Networks of space and identity: origin narratives and manifestations of the Itsukushima deity

PDF

Download

Open document

Transcript (if available)

Content NETWORKS OF SPACE AND IDENTITY: ORIGIN NARRATIVES AND MANIFESTATIONS OF THE ITSUKUSHIMA DEITY by Jesse Lee Drian A Dissertation Presented to the FACULTY OF THE USC GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY EAST ASIAN LANGUAGES AND CULTURES December 2020 Copyright 2020 Jesse Lee Drian ii Acknowledgements I thought at first that research and writing are solitary work. I quickly learned that I was mistaken. This dissertation would not have been possible without the limitless support of my dissertation committee: Lori Meeks, David Bialock, Duncan Williams, and Satoko Shimazaki. I could not have been more fortunate to have Lori Meeks as my advisor and mentor. She always made me want to be better, not because my work or abilities needed to be, but because she believed in my potential. David Bialock has provided countless incisive feedback on my work over the years, and he has profoundly influenced how I engage with texts through close readings, interdisciplinary perspectives, and theoretical thinking. Duncan William’s advice, first during my qualifying exam preparation and afterwards on my dissertation committee, helped me to reshape how I read and write while always keeping in mind the bigger picture. Seminars with Satoko Shimazaki taught me the value of expanding the scope of my research and allowing seemingly peripheral subjects to lead my ideas in new directions. I cannot thank them enough for all the time and energy they dedicated to reading the dissertation, offering crucial advice, and always being there for me. I owe a special debt of gratitude to Uejima Susumu, my advisor during my year of research at Kyoto University. His graciously invited me into his seminars, symposiums, and research excursions, and introduced me to other scholars. He taught me how to conduct research in Japan and how to be a better scholar. My time in Kyoto would not have been anywhere near as productive or memorable if it was not for Uejima Susumu’s immense support. I am also thankful to professors Yoshikawa Shinji, Tanigawa Yutaka, and all the graduate students in the Department of Japanese History who welcomed me into the department and made me feel at iii home. Special thanks to Teshigahara Takuya, Matsui Naoto, Hyunkyung Kim, and Takai Ryu for their help reading manuscript documents, invitations to events, and comradery. I am truly lucky to have been able to work with so many people over the course of my studies. I would not be here today if not for Hank Glassman. His classes first inspired me to think about religion and literature together, and in many ways, his teachings set me on the intellectual path that I have continued to walk since. provided a foundation for all my work since. During my masters, Ryūichi Abé introduced me to the Heike nōkyō scrolls offered to Itsukushima by the Taira, and it was this project that led me to start wondering about the history of Itsukushima Shrine after the fall of the Taira. Even before I arrived at USC, Joan Piggott has always been a constant source of support and has gone out of her way to invite me to participate in symposiums, kambun workshops, and social gatherings. Sonya Lee, Rongdao Lai, James McHugh, Chris Callahan, and Paul Groner all provided insights in classes and conferences at USC that advanced my research and energized me to explore new ideas. I also want to thank Christine Shaw, who has been a guiding light for navigating the requirements for the degree, as well as a constant source of encouragement. Funding from the Japan Foundation, Robert H.N. Ho Family Foundation in Buddhist Studies, and the USC Graduate School allowed me to conduct research in Japan and then devote myself to writing the dissertation. I would not have been able to complete my dissertation without their generous support. Additionally, events organized by the Japan Foundation and the Robert H.N. Family Foundation in Buddhist Studies introduced me to some of my closest friends and collaborators. At USC, the Shinso Ito Center for Japanese Religion and Culture, ACE- Nikaido Fellowship, and the Noboru and Barbara Inamoto Graduate Fellowship generously helped fund travel for research, conferences, and workshops. iv I could not have completed this without the support of friends and colleagues. Our conversations, trips to sacred sites and museums, and shared meals more than make up for all of the hard work, stress, and worries. Ori Porath and Andrew Macomber have been constant sources of academic and moral support. I knew that I could always count on them to do anything they could to help, and that has meant the world to me. My friends at USC have had to endure countless discussions about Itsukushima Shrine in seminars, symposiums, writing groups, and over dinners and coffees. My sincere thanks go out to Tori Montrose, Matthew Keller, Lisa Kochinski, Tatyana Kostochka, Nadia Kanagawa, Sachiko Kawai, Dan Sherer, Jillian Barndt, Emily Warren, Haiwei Liu, Angelica Stoddard, Carlos Parra, Sari Siegel, Chad Walker, Melissa Chan, Yunwen Gao, Lipeng Chen, Rio Katayama, and Xuejing Sun. I am fortunate to have been able to meet so many other close friends and collaborators: Andrea Giolai, Kevin Buckelew, Julia Cross, Emily Simpson, Kyle Bond, Eric Swanson, Adam Lyons, Deborah Price, Casey Martin, Tori Davis, and Eric Tojimbara. Your support and encouragement has kept me sane and motivated through the busy times, and helped me to make the most of the more relaxing moments. Finally, I am eternally grateful to my parents, Allan and Marjorie. I will leave this section short, because I feel like the only alternative would be to write multiple pages. I hope you know just how important your endless encouragement and compassion have been. You have been there for me to celebrate all my successes (no matter how small) and to help me get back on feet when things got difficult. I will never be able to thank you enough. Thank you. v Table of Contents Acknowledgments...........................................................................................................................ii List of Tables ................................................................................................................................vii List of Figures ...............................................................................................................................vii Abstract ...........................................................................................................................................x Introduction .....................................................................................................................................1 Comparison in the Realm of the Gods ....................................................................4 Dissertation Organization and Chapter Summaries ..............................................16 Resituating the Historical Identity of Itsukushima ...............................................21 Blurring the Place of Local Deities in Buddhism and Kami Worship ..................32 Origin Narratives for Connecting Intellectual and Vernacular Religion ..............38 Itsukushima Origin Narrative Manuscripts ...........................................................45 Conclusion ............................................................................................................54 Chapter 1: The Production of Origin Narratives and the Networking of Dragon Sisters .............58 Networks for Mediating Connections between the Gods .....................................62 Connection rather than Contestation for Affirming Origin Narratives .................68 Strategic Networks in the Itsukushima Daimyōjin nikki .......................................72 Incorporation and Standardization of the Jingū Association ................................80 Networks for the Protection of the Realm of Japan ..............................................84 Assembly with and in Relation to other Networks ...............................................89 Popular Reception of the Itsukushima Origin Narrative Associations .................95 Conclusion ..........................................................................................................102 Chapter 2: Benzaiten Outside of Itsukushima: Spatial Networks and Deity Multiplication ......105 Honji-suijaku and the Systematization of Relations ...........................................109 Standardizing and Hierarchizing the Multiple Benzaiten ...................................131 Networked Spaces for Mapping Relations in Space ...........................................142 Fluid Symbols and the Blurred Boundaries of Identity ......................................148 Conclusion ..........................................................................................................155 Chapter 3: From the Origins of the Shrine to the Original Form of the Deity: Personalizing the Worship of Itsukushima ............................................................................................159 Even the Gods Have Favorites: Benefits and Anxieties of Personal Connections ......................................................................................................163 Honji monogatari: The Original Form Narrative as Another Form of Origins ..............................................................................................................167 Humanizing the Original Form of the Deities ....................................................173 Explication and the Reinterpreting of the Narrative and the Deity ....................177 Summary of Itsukushima no honji ......................................................................187 Unifying the Travel of the Deity and the Devotee to Itsukushima .....................198 vi Ritual Re-Production of the Deity’s Travels ......................................................202 Honji monogatari as Narrative and Knowledge .................................................212 Conclusion ..........................................................................................................215 Chapter 4: The Rise and Fall of the Heike and the Surviving Association with Itsukushima Shrine ........................................................................................................................217 Sources ................................................................................................................221 Regional Perspectives on the Tale of the Heike ..................................................223 Itsukushima within the Nagatobon but Outside the Heike ..................................232 The Problematic History between Itsukushima and the Taira ............................236 Questioning Itsukushima’s Identity as the Taira’s Tutelary Deity .....................241 Shifting Perspectives from the Early to Late Medieval ......................................248 Reframing Itsukushima within the Nagatobon Tale of the Heike .......................254 Dream of the Council of the Deities ...................................................................255 The Origins of Kiyomori’s Worship of Itsukushima ..........................................261 Conclusion ..........................................................................................................275 Conclusion ..................................................................................................................................280 Topics for Future Study: Connections through Difference ................................288 The Importance of Simple Identities/Connections .............................................298 Bibliography ...............................................................................................................................303 vii List of Tables Table 1: Itsukushima Origin Narrative Variants ...........................................................................50 Table 2: Mentions of Itsukushima as Daughter of Dragon King Sagara ......................................76 Table 3: Benzaiten Groupings ....................................................................................................135 Table 4: Medieval Texts Associating Itsukushima and the Taira ...............................................249 viii List of Figures Figure 1: Map of Sacred Sites included in Table 2 .......................................................................77 Figure 2: Sannō nijūissha tōezu 山王二十一社等絵図. Image from Kageyama Haruki, ed., Shintō taikei. Jinja hen, Hie 神道大系. 