo chūshin ni.” Shirin 56, no. 6: 28–66. ———. 2001. Kodai Tōdaiji shōen no kenkyū. Keisuisha. Mass, Jeffrey P. 1974. “Jitō Land Possession in the Thirteenth Century: The Case of Shitaji Chūbun.” In Medieval Japan, ed. John W. Hall and Jeffrey P. Mass, 157–183. Yale University Press. Republished 1988, Stanford University Press. ———. 1976. The Kamakura Bakufu: A Study in Documents. Stanford Univer- sity Press. ———, ed. 1982. Court and Bakufu in Japan: Essays in Kamakura History. Yale University Press. Republished 1995, Stanford University Press. ———. 1989. Lordship and Inheritance in Early Medieval Japan: A Study of the Kamakura Sōryō System. Stanford University Press. ———. 1990. “The Kamakura Bakufu.” In Cambridge History of Japan, vol. 3 (Medieval Japan), ed. Kozo Yamamura, 46–88. Cambridge University Press. ———. 1992. Antiquity and Anachronism in Japanese History. Stanford Uni- versity Press. ———. 1995. “The Kamakura Bakufu.” In Warrior Rule in Japan, ed. Marius Jansen, 1–43. Cambridge University Press. ———, ed. 1997. The Origins of Japan’s Medieval World: Courtiers, Clerics, Warriors, and Peasants in the Fourteenth Century. Stanford University Press. ———. 1999. Yoritomo and the Founding of the Kamakura Bakufu. Stanford University Press. Matsubara Hironobu. 1976. “Kodai ni okeru tsu no seikaku to kinō.” In Kodai kokka no keisei to tenkai, ed. Ōsaka Rekishi Gakkai, 464–504. Yoshikawa Kōbunkan. ———. 2004. Kodai kokka to Setonaikai kōtsū. Yoshikawa Kōbunkan. Matsuda Fumio. 1984. Kii-no-kuni Ategawa no shō shi. Matsuda Fumio. Matsumura Keiji. 1998. “Mumon ginsen to Wadō ginsen—Asuka Fujiwara chiiki shutsudo ginsen wo chūshin ni).” In Shutsudo senka, vol. 9, 17–28. Shutsudo Senka Kenkyūkai. Matsunobu Yasutaka. 1989. “Zeni to kahei no kannen—Kamakuraki ni okeru kahei kinō no henka ni tsuite.” Rettō no bunkashi 6: 177–210. Matsushima Ken. 1992. “Tōdaiji Nandaimon Kongō Rikishi zō (agyō) zōnai shiryō.” Nanto bukkyō 67: 95–164. Matsushita Masashi. 1994. Umoreta minatomachi Kusadosengen, Tomo, Onomichi. Heibonsha. 496 Bibliography

Matsuyama Tomoko. 2003. “Chūsei sankan sonraku no nariwai keitai: Ōmi no kuni Katsuragawa・Ikada-tsu no shō wo chūshin toshite.” Shisō 44: 93–114. Matto-shi Kyōiku Iinkai, ed. 1990. Yokoe no shō iseki hakkutsu chōsa gaihō. Mattō-shi. Mauclaire, Simone. 2009. Le Nom, l’Écrit, le Non-dit: Mentalités rurales et culture intermédiaire dans le Japon médiéval. Collège de France Institut des Hautes Études Japonaises/Bocard. Minegishi Sumio. 2001. “Asama-san no bakuhatsu to shōen no keisei.” In Chūsei saigai, senran no shakaishi. Yoshikawa Kōbunkan. Miura Keiichi. 1990. Nihon chūsei senminshi no kenkyū. Buraku Mondai Kenkyūjo. ———. 1993. “Chūsei no doboku to shokunin shūdan.” In Nihon chūsei no chiiki to shakai, 293–327. Shibunkaku Shuppan. Miura Yuzuru and Zenkoku Jinja Meikan Kankōkai Shigaku Sentā, eds. 1977. Zenkoku jinja meikan, vol. 1. Zenkoku Jinja Meikan Kankōkai Shigaku Sentā. Miyagawa Mitsuru. 1977. “From Shōen to Chigyō.” In Japan in the Muroma- chi Age, ed. John W. Hall and Toyoda Takeshi, 89–105. University of California Press. Miyasaka Yūshō and Satō Tamotsu. 1984. Kōyasanshi. Shinbunsha. Mizubayashi Takeshi et al. 2001. Hō shakaishi. Yamakawa Shuppansha. Mizuno Keizaburō. 1977. “Sōdai bijutsu to Kamakura chōkoku.” Kokka 1000: 53–61. ———. 1997. Nihon chōkokushi kenkyū. Chūō Kōron Bijutsu Shuppan. Mizuno Keizaburō and Soejima Hiromichi. 1996. “Jōdoji Jōdodō Amida Nyoraizō no meiki ni tsuite.” In Nihon chōkokushi kenkyū, ed. Mizuno Keizaburō. Chūō Kōronsha. Mōri Hisashi. 1961. Busshi Kaikei ron. Yoshikawa Kōbunkan. ———. 1976. “Shunjōbō Chōgen to busshi Kaikei.” Bukkyō geijutsu 105: 61–75. Mōri Kazunori. 1974. “Bitasen no kachi hendō ni kansuru kenkyū: Jō, ge.” Nihon rekishi 310 and 311. Mori Kimiyuki. 2000. Kodai gunji seido no kenkyū. Yoshikawa Kōbunkan. Mori Yasuhiro. 1977. “Kinsei no toshi to ryūtsū kikō.” In Kiso keizaigaku taikei, vol. 5 (Nihon keizaishi), ed. Miyamoto Mataji. Seirin Shoin Shinsha. Mori Yukio. 2005. Rokuhara tandai no kenkyū. Zoku Gunsho Ruijū Kanseikai. Morillo, Stephen. 2014. Frameworks of World History: Networks, Hierarchies, Culture, vol. 1 (To 1550). Oxford University Press. Morimoto Masahiro. 2008. Zōtō to enkai no chūsei. Yoshikawa Kōbunkan. Morita Hayato. 2003. “Tōdaijiryō Iga no kuni Kuroda no shō dezukurichi no tenkai: Tengi jikken saikō.” Kamakura ibun kenkyū 12: 51–72. Bibliography 497

———. 2006. “Shōen kōryōsei no tenkai to zaichiryōshu no keiei.” Rekishi hyōron 674: 16–26. ———. 2013. “Inseiki no shōen.” In Shōenshi kenkyū handobukku, ed. Shōenshi Kenkyūkai, 65–94. Tōkyōdō Shuppan. Morita Tei. 1978. “Heian chūki gunji ni tsuite no ichi kōsatsu.” In Heian jidai seijishi kenkyū, 79–98. Yoshikawa Kōbunkan. Moriuchi Shūzō. 2010. “Suzuyaki to Tōbankei sueki: Seisanchi kara mita Suzu to Tōban no kankei.” In Suzuyaki Tanjō! shiseki Suzu tōki kama ato kuni shitei kinen shinpojiumu hōkokusho, ed. Suzu-shi Kyōiku Iinkai, 3–13. Suzu-shi and Suzu-shi Kyōiku Iinkai. Motoki Yasuo. 1996. Inseiki seijishi kenkyū. Shibunkaku Shuppan. Murai Yasuhiko. 1965. Kodai kokka kaitai katei no kenkyū. Iwanami Shoten. Murakami Kazuhiro and Nishimura Yasuhiro, eds. 2016. Shiryō de yomu Nihon hōshi. 2nd ed. Hōritsu Bunkasha. Murata Shūzō. 2012. “Tsuchimaru・Ameyama jōseki to yamairi・yamagoya ni tsuite.” In Tsuchimaru・Ameyama jōseki, ed. Izumisano-shi Kyōiku Iinkai and Kumatori-chō Kyōiku Iinkai, 92–93. Murayama Shūichi. 1962. Fujiwara no Teika. Yoshikawa Kōbunkan. Nabari-shi Kyōiku Iinkai, ed. and pub. 1983. Mie-ken Nabari-shi iseki chizu. ———. ed. and pub. 1985. Natsumi haiji, daiichiji hakkutsu chōsa gaiyō. Nagahara Keiji. 1955. Nihon hōken shakairon. Tōkyō Daigaku Shuppankai. ———. 1958. “Kugeryō shōen ni okeru ryōshuken no kōzō.” Hitotsubashi ronsō 40, no. 6: 532–560. ———. 1975. “Landownership under the Shōen-Kokugaryō System.” Jour- nal of Japanese Studies 1, no. 2: 269–296. ———. 1977. “Village Communities and Daimyo Power.“ In Japan in the Muromachi Age, ed. John W. Hall and Toyoda Takeshi, 107–123. Uni- versity of California Press. ———. 1990. “The Decline of the Shōen System” and “The Medieval Peas- ant.” In Cambridge History of Japan, vol. 3 (Medieval Japan), ed. Kozo Yamamura, 260–343. Cambridge University Press. ———. 1998. Shōen. Yoshikawa Kōbunkan. Republished 2005. ———. 1999. “Shōen kaitaiki ni okeru ukeou daikansei.” In Kōza Nihon shōenshi, vol. 4 (Shōen no kaitai), ed. Amino Yoshihiko, Ishii Susumu, Inagaki Yasuhiko, and Nagahara Keiji, 65–104. Yoshikawa Kōbunkan. ———. 2004a. Choma, kinu, momen no shakaishi. Yoshikawa Kōbunkan. Republished 2008, Nagahara Keiji chosaku senshū, vol. 8. Yoshikawa Kōbunkan. ———. 2004b. Nihon hōkensei seiritsu katei no kenkyū. In Nagahara Keiji chosakushū, vol. 2. Iwanami Shoten. Nagai Susumu. 2005. “Hachijō-in: Tennōke wo meguru nyoin gunzō.” Reki- shi dokuhon 35, no. 6: 118–121. 498 Bibliography

Nagamura Makoto. 1981. “Tōdaiji daikanjinshiki to ‘zenritsusō.’” Nanto bukkyō 47: 63–102. ———. 1989. Chūsei Tōdaiji no soshiki to keiei. Hanawa Shobō. ———. 2000. Chūsei jiin shiryō ron. Yoshikawa Kōbunkan. Naka Sadao, ed. 1974. Nabari-shi shi: Nabari-shi sei nijūshūnen kinen. Rinsen Shoin. Nakada Kaoru. 1948 [1906]. “Ōchō jidai no shōen ni kansuru kenkyū.” In Shōen no kenkyū, 58–301. Shōkō Shoin. Nakajima Keiichi. 2008. “The Establishment of Silver Currency in Kyoto.” International Journal of Asian Studies 5, no. 2: 219–234. Nakamura Ken. 1976. Kii no kuni Ategawa no shō. Yoshikawa Kōbunkan. Nakamura Naokatsu. 1976 [1939]. “Iga no kuni no Kuroda no shō.” In Shōen no kenkyū, 370–545. Bōchō Shiryō Shuppansha. ———. 1978a. Shōen no kenkyū. Tankōsha. ———. 1978b. “Kajūji-keryō ni tsuite.” In Nakamura Naokatsu chosakushū, vol. 4, 501–560. Tankōsha. Nakano Genzō. 1985. Raigōzu no bijutsu. Dōhōsha. Nakao Takashi. 1977. “Chōgen wo chūshin to suru kanjin hijiri shūdan no soshikika to teichaku—Tōdaiji shōryō dōsha bessho no keisei to un’ei wo megutte.” In Risshō Daigaku Shigakkai sōritsu gojūshūnen kinen, Shūkyō shakaishi kenkyū, ed. Risshō Daigaku Shigakkai, 175–192. Yūzankaku. Nakao Takashi and Imai Masaharu. 1983. Chōgen ・ Eison ・ Ninshō. Yoshikawa Kōbunkan. Naoki Kōjirō. 1987. Nihon kodai kokka no seiritsu. Shakai Shisōsha. Nara Kokuritsu Bunkazai Kenkyūjo, comp. 1955. Namuamidabutsu sazenshū. Collotype edition. Nara Kokuritsu Bunkazai Kenkyūjo shiryō, vol. 1. Shinyōsha. Nara National Museum. 2006. Daikanjin Chōgen: Tōdaiji no Kamakura fukkō to arata na bi no sōshutsu. Nara Kokuritsu Hakubutsukan. ———. 2013. Taimadera: Gokuraku Jōdo e no akogare. Nara Kokuritsu Hakubutsukan. Negayama Yasufumi. 2011. “Kamakuraki ni okeru ‘akutō ’ no gogi henka.” Shirin 94, no. 2: 72–73. Nelson, Thomas. 2004. “Slavery in Medieval Japan.” Monumenta Nipponica 59, no. 4: 463–489. Nihon Engyō Taikei Henshū Iinkai, ed. and pub. 1975. Nihon engyō taikei, Shiryō hen, vol. 1. Nihon Engyō Kenkyūkai. Nihon Kōkogaku Kyōkai, ed. 2000. Hatake no kōkogaku. Proceedings of the 2000 Meeting in Kagoshima, vol. 1. Nihon Komonjo Kenkyūkai, comp. 2017. Komonjo kenkyū 83 (special issue). Yoshikawa Kōbunkan. Bibliography 499