神社編：日吉. Vol. 29, 379-399. Tokyo: Shintō Taikei Hensankai, 1983. ...................................................................................117 Figure 3: Three Benzaiten as wish fulfilling jewels. Image from Keiran shūyōshū 渓嵐拾葉集, T2410, 76:625a23-625a24............................................................................................119 Figure 4: Enoshima engi emaki 江島縁起絵巻, third scroll of five, privately owned. Image from Kanagawa Kenritsu Rekishi Hakubutsukan 神奈川県立歴史博物館, ed., Kamigami to deau: Kanagawa no Shintō bijutsu 神々と出逢う：神奈川の神道美術. Yokohama: Kanagawa-ken Jinjachō , 2006, 99. ..............................................................................129 Figure 5: Map of Sacred Sites of Benzaiten from Table 3 .........................................................136 Figure 6: Kōya shisho myōjin zō 高野四所明神像 (The Four Deities of Kōyasan Temple Complex), Metropolitan Museum of Art. Late 14 th Century. Licensed under CC 1.0. .........................................................................................................................150 Figure 7: Shisho myōjin zō 四社明神像, Saizen’in, 15th century. Image from Ō saka Shiritsu Bijutsukan 大阪市立美術館, ed., Yama no shinbutsu: Yoshino, Kumano, Kōya: tokubetsuten; "Kii-Sanchi no reijō to sankeimichi" sekai isan tōroku 10-shūnen kinen 山の神仏：吉野・熊野・高野 : 特別展：「紀伊山地の霊場と参詣道」世界遺産登録 10 周年記念. Osaka: Ō saka Shiritsu Bijutsukan, 2014, 122...............................150 Figure 8: Hachimangu zu narabi ni sha rei 八幡宮図并社例, Kanazawa Bunko, 14th Century. Image from Sasaki Yasuyuki 佐々木康之, ed., Mizu: Shinpi no katachi = Prayers to water 水 : 神秘のかたち. Tokyo: Santorī Bijutsukan, 2015, 97. ...............................153 Figure 9: Itsukushima go honji いつくしま御ほん地, volume 3 of 3. Published by Fuyajinbei 婦屋仁兵衛, 1656. National Diet Library Digital Collection ....................................185 Figure 10: Kibune no honji きふねの本地, volume 3 of 3. Undated. National Diet Library Digital Collection ......................................................................................................185 Figure 11: Kibune no honji きふねの本地, Daitō kyū Memorial Library. Image from Shimazu Tadao 島津忠夫, ed., Monogatari sōshi II 物語草子 II. Daitōkyū Kinen Bunko zenpon sōkan: Chūko chūsei he n, v. 3. Tokyo: Daitōkyū Kinen Bunko, 2005, 362 ............................................................................................................................ 186 ix Figure 12: Tenjin engi emaki 天神縁起絵巻, Late Muromachi period, Tenri Library. Image from Ishikawa Tōru 石川透, Kanamitsu Keiko 金光桂子, Koida Tomoko 恋田知子, and Saitō Maori 齋藤真麻理, eds. Nara ehonshū 奈良絵本集, Shin Tenri Toshokan zenpon sōsho 23 . Tenri-shi: Tenri Daigaku Shuppanbu, 2018, 36................................................................................................................................186 Figure 13: Itsukushima engi 厳島縁起, Kō jō B unko 高乗文庫. Image from National Institute of Japanese Literature Database of Pre-Modern Japanese Works, coda 7. Licensed under CC 4.0 .......................................................................................................................194 Figure 14: Map of the Movements of Saeki Kuramoto and the Itsukushima Deity. ..................204 Figure 15: Itsukushima zue 厳島図絵, volume 1 of 10. Published by Yonamiyaihei 河内屋儀助, 1842. National Diet Library Digital Collection ........................................................206 Figure 16: Ippen Shōnin eden 一遍聖人絵伝, scroll 10, Shōjōkōji 清浄光寺 1299. Images from Shō kai 聖戒, and En'i 円伊. Ippen Shōnin eden 一遍聖人絵伝. Nihon emaki taisei 27, edited by Komatsu Shigemi 小松茂美, Murashige Yasushi 村重寧 and Furuya Minoru 古谷稔. Tokyo: Chūō Kō ronsha, 1978, 271-273. .......................................290 Figure 17: Yūzū dainenbutsu hon engi 融通大念仏本縁起, 1691, Ukai Bunko 鵜飼文庫. Grey. Image from National Institute of Japanese Literature Database of Pre-Modern Japanese Works, coda 44. Licensed under CC 4.0. ..................................................292 Figure 18: Shōtoku taishi eden 聖徳太子絵伝, scroll 5, Honshōji, 1322 -1323. Image from Amano Shinji 天野信治, ed., Honshōji shozō Shōtoku taishi eden kenkyū shiryō 本證寺所蔵「聖徳太子絵伝」研究資料. Nagoya: Nagoya Daigaku daigakuin bungaku kenkyūka fuzoku jinrui bunka isan tekusutogaku kenkyū senta, 2016. ....................294 x Dissertation Abstract This dissertation examines how scholarly monks and worshippers in medieval Japan (1200-1600) negotiated the meaning of local sacred space vis-à-vis deities that were simultaneously particular, universal, and otherworldly. Taking Itsukushima Shrine, a prominent island shrine off the shore of Hiroshima, as a case study, I trace how local deities and sacred sites gained prominence through identities that connected them over with networks of divinities enshrined across the Japanese archipelago. In this manner, I provide a new way of understanding local deities, and especially the Japanese kami, that balances their spatial heterogeneity with their translocal relationality. Additionally, I show how scholarly monks employed temple and shrine origin narratives (jisha engi) as a medium capable of harmonizing Buddhist logic with additional interpretive frameworks and knowledge that extended beyond the reach of established modes of Buddhist thought. The work of scholarly monks to understand the gods did not always subsume them within Buddhist intellectual frameworks. Instead, the confounding interrelations between deities and sacred places expanded the potential local, translocal, and universal significances of Buddhist and kami deities alike. My dissertation consists of four chapters plus an introduction and conclusion. The introduction situates my arguments and methodology in relation to previous scholarship on individual cultic sites and origin narratives. As a complementary alternative to the site-based study of sacred spaces and deities, I argue for the translocal site-based study as an approach for investigating how origin narratives and knowledge of local deities may develop outside the bounds of a single religious institution. Additionally, the introduction provides a historical overview of Itsukushima Shrine to highlight the connections between sacred spaces in centers and peripheries. xi The first chapter introduces how origin narratives produced after the Mongol invasions employed networks of association to link the Itsukushima Deity with a select group of deities recognized for their power to defeat the Mongols. Previous studies have emphasized the self- promotional constructed nature of post-Mongol invasion origin narratives. However, I explain how the proliferation of competing arguments instead led to increased emphasis and standardization of knowledge about local gods and their interrelations. Specifically, I analyze how the fourteenth century Itsukushima Daimyōjin nikki (Record of the Itsukushima Deity) origin narrative strategically identifies the Itsukushima Deity and Queen Consort Jingū as sisters of the Dragon Princess from the “Devadatta” chapter of the Lotus Sutra. This association stimulated Hachiman shrine origin narratives to incorporate and disseminate the image of the Itsukushima Deity as one of the elite deities who helped Jingū subjugate foreign enemies. The second chapter examines how scholarly monks resolved questions of spatial particularity which arose from comparisons of Benzaiten deities enshrined across the archipelago. The chapter examines how monks harmonized the universalizing logic of honji- suijaku (original form and local traces) and localizing details within origin narratives to explain local sacrality as simultaneously unique and interconnected with other sacred places. The writings of scholarly monks problematize the conventional view of local deities as defined solely by their spatial emplacement and fully appropriated within the Buddhist intellectual framework of honji-suijaku. Instead, the chapter demonstrates how scholarly monks engaged with Buddhist thought and origin narratives to provide interpretative frameworks capable of elucidating the deities’ translocal nature. The third chapter shifts focus to the relationship between worshippers and deities, and concerns about whether a deity will answer one’s individual’s prayers. The chapter examines xii how original form narratives (honji monogatari) reconstruct origin narratives to encourage audiences to develop personal bonds with a deity and sacred space. The Itsukushima no honji narrative combines an emotional tale of the Itsukushima Deity past life as a princess in India and her subsequent deification with information about the deity and ritual practices to provide devotees with a means to connect with the deity through their emotional and devotional responses to the narrative. Through this analysis, I show how worshippers could feel close to the Itsukushima Deity regardless of geographical proximity, and reconsider locality as a sense of belonging held by worshippers. The conclusion reflects on the dissertation’s translocal approach and the importance of methodologies which bring together micro and macro level perspectives. Recent scholarship has demonstrated the benefits of focusing on micro level subjects and particular historical and cultural contexts. I advocate for translocal site-based studies as an approach with the potential to expand methodological diversity and illuminate the interpretive viewpoints of the subjects we study. 1 Introduction It is certainly apt that people have called it one of the three scenic places within the four seas. However, the mountain seen from the seaside, apart from the buildings and corridors of the shrine, is nothing special. It is because of the shrine buildings constructed through Taira no Kiyomori’s efforts to be as beautiful and great as possible that the people of the world are awed by the sight of it and believe it should be called one of the three scenic places within the four seas. When I traveled throughout the provinces and caught glimpses of the scenic spots, I was surprised to see how different and vast the world is compared to how people suppose it to be. There are exceedingly many sites of unparalleled beauty, and the places superior to Itsukushima are too many to count. 1 Itsukushima Shrine is still known today as one of the three scenic places of Japan (Nihon sankei 日本三景), and it is one of, if not the most, famous world heritage site in Japan. 2 When one takes a boat to its location on the island of Miyajima, south of Hiroshima City, it feels like you are entering an otherworldly seascape. It is easy to forget that the archipelago is full of magnificent landscapes. Geographer Furukawa Koshōken ( 古川古松軒, 1726-1807) traveled extensively and had the rare chance to see many of its most renowned scenic spots with his own eyes. His experiences stimulated questions about the established acceptance of Itsukushima as one of the three scenic places in Japan and the absense of comparisons with other places. 3 Itsukushima 1 Furukawa Koshōken 古川古松軒, "Saiyū zakki 西遊雑記," in Nihon shomin seikatsu shiryō shūsei 日本庶民生活史料集成, ed. Miyamoto Tsuneichi 宮本常一, Haraguchi Torao 原口虎雄, and Higa Shunchō 比嘉春潮 (Tokyo: San'ichi Shobō , 1969), 332. 2 Joseph Cali, John Dougill, and Geoff Ciotti, eds., Shinto Shrines: A Guide to the Sacred Sites of Japan's Ancient Religion (Honolulu: University of Hawaiʻi Press, 2013), 235. 3 The standard three scenic places of Japan are Matsushima, Amanohashidate, and Itsukushima. Hayashi Gahō ( 林鵞峰, 1618-1680), son of Hayashi Razan 林羅山 and a notable literatus himself, was the first to give Itsukushima, Amanohashidate, and Matsushima as the three most beautiful places in 1643. Significantly though, there were alternative lists of scenic places in the seventeenth and eighteenth centuries and they did not always include Itsukushima or position it as prominently. The poet Ōyodo Michi kaze ( 大淀三千風, 1639−1707) only listed Itsukushima as seventh in his ranking of the twelve most scenic places of the realm. Additionally, Amano Sadakake ( 天野信景, 1663-1733) gives a list of the three most scenic places of Japan in his 1782 encyclopedic work Shiojiri 塩尻 but replaces Itsukushima with Dewa Kisakata 出羽象潟. Hasegawa Seiichi 長谷川成一, "Nihon sankei no seiritsu to meisho kan no tenkai 日本三景の成立と名所観の展開," in Nihon sankei e no izanai: Matsushima, Amanohashidate, Itsukushima 日本三景への誘い : 松島・天橋立・厳島, ed. Shimao Arata 島尾新 and Hasegawa Seiichi 長谷川成一 (Osaka: Seibundō Shuppan, 2007), 25-29. 2 Shrine is certainly beautiful, but what makes it better than the countless other “sites of unparalleled beauty” (zekkei no chi 絶景の地)? 4 Furukawa raises an oft-overlooked disjuncture within discussions of famous places (meisho 名所). On the one hand, many descriptions of famous places were devoid of comparison. On the other hand, there was a prevalent practice, especially during the early modern period, of ranking or categorizing places. Furukawa’s critique might appear rational and modern (and it was), but it is not so distant from the premodern as one might imagine. In medieval Japan, there was a vast array of texts, images, and performances which depicted Itsukushima as incomparably wondrous. Yet, there was also a surprising number of sources which identified Itsukushima Shrine or its deity through relations with other sacred spaces and divinities. And not unlike Furukawa Kōshoken, there were medieval Buddhist monks and intellectuals who examined the various descriptions together and could not help being left with questions. When we think of the hard scholarly work of scrutinizing texts and pondering the possible meanings of their similarities and discrepancies, it is easy to see the appeal in just trusting a list of places, no questions asked. My dissertation examines how Itsukushima Shrine’s renown as a sacred space was maintained through connections with other places, deities, and people. Itsukushima is famous for its distinctive elements, such as its placement above the water and its historical association with the Taira household. However, people in medieval Japan often stressed these characteristics less than the commonalities which aligned Itsukushima with popular discourses of the time. If this 4 In the later Tōyū zakki ( 東遊雑記, 1788), Furukawa Kōshoken describes his own preference for the s cenery of Mt. Fuji, Tago shore 田子浦, Kiyomigaseki 清見ヶ関, and Cape Miho 三穂ヶ崎 over the conventional three most scenic places. Koshōken 古川古松軒 Furukawa, "Tōyū za kki 東遊雑記," in Nihon shomin seikatsu shiryō shūsei 日本庶民生活史料集成, ed. Miyamoto Tsuneichi 宮本常一, Haraguchi Torao 原口虎雄, and Higa Shunchō 比嘉春潮 (Tokyo: San'ichi Shobō , 1969), 571. 3 were not the case, how could we explain the survival of Itsukushima’s image after the demise of the Taira in the Genpei War (1180-1185)? Kiyomori led the Taira and Itsukushima to center- stage in court society, culminating with the accession of Kiyomori’s grandson, Antoku Tennō ( 安徳天皇, 1178-1185) to the throne in 1180. However, as famously recounted in The Tale of the Heike (Heike monogatari 平家物語), the Taira’s successes were short lived. Kiyomori died in 1181 and the Taira lost the war to the Minamoto in 1185. This left Itsukushima without the Taira’s economic and cultural capital, and it challenged the shrine’s divine power as their protector. Why would anybody choose to worship the Itsukushima Deity if it was primarily known as the failed tutelary deity of the Taira? This question remains unexplored. Scholars have tended to examine Itsukushima’s “golden age” during the era of the Taira and ignore its post thirteenth century significances. This was the time when Itsukushima emerged within contexts central to Japan’s national history (e.g., Insei period and Genpei War) and literary canon (e.g., The Tale of the Heike). Local historians in Hiroshima Prefecture continue to study Itsukushima’s latter history, but the specificity of local history limits their readership. 5 Once we examine Itsukushima beyond the boundaries of the local, we find new translocal understandings of Itsukushima, and the interrelations between sacred spaces. The disproportionate body of research on Itsukushima and the Taira makes their 5 Many of the studies on Itsukushima Shrine were written by scholars at universities in Hiroshima or for publications affiliated with local historical institutes. Notably, the Research Center for the Study of the History and Culture of Itsukushima and the Inland Sea was established at Hiroshima University in 2004 and began publishing its findings in 2005. See Itsukushima kenkyū: Hiroshima daigaku sekai isan Itsukushima naikai no rekishi to bunka purojekuto kenkyū senta kenkyū seika hōkokusho 厳島研究 : 広島大学世界遺産・厳島- 内海の歴史と文化プロジェクト研究センター研究成果報告書, (Hiroshima: Sekai isan Itsukushima naikai no rekishi to bunka purojekuto kenkyū senta, 2005-); Matsuoka Hisato 松岡久人, Aki Itsukushimasha 安芸厳島社 (Kyoto: Hō zō kan, 1986); Matsui Teruaki 松井輝昭, Itsukushima monjo denrai no kenkyū: Chūsei monjo kanri shiron 厳島文書伝来の研究: 中世文書管理史論 (Tokyo: Yoshikawa Kōbunkan, 2008); Harada Yoshiko 原田佳子, Itsukushima no sairei to geinō no kenkyū 厳島の祭礼と芸能の研究 (Tokyo: Fuyō Shobō Shuppan, 2010); Hiroshima kenshi: Chūsei (Tsūshi hen) 広島県史. 中世（通史編）, 7 vols., vol. 2 (Hiroshima: Hiroshima-ken, 1984); Hiroshima daigaku miyajimagaku sentā 県立広島大学宮島学センター Kenritsu, ed. Miyajimagaku 宮島学 (Hiroshima: Keisuisha, 2014). 4 connection appear as the central aspect of Itsukushima’s identity, but this was not the full picture. Moreover, it was not the image of Itsukushima as known by medieval worshippers or scholarly monks. I argue that the identities of the Itsukushima Deity developed and gained meaning through commonalities with other gods and sacred spaces. The Itsukushima Deity did not have to vie against the numerous gods of Japan, it was enough of a success to be mentioned as comparable. As we shall see, writings by and about other temples and shrines strengthened and spread Itsukushima’s importance by drawing associations between their deities. Itsukushima did not need to command the spotlight as long as it was shining underneath it. Comparison in the Realm of the Gods In order to examine the importance of a sacred space like Itsukushima Shrine in premodern Japan, we must strive to work from contemporary perspectives towards that site. This is difficult, because people held varying viewpoints, and it is hard to know which ones deserve emphasis. Studies of sacred space often privilege records from people at the site itself—the institution’s clergy, visiting pilgrims, and surrounding communities. However, it was easier to skip the journey and hear about a temple or shrine from textual or oral transmissions. For example, Shōkokuji abbot Zuikei Shūhō ( 瑞渓周鳳, 1391-1473) wanted to hear more about Itsukushima Shrine after listening to Hakushun Zenju 伯春全壽 describe his trip there, but this did not lead him to make a pilgrimage himself. Zuikei stayed within the capital and heard an Itsukushima origin narrative from an acquaintance, the blind biwa lute performer Jōro 城呂. 6 6 Jōro explains that he learned the It sukushima origin narrative when he visited the shrine seven years earlier. Zuikei Shūhō’s diary also mentions several other narratives told to him by Jōro: sections from The Tale of the Heike; the performance of a prayer for rain at Shinsen’en garden by Kūkai; and a narrative about the origins of the smoke on 5 Narratives and information about Itsukushima Shrine and its deity spread widely. If we only listen to the voices from Itsukushima Shrine, we risk overlooking the tales and information that reached the many people who never traveled to the site. The core of my dissertation rests on two simple arguments: knowledge of local sacred spaces and deities often developed within translocal contexts; and sacred spaces and deities were compared and understood through their interrelationships. Medieval comparative arguments might sound imprecise or illogical to modern readers, but we must account for our epistemological assumptions. The epistemology of medieval Japan was based in analogical reasoning. Influenced by esoteric Buddhist ideas of the nondual relationship between signifier and signified, analogical reasoning enabled a variety of semiotic strategies and led to an ever- expanding number of identificatory associations between sacred entities. 7 Research on the local kami deities has increasingly stressed how networks of association connect and identify multiple divinities in relation to one another. Nevertheless, these scholars tend to shy away from comparative approaches, despite the fact that they are discussing comparative modes of thought. 8 Once we broaden our scope beyond a single deity or place, we can begin to appreciate why Takahashi Miyuki called the world of medieval Shinto “the age of the low deities overcoming the high” (kamigami no gekokujō no jidai 神々の下剋上の時代). 9 Building on this idea, Satō Hiroo descr ibed the medieval period as the “warring deities period” (kamigami no Mt. Fuji. Zuikei Shūhō 瑞溪周, Gaun nikkenroku batsuyū 卧雲日件錄抜尤, ed. Ikō Myō an 惟高妙安, Dai Nihon kokiroku (Tokyo: Iwanami Shoten, 1961), 11-15, 38, 198. 7 Bernard Faure, Gods of Medieval Japan: Volume 1, The Fluid Pantheon (Honolulu: University of Hawai`i Press, 2016), 46; Susan Blakeley Klein, "Esotericism in Noh Commentaries and Plays: Konparu Zenchiku's Meishuku shū and Kakitsubata," in The Culture of Secrecy in Japanese Religion, ed. Bernhard Scheid and Mark Teeuwen (London: Routledge, 2006), 229-30. 8 One notable exception is Bernard Faure’s recent studies on the pantheons of the gods. See Faure, Gods of Medieval Japan: Volume 1, The Fluid Pantheon; Gods of Medieval Japan: Volume 2, Protectors and Predators (Honolulu: University of Hawai`i Press, 2016). 9 Takahashi Miyuki 高橋美由紀, Ise Shintō no seiritsu to tenkai 伊勢神道の成立と展開 (Tokyo: Taimeidō , 1994), 78. 6 sengoku jidai 神々の戦国時代). 10 During this time, the kami deity Amaterasu lost her unrivaled position as the gods became increasingly comparable. The myths about the gods in the eighth century Nihon shoki 日本書記 and Kojiki 古事記 had genealogically privileged Amaterasu as the ancestral deity of the royal family. Then, the Buddhist logic of honji-suijaku (original form and local traces, 本地垂迹) replaced the myths as the dominant interpretive framework on the gods. Honji-suijaku worked as a combinatory system identifying deities manifesting in our world as local traces (suijaku) of transcendent Buddhas as their original forms (honji). This weakened Amaterasu’s hierarchical advantage by identifying Amaterasu as a manifestation of Dainichi Buddha ( 大日, Skt. Mahāvairocana ), and therefore, theoretically equivalent with the other deities who took Dainichi as their original form. 11 Subsequent scholarship on kami worship has paid surprising little attention to their arguments, but I believe that their ideas about the competitive comparisons of the kami deserve further consideration. Honji-suijaku logic facilitated a range of comparative issues concerning religious, spatial, and political matters. The challenge to the sacred authority of the religio-political elite temples and shrines calls attention to the need for more research on regional sacred spaces and their connections with the capital. The scholarly focus on the religious culture of the capital can be seen as a result of a politically oriented historiography together with the scarcity of primary sources about regional sites. The latter might be unavoidable, but the former can be amended by exploring understudied topics and approaches. Kuroda Toshio’s theory of the kenmon taisei (gates of power, 権門体制) system of shared rulership by elite power blocs of court nobles, warrior aristocrats, and temples and shrines has demonstrated the religious and political influence of the central elite temples and 10 Satō Hiroo 佐藤弘夫, Kami, hotoke, ōken no chūsei 神・仏・王権の中世 (Kyoto: Hōzōkan, 1998), 313. 11 Ibid., 313-14. 7 shrines. 12 Nevertheless, the resultant spotlight on the central power blocs has led to a neglect of other regional political and religious powers. Uejima Susumu argued that the kenmon taisei system began to deteriorate at the time of the Mongol invasions of 1274 and 1281, when the Kamakura Bakufu and the court began to directly request prayers to temples and shrines throughout the archipelago. This weakened the position of the elite religious institutions as the primary institutions for rites to protect the realm, and thus, as the recipients of political patronage. The role of the elite temples continued to diminish, leading Uejima to date the collapse of the kenmon taisei and kenmitsu taisei (exoteric-esoteric system, 顕密体制) to the subsequent Nanbokuchō period (1336 -1392). 13 Uejima’s arguments call on scholars to reconsider the connections between the center and regional sacred sites in the fourteenth to seventeenth centuries. The Itsukushima Deity provides an ideal case study for examining the forms of contestation and comparison in the warring deities period. If we are to take Takahashi Miyuki’s phrasing of “the age of the low deities overcoming the high” seriously, we need to look beyond the more widely studied elite temples and shrines around the capital. This might sound obvious, but Mark Teeuwen and Fabio Rambelli are among the only scholars to note the potential for lower ranking individuals and institutions to employ honji-suijaku relations to their advantage. They concluded that the subject requires further study, and the same can still be said today. 14 12 See Kuroda Toshio 黒田俊雄, "Chūsei jisha seiryoku ron 中世寺社勢力論," in Iwanami kōza Nihon rekishi: Chūsei 2 岩波講座日本歴史：中世 2, ed. Asao Naohiro 朝尾直弘 and Naoki Kō jirō 直木孝次郎 (Tokyo: Iwanami Shoten, 1975), 255-62.; Mikael S. Adolphson, The Gates of Power: Monks, Courtiers, and Warriors in Premodern Japan (Honolulu: University of Hawai`i Press, 2000), 10-21. 13 Uejima Susumu 上島享, Nihon chūsei shakai no keisei to ōken 日本中世社会の形成と王権 (Nagoya: Nagoya Daigaku Shuppankai, 2010), 460-63. 14 Mark Teeuwen and Fabio Rambelli, "Introduction," in Buddhas and Kami in Japan: Honji Suijaku as a Combinatory Paradigm, ed. Mark Teeuwen and Fabio Rambelli (New York: RoutledgeCurzon, 2003), 45-46. Recently, Anna Andreeva’s study of Miwa and Sujung Kim’s study of the Shinra Deity (primarily identified with the temple of Onjōji) provide novel approaches for decentering kami worship from the spaces and institutions concentrated around the capital. However, they still held closer location and relations with the politico-religious 8 Takahashi Miyuki and Satō Hiroo themselves focused their analysis on assertions of superiority for deities associated with elite institutions, and neglected more peripheral sites. Satō approached the contestations from the context of historical conflicts between prominent religious institutions. As a result, he inferred that it was primarily the power-bloc temples and shrines who were vying for power. Satō’s extensive textual examples present each of the Sannō Deity of Mt. Hiei, Hachiman, and the deities of Kasuga, Kumano, and Nikkō as the foremost deity in Japan. 15 These passages are undeniably self-promoting, but they can also be read as defenses against having one’s own authority challenged. To give an example cited by Satō, the 1366 Sakakiba no nikki さかき葉の日記 by the court noble Nijō Yoshimoto ( 二条良基, 1320-1388), states, “The notion that this Kasuga Deity is just a god, the same as all the others, is most regrettable. Within this realm of Japan, [everything] rests on the decisions of this shrine.” This pronouncement of Kasuga’s position of distinction is more defensive than opportunistic. It acknowledges the real possibility of the Kasuga Deity being likened to the various other gods and not above them. 16 This passage is more overtly wary of comparison than some of the others, but it is telling that sphere surrounding the capital than Itsukushima Shrine, and this led to different opportunities for engaging in competitive comparisons of the gods. Anna Andreeva, Assembling Shinto: Buddhist Approaches to Kami Worship in Medieval Japan (Cambridge, MA: Harvard University Asia Center, 2017); Sujung Kim, Shinra Myojin and Buddhist Networks of the East Asian Mediterranean (Honolulu: University of Hawai`i Press, 2019), 115. 15 Aside from Nikkō (located in modern -day Tochigi Prefecture in eastern Japan), these deities were all located at powerful institutions within pilgrimage distance of Kyoto. Considering Nikkō’s different location and stature, closer analysis of the Mt. Nikkō o rigin narrative (Nikkōsan engi 日光山縁起) might reveal significant differences concerning the portrayal of the Nikkō Deity, the intended audience, and its reception. S ee Satō, Kami, hotoke, ōken no chūsei 神・仏・王権の中世, 312-13; "Nikkōsan engi 日光山縁起," in Jisha engi 寺社緣起, ed. Sakurai Tokutarō 桜井徳太郎, Hagiwara Tatsuo 萩原龍夫, and Miyata Noboru 宮田登 (Tokyo: Iwanami Shoten, 1975), 276. 