Nihon shōen dētabēsu. 1993. Ed. Kokuritsu Rekishi Minzoku Hakubutsukan, www.rekihaku.ac.jp. Nishikawa Shinji et al., eds. 1987–1992. Byōdōin taikan. 3 vols. Iwanami Shoten. Nishimura Kōchō et al. 1982. “Jōdoji Amida sanzonzō to sono shikakuteki kankyō ni tsuite.” Tōkyō Geijutsu Daigaku Bijutsu Gakubu kiyō 17: 11–45. Nishitani Masahiro. 1998a. “Kugeryō shōen no henyō: Kujō-keryō shōen no kobetsuteki kenkyū wo chūshin ni.” Fukuoka Daigaku jinbun ronsō 29, no. 4: 2771–2833. ———. 1998b. Kamakuraki ni okeru kizoku no ie to shōen.” Nihonshi kenkyū 428: 28–60. ———. 2011. “Shōensei to saigoku shakai.” In Saigoku ni okeru seisan to ryūtsū, ed. Kawaoka Tsutomu and Koga Nobuyuki. Seibundō. Nishiyama Masatoshi. 1997. Umi wo tabi shita Suzuyaki. Ed. and pub. Suzu shiritsu Suzuyaki Shiryōkan. Nishiyama Ryōhei. 1984. “Ritsuryōsei shakai no henyō.” In Kōza Nihon rekishi, vol. 2, comp. Rekishi Kenkyūkai and Nihonshi Kenkyūkai, 129– 168. Tōkyō Daigaku Shuppankai. Noguchi Hanayo. 2005. “Chūsei zenki no ōke to Anrakujuin: ‘Nyoinryō’ to nyoin no honshitsu.” Hisutoria 198: 225–252. ———. 2006. “Nyoin wo meguru kenkyū to kadai.” Sōgō joseishi kenkyū 23: 89–91. Noguchi Minoru. 2002. “Rettō nettowāku no naka no Hiraizumi.” In Hirai- zumi no sekai, ed. Irumada Nobuo and Honzawa Shinsuke, 39–61. Ōshū kenkyū sōsho. Kōshi Shoin. ———. 2010. “Kyōto Shichijōchō kara rettō shochiiki e: Bushi to seisan・ryūtsū.” In Tsuwamonotachi no seikatsu bunka, ed. Iru- mada Nobuo. Tsuwamonotachi no jidai 2, 186–200. Koshi Shoin. Nomura Ikuyo. 1989a. “Nyoinron.” In Shinjin to kuyō, ed. Ōsumi Kazuo and Nishiguchi Junko, 135–166. Heibonsha. ———. 1989b. “Chūsei ni okeru tennō-ke.” In Kazoku to josei no rekishi: kodai chūsei, ed. Zen Kindai Joseishi Kenkyūkai, 294–319. Yoshikawa Kōbunkan. ———. 1992. “Ōken no naka no josei.” In Chūsei wo kangaeru: Kazoku to josei, ed. Minegishi Sumio, 7–38. Yoshikawa Kōbunkan. ———. 2006. Kazokushi toshite no nyoinron. Azekura Shobō. Noritake Yūichi. 2002. “Ōga no shō.” In Kinokuni shōen no sekai, vol. 2, ed. Yamakage Kazuo, 1–29. Seibundō. Ogawa Hirokazu. 2007. “Fujiwara Kiyokada, Sanetō no kanmotsu ukeou to zuryōsei.” Nihon rekishi 710: 1–17. 500 Bibliography

Oka Yōichirō. 2010. “Hiraizumi Fujiwarashi to kōtsū.” In Tsuwamonotachi no seikatsu bunka, ed. Irumada Nobuo. Tsuwamonotachi no jidai 2, 128– 155. Koshi Shoin. Okano Tomohiko. 1988. “Koga-keryō shōen no denryō to sono sōzoku.” Shigaku zasshi 96, no. 4: 458–478, 572–573. ———. 2002. Chūsei Koga-ke to Koga-keryō shōen. Zoku Gunsho Ruijū Kanseikai. Ōmura Takuo. 2000. “Chūsei zenki no Toba to Yodo.” Nihonshi kenkyū 439: 1–29. Ono Kōji. 1989. Nihon chūsei shōgyōshi no kenkyū. Hōsei Daigaku Shuppankyoku. Ono Masatoshi. 2001. Zukai, Nihon no chūsei iseki. Tōkyō Daigaku Shuppankai. Ono-shi Hensan Senmon Iinkai, comp. Ono-shi shi (OS) vol. 1, Honben I, 2002; vol. 2, Honben II, 2003; Bekkan (Bunkazai hen), 1996. Ono City. Ono-shi Kyōiku Iinkai, ed. and pub. 1997. Ono-shi Tōdaiji Summit in Ono, Symposium: Kokuhō Jōdodō wo kataru, Kirokushū. Ōsaka Rekishi Gakkai Kengaku Kentōkai. 2006. “Chūsei shōen no keikan: Hine no shō Ōgi.” Hisutoria 202: 2–6. Ōsaka-furitsu Sayama Ike Hakubutsukan, comp. and pub. 2002. Chōgen, Sayama Ike kaishū happyakunen kinen: Chōgen to sono jidai no kaihatsu. Ōsaka-furitsu Sayama Ike Hakubutsukan zuroku 4. Ōshima Nobujirō. 1964. Sekisho: Sono rekishi to jittai. Jinbutsu Ōraisha. Ōsone Shōsuke and Kubota Jun, eds. 2000. Kamo no Chōmei zenshū. Kichōhon Kankōkai. Ōta Yukio. 1994. “12—15 seiki shotō higashi Ajia ni okeru dōsen rufu— Nihon, Chūgoku wo chūshin toshite.” Shakai keizai shigaku 61, no. 2: 20–48. Ōtake Masami. 2005. “Akutō shiryō ichiran” and “Akutō shiryō ni mirareru ‘hyakushō’ ni tsuite.” In Akutō to nairan, ed. Akutō Kenkyūkai, 339– 377. Iwata Shoin. Ōtsu Tōru. 2001. “Kyaku shiki no seiritsu to sekkanki no hō.” In Hō shakaishi, ed. Mizubayashi Takeshi, Ōtsu Tōru, Nitta Ichirō, and Ōtō Osamu, 69–102. Yamakawa Shuppansha. Ōtsu-shi, ed. 1963. Ōtsu-shi shi, Bekkan. Ōtsu Shiyakusho. ———. Shinshū Ōtsu-shi: Kodai, vol. 1 (1978); Chūsei, vol. 2 (1979). Ōtsu-shi. Ōtsu-shi Shiritsu Kyōikukai, ed. 1911. Ōtsu-shi shi. Junpūbō. Overfield, Andrea. 2012. The Human Record: Sources of Global History, vol. 1 (To 1500). 7th ed. Wadsworth Cengage Learning. Oxenboell, Morten. 2005. “Images of Akutō.” Monumenta Nipponica 60, no. 2: 235–262. ———. 2006. “The Mineaiki and Discourses on Social Unrest in Medieval Japan.” Japan Forum 18, no. 1: 1–21. Bibliography 501

———. 2009. “Akutō. Images of Medieval Japanese Banditry.” PhD dissertation, University of Copenhagen. Ōyama Kyōhei. 1978. Nihon chūsei nōsonshi no kenkyū. Iwanami Shoten. ———. 1990. “Medieval Shōen.” In Cambridge History of Japan, vol. 3 (Medi- eval Japan), ed. Kozo Yamamura, 89–127. Cambridge University Press. ———. 1993. “Shōensei.” In Iwanami kōza Nihon tsūshi, vol. 7 (chūsei 1), ed. Asao Naohiro, 141–182. Iwanami Shoten. ———. 1996. Chūsei shōen no sekai: Tōjiryō Tanba no kuni Ōyama no shō. Shi- bunkaku Shuppan. ———. 1999. “Shunjōbō Chōgen no hisezokuteki keizai katsudō.” Ōtani gakuhō 78, no. 1: 1–22. ———. 2002. “Rekishi jūjutsu toshite no Mineaiki.” Nihonshi kenkyū 473: 1–35. ———. 2003. Yuruyakana kāsuto shakai, chūsei. Azekura Shobō. ———. 2012. Nihon chūsei no mura to kamigami. Iwanami Shoten. Payne, Richard, ed. 1998. Re-Visioning Kamakura Buddhism. University of Hawai‘i Press. Piggott, Joan R. 1982. “Hierarchy and Economics in Early Medieval Tōdaiji.” In Court and Bakufu in Japan: Essays in Kamakura History, ed. Jeffrey P. Mass, 45–91. Yale University Press. ———. 1987. “Tōdaiji and the Nara Imperium.” PhD dissertation, Stanford University. ———. 1993. Review of Thomas Keirstead’s The Geography of Power. Bulletin of the School of African and Oriental Studies LVI, no. 3: 41–42. ———. 1997. The Emergence of Japanese Kingship. Stanford University Press. ———, ed. 2006. Capital and Countryside in Japan, 300–1180: Japanese Histo- rians Interpreted in English. Cornell East Asia Series. ———. 2007. Shinsarugakuki, Introduction and Selections. In Haruo Shi- rane, Traditional Japanese Literature: An Anthology, Beginnings to 1600. Columbia University Press, 491–497. ———. Forthcoming. Shōen Documents and Materials in English Trans- lation, http://www.uscppjs.org/resources/#/estate-translation-project. Piggott, Joan R., and Yoshida Sanae. 2008. Teishinkōki: What Did a Regent Do? Cornell East Asia Series. Quinter, David. 2015. From Outcasts to Emperors: Shingon Ritsu and the Mañjuśri Cult in Medieval Japan. Brill. Rekishikan Izumisano Historical Museum home page, 2007. http://www .city.izumisano.lg.jp. Rekishikan Izumisano jōsetsu tenshi annai (Guide to the permanent exhibition at the Izumi Sano Historical Museum), 1996. Rekishikan Izumisano and Shōen Kenkyūkai, comp. and pub. 2005. Hine no shō no iseki to shiryō. 502 Bibliography

———, comp. and pub. 2010. Hineno-mura Yugawa yōsui ezu no suiro to genkyō. ———, comp. and pub. 2012. Hine no shō no shōen ezu no kenkyū. Rosenfield, John M. 2011. Portraits of Chōgen: The Transformation of Buddhist Art in Early Medieval Japan. Brill. Ryō Susumu. 1957. Kamakura jidai. 2 vols. Shunjūsha. Sakaehara Towao. 1980. “Coinage in the Nara and Heian Periods.” Acta Asiatica 39: 1–20. Sakai Kimi. 1999. Nihon chūsei no zaichi shakai. Yoshikawa Kōbunkan. Sakata Akihiro et al. 2006. “Zukeikagakuteki shuhō ni yoru Jōdoji Jōdodō nai no hikari kankyō no bunseki.” Nihon Kenchiku Gakkai kankyōkei ronbunshū 603: 7–23. Sakata Mie. 2005. “Jisha kōsō ni miru Kamakura kōki no chōbaku kankei.” Kanazawa Daigaku Bungakubu Nihonshigaku kenkyūshitsu kiyō 1: 33–44. Sakuma Takashi. 1993. “Hakkutsu sareta chūsei no mura to machi.” In Iwa- nami kōza Nihon tsūshi, vol. 9, ed. Asao Naohiro, 175–206. Iwanami Shoten. Sakurai Eiji. 1987. “Mittsu no shūrishiki.” Harukanaru chūsei 8: 15–28. ———. 1994. “Shōnin no ie, shokunin no ie.” In Shūkan Asahi hyakka Nihon no rekishi, Bessatsu 13, 32–35. Asahi Shinbunsha. ———. 1996. Nihon chūsei no keizai kōzō. Iwanami Shoten. ———. 2004. “Chūsei ni okeru bukka no tokusei to shōhisha kōdō.” In Kokuritsu rekishi minzoku hakubutsukan kenkyū hōkoku 113: 55–80. ———. 2007. “Ashikaga Yoshimitsu to chūsei no keizai.” ZEAMI 4: 68–82. ———. 2013. “Chūseishi e no shōtai.” In Iwanami kōza Nihon rekishi, vol. 6 (chūsei 1), ed. Ōtsu Tōru, Sakurai Eiji, Fuji Jōji, Yoshida Yutaka, and Yi Song-si, 3–28. Iwanami Shoten. Sakurai Eiji and Nakanishi Satoru. 2002. Shintaikei Nihonshi 12: Ryūtsū keizaishi. Yamakawa Shuppansha. Sasaki Ginya. 1972. Chūsei shōhin ryūtsūshi no kenkyū. Hōsei Daigaku Shuppankyoku. Sasaki Muneo. 2006. “The Court-Centered Polity.” Translated by Joan R. Pig- gott. In Capital and Countryside in Japan, 300–1180: Japanese Historians Interpreted in English, ed. Joan R. Piggott, 227–244. Cornell East Asia Series. Sato, Elizabeth. 1974. “The Early Development of the Shōen.” In Medi- eval Japan, Essays in Institutional History, ed. John W. Hall and Jeffrey P. Mass, 91–108. Yale University Press. ———. 1979. “Ōyama Estate and Insei Land Policies.” Monumenta Nipponica 34, no. 1: 73–99. Satō Kazuhiko. 1969. “Akutō ni tsuite nōto.” Gekkan rekishi 13: 1–6. ———. 1996. Nihon chusei no nairan to minshū undō. Azekura Shobō. Satō Kazuhiko and Kobayashi Kazutake, eds. 2000. Nanbokuchō nairan. Tōkyōdō Shuppan. Bibliography 503

Satō Kei. 1993. “Heian makki no Echizen no kami ni tsuite.” Fukui Kenritsu Hakubutsukan kiyō 5: 49–73. Satō Makoto. 2007. Kodai no chihō kanga to shakai. Yamakawa Shuppansha. Satō Makoto and Gomi Fumihiko, eds. 1996. Tochi to zaichi sekai wo meguru. Yamakawa Shuppansha. Satō Shin’ichi. 1971. Komonjogaku nyūmon. Hōsei Daigaku Shuppankyoku. ———. 2003. Shinpan komonjogaku nyūmon. Hōsei Daigaku Shuppankyoku. Satō Yasuhiro. 2001. “Shōen no kodai to chūsei.” Atarashii rekishigaku no tame ni 242–243: 2–15. Satō Yūki and Itō Rumi. 2013. “Kamakuraki no shōen.” In Shōenshi kenkyū handobukku, ed. Shōenshi Kenkyūkai, 95–126. Tōkyōdō Shuppan. Schirokauer, Conrad, David Lurie, and Suzanne Gay. 2013. A Brief History of Japanese Civilization. 4th ed. Wadsworth Cengage Learning. Segal, Ethan. 2009. “Awash with Coins.” In Currents in Medieval Japanese His- tory, ed. Gordon M. Berger, Andrew Edmund Goble, Lorraine F. Har- rington, and G. Cameron Hurst III, 331–361. Figueroa Press. ———. 2011. Coins, Trade, and the State: Economic Growth in Early Medieval Japan. Harvard University Asia Center. ———. 2012. “The Shōen System.” In Japan Emerging: Essays in Premodern His- tory, ed. Karl F. Friday, 167–177. Westview Press. Sekine Tatsuhito. 2004. “Hirosaki-shi Dōgatai shutsudo no Suzuyaki kyōyōki to sono igi.” Tōō bunka 75: 67–74. Shapinsky, Peter D. 2007. “With the Sea as their Domain: Pirates and Mar- itime Lordship in Medieval Japan.” In Seascapes: Maritime Histories, Littoral Cultures, and Trans-Oceanic Exchanges, ed. Jerry H. Bentley, Renate Bridenthal, and Kären Wigen, 221–238. University of Hawai‘i Press. ———. 2014. Lords of the Sea: Pirates, Violence, and Commerce in Late Medieval Japan. Center for Japanese Studies, University of Michigan. Shiga-ken, ed. 1927–1928. Shiga-ken shi. 6 vols. Shiga Prefecture. Shimizu Mitsuo. 1974. “Sekkanke ōban ho.” In Shimizu Mitsuo chosakushū, vol. 2, 246–256. Azekura Shobō. ———. 1975. “Tōdaijiryō Iga no kuni no Kuroda no shō.” In Shimizu Mitsuo chosakushū, vol. 3, 185–201. Azekura Shobō. Shimizu Ryō. 2013. “Nanbokuchōki, Sengokuki no shōen.” In Shōenshi kenkyū handobukku, ed. Shōenshi Kenkyūkai, 153–190. Tōkyōdō Shuppan. Shimozawa Atsushi. 2003. “‘Nanbokuchōki ni okeru akutō no hōteki ichi tsuke.” Kyōei Gakuen Tankidaigaku kenkyū kiyō 19: 13–45. Shinjō Tsunezō. 1994. Chūsei suiunshi no kenkyū. Hanawa Shobō. Shinshū Izumisano-shi shi. 2001–2009. Tsūshihen, 3 vols.; Bekkan, 4 vols. Comp. Izumisano-shi. Seibundō. 504 Bibliography

Shirane, Haruo. 2007. Traditional Japanese Literature: An Anthology, Begin- nings to 1600. Columbia University Press. Shively, Donald, and William McCullough, ed. 1999. Cambridge History of Japan, vol. 2 (Heian Japan). Cambridge University Press. Shōenshi Kenkyūkai, ed. 2013. Shōenshi kenkyū handobukku. Tōkyōdō Shuppan. Shūkan Asahi Hyakka. 1993. Nihon no rekishi, Bessatsu: Rekishi wo yominaosu, vol. 10 (Chūsei no tabi suru hitobito). Asahi Shinbunsha. ———. 1995. Nihon no rekishi, Bessatsu: Rekishi wo yominaosu, vol. 8 (Chūsei no mura wo tazuneru). Asahi Shinbunsha. Shūkan Asahi hyakka Nihon no rekishi, chūsei vol. 1.2. 1986. Chūsei no mura wo aruku. Asahi Shinbunsha. Shūkan Asahi hyakka Nihon no rekishi, chūsei vol. 2.20. 1986. Biwako to Yodo no suikei. Asahi Shinbunsha. Sotoyama Mikio. 2011. Chūsei Nagasaki no kisoteki kenkyū. Shibunkaku Shuppan. Souyri, Pierre-François. 2001. The World Turned Upside Down: Medieval Japa- nese Society. Columbia University Press. Steenstrup, Carl. 1980. “Sata Mirensho: A Fourteenth-Century Law Primer.” Monumenta Nipponica 35, no. 4: 405–435. ———. 1987. “The Legal System of Japan at the End of the Kamakura Period from the Litigants’ Point of View.” In Law and the State in Tra- ditional East Asia: Six Studies on the Sources of East Asian Law, ed. Brian E. McKnight, 73–100. University of Hawai‘i Press. Sugawara Masako. 1998. Chūsei kuge no keizai to bunka. Yoshikawa Kōbunkan. Suitō Toshio. 1989. “Kinsekibun.” In Kōza Nihon shōenshi, vol. 1, ed. Amino Yoshihiko, Ishii Susumu, Inagaki Yasuhiko, and Nagahara Keiji, 254– 281. Iwanami Shoten. Sukigara Toshio. 1999. “Shichijō chō to Hachijō-in chō.” In Kōkogaku ni manabu: Ikō to ibutsu, ed. Matsufuji Kazuto and Mori Kōichi. Dōshisha Daigaku. Suzuki Yōichi. 2010. Hine no shō iseki: Izumi ni nokoru chūsei shōen no keikan. Dōseisha. Suzu-shi Kyōiku Iinkai, ed. 2010. Suzuyaki Tanjō!: shiseki Suzu tōki kama ato kuni shitei kinen shinpojiumu hōkokusho. Suzu-shi and Suzu-shi Kyōiku Iinkai. Tabata Yasuko. 1989. “Shōen kenkyūshi.” In Kōza Nihon shōenshi, vol. 1, ed. Amino Yoshihiko, Ishii Susumu, Inagaki Yasuhiko, and Nagahara Keiji, 312–384. Iwanami Shoten. Taira Masayuki. 1996. “Kuroda Toshio and the Kenmitsu Taisei Theory.” Translated by Thomas Kirchner. Japanese Journal of Religious Studies 23, no. 3–4: 427–448. Bibliography 505

Takahashi Hiroaki. 1987. “Iga no kuni Komō maki ronsō to jūseiki no gun- jisei.” Kokushigaku 131: 37–62. Takahashi Kazuki. 2010. “Chūsei kenmon jisha no zaimoku chōtatsu ni miru gijutsu no shakaiteki haichi—chūsei zenki wo chūshin ni.” Koku- ritsu rekishi minzoku hakubutsukan kenkyū hōkoku 157: 31–50. ———. 2011. “Hokuriku shakai no kōtsū to chiiki kubun.” In Rettō no Kama kura jidai: Chiiki wo ugokasu bushi to jisha, ed. Takahashi Shinichirō, 230–260. Koshi Shoin. Takahashi Masaaki. 1981. “Saigoku jitō to ōchō kizoku: Aki no kuni Nuta no shō jitō Kobayakawa-shi no baai.” Nihonshi kenkyū 231: 1–33. Takahashi Noriyuki. 2011. A review of “Nihon chūsei shōensei seiritsushiron.” Shigaku zasshi 120, no. 4: 86–94. Takahashi Shinichirō. 1989. “Rokuhara tandai hikan to Hōjō-shi no Saikoku shihai.” Shigaku zasshi 98, no. 1: 374–388. Takahashi Toshiko. 1990. “Kamakuraki no jitō ryōshusei.” In Jinbutsu de tadoru Nihon shōenshi, ed. Abe Takeshi and Satō Kazuhiko, 125–140. Tōkyōdō Shuppan. ———. 2012. “Shōen no arikata wa dono chiiki demo mattaku onaji datta no deshō ka?” Nihon rekishi 764: 26–30. ———. 2014. Chūsei shōen no kankyō, kōzō to chiiki shakai: Bitchū no kuni Niimi no shō wo hiraku. Benseisha. Takamatsu Momoka. 2004. “Heian jidai no joseizō no tenkan wo: Inseiki no kokumo・nyoinron.” Rekishi chiri kyōiku 678: 70–75. Takebayashi Eiichi. 1994. “Chūsei Setouchi no shōhin ryūtsū.” Kenkyū hōkoku, Okayama Kenritsu Hakubutsukan 15: 2–30. Takefuji Nao. 1981. “Chūsei no Hyōgo-tsu to Setonaikai suiun: Irifune nōchō no fune sekichi hitei ni kanren shite.” In Hyōgo Kitaseki irifune nōchō, ed. Hayashiya Tatsusaburō, 232–271. Chūō Kōron Bijutsu Shuppan. Takeuchi Rizō. 1932. Nihon jōdai jiin keizaishi no kenkyū. Okayama Shoten. Republished 1997, Takeuchi Rizō chosakushū, vol. 2. Kadokawa Shoten. ———. 1942. Jiryō shōen no kenkyū. Unebi Shobō. Republished 1999, Take uchi Rizō chosakushū, vol. 3. Kadokawa Shoten. ———. 1960–1961. “Shōen to kizoku” series, in 14 parts. Nihon rekishi 143–162. ———, comp. 1975–1976. Shōen bunpuzu. 2 vols. Yoshikawa Kōbunkan. ———. 1982. “Old and New Approaches to Kamakura History.” In Court and Bakufu in Japan, ed. Jeffrey P. Mass, 268–283. Yale University Press. ———. 2000. “Shō no hatten to shōmin no seikatsu: Tōdaijiryō Iga no kuni Kuroda no shō.” In Takeuchi Rizō chosakushū, vol. 2, 241–306. Kado- kawa Shoten. Tamai Kōsuke, ed. 1954. Kenjugozen nikki. Nihon koten zensho, vol. 41. Asahi Shinbunsha. 506 Bibliography

Tamura Noriyoshi. 1994. Nihon chūsei sonraku keisei no kenkyū. Azekura Shobō. Tanabe Saburōsuke. 1972. Unkei to Kaikei. Nihon no bijutsu 78. Shibundō. ———. 1976. “Iga Bessho honzon kō.” Bukkyō geijutsu 105: 75–91. Tanaka Bun’ei. “Kanzaki-kawa ryūiki no hattatsu to kōshin. http://www .archives.city.amagasaki.hyogo.jp/chronicles/visual/02chuusei /chuusei1–1.html. Accessed August 11, 2014. Tanaka Fumihide. 1984. “Chūsei zenki no jiin to minshū.” Nihonshi kenkyū 266: 4–23. ———. 1994. Heishi seiken no kenkyū. Shibunkaku. Tanaka Hiroshi. 2008. Ritsuryō taisei wo sasaeta chihō kanga—Mirokuji iseki- gun. Shinsensha. Tanaka Jurō. 1984. “Mizuguchi ni tsuite no oboegaki, Iga Kuroda no shō wo chūshin ni.” In Takeuchi Rizō Sensei kiju kinen ronbunshū, vol. 2 (Shōensei to chūsei shakai), 104–115. Tōkyōdō Shuppan. Tanaka Kōji. 1996. “Nihon chūsei ni okeru zeni no shakaiteki kino wo megutte.” In Nōgaya Shutsudosen chōsa hōkokusho, ed. Nōgaya Shutsu- dosen Chōsakai, 102–123. Machida-shi Kyōiku Iinkai. Tanaka Tan. 1973. “Harima Jōdoji shiryō no saikentō.” Bukkyō geijutsu 93: 84–102. ———. 1976. “Chōgen no zōei katsudō.” Bukkyō geijutsu 105: 20–50. Tashiro Osamu. 1977. “Kōyasan kenryoku to nōmin no dōkō.” In Kōyasanryō shōen no shihai to kōzō, ed. Toyoda Takeshi, 215–266. Gannandō. Tatsuta Yoshio. 2003. Chūsei Tōjiryō shōen no shihai to zaichi. Azekura Shobō. Tauger, Mark B. 2011. Agriculture in World History. Routledge. Teishitsu Rinya-kyoku, ed. and pub. 1937. Goryō chishikō. Tignor, Robert, et al. 2014. Worlds Together, Worlds Apart. W. W. Norton & Co. Tōchika Shin. 2009. “Chūsei Kongōfukuji no kanjin katsudō to Hata no shō.” In Bukkyō Daigaku daigakuin kiyō 37, ed. Bungaku Kenkyūka, 55–72. Toda Yoshimi. 1956. “Kuroda no shō ni okeru jiryō to shōmin—shōensei no hikodaiteki seikaku ni tsuite.” Nihonshi kenkyū 30: 31–58. ———. 1967 [1959]. “Chūsei shoki nōgyō no ichi tokushitsu.” In Nihon ryōshusei seiritsushi no kenkyū, 168–191. Iwanami Shoten. ———. 1967. Nihon ryōshusei seiritsushi no kenkyū. Iwanami Shoten. ———. 1979. Chūyūki. Soshiete. ———. 1991. Shoki chūsei shakaishi no kenkyū. Tōkyō Daigaku Shuppankai. ———. 2006. “Kyoto and the Estate System in the Heian Period.” Translated by Janet R. Goodwin. In Capital and Countryside in Japan, 300–1180: Japanese Historians Interpreted in English, ed. Joan R. Piggott, 245–279. Cornell East Asia Series. Togawa Tomoru. 2013. “Kodai shōen, shoki shōen.” In Shōenshi kenkyū handobukku, ed. Shōenshi Kenkyūkai, 1–30. Tōkyōdō Shuppan. Bibliography 507

Tōkyō Daigaku Shiryō Hensanjo. 1988–2003. Nihon shōen ezu shūei. 7 vols. Tōkyō Daigaku Shuppankai. ———. 2007. Nihon shōen ezu shūei, Shakumon, vol. 1. Tōkyō Daigaku Shuppankai. Tokyo National Museum. 2000. Visions of the Pure Land: The Treasures of Byōdōin Temple. Asahi Shinbunsha. Tonomura, Hitomi. 1992. Community and Commerce in Late Medieval Japan: The Corporate Villages of Tokuchin-ho. Stanford University Press. Torao Toshiya. 1993. “Nara Economic and Social Institutions.” Translated by William Wayne Farris. In Cambridge History of Japan, vol. 1 (Ancient Japan), ed. Delmer M. Brown, 415–452. Cambridge University Press. Toyoda Takeshi. 1977. Kōyasanryō shōen no shihai to kōzō. Gannandō. Tsuda Nobukatsu. 1972. “Heian chūki ni okeru nōmin tōsō, Iga no kuni no Kuroda no shō no Tengi jiken wo megutte.” Nihonshi kenkyū 125: 20–39. Uejima Tamotsu. 1989. “Shōen monjo.” In Kōza Nihon shōenshi 1, ed. Amino Yoshihiko, Ishii Susumu, Inagaki Yasuhiko, and Nagahara Keiji, 89–147. Iwanami Shoten. Uejima Tamotsu, Ōyama Kyōhei, and Kurogawa Naonori, eds. 2015. Tōji hyakugō monjo wo yomu. 2nd ed. Shibunkaku Shuppan. Uesugi Kazuhiko. 1996. Nihon chūsei hō taikei seiritsushiron. Azekura Shobō. Ujita Kazuo. 1991. “Kawachi no kuni ・ Kusu no ha no maki ni okeru doki seisan no tenkai.” Hisutoria 133: 86–104. Umekawa Mitsutaka. 1997. “Shiryō ni mieru kokushoku no doki.” Ritsumei- kan Daigaku kōkogaku ronshū 1: 411–421. Uno Takako. 1984. “Kōhanki no sueki: Heiankyō・Kyōto shutsudohin ni miru chūsei yōsō no keisei.” Shirin 67, no. 6: 66–104. Uno Takao. 2001. Shōen no kōkogaku. Aoki Shoten. Usui Nobuyoshi. 1986. “Chisei no kotai to teishin shoryō no henten: Yamashina-ke no keisō.” Nihon rekishi 460: 45–54. Van de Veere, Hendrik. 2000. Kōgyō Daishi Kakuban. Hotei Publishing. Varley, H. Paul. 1966. The Ōnin War: History of Its Origins and Background with a Selective Translation of The Chronicle of Ōnin. Columbia University Press. Veccia, Susan H. 2004. Uncovering Our History: Teaching with Primary Sources. American Library Association. von Verschuer, Charlotte. 2006. Across the Perilous Sea: Japanese Trade with China and Korea from the Seventh to the Sixteenth Centuries. Translated by Kristen Lee Hunter. Cornell East Asia Series. ———. 2007a. “Life of Commoners in the Provinces: The Owari no gebumi of 988.” In Heian Japan: Centers and Peripheries, ed. Mikael Adolph- son, Edward Kamens, and Stacie Matsumoto, 305–328. University of Hawai‘i Press. 508 Bibliography