16 The following sentences explain the Kasuga Deity’s importance by associating it, as the tutelary deity of the courtly Fujiwara, with the Ise Deity and Hachiman as the ancestral deities of the royal family. Once linked into this prominent grouping, Nijō Yoshimoto then differentiates Kasuga’s importanc e over matters of governance from the royal character of Ise and Hachiman. While separate, he brings them together as both crucial for the protection and flourishing of the realm. "Sakakiba no nikki さかき葉の日記," in Gunsho ruijū. Jingibu 群書類従. 神祇部, ed. Hanawa Hokinoichi 塙保己一 (Tokyo: Yagi Shoten, 2013), 70-71. 9 almost all of the entries cited by Satō similarly acknowledge the myria d gods. 17 The influential temples and shrines seemed to realize the dangers of a comparable multitude. If every deity can claim to be number one, the statement becomes meaningless. When all the gods are special, nobody is. Medieval concerns over the relations between the gods have been easy for modern-day scholars to overlook because religious sites did not want to appear defensive. Especially when studied from the perspective of elite temples and shrines, we see the power of honji-suijaku to support the status quo, and the potential for destabilizing their privileged positions remains out of sight. As Fabio Rambelli and Mark Teeuwen have argued, the deities linked as the original form and the local trace were metaphysically identical, but worldly power relations could be overlaid 17 Regarding the Sannō Deity, the Yotenki 耀天記 proclaims, “Sannō is the peerless sacred shrine of Ja pan. It is the number one greatest deity under heaven. Among the various deities, it is the fundamental one. Among the ten thousand shrines, it is the original one.” "Yōtenki 耀天記," in Shintō taikei. Jinja hen, Hie 神道大系. 神社編：日吉, ed. Kageyama Haruki 景山春樹 (Tokyo: Shintō Taikei Hensankai, 1983), 613. The Hachiman gudōkun, otsu (latter) variant 『八幡愚童訓』乙本 describes the Hachiman deity, “In sum, Hachiman Daibosatsu surpasses in sanctity all the buddhas in the ten directions and rises above all the three thousand kami.” (Translation from Royall Tyler.) "Hachiman gudō kun (otsu) 八幡愚童訓 ( 乙)," in Jisha engi 寺社緣起, ed. Sakurai Tokutarō 桜井徳太郎, Hagiwara Tatsuo 萩原龍夫, and Miyata Noboru 宮田登 (Tokyo: Iwanami Shoten, 1975), 262; Royall Tyler, trans., "Hachiman gudō kun: A Hachiman Primer for Untutored Youth," in Iwashimizu Hachiman in War and Cult (Charley's Forest, NSW, Australia: Blue-Tongue Books, 2017), 113. The Nikkōsan engi 日光山縁起 introduces the deity of Nikkō by begi nning with the cosmological origins of Japan and the manifesting of the various deities and lands, “Even though all of the over three thousand [deities] became the protectors of the five Kinai provinces and the seven circuits (goki shichidō 五畿七道) and serve as ancestral deities protecting the realm as expedient means bringing salvific benefits, the benefits of the Great Mangan Bodhisattva (Mangan dai bosatsu 満願大菩薩) of Mt. Nikkō in Shimotsuke Province, Tōsandō circuit are superior to those of all the other deities.” "Nikkōsan engi 日光山縁起," 276. Lastly, the title of the Kumano Deity as “Japan’s number one most miraculously efficacious, the provisional manifestation of the three deities of Kumano” (Nihon daiichi dai reigen Kumano sansho gongen 日本第一大霊験熊野三所権現) itself does not allude to other deities, but several medieval Kumano texts such as the Kumano-san ryakki ( 熊野山略記, 1430) relate Kumano’s position through comparison with other deities. It explains, “Ise is named Japan’s number one ancestral deity and Kumano is called Japan’s number one miraculously efficacious. Both deities can be considered to be the same in essence (dōtai 同躰). As for the name of “Japan’s number one,” it is hard to say that [the two titles] are completely identical, [similarly,] there are differences between the real and provisional bodies [of a buddha]. As named, Kumano’s superiority is its miraculous efficacy. It is not the case that its nature as a provisional manifestation is better than that of the other deities or that its miraculous efficacy exceeds that of the ancestral deities of the royal family.” "Kumano-san ryakki 熊野山略記," in Kumano 熊野, ed. Takikawa Masajirō 瀧川政次郎 (Tokyo: Chihō shi kenkyūjo, 1957), 439; Kondō Yoshihiro 近藤喜博, Nihon no kami: Shintō shigaku no tame ni 日本の神：神道史学のために (Tokyo: Ō fūsha, 1968), 52. 10 onto honji-suijaku relations. 18 Dichotomies of center and periphery, original and copy, high and low were mapped onto “original form and local trace” to assert authority over others. 19 More abstractly, Jason Josephson uses the term “hierarchical inclusion” to describe how honji-suijaku acted as discursive practices “subordinating marks of difference into a totalizing ideology, while still preserving their external signs.” 20 Hierarchical inclusion allows for the simultaneous assertion of shared identity while retaining distinctions of status and power. He argues that the honji-suijaku system itself is inherently hierarchical and privileges the Buddhist deities of the original form. While this was often the case, increasing comparisons between local traces during the medieval period necessitated additional means of differentiation beyond the symbolic potential of the original form and local trace dichotomy. I contend that the hierarchical interpretations were adjusted through other discursive frameworks, allowing for the incorporation of new systems of hierarchical distinctions and homogenous connections. Regional sites like Itsukushima might have had the most to gain through comparative contestations, but they also had fewer resources for collecting information about others and transmitting information about their deity. This raises the question: how was Itsukushima able to engage in this battle of information and construct effective associations with other deities? In short, the shrine had to rely on the support of scholarly monks at other temples and shrines to spread texts and information beyond the individual reach of Itsukushima’s clergy. The extent to which such external transmissions resulted from Itsukushima’s agency or from contingency is difficult to ascertain. The scarcity of medieval documents from Itsukushima Shrine limits our 18 Teeuwen and Rambelli, "Introduction," 44. 19 This was also true in cases of landed estates (shōen 荘園) and subsidiary temples and shrines (matsuji, massha 末寺・末社) under the authority of another religious institution. Satō, Kami, hotoke, ōken no chūsei 神・仏・王権の中世, 314-16. 20 Jason Ananda Josephson, The Invention of Religion in Japan (Chicago: University of Chicago Press, 2012), 24- 27. 11 ability to uncover the specific actions taken by its priests to promote the shrine. 21 Instead, I found discussions of the identity of the Itsukushima Deity within compilations produced and copied at other temples and shrines to be more useful. While acknowledging the gaps in the extant sources, I take this as an opportunity to reevaluate assumptions about the independent power of temple and shrine clergy to construct the identity of the site and its deities. Rather than telling a story about how specific persons or institutions bolstered Itsukushima’s image, I highlight the strategies employed in texts to sway readers to further circulate and reproduce the Itsukushima Deity’s identity. Considering the asymmetrical power relations favoring the elite temples and shrines, textual maneuvers to promote the spread of information about the Itsukushima Deity could be more successful than anything Itsukushima’s clergy could do on their own. Methodology: Following the Networked Traces My study of Itsukushima Shrine builds on recent scholarship on individual sacred spaces by expanding the field of inquiry beyond a single site. The study of a single cultic site has become an established and fruitful approach for examining Japanese religions in situ, revealing the multidimensionality of lived religious practices and beliefs. However, the focus on a single place and their collections of sources can risk leading scholars to consider the site as a bounded space wherein practices and beliefs develop locally and apart from external influences. Allan Grapard’s pioneering study of the Kōfukuji -Kasuga temple-shrine multiplex, for example, argues 21 There is a relatively high number of medieval documents from Itsukushima Shrine compared to other regional temples and shrines (approximately 510 documents from the Heian to Muromachi periods, and 2,600 from the Warring States period). However, this is often less than more central power-bloc temples. Matsui Teruaki compared the management of historical documents at Itsukushima Shrine and Tōdaiji in the medieval period, and argued th at Tōdaij i’s system was more capable than that at Itsukushima for preserving and cataloguing the documents held, lent out, and returned. The importance of precedence within court activities required that documents be recorded and kept accessible, but the records of warrior families and regional religious institutions show that they were not usually as stringent in managing their documents. Matsui, Itsukushima monjo denrai no kenkyū: Chūsei monjo kanri shiron 厳島文書伝来の研究: 中世文書管理史論, 18-23, 30. 12 that “Japanese religiosity is grounded in specific sites at which beliefs and practices were combined and transmitted exclusively within specific lineages before they were opened to the general public.” 22 This powerful statement illuminates the significance of local particularity, but it comes at the expense of placing relationships between places as inherently secondary. Jonathan Z. Smith describes comparison as “a disciplined exaggeration in the service of knowledge. It lifts out and strongly marks certain features within difference as being of possible intellectual significance, expressed in the rhetoric of their being 'like' in some stipulated fashion.” 23 From this perspective, the site-based study stresses emplaced practices and beliefs as a means to avoid working from preconceived religious categories (e.g., doctrine, ritual, etc.). It is not comparison that is the problem, but undisciplined comparison. As an example demonstrating the merits of disciplined parameters, Lori Meeks’s study of Hokkeji informs the reader from the first page that she is relating the convent’s medieval revival movement with broader changes regarding female monasticism. 24 This provides a methodological foundation for bringing together various types of sources and perspectives while remaining grounded in specific historical contexts. The parameters can become murky though, when employing a site-based approach to examine sacred space or religious institutions. Without an external object of comparison, the site-based study is inclined to highlight local significance and meaning. This does not lessen the value of such studies, but it is worth reflecting on the biases within our methodologies. 22 Allan G. Grapard, The Protocol of the Gods: A Study of the Kasuga Cult in Japanese History (Berkeley: University of California Press, 1992), 4. 23 Jonathan Z. Smith, Drudgery Divine: On the Comparison of Early Christianities and the Religions of Late Antiquity (Chicago: University of Chicago Press, 1990), 52. 24 Lori Rachelle Meeks, Hokkeji and the Reemergence of Female Monastic Orders in Premodern Japan (Honolulu: University of Hawaii Press, 2010), 1. 13 In more theoretical terms, there remains a need for approaches that transcend the conventional divides between macro/micro analysis and global/local scales. Seen as separate, the macro-level approach works down from concepts and categories to see how they applied to people in different places and times, while the micro-level approach works up from the contexts of a specific place and time to uncover how concepts and categories were conceived and understood. However, this dichotomy assumes that the micro-approach only applies for the study of a single, and usually local-level, place. Moreover, the selection of the place to study risks working from preconceived notions about individual sites, and consequentially, can lead scholars to overlook lesser known sites, connections with other places, or additional spatial frameworks. As an alternative, historians Christian De Vito and Anne Gerritsen propose a “micro-spatial perspective” to apply the micro-level approach’s close contextualization to translocal interconnected spaces. They propose putting these ideas into practice by “following the traces” of moving people, objects, and ideas, and examining the contexts that connect them. 25 Anna Andreeva’s study of Mt. Miwa applies a micro-spatial perspective to the site-based study, expanding its scope by exploring the networks connecting Miwa with other sacred places. Her methodological innovations allow her to illuminate the shared agency between the clergy at Mt. Miwa with mobile religious practitioners and lineages linking Miwa with other nearby places. Regardless of the power of a site’s clergy, non-elite itinerant Buddhist practitioners and holy-men, often ignored by scholars, were able to access ideas, rites, and texts about the kami, transmit and use them elsewhere, and therefore, remove them from the context or control of a particular sacred site. 26 25 Christian G. De Vito and Anne Gerritsen, "Micro-Spatial Histories of Labour: Towards a New Global History," in Micro-Spatial Histories of Global Labour, ed. Christian G. De Vito and Anne Gerritsen (Cham, Switzerland: Palgrave Macmillan, 2018), 3-9. 26 Andreeva, Assembling Shinto: Buddhist Approaches to Kami Worship in Medieval Japan, 11-15. 14 Sujung Kim’s study of the Shinra Deity similarly stresses networks linked through the movements and interactions of people and gods but shifts her analysis to the networks influencing the identities of a deity rather than a cultic site. She examines the historical human networks through which ideas traveled and changed, together with the symbolic networks enabling the Shinra Deity to be fluid and hybrid, and not just the local god of Onjōji. 27 Considered together with Andreeva, they both highlight the various actants and diffused agency which emerge when we follow the traces which moved to and from the primary site of study. I follow in the footsteps of Andreeva and Kim and take their insights in a new direction by examining networks of mobile texts and discourses rather than people. Exploring the textual and interpretive practices mediating understandings of the Itsukushima Deity, I aim to balance Andreeva and Kim’s examination of the diffusion of power among actants with the regulatory and stabilizing frameworks which led some viewpoints to gain prominence and others to be forgotten. The power of texts to disseminate beyond the reach of the institution’s human networks was crucial for the incorporation of Itsukushima into discursive connections with other divinities. But there were still power asymmetries that promoted information about Itsukushima processed from writings and discourses about elite temples and shrines. Given the peripheral position of Itsukushima Shrine, the multiple successful connections made between the Itsukushima Deity and prominent deities and religious discourses is remarkable. The work needed to formulate and then accept associations with the Itsukushima Deity helps to shine a light on the larger work being done to formulate new bodies of religious knowledge and interpretative frameworks. Bruno Latour’s work on actor-network theory has inspired scholars in various fields to examine networks as a means for transcending boundaries 27 Kim, Shinra Myojin and Buddhist Networks of the East Asian Mediterranean, 2-11. 15 and finding connections. However, his arguments about the power of networks to regulate agency (e.g., to prevent controversy) in the production of knowledge tend to garner less attention. 28 Examining fluidity or regulation alone can lead to views of openness or domination respectively and overlook the complex processes and negotiations involved in making an idea appear to be true. As Latour demonstrates, it is often the defenses against controversy in academic writing which reveal the hidden work involved in the production and reception of knowledge. Therefore, he advises practitioners of Actor-Network Theory “not to look for the intrinsic qualities of any given statement but to look instead for all the transformations it undergoes later in other hands. 29 I apply Latour’s method to the production of the Itsukushima Deity’s image by analyzing how information from Itsukushima Shrine origin narratives (jisha engi 寺社縁起) was reproduced and changed in external texts. Origin narratives are mythic-histories of a certain temple or shrine, often including information about the site’s founding and later history, its deities, and tales of their miraculous deeds. 30 Scholars of Japanese religion and literature usually study origin narratives individually, or within a larger study of a cultic site. As discussed above for the site-based study, centering analysis on the temple or shrine in the origin narrative can risk 28 Manuel Vasquez argues that religious studies approaches based in network theory often emphasize fluidity and openness, and neglect powers that limit or exclude access to the networks. Manuel Vásquez, "Studying Religion in Motion: A Networks Approach," Method & Theory in the Study of Religion 20, no. 2 (2008): 172-73, 79. 29 Bruno Latour, Science in Action: How to Follow Scientists and Engineers through Society (Cambridge, MA: Harvard University Press, 1987), 25-26, 43-46, 59. 30 The explanation of origin narratives here reflects common characteristics of the genre but should not be seen as a strict definition. Heather Blair and Kawasaki Tsuyoshi stress that there is no defining characteristic that can be found in all origin narratives. They instead argue for inclusively considering a variety of characteristics that are often, but not necessarily found in any one text. Hashimoto Akihiko proposes an abstract definition of origin narratives as “discourses which construct religious value through the relationality between objects and events,” to allow for a variety of forms while retaining a shared function. Heather Blair and Tsuyoshi 川崎剛志 Kawasaki, "Editors' Introduction: Engi: Forging Accounts of Sacred Origins," Japanese Journal of Religious Studies 42, no. 1 (2015): 2; Hashimoto Akihiko 橋本章彦, "Engi gaku he no shō tai 縁起学への招待," in Yūraku to shinkō no bunkagaku 遊楽と信仰の文化学, ed. Tsutsumi Kunihiko 堤邦彦 and Tokuda Kazuo 徳田和夫 (Tokyo: Shinwasha, 2010), 15. 16 overlooking translocal contexts. Origin narratives of faraway and seemingly unrelated temples and shrines reproduced and altered information from Itsukushima Shrine origin narratives, but this is hard to notice, much less explain, within a local-level study. I first follow the movements of information from Itsukushima origin narratives to externally produced texts (many but not all of which are origin narratives). The abundant connections reveal how origin narratives were produced and received in relation to one another. I then analyze discrepancies with the Itsukushima origin narratives to reveal questions and controversies about the comparative relations between the deities. Dissertation Organization and Chapter Summaries Each of the following chapters explores a different variant of the Itsukushima Shrine origin narrative and examines how the information within influenced and was influenced by external texts and discourses. The primary Itsukushima origin narratives of each chapter can be traced back to the fourteenth century at the latest, although in some cases the earliest extant manuscript can only be dated to the sixteenth century. They are: 1. Itsukushima Daimyōjin nikki 厳島大明神日記 (Record of the Itsukushima Deity, manuscript dating estimated to 1323-1343) 31 2. Discussion of Itsukushima within the Benzaiten section of the Keiran shūyōshū 溪嵐拾葉集 (Collection of Leaves Picked from Mountain Streams, Benzaiten section written prior to 1318) 32 3. Itsukushima no honji 厳島の本地 (Original Form of the Itsukushima [Deity], earliest manuscript titled Itsukushima no engi いつくしまのゑんぎ dated 1346) 33 31 Kanazawa Bunko 金沢文庫, ed. Kanazawa Bunko no chūsei shintō shiryō 金沢文庫の中世神道資料 (Yokohama: Yokohama kenritsu Kanazawa Bunko, 1996), 51-53. 32 T 2410:76.0625b09-76.0625b26 33 Transcribed within "Kaidai: Itsukushima no engi 解題：いつくしまのゑんぎ," in Muromachi jidai monogatari taisei 室町時代物語大成, ed. Yokoyama Shigeru 横山重 and Matsumoto Ryūshin 松本隆信 (Tokyo: Kadokawa Shoten, 1973). 17 4. Itsukushima origin narrative within the Nagatobon variant of the Tale of the Heike 長門本平家物語 (estimated to have been produced in its current form between the late fourteenth to sixteenth centuries). 34 As shown here, the Itsukushima origin narratives were not always transmitted as independent texts. The Itsukushima origin narrative within the Nagatobon variant of the Tale of the Heike is almost identical to the Itsukushima Daimyōjin nikki, but the context of the Tale of the Heike and the organizational framing of the origin narrative within the larger Heike narrative are significant enough to discuss separately. I organize the chapters around these versions of the Itsukushima origin narrative because of their contextual significance rather than individual content. In each case, I find a different discursive controversy and investigate its negotiation through changes in the Itsukushima origin narrative and its usage in other texts. In this manner, I am not merely examining the production of the Itsukushima Deity’s identities, but also the production of bodies of knowledge and interpretive frameworks needed for making sense of translocal relations between local deities. In short, the chapters examine the controversies over (1) which deities deserve praise for defeating the Mongols; (2) how local manifestations of the same transcendent Buddha can be differentiated; (3) why a deity will answer one’s individual prayers and therefore deserve to be worshipped over the various other deities; and (4) how the Itsukushima Deity can be depicted positively despite its inability to protect the Taira household. The controversies and their resolutions all involve comparative relations connecting Itsukushima with other deities, sacred spaces, individual worshippers, and historical memories. 34 Asahara Yoshiko 麻原美子 and Nanami Hiroaki 名波弘彰, eds., Nagato-bon Heike monogatari no sōgō kenkyū: Kōchū hen 長門本平家物語の総合研究：校注篇, 3 vols., vol. 1 (Tokyo: Benseisha, 1998), 392-402. 18 The first chapter, “The Production of Origin Narratives and the Networking of Dragon Sisters,” begins with the relations between fourteenth century temple and shrine origin narratives. Inspired by Bruno Latour’s study of the production of scientific knowledge, I trace how the copying and compilation of origin narratives within temple libraries led to the standardization of knowledge about the gods and their relations. As the origin narratives of various temples and shrines became accessible among scholarly monks, newly produced origin narratives increasingly emphasized commonalities and connections between the gods of different sites. Specifically, this chapter explores how the identity of the Itsukushima Deity as the sister of the Dragon Princess from the “Devadatta” chapter of the Lotus Sutra became widely known through the incorporation of this association into writings about the gods of other sacred spaces. The siblings of the Dragon Princess grew as other origin narratives made similar claims, and over time the frequently mentioned sisters became standardized while the others were forgotten. Analyzing the fourteenth century Itsukushima Daimyōjin nikki (Record of the Itsukushima Deity) origin narrative, I show how the text strategically connects the Itsukushima Deity and Queen Consort Jingū as sisters of the Dragon Princess, and how this relationship was then copied and disseminated in contemporary Hachiman shrine origin narratives. In the fourteenth century, when various temples and shrines were working to promote their importance in defeating the Mongols fleets, Queen Consort Jingū’s mythic subjugation of the Korean kingdoms became a symbolic precedent for divine power to defeat foreign enemies. This resulted in multiple sites working to present their deities as related to Jingū, and the Itsukushima Daimyōjin nikki serves as one example of how this could be successfully accomplished. The second chapter, “Benzaiten Outside of Itsukushima: Spatial Networks and Deity Multiplication,” builds on chapter one by investigating how the collection of knowledge about 19 the gods led scholarly monks to ask new questions about the connections between sacred spaces. Beginning from comparisons of regional Benzaiten deities within the writings of scholarly monks, such as the Tendai monk Kōshū’s Keiran shūyōshū (Collection of Leaves Picked from Mountain Streams) and Monkan’s Kinpusen himitsu den (Secret Transmissions of Mt. Kinpu), the chapter examines how the universalizing logic of the honji-suijaku paradigm and the localizing ability of origin narratives needed to be brought together to explain the Benzaiten sites as both unique sacred spaces and as interconnected with one another. Recent scholarship has rightfully emphasized the fluidity and flexibility of honji-suijaku associations to connect transcendent Buddhas as the original form (honji) and Japanese kami deities as the local traces (suijaku) through which they manifest in the world. However, the focus on the honji-suijaku paradigm as the primary framework for understanding the gods has limited scholarly consideration of particularizing characteristics like spatial-specificity. Detailing the translocal contexts and interpretive strategies used to make sense of the seeming conflicts between local and transcendent sacred spaces, the chapter highlights the complementary interactions between Buddhist thought and origin narratives. The third chapter, “From the Origins of the Shrine to the Original Form of the Deity: Personalizing the Worship of Itsukushima,” presents locality as a feeling of belonging in order to reconsider relations between worshippers and sacred space apart from territorial identification. Focusing on the origin narrative subgenre of original form narratives (honji monogatari), and the Original Form of Itsukushima (Itsukushima no honji), the chapter examines how the genre reconstructs origin narratives to encourage popular audiences to develop emotional and personal bonds with a deity. Awareness of the scores of deities and worshippers across Japan could raise concerns over which deity to pray to, and why that deity would answer one’s prayers over those 20 of countless others. Personal connections between devotee and deity provided an answer, and explained why it would be worth making an arduous journey to a far-off sacred space like Itsukushima Shrine. Original form narratives combine knowledge of the deity, devotional practices, and moving stories to make deities feel close and accessible to all, regardless of spatial proximity. The chapter demonstrates how the Original Form of Itsukushima connects the narrative of the Itsukushima Deity’s arrival at Itsukushima from India with rites of sea travel performed at Itsukushima. The experiences of hearing the narrative and watching the rites reproduce the travels of the deity in connection with the