———. 2007b. Le Riz dans la Culture de Heian, Mythe et Reality. College de France. English translation by Wendy Cobcroft, 2016: Rice, Agriculture, and the Food Supply in Premodern Japan. Routledge. ———. 2009. “Demographic Estimates and the Issue of Staple Food in Early Japan.” Monumenta Nipponica 64, no. 2: 337–362. Wakita Haruko. 1969. Nihon chūsei shōgyō hattatsushi no kenkyū. Ochanomizu Shobō. Republished 1977. ———. 1981. “Chūseishi kenkyū to toshiron.” In Nihon chūsei toshiron, 1–68. Tōkyō Daigaku Shuppankai. Wakita Osamu. 1963. Kinsei hōken shakai no keizai kōzō. Ochanomizu Shobō. Watanabe Hirochika. 2013. “Akutō kenkyūshi wo megutte: Kenkyūshi to shiryō.” In Chūsei shōen no kiso, ed. Akutō Kenkyūkai, 209–220. Iwata Shoin. Watanabe Makoto. 2010. “Jūniseiki no Nissō bōeki to sanmon, Hachiman, In no miumaya.” In Tsuwamonotachi no seikatsu bunka, ed. Irumada Nobuo, 228–252. Koshi Shoin. Watanabe Sumio. 1970. Kinai shōen no kiso kōzō, vol. 2. Yoshikawa Kōbunkan. Waterhouse, David. 1996. “Notes on the kuji.” In Religion in Japan: Arrows to Heaven and Earth, ed. P. F. Kornicki and I. J. McMullen, 1–38. Cam- bridge University Press. Weber, Max. 1978. Economy and Society. 2 vols. Ed. Guenther Roth and Claus Wittich. University of California Press. Winterstein, Prescott B. 1974. “The Muromachi Shugo and Hanzei.” In Medieval Japan, ed. John W. Hall and Jeffrey P. Mass, 210–220. Yale University Press. Yabe Yoshiaki. 1989. “Chūsei tōki.” In Kōza Nihon shōenshi, vol. 1, ed. Amino Yoshihiko, Ishii Susumu, Inagaki Yasuhiko, and Nagahara Keiji, 282– 311. Iwanami Shoten. Yagi Atsuru, ed. 1998. Zusetsu Yamaguchi-ken no rekishi. Kawade Shobō Shinsha. Yamada Akiko. 2003. “Tennō junbo naishinnō ni kansuru ichikōsatsu.” Nihonshi kenkyū 491: 56–72. ———. 2005. “Shijō-no-miya ikō no sekkanke josei ingū wo meguru kōsatsu: Kōkamon-in wo jirei toshite.” Bunka keishōgaku ronshū 2: 81–89. ———. 2010. Chūsei zenki josei ingū no kenkyū. Shibunkaku Shuppan. Yamaguchi Ryūsuke. 2010. “Kaikei no kiseki to zōzō.” Bessatsu Taiyō 176: 96–111. Yamakage Kazuo. 1977. “ ‘Akutō ’ ni kansuru kisoteki kōsatsu.” Nihonshi kenkyū 178: 1–41. ———. 1988. “Kongōbuji shuto to sono seika.” In Chūsei jiinshi no kenkyū, vol. 1, ed. Chūsei Jiinshi Kenkyūkai, 311–342. Hozōkan. Bibliography 509

———. 1997. Chūsei Kōyasan no kenkyū. Seibundō. ———. 2002. “Kōno-Makuni no shō.” In Kinokuni shōen no sekai, vol. 2, 171– 198. Seibundō. Yamamoto Tsutomu. 2002. “Matsunaga Jian kyūzō no butsuji—Kaikei no ichishiryō to shite.” Museum 576: 39–44. Yamamoto Yasukazu. 1990. “Shinshutsu no tōhon Amida sanzon raigōzu ni tsuite.” Kinko sōsho 17: 343–362. Yamamura, Kozo. 1981. “Tara in Transition: A Study of a Kamakura Shōen.” Journal of Japanese Studies 7, no. 2: 349–391. ———. 1988. “From Coins to Rice: Hypotheses on the Kandaka and Koku- daka Systems.” Journal of Japanese Studies 14, no. 2: 341–367. ———, ed. 1990. Cambridge History of Japan, vol. 3 (Medieval Japan). Cam- bridge University Press. Yamatoya Ikumi. 2006. “ ‘Hyōgo Kitaseki irifune nōchō’ kara miru bukka no hendō.” Hokudai shigaku 46: 1–23. Yamauchi Harutomo. 2012. “Eikyō Kakitsuki no Iyo Saionji-shi no kaku shitsu to bakufu kenryoku.” Chihōshi kenkyū 62, no. 2: 23–38. Yamauchi Shinji. 2003. Nara Heianki no Nihon to Ajia. Yoshikawa Kōbunkan. ———. 2009. Nisso bōeki to “Iō no michi.” Yamakawa Shoten. Yamauchi Yuzuru. 1985. Yugeshima no shō no rekishi. Yuge-chō. Yamazaki Takayuki. 1982. “Amida sanzonzō no kōsatsu ni tsuite.” Tōkyō Gei- jutsu Daigaku Bijutsu Gakubu kiyō 17: 11–45. ———. 1994. “Kyodai butsuzō no kōzō ni tsuite.” Nanto bukkyō 70: 64–80. Yasuda Motohisa. 1977. Nihonshi shōhyakka: Shōen. Kondō Shuppansha. Yiengpruksawan, Mimi. 1995. “The Phoenix Hall at Uji.” The Art Bulletin 77, no. 4: 647–672. Yokouchi Hiroto. 2008. Nihon chūsei no bukkyō to higashi Ajia. Hanawa Shobō. Yoshida Akira. 1980. Nihon kodai sonrakushi josetsu. Hanawa Shobō. ———, ed. 1998. Nihon kodai no kokka to sonraku. Hanawa Shobō. Yoshida Takashi. 1983. Ritsuryō kokka to kodai no shakai. Iwanami Shoten. Yoshida Yasuo. 1982. “Koremune Takatoki den kō.” Nihonshi kenkyū 20: 22–58. Yoshie Akio. 1967. “Sekkanke ryō sōzoku no kenkyū josetsu.” Shigaku zasshi 76, no. 4: 1–45. ———. 1970. “Shoki chūsei sonraku.” In Kōza Nihonshi, vol. 2, ed. Reki shigaku Kenkyūkai/Nihonshi Kenkyūkai, 105–130. Tōkyō Daigaku Shuppankai. ———. 1974. “Hō no keisei to sono tokushitsu.” Hokkaidō Daigaku Bun- gakubu kiyō 22, no. 1: 121–250. Yoshioka Yasunobu. 1989. Nihon kaiiki no toki・tōji, chūsei hen. Jinrui- shi Shōsho. Rokkō Shuppan. 510 Bibliography

———. 1990. “Suzuyaki kara Echizenyaki e.” In Nihonkai to kitakuni bunka, Umi to rettō bunka, ed. Amino Yoshihiko, 289–317. Shōgakukan. ———. 1994. Chūsei sueki no kenkyū. Yoshikawa Kōbunkan. ———. 2010. “Nihonkai no chinbotsusen wa kataru.” In Suzuyaki tanjō! Shiseki “Suzu tōki kama ato” kuni shitei kinen shinpojiumu hōkokusho, ed. Suzu-shi Kyōiku Iinkai, 22–42. Suzu-shi and Suzu-shi Kyōiku Iinkai. Zauhō Kankōkai and Narasaki Shōichi, ed. 1977. Sekai tōji zenshū, vol. 3. Shōgakukan. Contributors

Kristina Buhrman is an assistant professor in the Department of Religion at Florida State University, where she teaches Buddhism and Japanese religion. Her research focus is on the social history of knowledge in early medieval Japan. She has published on the history of science in premodern Japan (“History of Knowledge and Craft,” 2017), on the importation and transformation of astrology in Heian Japan (“Ptolemy and Sima Qian in Elev- enth-Century Japan: The Combination of Disparate Astrological Systems in Practice,” 2016), and on archives and the creation of the memory of disaster (“Remembering Future Risk: Considering Technologies of the Archive for Discussion of Tōhoku’s Seismological Past After 2011,” forthcoming).

Michelle Damian completed her PhD in premodern Japanese history at the University of Southern California in 2015, and after a year as a Reischauer Institute of Japanese Studies (Harvard University) postdoctoral fellow, she is now an assistant professor in the History Department of Monmouth Col- lege. She has published articles in Education About Asia and is working on a book manuscript about maritime trade networks in medieval Japan.

David Eason is an associate professor at Kansai Gaidai University in Osaka, Japan, where he teaches a wide assortment of Japanese history classes for both Japanese and foreign exchange students. He recently translated Fujita Hirotsugu’s essay “Geography and History and Geography in His- tory” (2017), and he continues to work on a research monograph centered on the shifting use of emotional rhetoric evident within fifteenth- and sixteenth-century legal texts.

Endō Motoo is an associate professor at the University of Tokyo Historio- graphical Institute. He is the editor and compiler of the document collection Tōdaiji monjo in the Dainihon komonjo series, and he also studies the early medieval court. Recent publications include Chūsei ōken to ōchō girei (2008), “Chūsei Tōdaiji monjo wo fukan suru” (2011), and “Go-Sanjō ・ Shirakawa In no nenjū gyōji” (2015).

511 512 Contributors

Philip Garrett is lecturer in Japanese history at Newcastle University. His primary interests are the institutional history of Japanese temple complexes, medieval crime and law, and tsunami history. Recent publications include “Crime on the Estates: Justice and Politics in the Kōyasan Domain” (2015) and “A Systematic Review of Geological Evidence for Holocene Earth- quakes and Tsunamis along the Nankai-Suruga Trough, Japan” (2016).

Janet R. Goodwin was a founding faculty member of the University of Aizu in Aizu-Wakamatsu, Japan. Now retired, she is a research associate of the East Asian Study Center at the University of Southern California. Publications include Selling Songs and Smiles: The Sex Trade in Heian and Kamakura Japan (2007) and “Memories and Strategic Silence in Jōdoji engi,” with Kevin Wilson (2015). An example of her most recent research is “Outcasts and Marginals in Medieval Japan: A Historiographical Over- view” (2017).

Hirota Kōji recently accepted a position on the staff of the Shizuoka City Cultural Promotion Foundation, where he will take charge of its museum. Formerly he was on the staff of the Izumisano City Education Committee in Ōsaka Prefecture. His research concerns the history of medieval communities. Recent publications include “Masamoto-kō tabi hikitsuke no nikki shiryōgaku,” Nihonshi kenkyū 48 (2013) and “Zaichi ryōshu shihai shita ni okeru chūsei sonraku—Higo no kuni Kamikura no shō no sonraku,” in Chūsei sonraku to chiiki shakai, (2016). In 2014 he curated an exhibit at the Wakayama City Museum titled “Shōen no keikan to ezu.”

Yoshiko Kainuma is an independent art historian who received her PhD from the University of California, Los Angeles. Her specialty is the Bud- dhist art of medieval Japan, particularly the relationship between works of art and social functions. Among her published articles is “Chōgen’s Jōdoji Amida Triad and Its Environment: A Theatrical Effect of the Raigō Form” (2014), and she is working on a book on the Buddhist sculptor Kaikei and his statues of Amida.

Rieko Kamei-Dyche is an assistant professor in the Faculty of Arts at Risshō University in Tokyo, where she teaches courses on Japanese history and culture. Her work often employs historical readings of literary sources, an issue she investigated in “Sailing Between Two Seas: A Discussion of Recent Japanese Writing on the Integration of Literature and History” (2011), among other studies. She is currently working on a study of courtier family capital aggrandizement strategies and a translation of Godai teiō monogatari. Contributors 513

Sachiko Kawai received her PhD in premodern Japanese history from the University of Southern California in 2015, and subsequently was a postdoctoral fellow at the Harvard College and then at the Reischauer Institute of Japanese Studies, also at Harvard University. Her publications include “The Land-Based Power of Nyoin in Early Medieval Japan” (2014), “Life-Cycle Rituals for Newborns in Heian Japan” (2018), and “Talking to a Deity: The Royal Lady Hachijō-in at Prayer” (2016). She is currently working on a book manuscript titled Uncertain Powers: Female Royals as Householders and Land- lords in Early Medieval Japan, 1100–1300.

Kimura Shigemitsu is professor emeritus at Tokyo Geijutsu University, a member of the Japan Academic Conference Association, and chair of the Japanese Historical Association Committee. His research focuses on the history of agriculture in medieval Japan and the society of the Japanese estate system. Recent publications include Nihon nōgyōshi (editor) (2010), Shoki Kamakura seiken no seijishi (2011), “Chūsei shōen no kōzō” (2013), and Nihon chūsei hyakushō seiritsu shiron (2014).

Nagamura Makoto is professor emeritus at Nihon Joshi Daigaku and head of the Kanagawa Prefectural Kanazawa Bunko library. His research concerns the history of medieval Japanese temples, the classification and functions of Buddhist texts, and perspectives on the Buddhist law in medieval monastic society. His 1989 book Chūsei Tōdaiji no soshiki to keiei is a seminal work in the study of the great Nara temple. Recent publications include “Kamakura no Kegon” (2016), “Sanbōin Kenshun to Takauji” (2016), and “Heian jidai ni okeru Tōdaiji no kyōgaku to hōe” (2017).

Nishida Takeshi graduated from Nara University in 1980 with a specialty in archaeology. While with the Education Committee of Ono City, Hyōgo Prefecture, he led the section called “Creating a Lively Society” for the Ono City Board of Education and was the assistant head of the Ono City Kokōkan Museum. His research concerns history as seen through archaeology, Jōdoji and Ōbe Estate, medieval castles in the Kako River basin, best practices for the preservation and use of regional cultural heritage, and using cultural artifacts to make a living city. Recent publications include “Chōgen shōnin to Jōdoji” (2011), Kanatsurubejō ato iseki hakkutsu chōsa hōkokusho (2015), and Ōshima Ōtsubo iseki hakkutsu chōsa hōkokusho (2016).

Noda Taizō is a professor at Tachibana University in Kyoto. His research focuses on the political history of the Kinai region of Japan in Muroma- chi and Sengoku times. He also studies documents of the Kamo Wakei- kazuchi Shrine from the same period, and from 2009 to 2012 he was a representative of a group that received a government grant-in-aid for 514 Contributors scientific research on that project. Recent publications include Sengokuki Miyoshishi kenryoku no kisoteki kenkyū (a report on research funded by a grant-in-aid) (2005), “Chūsei shōen to kinsei ezu” (2010), and “Sen- gokuki no Koderashi” (2014). He is also a coauthor of Himeji-shi shi, vol. 9 (2012).