worshipper’s own travels through which they make the space their own. In this manner, the chapter expands upon chapter three’s arguments on honji-suijaku and origin narratives to show how the localizing role of origin narratives also impacted devotional and ritual practices. The abstract thoughts of religious elites can seem divorced from the practices of the average worshipper, but the connections become clear when we examine how origin narratives were read, circulated, and transformed over time. Lastly, the fourth chapter, “The Rise and Fall of the Heike and the Surviving Association with Itsukushima Shrine,” reflects on the previous discussions of the interrelational networks negotiating the identities of the Itsukushima Deity to reevaluate the modern identity of the Itsukushima Deity as the tutelary god of the Taira household. I propose that the fall of the Taira did not mar the image of the Itsukushima Deity, because their relationship was rarely mentioned prior to the late medieval period. As the Heike’s discursive influence spread over the course of the medieval period, it became easier to positively associate the Taira with Itsukushima as an elite historical household. The chapter then examines how the understudied Nagatobon variant of the Tale of the Heike works to positively portray the Itsukushima Deity within the context of the Taira and the Tale of the Heike. The Nagatobon incorporated external narratives, including an 21 Itsukushima origin narrative, to alter Itsukushima’s image while remaining consistent with the other Heike variants. The addition of tangential narratives and associations enable the Nagatobon to present Itsukushima as a renowned sacred space of western Japan, and not just the god of the Taira. Tracing how some associations or narratives could be used to overshadow others, the chapter stresses the intertextual and referential connections between the gods, and within narratives, and therefore, the need for scholars to look for connections beyond any single discourse or body of sources. Resituating the Historical Identity of Itsukushima It should be clear from the discussions so far that this dissertation will not provide a comprehensive account of the history of Itsukushima Shrine. The local institutional circumstances of Itsukushima Shrine play relatively minor roles for the translocal usages of its origin narratives. Therefore, it is worth taking a little time here to consider how conventional understandings of the shrine’s history have worked outwards from the preconceived notion of Itsukushima as the tutelary shrine of the Taira. 35 The shrine’s survival after the fall of the Taira becomes easier to explain once we recognize that its medieval identity was not confined to that one historical moment. Instead, the ability of the Itsukushima Deity’s identity to adapt to changing circumstances, and the shrine’s translocal political and religious relations often played key roles in its continued successes. 35 Taira no Kiyomori likely developed connections with Itsukushima after becoming the governor of Aki Province in 1146, but his family already held regional stature in western Japan through their appointments as provincial governors since the time of his grandfather Masamori 平正盛. (Masamori’s family also held appointments in Ise Province, and this is the basis for the distinction of his lineage as the Ise Taira.) Therefore, while Kiyomori was the first to assert a special familial connection with Itsukushima, their relationship can also be seen as holding geopolitical meaning tracing back to earlier generations. Concerning the history of the Ise Taira through Masamori, Tadamori ( 平忠盛, 1096-1153, Kiyomori’s father), and Kiyomori, see Mikael S. Adolphson, "Fukuhara," in Lovable Losers: The Heike in Action and Memory, ed. Mikael S. Adolphson and Anne Commons (Honolulu: University of Hawai'i Press, 2015), 29-32. 22 Although most of the extant historical records regarding Itsukushima Shrine begin from the era of the Taira’s patronage (1160-1185), there are enough earlier sources to suggest that the shrine was already a prominent sacred space of Aki Province by the ninth century. Archaeological surveys conducted at Mt. Misen 弥山, the mountain behind Itsukushima Shrine, revealed pottery with the earliest being from the late sixth to early seventh century, as well as Buddhist ritual implements and traces of mountain ascetism from the late ninth century to the mid-Heian period (794-1185). 36 In the literary tradition as well, the poet Nōin ( 能因, 988-ca. 1050) listed Itsukushima as a famous place of Aki Province, demonstrating that it was known as an esteemed regional site prior to the Taira. 37 Written records from the Heian period further present Itsukushima as one of the two highest ranking shrines in Aki Province. The earliest record, a 7/17/811 entry in the Nihon kōki ( 日本後紀, 840), mentions both Itsukushima Shrine and Hayatani Shrine 速谷神社 as renowned shrines given offerings by the Ministry of Kami Affairs (jingikan 神祇官). Additionally, while the Engishiki ( 延喜式, 927) similarly lists both Itsukushima and Hayatani shrines as great shrines, it further elevates the Hayatani Deity as one of the gods to receive monthly offerings and 36 Considering that these objects were found midway up the mountain, Seo Shūzō suggests that the area was like ly a liminal place between the realm of the deities at the peak, and the realm of humans at the foot of the mountain. The midway point on the mountain would position religious practitioners close to the gods, while Itsukushima Shrine and Jigozen Shrine 地御前神社, the outer shrine across the shore from the island, provided more convenient locations for worship. Shūsō 妹尾周三 Seo, "Aki, Itsukushima ni okeru shin hakken no sairei iseki: Misen no chūfuku de hakken sareta gankai gun no kentō 安芸、厳島における新発見の祭祀遺跡：弥山の中腹で発見された岩塊群の検討," Museum, no. 639 (2012): 7, 12, 18-19, 35. 37 Nōin 能因, "Nōin utamakura (kōhon) 能因歌枕（広本）," in Nihon kagaku taikei 日本歌学大系, ed. Sasaki Nobutsuna 佐佐木信綱 (Tokyo: Kazama Shobō , 1973), 99. For later references to Itsukushima as an utamakura, see poem 3739 in Utamakura nayose ( 歌枕名寄, ca. 1303) and the section on famous islands in the Waka iroha (ca. 1198, 和歌色葉). Notably, the Waka iroha incorrectly locates Itsukushima in Sanuki Province (modern-day Kagawa Prefecture). Kuroda Akiko 黒田彰子, ed. Shinsen utamakura nayose 新撰歌枕名寄, 2 vols., vol. 2, Koten bunko 古典文庫 (Tokyo: Koten Bunko, 1989), 221; Jōkaku 上覚, "Waka iroha 和歌色葉," in Nihon kagaku taikei 日本歌学大系, ed. Sasaki Nobutsuna 佐佐木信綱 (Tokyo: Kazama Shobō , 1974), 164. 23 the feast of the first fruits (tsuki nami niiname 月次新嘗). 38 Overall, the pre-twelfth century records present both Itsukushima and Hayatani shrines as the two most prominent shrines of Aki Province. It was not until the fourth month of 1164 that a document recognized Itsukushima’s superiority as the ichinomiya 一宮 (principal shrine) of Aki Province, by which point Taira no Kiyomori had already begun to worship the Itsukushima Deity. 39 As Itsukushima’s fame continued to rise, Hayatani Shrine’s prominence in the written records diminished. From the Kamakura period (1185-1333) on, we most commonly see mentions of the Hayatani Deity within larger discussions of gods and shrines associated with Itsukushima Shrine. 40 Taira no Kiyomori’s worship of Itsukushima Shrine built on the site’s existing regional reputation and repositioned it within the geopolitical context of the central temples and shrines protecting the realm. As Heather Blair has argued, the Taira modeled their worship of Itsukushima Shrine on the politico-religious practices of the ruling elite to assert and distinguish the sacred authority of the Taira. On the one hand, Itsukushima’s distant maritime environment differentiated the shrine from the mountainous pilgrimage destinations of Kumano and Kinpusen favored by past sovereigns and members of the royal family. On the other hand, the dedication of the Heike nōkyō scrolls, and the formalization of two annual ceremonies, the Thousandfold Offering (senbu kuyō 千部供養 or sensō kuyō 千僧供養) of the third month and the Buddhist 38 In contrast to the Engishiki’s emphasis on Hayatani, a 1017 entry in the Sakeiki 左経記 only mentions Itsukushima as the representative shrine of Aki Province within a record of court ordered offerings to be made at powerful shrines in the provinces along the seven circuit routes (shichidō 七道). Matsuoka, Aki Itsukushimasha 安芸厳島社: 35. 39 This was four years after Kiyomori’s first recorded pilgrimage to Itsukushima in 1160. See the document for the commendation of the fields of Kiyohara Kiyosue ( 清原清末田畠等寄進状), dated 4/21/1164: doc. 37 in "Shinshutsu Itsukushima monjo 新出厳島文書," in Hiroshima kenshi: Kodai chūsei shiryō hen 広島県史. 古代中世資料編, ed. Hiroshima-ken 広島県 (Hiroshima: Hiroshima-ken, 1978); Kadoshige Hajime 角重始, "Aki no kuni 安芸国," in Chūsei shokoku ichinomiyasei no kisoteki kenkyū 中世諸国一宮制の基礎的研究, ed. Chūsei shokoku ichinomiya kenkyūkai 中世諸国一宮研究会 (Tokyo: Iwata Shoin, 2000), 482. 40 For example, there are Kamakura period records of the Hayatani Deity being enshrined at the Hirara estate (Hirara no shō 平良荘) possessed by Itsukushima Shrine. Kadoshige, "Aki no kuni 安芸国," 488. 24 Canon Assembly (issai-kyō e 一切経会) of the tenth month, matched the grandiose ritual and textual practices performed at Kumano and Kinpusen, presenting Itsukushima as their equal. 41 The Taira’s worship of Itsukushima within the religious-political culture of the Heian court helped the shrine to become recognized as one of the select religious institutions protecting the realm. Nevertheless, it was still identified by its location in western Japan, and thus, away from the court. In 1175, retired sovereign Go-Shirakawa selected the Itsukushima Deity as one of the gods to be enshrined in the comprehensive shrine (sōsha 惣社) he established at his temple, the Rengeōin 蓮華王院. The comprehensive shrine was designed to symbolically encompass all the deities throughout the Japanese archipelago by enshrining the deities of twenty-one of the twenty-two shrines protecting the realm (Ise being the exception) along with four regional deities representing the four directions: Hinokuma (south), Atsuta (east), Itsukushima (west), and Kehi (north). 42 While still being characterized as a regional deity, Itsukushima’s selection positioned it above the other deities of western Japan, and alongside the central deities of the twenty-two shrines, almost all of which were located close to the capital. 43 In 1179, the Council of State considered adding Itsukushima to the twenty-two shrines protecting the realm, despite the 41 After first being held by Retired Sovereign Shirakawa, Thousandfold Offerings became a standard rite sponsored by the standing or retired sovereign to protect the realm. Kiyomori then appropriated the rite by changing the location to Fukuhara or Itsukushima, the time to the third and tenth months, and the recited sutra to the Lotus Sutra. Heather Blair, "Rites and Rule: Kiyomori at Itsukushima and Fukuhara," Harvard Journal of Asiatic Studies 73, no. 1 (2013): 29-31. 42 Uejima, Nihon chūsei shakai no keisei to ōken 日本中世社会の形成と王権, 393; Heather Blair, "Rites and Rule: Kiyomori at Itsukushima and Fukuhara," Harvard Journal of Asiatic Studies 73, no. 1 (2013): 15-16. 43 Allan Grapard organizes the twenty-two shrines into three hierarchical levels based on location: the upper seven shrines are mostly located in the area surrounding the capital (with the only exceptions being the Ise and Kasuga shrines connected with the royal family and the powerful Fujiwara household respectively); the middle seven were mostly located in Yamato Province, the former political center (in modern day Nara Prefecture); and the lower eight included farther away shrines, such as Hirota Shrine in Ōmi Province. T hat being said, even the outermost shrines of the twenty-two shrine system were not as distant as Itsukushima Shrine. Allan G. Grapard, "Institution, Ritual, and Ideology: The Twenty-Two Shrine-Temple Multiplexes of Heian Japan," History of Religions 27, no. 3 (1988): 250- 51. 25 geopolitical preference shown towards shrines close to the capital. 44 In the end, though, the Council of State decided against adding Itsukushima Shrine. The Taira’s worship successfully incorporated Itsukushima into the framework of sacred sites particular to elite households, but it did not fully negate the shrine’s disparate location. Itsukushima’s distance from the capital could be used to draw comparisons with Kumano and Kinpusen as pilgrimage destinations, but it was harder to fully associate it with the sites for the protection of the realm based around the capital. Itsukushima Shrine did not lose its high standing after the Taira’s defeat in the Genpei War in 1185, because the shrine’s image and political importance as the representative shrine of western Japan was not reliant on the Taira. The newly established Kamakura Bakufu government did not punish Itsukushima for allying with the Taira, and instead worked to procure the shrine’s support through offerings for prayers for Minamoto no Yoritomo’s victory over the Northern Fujiwara in 1189. 45 Once we look beyond Itsukushima Shrine’s association with the Taira, we can find similarities between the Bakufu’s support towards Itsukushima and Munakata 宗像 shrines. Munakata Shrine also sided with the Taira, and there was even a petition sent to the Bakufu requesting that Munakata Shrine estates be seized as lands belonging to the Taira. 44 Heather Blair explains the timing of the discussion of adding Itsukushima to the twenty-two shrines by noting that Taira no Kiyomori’s grandson, Prince Tokihito ( 言仁, 1178-1185), was born the year prior and was next in line to become the sovereign (enthroned as Antoku Tennō 安徳天皇). Blair, "Rites and Rule: Kiyomori at Itsukushima and Fukuhara," 38. 