Ōyama Kyōhei is professor emeritus at the University of Kyoto. He was a professor in the University’s Faculty of Literature, was the head of the Fac- ulty of Literature Museum, and was a professor at Ōtani University. He spe- cializes in the history of medieval Japan. Major works include Nihon chūsei no mura to kamigami (2012), Yuruyakana caasuto shakai ・ chūsei Nihon (2003), and Nihon chūsei nōsonshi no kenkyū (1978). In English, see “The Fourteenth Century in Twentieth-Century Perspective” (1997) and “Medieval Shōen” in The Cambridge History of Japan, vol. 3 (1990).

Joan Piggott is Gordon L. MacDonald Professor of History and Director of the Project for Premodern Japan Studies at the University of South- ern California in Los Angeles. She is the editor of Capital and Country- side in Japan 300–1180 (2006), Teishinkoki, The Year 939 in the Journal of Regent Fujiwara no Tadahira (with Sanae Yoshida) (2008), and Birth and Death in the Royal House, Selections from Fujiwara Munetada’s Chūyūki (with Christina Laffin and Yoshida Sanae) (2018). Most recently she has published “Heian Kyō, from Royal Center to Metropole” in The Routledge Handbook of Premodern Japanese History (2017). She is currently completing an annotated translation and research concerning urban developments in the Kyoto capital described in Fujiwara Akihira’s New Monkey Music (Shinsarugakuki).

Sakurai Eiji is a professor in the faculty of comprehensive cultural research of the graduate school of the University of Tokyo. His research focuses on the history of medieval Japan, especially commercial economic history and the gift economy. His publications include Muromachi hito no seishin (2009), and Zōyo no rekishigaku (2011). He is an editor of the 2013 Iwanami kōza Nihon no rekishi series, to which he contributed two articles, “Chūseishi e no shōtai” and “Chūsei no gijutsu to rōdō.”

Ethan Segal is associate professor of history and chair of the Japan Council at Michigan State University. He is the author of Coins, Trade, and the State: Economic Growth in Early Medieval Japan, as well as articles and book chapters on topics including premodern economic thought, proto-nationalism, and the textbook controversy in history education. He has received numerous awards for his teaching and, in 2015–2016, served as president of the Mid- west Conference on Asian Affairs. Contributors 515

Dan Sherer received his doctorate in Japanese history from the University of Southern California in 2017. His research focuses on sixteenth-century political and religious institutions. His current work concerns the unification of the Nichiren sect in Kyoto and the sect’s negotiation of the violent era from the rise of the Lotus Leagues to the crises of the early Edo period.

Index

Numbers in bold refer to figures, tables, and plates. agriculture: expansion of, 150–153, also Kobayakawa warrior family; 154–156, 187n.3; on Kuroda Nuta Estate Estate, 164–165; on Ōbe Estate, akutō (“evil bands”): dress of, 407, 415, 156–161. See also dry field 422n.25, 425n.67; identification agriculture; land opening, land of, 404–405, 406, 422n.18; as ikki, reclamation; rice production; 408, 423n.31; in the Mineaiki, wet field agriculture 406–408, 415, 419; Nagahara Akamatsu family: Akamatsu Mitsu Keiji on, 13, 405; as precursors of masa, 432, 433; Akamatsu ikki, 408, 423n.31; residences of, Norihisa appointed as governor 416, 416 of Harima, 435–436; assassina- —Ōbe Estate cases: first case (1295), tion of Ashikaga Yoshinori, 25, 411–416, 420–421; second case 374n.37, 431–432; conscription (1322), 416–419, 420–421; table of akutō involved in, in Harima Province established 417 —shogunate laws and edicts concern- by, 428; fortress at Kinoyama, ing, 422n.17, 426n.80; akutō 428; incorporation of the Ogawa used as a legal term, 412, 413, family into their administration, 419–421 429; management of Iwa Shrine, Amino Yoshihiko: on the estate 429; replaced by Yamana as strategy of the Saionji, 322, 325, governors of Harima, 432–435; 339–340; on the range of liveli- support for Tōdaiji’s recovery of hoods supported by estates, xxv, control over Ōbe Estate, 95, 102 xxx, 17, 385; on rent payments, Akanabe Estate, 30–31n.29, 186, 42–44 190–191n.47, 192n.68; silk pro- Arai Takashige, 191n.54, 244, 407 duction, 44 Arakawa Estate in Kii Province: attacks Akashi: as a fortification against akutō from Tanaka Estate and Yoshinaka in the Mineaiki, 407, 417; as one Estate, 306–308; challenges from of Akamatsu Mitsumasa’s hold- Taira followers, 293; dispossession ings, 433 and exile of its reeve, 383; Hachijō- Aki Province: opening of land desig- in’s management of, 305–308 nated as a hereditary township Asakawa, Kan’ichi, 9, 25n.83, in, 8; warlord families of, 15. See 29–30n.18, 138, 448–449,

517 518 Index

450, 453, 455–456n.28, in the development of, 45–46; 456nn.30–31 shift from local salt to cash at Ashikaga Yoshinori, assassination of, Yugeshima, 355 25, 374n.37, 431–432 Chen He-qing: and Ōbe Estate, 203, azukaridokoro. See custodians 235; restoration of the head of the Great Buddha at Tōdaiji, Banse Akemi, 284, 305 260; and Tōdaiji, 89, 235, bessho (autonomous temples): ambigu- 248–249 ous status of, 237; and hijiri, 209 Chikuzen Province: Kanzeonji in, 6; —Chōgen’s establishment of, 204, Munakata Shrine and Munakata 223–227, 272, 273n.1; examples Estate in, 332, 337, 338, 340, of (see Jōdoji Temple); five tiered 349n.61 “triangular” stupas erected Chōgen, Plate 1, Plate 17; arrival at at, 218, 229n.23; Iga Bessho, Ōbe Estate, 73; autobiography of 276n.41; images of Amida (see Namuamidabutsu sazenshū); enshrined at, 237, 258; Jōhan construction of Jōdoji (see under as heir to them, 244, 269; Kōya Jōdoji Temple); his legend- Shin, 250n.23, 257, 259; Wata- ary creation of Nigori Pond, nabe Bessho, 213, 221, 222, 75–76; as a kanjin hijiri, 198, 224–225, 237, 258, 259, 276n.49 202–203, 205, 270; management Bifukumon-in: commendation of of Ōbe Estate, 203–206, 231, Arakawa Estate to Kongōbuji, 234–237, 256–267; reconstruc- 306–307; as provincial proprietor tions at Sayama (see Sayama of Echizen, 307 Pond); reconstructions at Bingo Province: estates of (see Ōta Uozumi Anchorage (see Uozumi Estate in Bingo); salt shipments Anchorage); Shingon beliefs of, from, 358–359, 361, 363; shugo 218–219; Tōdaiji rebuilding by, (constable or military governor) xxix, 75, 87, 198, 202–203, 212, of, 15, 372 232, 253. See also Ōbe Estate— Bourdieu, Pierre, 346n.4 management by Chōgen Byōdōin temple in Uji, instituted prop- —Amidism of: “amidabutsu” appella- erties donated to support it, 23 tions associated with, 241, 244, —Phoenix Hall (Hōōdō) of: the 246, 271; and the concept of design of the Jōdoji Jōdodō com- kechien, 214–215; and the design pared with, 261–263, 275n.28; of the Jōdoji Jōdodō, 263; and as the manifestation of Amida’s the design of Jōdoji’s Amida Western Paradise on earth, 261; triad, 268, 270, 272; establish- mudra of Amida at, 266 ment of autonomous temples (see under bessho); and his cash commutation, 41–49; and the reconstruction of Sayama Pond, autarky of the estate system, 10, 219; public works projects, 212, 17; and Chinese money, 40–42, 226–228; “the secular” and “the 50–51, 55n.6; and the growth of sacred” not viewed as totally commercial crops, 48; and new different spheres by, xxxiv, 228, kinds of products, 48–49; pay- 272; and Shingon esotericism, ments for shipments from ports 218, 257 at Onomichi and Setoda, 359; commendation of estates: and aggres- role of commodity exchange sive tax collection in the late Index 519

eleventh century, 6; of Ara- dōshū (service monks): and Kakunin’s kawa Estate to Kongōbuji by takeover of Yanase, 181; and Bifukumon-in, 306–307; and the Tōdaiji’s organization, 199–200, expansion of the estate system, 200 8, 22; Ikadachi Estate’s com- dry field agriculture: and agricul- mendation to Mudōji, 313n.27; tural productivity, 154, 158; instituted estates compared irrigation ponds and channels with, 35–36n.87; institution of for, 154–156; management by Ōbe Estate distinguished from, cultivators, 149–150; opening of xxxi; Kanokogi Estate in Higo fields in Hisatomi Township by Province, 6, 8–9; of Ōta Estate Hata no Tametoki, 151–152, 153, to Kōyasan by Go-Shirakawa, 12; 154–155; and “paddy-centric” his- transformation of commended toriography, xxxiii, 143, 145, 148 fields into temple estates at Kanzeonji, 6 Ebisawa Tadashi, 44 commercial crops: lack of written Echizen Province: Asakura warlords documents regarding, 49–50; in, 15; Chimori Estate in, 21; and the spread of cash commuta- Ippon Royal Grant Fields in, 307; tion, 48. See also rice shipping Kuwabara Estate in, 5; labor and custodians (azukaridokoro): on Ara- production in, 144; Ushigahara kawa Estate, 306; definition of, Estate in, 24, 29–30n.18, 35n.83 xxx, 315n.70, 443; duties of, 92, Enkyōji Temple on Mount Shosha, 293, 302, 315n.70; Fujiwara no 416, 429 Teika as, 308; on Hine Estate, estate managers. See geshi, gesu 107, 111; income of, 284; on estates (shōen): and the broader socio- Kuroda Estate, 192n.66; on political order, xxx, 283; of noble Kuroda New Estate, 200–201; on and religious overlords as epi- Munakata Estate, 337; on Nate centers of unrest, 430–431; and Estate, 381; on Nuta Estate, shared rights to income from 348n.41; on Ōbe Estate, 256, the land (shiki), xxvii, 283–284; 428; on Ōta Estate, 8, 284, 354; social and economic roles, xxv, Ōuchi family as, 369, 370, 371; xxx; spatial power of, 323–324; as rent collectors, 32n.50; tem- ties between social levels created ple monks as, 199; on Wakayama by, xxv, xxx, 35n.84. See also com- Estate, 302, 308 mendation of estates; instituted estates; supreme proprietors Daidenbōin, 397, 402n.67 —early estates (shoki shōen): Chimori daikan: as agents or proxies, 43, 89, Estate in Echizen, 21; failure of, 100, 409, 417, 428; as contrac- 5, 21, 34n.72; institution of, 20, tors, 15, 25, 32n.50, 95, 100, 113 21; structure of, xxxviin.2 documentary records: lack of written —management of. See proprietors records regarding commercial (ryōke); supreme proprietors crops, 49–50; pronouncements (honke). See also commendation concerning disputes, 441; regard- of estates; instituted estates ing Ōbe Estate, 90–92, 93, 96; —and the official land system (shōen- Tōdaiji monjo collection, 81, 101, kōryōsei ), xxviii, 8 412, 425n.76; types and features —as a template for merchant and of, 81–84 artisan guilds, xxx 520 Index