45 The Bakufu implementation of the offerings to Itsukushima worked to shift power away from the head priest, Saeki no Kagehiro, and promote lower ranking priests who had more to gain from building relations with Kamakura. In 1189, a lower fifth ranked Minamoto lord ordered Saeki Kiyomoto 佐伯清元 to arrange for kagura performances and prayers to be made for the sake of Minamoto no Yoritomo during the military campaign against the Northern Fujiwara. As a reward for these rites, the Minamoto offered tax exemptions for the various services and dues from the locally possessed rice fields and residences of the village of Takamiya. Addressing the order to Kiyomoto helped to shift power away from Kagehiro without removing him from his position as head priest. By 1196, Kiyomoto was the highest-ranking shrine official outside of Kagehiro. The bakufu also appointed Saeki Tamehiro 佐伯為弘 as a land steward (jitō 地頭) in 1216, further connecting Kagehiro’s relatives with Kamakura. See doc. 22 and 23 in "Shinshutsu Itsukushima monjo 新出厳島文書," 181-82; Matsuoka, Aki Itsukushimasha 安芸厳島社: 97; Jeffrey P. Mass, "Patterns of Provincial Inheritance in Late Heian Japan," Journal of Japanese Studies 9, no. 1 (1983): 89, note 78. 26 However, the Bakufu declined the petition and granted the Munakata shrine family authority over their estates. The editors of the Munakata city history argued that the Bakufu seized estates from powerful families that had supported the Taira but forgave less authoritative figures such as the Munakata Shrine family. Because the Bakufu’s primary goal was to secure ruling authority throughout the realm, it made sense to only punish those whose power was threatening to them, and to promote lower level officials who could help them gain influence in faraway areas like Kyushu. 46 The same explanation fits for Itsukushima Shrine; the Bakufu did not punish Itsukushima Shrine because there was more to be gained from acquiring the support of the shrine family. Itsukushima’s relations with both the Taira and the Minamoto Bakufu only makes sense when we move away from center-based modes of authority and consider the import of connections with regional sites. In this manner, the study of regional sacred sites like Itsukushima provide a lens for reexamining relations between the capital and the provinces. 47 From this perspective, it is helpful to consider David Spafford’s arguments against insider/outsider dichotomies for understanding spatial identity. Spafford similarly calls attention to the issues that arise from assuming oppositional relations between court culture and local identity. 48 Individuals in the province were 46 The powerful families punished by the Kamaura bakufu included: the shrine family of Usa Hachiman Shrine (especially the head priest Usa no Kimimichi 宇佐公通), the Harada 原田, Yamaga 山鹿, Itai 板井, and Kikuchi 菊池. Munakata shishi. tsūshi hen 宗像市史. 通史編, 2 vols., vol. 2 (Munakata: Munakata-shi, 1999), 317-19; Nakano Hatayoshi 中野幡能, Usagū 宇佐宮 (Tokyo: Yoshikawa Kō bunkan, 1985), 117-19. 47 For studies arguing for moving beyond the center and periphery model in Japanese studies, see Mikael S. Adolphson, Edward Kamens, and Stacie Matsumoto, eds., Heian Japan, Centers and Peripheries (Honolulu: University of Hawai'i Press, 2007); Joan R. Piggott, ed. Capital and Countryside in Japan, 300-1180 : Japanese Historians in English (Ithaca, NY: East Asia Program, Cornell University, 2006). 48 While working on a different time and place, Uejima Susumu similarly argues for a more complex and complementary relationship between the center and the provinces. In contrast to earlier Marxist arguments about the use of religion by the government to control the people, Uejima Susumu contends that the religious systems undergirding medieval ruling authority developed in no small part through the activities of custodial provincial governors (zuryō 受領) to connect the center and the provinces. In other words, the court only had the power to present models of behavior and thought and had to rely on people from the provinces to diffuse and participate in those practices. Uejima, Nihon chūsei shakai no keisei to ōken 日本中世社会の形成と王権, 229-31. 27 able, and often eager, to participate in court culture and present their own environs through literary tropes from the center. Additionally, writers from the center were capable of writing about provincial places with rhetoric and details outside of poetic conventions on famous places. 49 The realities of spatial relations are often more complicated and less coherent than our preconceived notions would lead us to believe. We need to find sources that bring together the court and the provinces and let them replace capital-centric theories as our guides. Itsukushima Shrine’s Medieval Period History The post-Taira Kamakura period history of Itsukushima Shrine can provide a brief glimpse into the connections between the provinces and the political centers of Kyoto and Kamakura. Although historical scholarship on Itsukushima often takes the estates (shōen 荘園) gained during the Taira’s patronage as the basis for its later economic survival, estate support was not always sufficient. 50 Regional institutions like Itsukushima often faced more difficulties than central elite temples and shrines in procuring resources and tax exemptions from estates. 51 Therefore, large scale projects, such as the reconstructions of shrine buildings after fires in 1207 and 1223, often required appeals for support to provincial, court, and shogunate officials. The Bakufu’s support helped it to strengthen its connections with Itsukushima Shrine and the region. The warrior government placed one of its officials, Fujiwara no Chikazane 藤原親実, as the head priest of Itsukushima after the Jōkyū disturbance in 1221, and then gave him a second 49 David Spafford, A Sense of Place: The Political Landscape in Late Medieval Japan (Cambridge, MA: Harvard University Press, 2013), 61-62. 50 Tamura Hiroshi 田村裕, "Itsukushima sharyō shōen no keisei to kurashi ki ni tsuite 厳島社領荘園の形成と倉敷について," in Naikai chiiki shakai no shiteki kenkyū 內海地域社会の史的研究, ed. Matsuoka Hisato 松岡久人 (Yamaguchi-ken Tokushima-shi: Matsuno Shoten, 1978), 65. 51 Adolphson, The Gates of Power: Monks, Courtiers, and Warriors in Premodern Japan, 62. 28 appointment as the provincial constable (shugo 守護) of Aki Province in 1235 to help compel the provincial government to fulfill their responsibilities for repairing the shrine. 52 Chikazane continued to reside in Kamakura, working together with the Saeki family who performed rites and administrative duties at the shrine. 53 This structure provided a direct connection between the shrine and the warrior government, providing support for the shrine and expanding the reach of the Bakufu in the provinces. The developing relationship with the Kamakura Bakufu led to ritual changes at Itsukushima positioning the shrine as a sacred site offering prayers for the Bakufu. Notably, the 1232 establishment of an annual ceremony opening the shrine hall (mitobiraki sechie 御戸開節会) to pray for the family of the shogun (shōgun go kitōsho 関東御祈祷), replaced the Thousandfold Offering established by Kiyomori. 54 Beyond the change in sponsorship, the move from a Thousandfold Offering, traditionally sponsored by the sovereign, to a rite for the shogunate, further emphasized the warrior government’s individual authority to protect the 52 Matsuoka, Aki Itsukushimasha 安芸厳島社, 99-103; Kadoshige Hajime 角重始, "Kamakura, Nanbōkucho ki Itsukushima sha shihai no tokushitsu 鎌倉・南北朝期厳島社支配の特質," in Itsukushima shinkō jiten 厳島信仰事典, ed. Nozaka Motoyoshi 野坂元良 (Tokyo: Ebisu Kōshō Shuppan, 2002), 266. 53 Chikazane exerted his power as head priest by issuing orders to the local administrative office, the sō-mandokoro 惣政所, run by Saeki Kagesada 佐伯景貞. See Kadoshige, "Kamakura, Nanbōkucho ki Itsukushima sha shihai no tokushitsu 鎌倉・南北朝期厳島社支配の特質," 266-67. 54 Yunoue Takashi mentions the phrase, “site of prayers for the family of the shogun” (shōgunke go kitōsho 将軍家御祈祷所) as one of several terms used interchangeably to describe prayers for the Bakufu from designated temples and shrines. While the records about the mitobiraki sechie rite at Itsukushima do not include the word “site” ( 所), the rest of the phrasing is identical (shōgunke go kitō). Itsukushima might not have been formally recognized as a shogunate prayer site, but Itsukushima presented itself as performing the same function, and the Bakufu recognized the rites with offerings of swords to Itsukushima between 1239-1246. Yunoue Takashi 湯之上隆, Nihon chūsei no seiji kenryoku to Bukkyō 日本中世の政治権力と仏教 (Kyoto: Shibunkaku Shuppan, 2001), 31. Regarding the introduction of the mitobiraki sechie as an annual rite for the shogunate see doc. 111 in "Shinshutsu Itsukushima monjo 新出厳島文書," in Hiroshima kenshi: Kodai chūsei shiryō hen 広島県史. 古代中世資料編, ed. Hiroshima- ken 広島県 (Hiroshima: Hiroshima-ken, 1978), 353. For records of the offerings of swords during the second months of 1239, 1241, 1242, and 1243, see docs. 30, 32, 33, and 36 in "Go hanmotsu chō 御判物帖," in Hiroshima kenshi: Kodai chūsei shiryō hen 広島県史. 古代中世資料編, ed. Hiroshima-ken 広島県 (Hiroshima: Hiroshima, 1978), 23-26. For a fifth document of donation from the second month of 1246 as well as a copy of a receipt of the donation, see docs. 81-82 in "Shinshutsu Itsukushima monjo 新出厳島文書," 290. 29 realm. In this manner, the prayers for the shogunate and the protection of the realm mirrored the prayers for the sovereign by royally designated temples (chokuganji 勅願時). This enabled the warrior government to appropriate the legitimizing practices of the court and emphasize the geographic scope of the Bakufu’s religious practices and authority. 55 The Kamakura Bakufu began establishing direct connections with Itsukushima Shrine and other regional sacred sites in the early thirteenth century, but it was the Mongol invasions that led them to request prayers from sites throughout the provinces. The court also issued prayers against the Mongols, but they were primarily issued to the twenty-two shrines protecting the realm, and elite power-bloc temples near the capital. 56 In contrast, the Bakufu issued the first orders for prayers to subjugate the Mongols to temples and shrines throughout the realm in the ninth month of 1275. 57 Itsukushima Shrine had already received the order on 12/2/1274, likely expediated by head priest Fujiwara no Chikisada’s 藤原親定 close connections with the Bakufu. 58 To simplify the process of conveying orders for prayers throughout the realm, the warrior government limited them to the kokubunji (state monasteries 国分寺) and ichinomiya of each province. Itsukushima Shrine’s rank as icihinomiya of Aki Province, close connections with Kamakura, and strategic location on the San’yōdō 山陽道 circuit along the Seto inland sea 55 Yunoue, Nihon chūsei no seiji kenryoku to Bukkyō 日本中世の政治権力と仏教, 32-36, 40-45. 56 Concerning requests for prayers against the Mongols from members of the court to the twenty-two shrines and elite temples (between 1268-1293), see Aida Shintarō 相田二郎, Mōko shūrai no kenkyū 蒙古襲来の硏究, Zō hoban ed. (Tokyo: Yoshikawa Kō bunkan, 1982), 59-77. 57 Kaizu Ichiro 海津一朗, "Ikoku kōfuku kitō taisei to shokoku ichinomiya kōgyō 異国降伏祈祷体制と諸国一宮興行," in Chūsei ichinomiyasei no rekishiteki tenkai. ge (sōgō kenkyū hen) 中世一宮制の歴史的展開. 下( 総合研究編), ed. Ichinomiya kenkyūkai (Tokyo: Iwata Shoin, 2004), 292-93. 58 This is the earliest record of an offering for prayers against the Mongols by the Bakufu. Kunihiko 伊藤邦彦 Itō, "Kamakura bakufu 'ikoku kōfuku' kitō to Ichinomiya: Shugo seido to no kankei wo chūshin ni 鎌倉幕府「異国降伏」祈禱と一宮：守護制度との関係を中心に," in Chūsei ichinomiyasei no rekishiteki tenkai. ge (sōgō kenkyū hen) 中世一宮制の歴史的展開. 下( 総合研究編), ed. Ichinomiya kenkyūkai (Tokyo: Iwata Shoin, 2004), 262. See doc. 11766 in Takeuchi Rizō 竹內理三, ed. Kamakura ibun 鎌倉遺文, 42 vols. (Tokyo: Tokyōdō Shuppan, 1971). For an English translation, see doc. 58 in Thomas Conlan, In Little Need of Divine Intervention: Takezaki Suenaga's Scrolls of the Mongol Invasions of Japan (Ithaca, NY: East Asia Program, Cornell University, 2001), 247-48. 30 helped it to receive regular offerings for prayers in 1274, 1289, and 1293. 59 The Kamakura warrior government’s religious response against the Mongols supported Itsukushima Shrine, but it also created similar connections with temples and shrines throughout the realm. The warrior government’s administration of prayers throughout the archipelago to defeat the Mongol demonstrated their power to protect the realm, but complications arose once the temples and shrines began to appeal for rewards. The gods were understood as fighting in the battles themselves, and therefore, deserved compensation resembling that given to warriors for military service. 60 By 1284, the Bakufu became directly involved in collecting information on the kokubunji temple and ichinomiya shrine of each province in order to properly issue lands and positions as rewards. 61 Nevertheless, the scale of the warrior government’s prayers made it difficult to respond in a timely manner. Takeo Shrine 武雄神社, for example, appealed to the Bakufu in 1309 because it still had not received rewards despite being specified as one of six shrines in Kyushu noted for its miraculous deeds against the Mongols (along with Sumiyoshi in Echizen Province, Kōra, Aso, Kag ami, and Kawakami shrines). 62 Even after limiting the sites that merited remunerations, the Bakufu still struggled with the task of dispensing rewards. 59 See docs. 83-85 in "Shinshutsu Itsukushima monjo 新出厳島文書," 291-92. Also see docs. 44-46 in "Go hanmotsu chō 御判物帖," in Hiroshima kenshi: Kodai chūsei shiryō hen 広島県史. 