Fujiwara Yorimichi: on the manage- Go-Shirakawa Tennō: commenda- ment of estates, 22; as the patron tion of Ōta Estate to Kōyasan, of Byōdōin temple in Uji, 23, 12; favorable treatment of 262; and the reign of Go-Reizei Tōdaiji, 181–184; house arrest Tennō, 191n.55 by Kiyomori, 308; Kuroda Estate Fukuda Township (ho), 416, 435 expansion recognized by, 186; Funaki Estate: location of, 286, 289, mother of (see Taikenmon-in); 290, 312n.22; shipbuilding asso- Ōbe Estate income donated ciated with, 290; Upper Kamo to Chen He-qing, 203, 235; Shrine, 291, 313n.30 orders related to Upper Kamo Shrine, 313n.30; as proprietor of gakuryo (scholar monks): Shōgen, 200– Yugeshima Estate, 353–354; sister 201; and Tōdaiji’s Kuroda cul- of (see Jōsaimon-in); as supreme tivators, 182, 183; and Tōdaiji’s proprietor of Ōta Estate, 8, organization, 199–200, 200 354; victory of his faction in the Genpei civil war: allegiance of the Hōgen coup, 181 Matsuura League during, Go-Toba Tennō, 248, 276n.49, 331; 349n.75; fundamental causes of, consort Shumeimon-in, 337; and 24–25, 197; lawsuits related to, the dismissal of Chen He-qing, 390; reconstructions by Chōgen 89; Jōdodō designated as a royal following, 75, 87, 212, 232 prayer chapel, 238; and the geshi, gesu (estate managers), 13, 107, management of Tōnan’in, 89; 175, 254, 381; the Nakahara as, Minase Palace of, 350n.90; and 111, 139n.6; Narisuke’s illegal Ōbe Estate, 89; and the victory holding of the title of, 306. See of Kamakura forces in the Jōkyū also shōkan War (1221), 10 gō (townships) and ho (new town- Goffman, Erving, 323 ships): and the expansion of the Gomi Fumihiko, 19, 284, 305, 316n.84, estate system, 8, 153–156; hitosato 329, 347n.30 or murazato as proper terms for medieval gō, 221; individual Hachijō-in: and Arakawa Estate, examples of (see Fukuda Town- 306–308; donation of an estate ship; Hisatomi New Township; to support nenbutsu practitioners, Koinumaru Township; Suchi 225; estates of, 295, 298 (see also Township) Arakawa Estate in Kii Province); Go-Daigo Tennō: court-in-exile at maritime trade networks of, 75, Yoshino, xx, 95, 108; estates 298–300; and Prince Mochihito’s donated to Tōji, 17; Kenmu rebellion, 281, 312n.4; properties Restoration attempted by, xx, in the capital of, 295, 296 xxxv, 13; and landholdings of the Hachiman shrines (Hachimangū): Saionji family, 326–327 Iwashimizu Hachiman Shrine, Go-Fushimi Tennō, edict pertaining to 48, 379, 380, 397, 400n.15, treason, 418 402n.66; Jōdoji Hachiman Go-Murakami Tennō, 396 Shrine, 77; Tōdaiji Hachiman Go-Saga Tennō: ivory baton presented Shrine (see under Tōdaiji); Yusu- to, 333; ox presented to, 343 hara Hachiman Shrine, 339 Go-Sanjō Tennō, failed effort to stop Hall, John Whitney, 450–451, 456n.31, the proliferation of estates, 446 456n.33 Index 521 hanzei (half-tax) system, 13–14, 25, 114, —Iriyamada Village: and Hine Estate 352–353, 441–442 village structure, 133–134; Harima Province, 416; estates of (see structural changes during the Hirano Estate; Ōbe Estate; Yano Muromachi and Warring States Estate); military governors of periods, 109–110 (see Akamatsu family; Hoso- —Tsuchimaru Village: archaeological kawa family; Yamana family); sites in, 115–116, 119, 123–125, Tendai complex establishments 127–128, 137; community of, on, 416, 429. See also Uozumi 109, 136; location of, 126, 127, Anchorage 130–131, Plate 12; mountain Harrington, Lorraine, 405, 426n.80 fortifications, 108, 112, 119, 124, Haruta Naoki, 144 126, 131; rice paddy landscape Hashimoto Masataka, 108, 112, 119, of, 132–133, Plate 12 124 Hirano Estate, 417, 418, 419–420, Hashimoto Michinori, 156 425nn.75–76 Higo Province, Kanokogi Estate, 6, 8–9 Hirata Tenshu, 303 hijiri (unofficial religious practi Hisatomi New Township (gō ), 151–152, tioners): Chōgen as a kanjin 153, 154–155, 191–192n.57 hijiri, 198, 202–203, 205, 270; historiography: on akutō, 405–406; cremation specialists (sanmai and English readers, 28, 36n.89, hijiri), 119; estate management 444–448; Kamakura Saho’s by, xxxiv, 203, 205–206; kanjin overview of current research on hijiri activities, 202, 254, 255 estates, 27–28, 36n.88; “medi- Hikoyoshi Mieko, 312n.4 eval” (chūsei ) applied to Japanese Hine Estate, Plate 8; chronology of, history, 448, 455n.23; “paddy- 111–115; Izumi military gov- centric” views of medieval society ernors and, 16, 108, 109, 113, and economy, xxxiii, 143–145, 114, 122, 130, 132, 134–135, 148; reassessment of the activities 136, 139n.7; Kujō Masamoto’s of yeomen, 385. See also Shōen residence at, 110; Ōgi District of, History Research Group 125–126, 126; self-government of Hōgen coup: abbot Jien on, 194n.97; villages during the Muromachi victory of Go-Shirakawa’s faction period, 133–134; terraced rice in, 181 fields on, 132–133, Plate 11 Hōjōji Temple: Kitayama villa built —archaeological sites: list of, 116–122, by Saionji Kintsune compared 127; as National Historic Sites, with, 347n.40; as the propri- 115–116; types of, 123–125, etor of Yoshinaka Estate, 306; 127–128 Tachibana no Tomonobu’s —Chōfukuji: Kujō Masamoto’s residence repairs of, 331 at, 118, 125, 130; as a National Hosokawa family: Izumi Province mili- Historic Site, 123, 125, 128 tary governors from, 108–109, —Hineno Village, 129, Plate 9, Plate 113, 114; as kanrei (deputy sho- 13; Hine Shrine (Ōizeki Shrine), guns), 360, 434; as military gov- 126–127; Masamoto’s manage- ernors of Awa Province, 372; as ment of, 110; Negoroji’s manage- military governors of Bizen Prov- ment of, 110–111; Yugawa Canal, ince, 371; as military governors 120, 123, 127, 131, 135, 136, 136, of Harima Province, 15, 430, 433; 155, Plate 14 as military governors of Sanuki 522 Index

Province, 361, 368, 369, 370, 372; illegal behavior of his retainer ports affiliated with, 368 Narisuke, 306; Kōfukuji fire set Hurst, G. Cameron, 35n.84 by, 197; Tōdaiji fire set by, 186, hyakushō. See yeomanry 197 Hyōgo checkpoint: registers of goods, Ishida Hisatoyo, 218 46–47, 48, 356–363. See also rice Ishimoda Shō, 3, 164–165, 187 shipping; salt shipping Itabae Soma (Forest), 166, 171, 188n.12 Iga Province: estates of. See Kuroda Iwashimizu Hachimangū: Ōga Estate Estate. See also Nabari District invaded by, 402n.66; Ōyamazaki Ikadachi Estate: and Hiei temples jinin affiliated with, 48; Suda and shrines, 290–291; location Estate of, 23, 379, 397; Tomobu- of, 286, 289, 290, 312n.22; and chi Estate of, 379, 380, 400n.15 Mudōji on Mount Hiei, 290–291, Iyo Province: estates of (see Uwa Estate; 313n.27; shipbuilding associated Yuge Estate; Yugeshima Estate). with, 290 See also Saionji family; Yuge port ikki (leagues): akutō as precursors of, in Iyo Province 408, 423n.31; debt relief move- Izumi Province, estates of. See Hine ments mobilized by, 430–431 Estate indigo, 46, 48, 53, 56n.29, 357; Kujō Indigo Guild, 49, 57n.35 Jigen-in, 120, 123, 127 instituted estates (risshō ): in the era of Jingoji Temple: “Mongaku-i” irrigation retired monarch Shirakawa, 24, system of, 155, 162n.24. See also 35–36n.87; example of Kanakogi Kaseda Estate in Kii Province Estate, 8–9; example of Suda Jōdoji Temple (Harima Bessho): Estate in Kii Province, 23; fiscal agricultural expansion centered and ideological contradiction for on, 157, 160–161; Amida triad court government, 35–36n.87; (see under Kaikei); artifacts from Kawabata Shin on, 35n.80, Kiyotani associated with, 74; 35–36n.87 “bare torso” Amida Nyorai by Inui Naoko, 347n.21 Kaikei, 269, Plate 22; building iron production, 46, 47; and goods site, 67; Chōgen’s founding of shipped on Onomichi boats to it on Ōbe Estate, 213, 236–239, Hyōgo, 367; and rent payments, 254, 256, 258–259; construction 42, 44; skill required for, 43 of, 237, Plate 18; documenta- Ise Taira family: control of Echizen, tion of a dispute involving, 90, 307; dominance at court of, 24, 93; Edo-period illustration of, 186; and the Genpei civil war, 24; Plate 6; and Hachijō-in, 225; Jōsaimon-in’s ties to, 292–293; Kan’amidabutsu’s grave, 66; Minamoto defeat of, 10; Taira no main image from Kōdo ruined Jishi (Kenshunmon-in), 292–293 temple installed in Yakushi Hall —Taira no Kiyomori: clash with forces at, 72–73, 259; masks by Kai- of Minamoto Yoritomo (see Gen- kei and others, 269, Plate 23; pei civil war); daughter Tokushi ruins of subtemples of, 66, 74; (Kenreimon-in), 308; trade con- sculptures at, 258; temple ruins ducted by, 333–334, 349n.60 nearby, 236, Plate 5 —Taira Shigehira: as custodian (azu- —Jōdodō, Plate 18; Chōgen’s estab- karidokoro) of Ōta Estate, 8, 284; lishment of, 204, 237, 273n.12; Index 523

designation as a royal prayer cha- denced by inscriptions, 270–271; pel, 238; as example of daibutsuyō as a raigō Amida triad, 264, 266, style, 260; interior design, 259– 268, 269, 270, 272, 276n.44; 261; and the religious needs of raigō triad at Kōdaiin compared local estate residents, 205–206, with, 266, 267, 268; Saihōji 260 Amida triad compared with, 268, —organization: and the management 276n.44, Plate 21; visual effects of Tōdaiji’s estates, 202–203; of, 239, 263–264 monks’ cottage ruins, 67, 74, Kamakura judicial system: akutō related Plate 6 cases on Ōbe Estate (see under Jōhan (Gan’amidabutsu): as Chōgen’s akutō); and disorder caused heir, 203–205, 225, 229n.5, 236, by the Genpei civil war, 390; 244, 259; and the Shōbō lineage overview of how cases advanced of Daigoji, 213, 236 through the system, 410–411 Jōsaimon-in: as a Kamo priestess, 291; —monchū (judicial hearings), 409, officials in service to, 314n.38; 410–411, 423n.38; and the parents (see also Taikenmon-in; Nate-Niunoya dispute, 389–390, Toba Tennō/In) 400n.30 —estates of: locations of, 287, 288, 289; Kamakura Saho, 22, 27–28, 36n.88 in Ōmi Province, 289, 290; strate- Kan’amidabutsu (Kan’a, Kan Amida- gic access to trade and trans- butsu): amidabutsu suffix of his portation circuits of, 286–287, name, 244, 271; and Chōgen’s 292–295; in Yamashiro Province, development of Ōbe Estate, 233, 288, 290–291. See also Funaki 236, 238–239, 256; grave of, 66 Estate; Ikadachi Estate Kan’ei tsūho coins, 40, 53–54 judicial system: akutō -related cases on Karikome Hitoshi, 237–238, 247, 271 Ōbe Estate (see under akutō ); Kaseda Estate in Kii Province: attack documentation of pronounce- on Kokawadera, 396; disputes ments concerning disputes, 441; with Shizukawa Estate, 380, 397; tōchigyō (long-term land man- Jingoji as its proprietor, 155; agement claims), 419. See also location of, 388, 398 Kamakura judicial system Katsumata Shizuo, 49–50 Kawabata Shin, 34n.74, 35n.80, Kaikei: amidabutsu suffix of, 244; 35–36n.87, 156, 234, 238, 239 Amida triad at Shin-Daibutsuji, Kawachi Province: Amida Nyorai at 263, 268, Plate 20; “bare torso” Kannonji, 254; Jōsaimon-in’s Amida Nyorai at Jōdoji by, 254, estates in, 292; metal cast- 259, 269, Plate 22; masks at ers from, 216, 217, 224, 242; Jōdoji Temple, 269, Plate 23 Saionji family landhholdings —Amida triad at Jōdoji Temple, 237, in, 326; and Sayama Pond, 215, 258, 265, 268, 273–274n.14, 216, 242 Plate 16; Amida triad by Unkei at Keirstead, Thomas E., 448, 451–452, Jōrakuji compared with, 266; and 453, 454n.1, 455n.23, 456n.37 Chōgen’s Amidist beliefs, 268, kenmon (“gates of power”) theory of 270, 272; Goryeo-era Korean Kuroda Toshio, xxvii, 18–19, Amida iconography compared 33n.63, 322, 378, 457 with, 268, 276n.45; iconography Kenshunmon-in, as Taira no Kiyomo- of, 269, 272; local support evi- ri’s sister-in-law, 292–293 524 Index

Kii Province: map of, Plate 29; control over, 394, 401n.57. See partitioning of Minabe Estate, also Uno Minamoto 401n.49; shugo (military gover- Kongōfukuji, sacred and secular roles nors) of, 377, 391; Suda Estate of, 255 in, 23, 397; upper Kinokawa Kōyasan. See Mount Kōya valley of, 398; yeoman activity Kuboki, Kuboki Village: grave complex in, 385. See also Jingoji Temple; on the middle terrace of Ōbe Kaseda Estate in Kii Province; Estate, 64; location of, 416; Nate-Niunoya dispute Shirai Hachirō (land steward of), Kiley, Cornelius J., xxv, 35n.84, 391 416–417, 417 Kimura Shigemitsu, xxxiii, 25, Kudō Keiichi, 34n.66, 180–181 28–29n.5, 165, 174, 187n.3, 190– Kujō house, estates of. See Hine Estate 191n.47, 449; on Ikeda Estate in —Masamoto, 114–115; Hineno and Tōtōmi, 26, 35n.85 Iriyamade delegated to Negoroji, Kiyomori. See under Ise Taira family 109; posthumous name Jigen- Kiyotani archaeological site, location in, 120; residence at Chōfukuji, on Ōbe Estate, , 73–74 65 118, 125, 130; residence at Hine Kobayakawa warrior family, 15; as Estate, 16, 110 military stewards of Nuta Estate, —as one of the one of the five 340, 359 branches of the regents’ (sesshō ) Kōdoji Temple: abandoned temple family, 105 ruins of, 64, 72–73; main image Kujō Indigo Guild, 49, 57n.35 from installed in Yakushi Hall at kumon (reeve, registrar): on Hine Jōdoji by Chōgen, 73, 259; model Estate, 107, 111, 217; on Kuroda of, Plate 5 Estate, 175; on Mount Kōya Koinumaru Township (ho) in Harima holdings, 400n.15, 401n.59; on Province, 149, 154 Kōkamon-in: independence of, 310– Nagataki Estate, 111, 139n.6; on 311; nyoin status of, 297; proper- Nate Estate, 381, 382–383, 384, 395 ties of, 296, 297, 299, 310 Kokawa(dera) Estate, border with Nate —on Ōbe Estate: akutō violations of Estate along the River Minase, fields belonging to, 416, 418, 381, 385, 388, 395 426n.78; documents regard- Kongōbuji of Mount Kōya: annexing of ing, 90, 91, 92, 93, 96; as estate Ōga and Shibuta, 402n.67; and managers, 103n.8, 415; monks Arakawa Estate, 306–307, 308, as, 97; Ō family as, 91, 93, 94, 400n.15; expansion of its estate 101–102, 246, 249, 271, 427, 436; holdings, 379, 399n.2; holdings in a quarrel with Jōdoji monks, of, xxx; participation in provin- 249; Toyofuku family appointed cial disputes, 379–382, 392–393, as, 95, 436; as warriors, 101–102, 395–396, 398–399; Raisen of 428, 436 Renjōin as the administrator of, Kuroda Akinobu, 50–51 400n.17; Renjōin (subtemple Kuroda Estate: commending process of), 394–395. See also Nate Estate of Kanzeonji compared with, 6; in Kii Province; Nate-Niunoya development in Nabari District, dispute; Shizukawa Estate 171–173, 184–185; Ishimoda —and local warrior elite: connections Shō’s seminal work on it, 3, with, 393–395, 398–399; warrior 164–165, 187; kyōdōtai communi- Index 525

ties formed on, 244; location of, in Kii, 382, 385, 387, 389, 390, 167 392; on Ōbe Estate, 427; on Kuroda Hideo, 187n.3, 190–191n.47, Ōyama Estate, 31n.35, 32n.53; 191n.50, 193n.79; on looking at Ōyama Kyōhei on, 31n.33; as old estate maps, 18, 26 rent collectors, 146; and the Kuroda Toshio: on akutō, 405, 407; on Saionji, 340; as silk producers, the decline of courtier authority, 44; and the stabilization of the 346n.6; definition of the estate as estate system, 12, 19, 25, 26; on a stratified structure, 9, 34n.66; Tachibana Estate, 370; textbook kenmon (“gates of power”; “influ- discussions of, 447; and Tōdaiji ence/authority”) theory of, xxvii, reconstruction, 242; on Yanai 18–19, 33n.63, 322, 378, 457 Estate, 371 military stewards’ contracts ( jitō uke- land opening, land reclamation: and sho), 11 agricultural expansion, 150–153; military stewards’ deputies ( jitōdai ): and agricultural technology, 25; at Niunoya in Kii, 382, 385, 387; and Chōgen, 75–76, 212, 213, and Ōbe Estate, 90, 93 221, 222–223, 228, 238–239, Minamiyama grave complex, 72 243, 247, 253, 256, 257, 270, Minamoto no Yoshiharu. See under Uno 271; and classical estates, 21; Minamoto and dry fields, 191n.52, 191n.57; Minase (river), 381, 384, 386–387, 388, on Hine Estate, 107, 109, 112, 389, 390, 391, 395–396 139n.13, 139n.15; and iron tools, Minase Palace: Go-Toba’s fondness for, 191n.49; and irrigation, 154–156; 350n.90; and the Saionji family’s on Kuroda Estate, 168, 173, network of trade and transporta- 175, 183, 185; by local elites, 24, tion, 344 35n.85; methods of, 155; and Mineaiki: akutō in, 403, 406–408, 415, warriors, 19 418, 419; authorship and dating of, 406; fortifications against Mass, Jeffrey P., 187n.5, 451, 453 akutō at Akashi and Nagaishi, Matsubara Estate in Harima Province, 407, 417 340 misshū monks (esoteric monks), at Matsuura League, 339, 349n.75 Tōdaiji, 199, 200 Meigetsuki, 335, 350n.90 Miyoshi Nagahira, as a house manager metal casters in Kawachi Province, 216, for Saionji Kintsune, 330 217, 224, 242 monetary history of premodern Japan, military stewards ( jitō ): and akutō, 37–41. See also cash commutation 416–417, 419; confiscation of Mongols: estate system endangered by, their rights, 13; conflict with 12–13; as kyōto, 405; prayers for proprietors, 26, 93; and control the defeat of, 255; Yuan mon- over estates, 11–12; and criminal etary policies of, 42 matters, 408; cruelty to cultiva- monks: types of (see dōshū; gakuryo; tors, 31n.31; as estate overseers misshū monks; ritsu monks). See on behalf of the Kamakura sho- also hijiri gunate, xxvii, 10; on Hine Estate, Morita Hayato, 24–25, 35n.84 107, 111, 113; Kamakura house- Moriuchi Shuzo, 301–302 men as, 19; Kobayakawa family Mount Hiei—Tendai complex: as, 340, 350n.79, 370; at Niunoya abbot Jien on the Hōgen coup, 526 Index

194n.97; contract deputies from, Nabari District: in ritsuryō times, 32n.50; Mudōji’s relationship 165–166; wet field agriculture with Ikadachi Estate, 290–291, at, 168–171, 174. See also Kuroda 313n.27 Estate; Yagawa; Yagawa Estate Mount Kōya (Kōyasan): location Nagahara Keiji: on akutō, 13, 405; of, 388, Plate 29. See also on arable land in Heian times, Daidenbōin; Kongōbuji of 187n.3; on estate structure and Mount Kōya; Uno Minamoto history, 4–16, 26–27, 283–284; movement of goods and people, 17; on Fujiwara no Teika, 322, 325; and the demise of cash commuta- Genpei civil war viewed as a tion, 50–54; geographic informa- movement of provincial land- tion system (GIS) analysis of, 357, opening elites by, 10, 24; on 374n.36; maritime trade networks military stewards, 11, 12; on the of Hachijō-in and Kōkamon-in, shōen-kokugaryō system, xxviii; on 295–300; and the relationship silk production, 43; three-stage between nyoin power and estate model of estate development, locations, 295; and Suzuyaki pro- 5, 20 duction, 300–305; water transport Nakahara (later called the Hineno) network built by the Hōjō sho- family, as estate managers, 111, gunal regents, 10. See also Hyōgo 112, 139n.6 checkpoint; indigo; iron produc- Nakahara Yorisada, transference of his tion; rice shipping; salt shipping; inherited rights to Shigetomo vertical estate structure Village, 153 Muromachi shogunate, xxxv; conflict Namuamidabutsu sazenshū (The Good in Harima Province, 95, 112; Deeds of Namuamidabutsu): on and lawsuits filed by temples, austerities Chōgen performed 206–207; Nishi Hachijō legal in his youth, 198; on bessho, 237; complaint, 49; tōchigyō (long- concept of kechien in, 245; on the term land occupation claims) construction of Nagaodera, 76; during, 419; villages on Hine dating of, 213–214, 229n.7, 237, Estate during (see Hine Estate). 246; evidence of Chōgen’s activi- See also Hashimoto Masataka ties recorded in, 213–215, 216, —Nanbokuchō period of, 436n.1; 218, 219, 226–228; on images of involvement of military gover- Amida enshrined at bessho, 237 nors in shrine and temple hold- Nate Estate in Kii Province: border ings during, 428–430. See also with the Kokawa Estate along Ogawa family the River Minase, 381, 384, 388, myō, myōden (named fields): definition 395; conflict with Kokawadera, of, xxvii, 55n.15, 175, 210n.8; 379; dispute with Niunoya Vil- development of, 11, 34n.67; lage (see Nate-Niunoya dispute); rents from, 10, 44, 95, 96, 97, 201 Kongōbuji’s administration of, myōshu (named field holders): added 398–399, 401n.59, 401–402n.61 rent and, 14, 19; definition of, Nate-Niunoya dispute: class and xxvii, xxxviin.4, 443; duties of, regional tensions exposed by, 10, 11; landholdings of, 451; and 377, 382–387; disputed area of Muromachi-period military lead- Mount Shiio, 383–387, 388, 389, ers, 433; sources on, 454n.4; as 390; and Kongōbuji’s involve- temple donors, 247, 271 ment, 380–381, 395–396; monchū Index 527

(judicial hearings), 389–390, Kuboki; Minamiyama grave com- 400n.30; social and political ten- plex; Takada grave complex); sions underpinning the balance Tōdaiji’s institution of, xxxi of power in northern Kii, 377– —management by Chōgen: Jōhan 378, 396–399; and water rights, (Gan’amidabutsu) chosen as his 377, 382–383, 384, 386–387, 389, successor, 203–205, 225, 229n.5, 390, 391, 396 236, 244; reorganization of culti- (New) Nuta Estate in Aki Province, vation efforts on the terraces of, Kobayakawa warrior family as 234; restoration of, xxix, 87 stewards of, 340, 359 —management by the Ordination Niimi Estate of Bitchū Province: Platform–Oil Storehouse, 95, 96; Hosokawa Katsumoto’s confisca- shared rights to income during tion of, 15; villages (see Yoshino the Muromachi period, 101–102 Village) —management—documentary Nishiyama Ryōhei, 189n.24 records, types and features of, Niunoya family, border dispute with 81–84 Nate. See Nate-Niunoya dispute —Nigori Pond, 238, Plate 7; Chōgen’s Noguchi Hanayo, 285 legendary creation of it, 75–76 Noguchi Minoru, 297, 299 —Ō family as reeves of, 91, 93, 94, Nuta Estate in Aki Province, 328; 101–102, 246, 249, 271, 427, 436 Saionji control of, 331–332, —Ōji area: grave complex, 64, 72, 76; 340–341, 348n.41 Ōji Shiro-no-shita settlement, nyoin: changing criteria for, 283; honke 69–70, 69; Ōji Tsuji-no-uchi shiki (supreme proprietorship) settlement, 70; Yayoi pottery held by, 283–284; political and from Ōji Shiro-no-shita, 79 economic power of, 284–285. —Ōzuka mounded tomb: location on See also Hachijō-in; Jōsaimon-in; Ōbe Estate, 63–64, 64; tumulus Kōkamon-in period artifacts unearthed from, 71–72, 71 Ōbe Estate: cases of akutō on (see under —terraces: Chōgen’s reorganization akutō [“evil bands”]); location in of cultivation efforts on, 234; Harima Province, 62, 231, 416; growth of settlements on, 76–77; ties with the warrior establish- maps of, Plate 2, Plate 3, Plate ment, 427 4; overview of three levels of, —agricultural conditions at: cultiva- 62–67; Tera-yu (“Temple Canal”) tion in the Nanbokuchō period, running through the lower ter- 158, 159; irrigation practices, race of, 222 156–161, 159 Ōga Estate, 397, 398, 402n.66, 402n.67 —lands of: boundaries of, 62–63, 68; Ogawa family, 429 Kamakura-period archaeological Ogawa Hirokazu, 190n.37 sites, Plate 4; modern districts Ōi Estate, 30–31n.29, 186, 192n.68 of, 77–78n.6; pre-Kamakura Ōi family, management of Iwa Shrine, archaeological sites, Plate 3; sites 429 within its territory that post- Ōnin War, 15, 25, 114, 370, 371, 436; date its founding (see Kiyotani Kyoto’s destruction during, 16; archaeological site); sites within rivalry between warrior families its territory that predate its estab- as the seeds of, 434 lishment (see Kōdoji Temple; Ono Takashi, 22 528 Index

Ōta Estate in Bingo: Bingo constable’s 5; grain used to fill obligations, control over, 15; commendation 42, 52, 140n.19; in iron, 44; lists to Kōyasan by Go-Shirakawa, 12; of rent payers (nayosechō), 84; institution of, 8; Onomichi as payments in cash, 10, 41–42, 51; the “warehouse” port for, 352, in salt at Yugeshima Estate, 353, 353–355, 359–360, 365–366, 354, 364; variety of goods used 367, Plates 25–28; provincial to fill obligations, 42–43, 52. See constable’s control over, 15; rent also Ōbe Estate—management payment delays, 364, 365–367 by the Ordination Platform–Oil Oxenboell, Morten, 406, 408, 415, 419, Storehouse 422n.25, 426n.80 rice production: and the annual labor Ōyama Estate of Tanba Province, pro- cycle of medieval fishing villages, tests at, 15 144; and rent payments, 42–43 Ōyama Kyōhei: on the financial inde- rice shipping: and the Hyōgo check- pendence of bessho, 237; on mili- point, 357, 374n.32, Plate 27, tary stewards, 31n.33; Ōbe Estate Plate 28; “Sanuki rice,” 357, 361, Survey Committee headed by, 62; 374n.32; and Uozumi Anchor- research on Nishidai Village, 46 age, 241 Ōzu Estate in Kii Province, location of, risshō. See instituted estates 388, 398 ritsu monks (risshū, world-renouncing ascetics): at Tōdaiji, 199, 200, pirates: akutō as, 417; as a problematic 202. See also Tōdaiji—Ordination term, 374n.33; encounters with Platform (Kaidan’in) used as an excuse for not paying royal family: genealogy, 282; royal rents, 364; Fujiwara no Sumi- grant fields (chokushiden), 21, tomo, 339, 349n.72; Uwa and 307; tennō (Heavenly Sovereigns) Uno as bases for, 339–340; wakō (see Go-Daigo Tennō; Go-Fushimi stronghold of Zhangzhou, 50 Tennō; Go-Murakami Tennō; proprietors (ryōke): Saionji as, 325; Go-Saga Tennō; Go-Sanjō Tennō; and the vertical estate structure, Go-Shirakawa Tennō; Go-Toba xxvii, 4. See also supreme propri- Tennō; Toba Tennō/In). See also etors; vertical estate structure vow temples —royal ladies: royal grant fields Raisen of Renjōin: as the adminis- donated by daughters of Kanmu trator of Kongōbuji of Mount Tennō, 21. See also Bifukumon- Kōya, 400n.17; identified as in; Hachijō-in; Jōsaimon-in; Renjōin Sōsenbō, 401n.59; and Kenshunmon-in; Kōkamon-in; Kongōbuji’s administration of nyoin; Taikenmon-in Nate, 401n.57 Ryō Susumu, 330 reeves. See kumon rent payments: “added rent” (kaji- sacred and secular integration: and shi ), 14, 171; and agricultural Chōgen’s Amidism, xxxiv, 272; productivity, xxxiii, 154, 158; and Chōgen’s public works collection by custodians (azu- projects, 212, 226–228; and karidokoro), 32n.50; collection Chōgen’s restoration of Ōbe by Zen monks, 32n.50; and Estate, 231–232, 235–236, 242, “cooking the books,” 49–50; 243–245; and Chōgen’s restora- and early estates (shoki shōen), tion of Sayama Pond (see Sayama Index 529

Pond); and estates, 206–207, communities evidenced by, 255–256, 272, 458; involvement 219–221, 241–242; stone plaque of military governors in shrine inscription detailing his efforts, and temple holdings, 429–430; at xxix–xxx, 217, 241–242, Plate 15; Kongōfukuji, 255; and the prac- Uozumi project compared with, tice of “imprisoning a name,” 241–243 207–209. See also Sayama Pond; Sayama Pond Museum, 229n.10 Uozumi Anchorage Shapinsky, Peter D., 374 Saionji family: landholdings, 324, Shibuta Estate, 397, 402n.67 326–328, 329–330; as ryōke for shiki (rights to income, officerships): royal estates, 325; spatial power Chōgen’s willing of proprietor’s of, 335, 339–345 rights (ryōke shiki ) to Tōnan’in, Saionji Kintsune: Kitayama villa built 88; and estate diversity, 34n.74; by, 319–320, 345, 347n.40; increased complexity of, 25, 28; Makishima villa built by, 342, and the Ise Taira, 24; military 350n.83; power and influence of, stewardships ( jitō shiki ), 10, 321, 329; Uwa Estate controlled 11–12; and the structure of by, 329–330, 338–339 estates, xxv, 4, 5, 8, 9, 18, 26, salt production: and the annual labor 283–284, 457; weakening of, 14, cycle of medieval fishing villages, 15–16, 19, 34n.66, 352, 368. See 144; and estates, 5, 10; and rent also vertical estate structure payments, 42; skill required for, 43 Shimizu Ryō, 25, 374n.37 salt shipping, 358–361, 363, 367; and Shin-Daibutsuji Amida triad, 263, 268; the Hyōgo checkpoint, 46–47, head of Amida, Plate 20 358–359, 360, 367, 367; Plate 25, Shinsarugakuki (New monkey music) by Plate 26 Fujiwara Akihira, 191n.48, 334, sangō (Tōdaiji’s monastic council; 443; description of an agricul- three deans), 172, 199, 204, 411 tural specialist in, 173 Sanuki Province: and salt shipping, Shirakawa Tennō: instituted estates 360–361; and “Sanuki rice,” 357, during the time of, 24, 374n.32 35–36n.87; and Tōdaiji, 178 Sanzen’in Temple Amida triad, 266, Shizukawa Estate: attack on Plate 19 Kokawadera, 396; disputes with Sasaki Ginya, 34n.66 Kaseda Estate, 380, 397; location Sato, Elizabeth, 30–31n.29 of, 388, 398 Satō Kazuhiko, 405, 407 Shōen History Research Group: call Satō Makoto and Gomi Fumihiko, for clearer historicity by, 27; 33n.64 estate development traced Satō Shin’ichi, 34n.65 through five stages, 20–27; Satō Yuki and Itō Rumi, 25 Handbook for Research in Shōen His- Sayama Pond: construction and tory (Shōenshi kenkyū handobukku) repairs of, 215; discovery of edited by, 4, 20 Chōgen’s plaque in it, 215; vil- shōkan (estate management staff): defi- lages drawing irrigation water nition of, 443; eleventh-century from, 219–220, 220 organization of, 23; at Kuroda, —Chōgen’s restoration of: and the 175; and production of non-rice centrality of his Shingon beliefs, agricultural goods, 43. See also 219; his relationship with local geshi, gesu; kumon 530 Index

Shōken (Nyoamidabutsu): as Chōgen’s Suzuyaki ware: emulation of lacquer- initial choice as heir to the man- ware, 303; pot excavated from agement rights of Ōbe and other the remains of Prince Mochi- Tōdaiji estates, 203–204, 205, hito’s residence, 303, 316n.73; 229n.5; and the Shōbō lineage of production on Wakayama Estate, Daigoji, 203, 236 300–305 shugo (constables, military governors), 15, 95, 102; and Akamatsu- Tabata Yasuko, 18–19, 34n.67 Yamana rivalry, 435–436; and Tai Estate in Awa Province, 340 criminal matters, 408; docu- Taikenmon-in: children of (see Go- ments regarding, 99; dues owed Shirakawa Tennō; Jōsaimon-in); to, 428; estate management by, as consort of Go-Toba Tennō, 14, 31n.44; in Harima Province 24; lumber for building Kasuga (see Akamatsu family; Hosokawa Shrine supported by, 193n.81; family; Yamana family); increased royal vow temple supported by, authority of, 12; Izumi Province 24 military governors and Hine Taira family. See Ise Taira family Estate, 16, 108, 109, 113, 114, Takada grave complex, 64, 68–69 122, 130, 132, 134–135, 136, Takahashi Hiroaki, 189n.19, 189n.25 139n.7; in Kii province, 377, 391; Takahashi Kazuki, 49 levy (tansen) imposed on fields, Takahashi Toshiko, 36n.92 15, 113; list of names (1445), 372; Takeuchi Rizō, 9 and Ōbe Estate, 206; Ōuchi fam- Tale of Genji, 39, 164, 443 ily as, 369, 370, 371; punishment Tamura Noriyoshi, 191n.56 of tax rebels by, 174; and the Tanaka Estate, 306, 307, 308 shogunal hanzei (half-tax) system, Tanaka Tan, 273–274n.14, 274n.19 13–14, 25, 114, 352–353, 441–442; Toba Tennō/In: children of (see Go- and Yano Estate, 14, 428 Shirakawa Tennō; Jōsaimon-in); shugodai (shugo’s deputy): and Kujō consort of (see Taikenmon-in); control of the Hine Estate exemption of Yugeshima from (1234), 111; mobilization of taxation requested of, 353; shogunal retainers (gokenin) in expansion of estates, 7–8, 24; Kii, 391; and the Yamana family’s involvement in Tōdaiji’s disputes control of Ōbe Estate, 432–433 with Kuroda cultivators, 180 Shunjōbō Chōgen. See Chōgen Toda Yoshimi, 9, 145–146 silk production: skill required for, 43; Tōdaiji: destruction of its Great Bud- and trade with rice field owners, dha Hall, 186; disputes with Iga 44–45 provincial officials, 178–182, silkworm raising: and the annual labor 184–186; estates of (see Akanabe cycle of medieval fishing villages, Estate; Ōbe Estate; Ōi Estate); 144 map of, 85; organizational Souyri, Pierre-François, 405 structure of, 84, 86; rebuild- Suchi Township (gō ), 165, 188n.6. See ing by Chōgen, xxxiv, 75, 87, also Yagawa 198, 202–203, 212, 232, 253; supreme proprietors (honke): nyoin as, Sōnshōin, 85, 86, 200; subtem- xxxiv, 283; and the vertical estate ples (see Tōnan’in) structure, xxv, 4. See also com- —Great Buddha Hall: casting master mendation of estates of the Great Buddha in (see Chen Index 531

He-qing); location of, 85; rev- Tōji hyakugō monjo, 17, 33n.58, enue from Ōbe and other estates 57n.35, 314n.46 used to support its reconstruc- Tōnan’in, location of, 85 tion, 203–205 —as cloister within Tōdaiji, 84; docu- —Hachiman Shrine, 247–248, 356, ments associated with, 90–92 411; “forcible petition” (gōso) —management of Ōbe Estate: alterna- using the carriage of its kami, tion with corporate Tōdaiji and 206–207; location of, 85; revenue the Ordination Platform––Oil from estates used to support it, Storehouse, 87, 88, 101; in 99, 205, 221 Chōgen’s testament, 88, 88, —Hokkedō (Lotus Hall): Chōgen’s 89–90; Go-Toba’s order, 89 rebuilding of, 202–203; location Toyotomi Hideyoshi: cadastral survey of, 85; monastic factions at, 182; of, xxxvi; invasion of Korea, residences for service monks 51–52; Izumi Province subdued attached to, 199, 200 by, 111, 115; unification of Japan, —Inzō (Archives), location of, 85 51 —Nigatsudō: curse ritual described in the Nigatsudō-e engi, 207–208; Uno Minamoto: and the monks of location of, 85; Shūnie ceremony Renjōin, 394–395; settlement in held at, 207, 210n.16 the Kokawa area of Kii, 382–383, —Oil Storehouse (Aburakura): dispo- 384 sition of its documents, 86, 87; —Yoshiharu: expulsion from Nate, management by the Ordination 382–384; his claim to Mount Platform, 84; and the manage- Shiio, 385; water from Nate ment of Ōbe Estate, 88, 95, 96, diverted to irrigate his fields in 97–101; ritsu monks based at, Niunoya, 382, 384, 387 200 Uozumi Anchorage: Chōgen’s restora- —Ordination Platform (Kaidan’in), tion of, 221, 227, 228; Chōgen’s 85, 87; Oil Storehouse (Abura solicitation of support from kura) under its authority (see the local community, 240–241; Tōdaiji—Oil Storehouse); Sayama Pond restorations com- revenue from the estate used pared with, 241–243 to support it, 205; ritsu monks Uwa Estate: coastal location of, 335, based at, 199, 200 339–340; Nagahira’s negotiations —reconstruction of: and Chōgen’s over the control of, 331; Saionji organizational skills, 239–241, Kintsune’s control over, 329–330, 244, 247–248; Chōgen’s religious 338–339; and the Tachibana ideology, 225–226 family, 329 —three deans. See sangō —Tōdaiji monjo collection, 81, 101, 412, vertical estate structure: and the 425n.76 broader sociopolitical order, xxx, Togawa Tomoru, 21 283; Nagahara Keiji on, 4; and Tōji: estate holdings (see Niimi Estate shared rights to income from of Bitchū Province; Yano Estate the land (shiki), xxvii, 283–284; of Harima Province); estates ties between social levels created donated by Go-Uda and Go- by, xxv, xxx, 35n.84. See also Daigo, 17; as proprietor of commendation of estates; shiki; Yugeshima Estate, 354, 364; supreme proprietors 532 Index von Verschuer, Charlotte, 187n.3, 349 Yano Estate of Harima Province: imposi- vow temples (goganji): Eigen’an at tion of taxes and duties on, 428; Kenninji, 109; Enkōin, 24; ties with the warrior establish- Hachijō-in’s Anrakuju’in, 285; ment, 427; unrest in, 14, 431 Hōkongō’in (of Jōsaimon-in’s Yasuda Motohisa, 28, 36n.99 mother), 291; Jōdodō designated yeomanry (hyakushō): as a term, as a royal prayer chapel, 238; 424n.46; and class tensions, 381, support by nyoin of, xxxiv, 8; on 383; and communal conflict, Ushigahara Estate in Echizen, 387–393, 397–398; reassess- 29–30n.18; Zanmaiin, 23 ment of the occupations of, 385; record of payments to, 99; wet field agriculture: conversion of relationship with local elites rice fields into commercial and external authorities, 384, crop production, 49; dry field 392; rights to land claimed and agriculture compared with, defended by, xxvii, 411–415, 417, 148; irrigation practices at Ōbe 418, 420–421, 425n.65 Estate, 156–161, 159; in Nabari, Yoshimura Takehiko, 21 168–171, 174; “paddy-centric” Yoshinaka Estate, 306, 308 views of medieval society and Yoshino: conflict with Kyoto, 95, 396, economy, xxxiii, 143–145, 148, 436n.1; Go-Daigo Tennō’s court- 150; tax and rent payments in exile at, xx, 95, 108 calculated in terms of, xxxiii. See Yoshino Village (Niimi Estate), rent also rice production payments made in iron by, 44 Yoshioka Yasunobu, 301, 303, 304 Yagawa (or Yakawa), 171, 176, 178, Yuge Estate, 146–147 183; confiscation of crops from, Yuge port in Iyo Province, 352, Plates 181; Jōsaimon-in’s report of theft 25–26; salt shipped from, 359, of lumber by agents of, 293; wet 361, 365 and dry fields in, 170, 179 Yugeshima Estate: Go-Shirakawa as Yagawa Estate, land tax and rent proprietor of, 353–354; rent demands, 170–171, 175 payment delays at, 366; Yamamura, Kozo, 442, 451 rent payments in cash, 355; rent Yamana family: and conflict with the payments in salt, 353, 354, 359, Akamatsu in Harima, 435–436; 364; salt production on, 11, 44, as governors of Harima, 430, 353–354, 358–359, 364–365; Tōji 432–434 as proprietor of, 354, 364 Plate 1. Seated image of Chōgen. Jōdoji, Hyōgo Prefecture. Photo: Ono-shi Kyōiku Iinkai.

Plate 2. Terraces of Ōbe Estate. Drawn by Nishida Takeshi. Plate 3. Pre-Kamakura archaeological sites on Ōbe Estate. Drawn by Nishida Takeshi.

Plate 4. Kamakura-period archaeological sites on Ōbe Estate. Drawn by Nishida Takeshi. Plate 5. Model of the abandoned Kōdoji Temple, Hyōgo Prefecture. 1/20 scale, ceramic. Photo: Ono-shi Kyōiku Iinkai.

Plate 6. Edo-period illustration of Jōdoji buildings. Photo: Ono-shi Kyōiku Iinkai. Plate 7. Nigori Pond, Ōbe Estate. Photo by Nishida Takeshi.

Plate 8. The site of Hine Estate. Drawn by Hirota Kōji. Plate 9. Hineno Village on Hine Estate, dated 1316. Courtesy of the Imperial Household Agency (Kunaichō). Plate 10. Site of three inner villages, Hine Estate. Photo by Hirota Kōji.

Plate 11. Terraced rice fields on the site of Hine Estate. Photo by Hirota Kōji. Plate 12. Location of Tsuchimaru Village, Hine Estate. Photo by Hirota Kōji.

Plate 13. Location of Hineno Village, Hine Estate. Photo by Hirota Kōji. Plate 14. Yugawa Canal, Hineno Village, Hine Estate. Photo by Hirota Kōji.

Plate 15. Plaque with account of Chōgen’s repairs of Sayama Pond, dated 1202. Photo: Osaka Prefectural Sayama Pond Museum. Plate 16. Amida triad by Kaikei, c. 1195. Wood with gold foil. Amida, height 530 cm; Kannon and Seishi, height 371 cm. Jōdoji, Hyōgo Prefecture. National Treasure. Photo: Ono-shi Kyōiku Iinkai. Plate 17. Chōgen, ﬁrst half of thirteenth century. Wood with polychromy. Height 81.8 cm. Shunjōdō, Tōdaiji, Nara Prefecture. National Treasure. Photo: Nara National Museum; Kyousuke Sasaki. Reproduced in Kainuma 2014, Artibus Asiae 74‑1.

Plate 18. Jōdodō, 1194. Jōdoji, Hyōgo Prefecture. National Treasure. Photo: Ono-shi Kyōiku Iinkai. Plate 19. Amida triad, 1148. Wood with gold foil. Amida, height, 233 cm; Kannon, height 131.8 cm; Seishi, height, 130.9 cm. Sanzen’in, Kyoto Prefecture. National Treasure. Photo courtesy of Sanzen’in Temple.

Plate 20. Head of Amida (before restoration) by Kaikei, c. 1202. Wood. Height 107.5 cm. Originally part of an Amida triad. Shin-Daibutsuji, Mie Prefecture. Important Cultural Property. Photo: Nara National Museum; Kinji Morimura. Reproduced in Kainuma 2014, Artibus Asiae 74‑1. Plate 21. Amida triad, Ming period, China. Hanging scroll, pigment and gold on silk, 121.0 × 60.8 cm. Saihōji, Aichi Prefecture. Photo: Nara National Museum; Kinji Morimura. Reproduced in Kainuma 2014, Artibus Asiae 74‑1.

Plate 22. Amida Nyorai by Kaikei, 1201. Wood with gold foil. Height 266.5 cm. Jōdoji, Hyōgo Prefecture. Important Cultural Property. Photo: Nara National Museum; Kinji Morimura. Plate 23. Masks by Kaikei and others, c.1201. Wood with polychromy. Height c. 18.9 cm. Jōdoji, Hyōgo Prefecture, Important Cultural Property. Photo courtesy of Ono-shi Kyōiku Iinkai.

Plate 24. Kōkamon-in’s estates. Template from Jeffrey P. Mass, ed., The Origins of Japan’s Medieval World, Stanford University Press, 1997. Originally drawn by Arnie Olds. Used with permission. Plate 25. All ports shipping salt to the Hyōgo checkpoint. Created by Michelle Damian.

Plate 26. Major ports shipping salt to the Hyōgo checkpoint. Created by Michelle Damian. Plate 27. All ports shipping rice to the Hyōgo checkpoint. Created by Michelle Damian.

Plate 28. Major ports shipping rice to the Hyōgo checkpoint. Created by Michelle Damian. Plate 29. Kii Province. Drawn by Philip Garrett.