古代中世資料編, ed. Hiroshima-ken 広島県 (Hiroshima: Hiroshima, 1978), 29-31. In English, see docs. 58, 60-63 in Conlan, In Little Need of Divine Intervention: Takezaki Suenaga's Scrolls of the Mongol Invasions of Japan, 247-50. 60 Thomas Conlan, State of War: The Violent Order of Fourteenth-Century Japan (Ann Arbor: Center for Japanese Studies, University of Michigan, 2003), 174, 79. 61 Writings to the Bakufu with the information sometimes began by repeating the specific information requested by the Bakufu as: “historical details, current circumstances, superintendent, and tax-exempt lands” (ōko shisai, tōji shidai, kanrei ni oyobi menden 往古子縮、当時次第、管領仁及免田). For one example, see the letter sent to the Bakufu from Satsuma Tenmangū Shrine copied in a later 1321 petition: doc. 27819 in Takeuchi, Kamakura ibun 鎌倉遺文, v. 36, pp. 112-15. 62 Doc. 23721 in ibid.; Aida, Mōko shūrai no kenkyū 蒙古襲来の硏究, 85; Kyotsu Hori, "The Mongol Invasions and the Kamakura Bakufu" (Ph.D. diss., Columbia University, 1967), 175-76. 31 The opportunity to receive rewards and prestige led to numerous petitions to the governments, and competition for recognition for defeating the Mongols. 63 So many temples and shrines sent appeals to the Bakufu proclaiming the superior efficacy of their prayers and the martial power of their deities that Kaizu Ichirō described them as sounding clichéd. 64 The “divine winds” that sank the Mongol ships were a clear sign of the gods’ power to protect the realm, but how could any sacred site demonstrate the power of its deities apart from those of the countless other temples and shrines? One strategy was to present the appeals as longstanding requests for restoration assistance and try to make the requests sound less opportunistic. 65 Some religious institutions used the information about the site sent to the Bakufu to try and distinguish the site’s efficacy against the Mongols, or to argue for their proper place as the province’s ichinomiya shrine. 66 Itsukushima Shrine held the advantages of being the established ichinomiya of Aki Province since 1164, and possessing close connections with the Kamakura Bakufu, but as we shall see, Itsukushima still became involved in comparisons with other temples and shrines 63 Religious institutions made so many legal appeals to the Bakufu that Inaba Nobumitsu describes the warrior government becoming “a ticket window for temple and shrine litigations.” Inaba Nobumichi 稲葉伸道, "Chūsei no kokka to jisha: Ōchō to bakufu no jisha seisaku 中世の国家と寺社：王朝と幕府の寺社政策," Nenpō chūsei shi kenkyū 28 (2003): 21. 64 Kaizu Ichirō 海津一朗, Chūsei no henkaku to tokusei: Shinryō kōgyōhō no kenkyū 中世の変革と徳政 : 神領興行法の研究 (Tokyo: Yoshikawa Kō bunkan, 1994), 226-27. 65 Itsukushima Shrine, for example, requested tax exemptions to fund repairs on 5/9/1292, by stressing that they did not receive enough support to complete restorations when the Bakufu agreed to their previous appeal in 1235. The document hints at their service in defeating the Mongols, noting their faithfulness in performing prayers, but uses the precedent of the 1235 appeal to distinguish their case from the others being sent to the court at the same time. This appeal was accepted two years later (1294) and the repairs were completed in 1300. See docs. 17889, 18511, and 20419 in Takeuchi, Kamakura ibun 鎌倉遺文, v. 23, pp. 237-38; v. 24, p. 125; v. 27, p. 28-32. 66 Ihara Kesao uses the case of Satsuma Tenmangū to describe the complex maneuvers in seeking rewards. While Satsuma Tenmangū itself was not the ichinomiya, it presented itself as part of the kokubunji institution, and presented their histories and accomplishments together in a document sent to the warrior government. Ihara Kesao 井原今朝男, "Chūsei no kokuga jisha taisei to minshū tōgō girei 中世の国街寺社体制と民衆統合儀礼," in Chūsei ichinomiyasei no rekishiteki tenkai. ge (sōgō kenkyū hen) 中世一宮制の歴史的展開. 下( 総合研究編), ed. Ichinomiya kenkyūkai (Tokyo: Iwata Shoin, 2004), 230, 32-33. For another example, see Kaizu Ichiro’s discussion of how Amano Shrine contested Hinokuma Shrine’s position as ichinomiya of Kii province and presented its history and recent importance in performing prayers against the Mongols to argue for its own rightful place as ichinomiya. Kaizu, "Ikoku kōfuku kitō taisei to shokoku ichinomiya kōgyō 異国降伏祈祷体制と諸国一宮興行," 306-14. 32 throughout the Japanese archipelago. As a regional and pilgrimage site, Itsukushima needed to not only construct but also maintain an image of power and importance vis-à-vis other sacred places. Blurring the Place of Local Deities in Buddhism and Kami Worship My translocal study of Itsukushima Shrine challenges the preconceptions within modern scholarly categorizations by investigating how people in premodern Japan raised and resolved questions about their religious knowledge. When we look at the production of knowledge rather than the knowledge itself, we can better pinpoint the places where our modern viewpoints can lead us astray. The following chapters will argue for the need to blur conventional dichotomies, such as: Buddhism/Shinto, local/universal, center/periphery, elite/popular knowledge, knowledge/culture. 67 The imprecise boundaries between these paired concepts were all concerns for understanding the identity of Itsukushima Shrine and its deity in relation to the other divinities in Japan, but not always in the ways we would expect. The relationship between Buddhism and kami worship provides a suitable starting point for blurring boundaries. Building on the ideas of Fabio Rambelli, I will argue that comparisons of kami deities forced religious thinkers to expand the conceptual reach of Buddhist lines of reasoning by incorporating alternative discourses about local sacrality. Rambelli reveals how several typologies for kami (e.g., primordial deities) raised issues which confounded Buddhist logic, and led to “quasi-Buddhist kami” who were increasingly understood outside of Buddhist 67 These are all known topics in the study Japanese religions, and new publications are constantly providing introducing nuanced perspectives. However, it is sometimes easier for us to think about our sources in regard to our fields’ theories and debates rather than contemporary perspectives. Tracing the connections between networks (e.g., of information about associated gods) is not a solution, but its ability to distance methodology from scholarly motivations is significant. 33 frameworks. 68 His discussion briefly mentions the regional significance of the quasi-Buddhist kami, and I further raise the spatial specificity of the kami as a similar typology crossing the limits of the Buddhist honji-suijaku framework. Descriptions of a single deity could overlook questions about the relationship between the universal equivalences of worldly deities as manifestations of a transcendent Buddhas, and their spatial particularity as the deities of individual temples and shrines. However, such issues became unavoidable once scholarly monks began comparing origin narratives and tracing the relationships between the deities. Additionally, this matter of worldly specificity was not unique to the kami, but also applied to the individual manifestations of Buddhas and bodhisattvas at a sacred place. In this manner, I shift focus away from a presumed Buddhist-kami division, and towards specific controversies about shared identity and individuality for sacred entities in the world. Moving away from the Buddhist-kami binary could demonstrate the broader relevancy of studies on the kami for scholars of Buddhism, regional cults, translocal culture. From the perspective of religious knowledge, questions about the kami and their spatial significance could also influence Buddhist beliefs and practices. The movement of people, sacred objects, and deities can raise related questions about the relationship between sacred spaces in the world, the transcendent, and persons located near and far. Despite increasing recognition of the complicated 68 Notably, Rambelli’s discussion of the threefold kami typology first found in the Nakatomi no harae kunge (late twelfth century) shows that tensions between the unifying tendencies of original enlightenment thought and the need for hierarchical relations between the kami deities existed prior to the Mongol invasions. Its typology gives the two Ise deities as the kami of original enlightenment, other elite deities as the “kami actualizing enlightenment” (shikakushin 始覚神), and dangerous and malevolent deities as the “kami of non-enlightenment” (fukakushin 不覚神). In this manner, it preserves the superiority of the Ise deities, while still recognizing the inherent enlightenment of other kami deities. See Fabio Rambelli, "Re-positioning the Gods: "Medieval Shintō" and the Origins of Non- Buddhist Discourses on the "Kami"," Cahiers d'Extrême-Asie 16 (2006): 307-09, 21; "Before the First Buddha: Medieval Japanese Cosmogony and the Quest for the Primeval Kami," Monumenta Nipponica 64, no. 2 (2009): 244- 47. 34 relations between Buddhism and kami worship, the categorical separation of Buddhist and kami is still deeply rooted in the study of Japanese religions. Kuroda Toshio transformed the study of the kami when he argued that medieval Shinto was not an independent religious tradition, but rather, existed as a part of Buddhism. 69 However, as Helen Hardacre succinctly stated, “Kuroda may have succeeded too well.” 70 The historicization of Shinto within the context of medieval Buddhist culture countered conventional notions of Shinto as the indigenous religion of Japan, and as the basis of nationalistic interpretations of Japan as the realm of the gods (shinkoku 神国) ensuing from the Mongol invasions. 71 Recent scholarship has tempered Kuroda’s theory of medieval Shinto as subsumed within Buddhism by stressing that the diversity of religious practices and imaginaries concerning the kami, and thus showing how the kami could transcend Buddhist theorizations. These positions highlight the importance of studying kami worship as it was practiced, but it preserves Kuroda’s notion of a Buddhist “intellectual hegemony,” that remained unchallenged until the development of the Yoshida Shinto school in the late fifteenth century. 72 Therefore, scholars read treatises on the kami as either based in Buddhist thought or arguing against it. Either way, kami worship remains secondary and tied to Buddhism. John Holt’s study of Buddhism and local cults in modern-day Lao argues against the assumption of a unidirectional Buddhist subjugation on 69 Toshio Kuroda, James C. Dobbins, and Suzanne Gay, trans., "Shinto in the History of Japanese Religion," Journal of Japanese Studies 7, no. 1 (1981): 11-14. 70 Helen Hardacre, Shinto: A History (New York: Oxford University Press, 2017), 5. 71 Toshio Kuroda and Fabio Rambelli, "The Discourse on the "Land of Kami" (Shinkoku) in Medieval Japan: National Consciousness and International Awareness," Japanese Journal of Religious Studies 23, no. 3/4 (1996): 371, 79-81, 83; Hiroo Satō , "The Emergence of Shinkoku (Land of the Gods) Ideology in Japan," in Challenging Paradigms, Buddhism and Nativism: Framing Identity Discourse in Buddhist Environments, ed. Henk Blezer and Mark Teeuwen (Leiden: Brill, 2013), 46. 72 See Kuroda, Dobbins, and Gay, "Shinto in the History of Japanese Religion," 18. Hardacre, Shinto: A History, 150; Rambelli, "Re-positioning the Gods: "Medieval Shintō" and the Origins of Non -Buddhist Discourses on the "Kami"," 306, 12; William R. LaFleur, The Karma of Words: Buddhism and the Literary Arts in Medieval Japan (Berkeley: University of California Press, 1983), 13. 35 local cults, by showing how the local religious culture influenced Buddhist thought in the area. 73 This is a step in the right direction, but we can go even further to reconsider Buddhist and kami deities within the same ontological or interpretive frameworks. Conceptualizing the Buddhist and kami deities in terms of their spatial relations can provide a perspective for examining the various gods from the same level. Satō Hiroo challenged the Buddhist/kami binary relation with an alternative paradigm of “otherworldly ‘saving deities’ and this-worldly ‘wrathful deities.’” Seen as a spatial framework wherein worldly Buddhist and kami deities function on the same grounds, Satō has provided scholars with a roadmap for distancing ourselves from assumed categorical divisions. The medieval oaths (kishōmon 起請文) examined by Satō shows a hierarchical distinction ra ising the otherworldly Buddhist deities over the worldly deities, but they cannot say anything about the differentiations (or lack thereof) between local deities. He briefly considers the possibility for specialized powers to differentiate worldly deities, but he leaves us more questions than answers. Notably, he mentions in passing, “The idea that it was up to the people which deity they chose to address with their problems quite naturally led to a pronounced weakening of the authority of the deities.” 74 This is an issue that would have been of concern to religious clergy who needed to be able to explain why their deity deserves to be worshipped over the countless others, but how could this be done if there was no hierarchical framework for comparing worldly deities? 73 Bernard Faure briefly remarks along the same lines that honji-suijaku associations developed earlier to assimilate kami into the Buddhist pantheon, but ended up providing new means for local cults to influence Buddhism. See John Holt, Spirits of the Place: Buddhism and Lao Religious Culture (Honolulu: University of Hawai'i Press, 2009), 7; Faure, Gods of Medieval Japan: Volume 2, Protectors and Predators, 341. 74 Hiroo Satō , "Wrathful Deities and Saving Deities," in Buddhas and Kami in Japan: Honji Suijaku as a Combinatory Paradigm, ed. Mark Teeuwen and Fabio Rambelli (New York: RoutledgeCurzon, 2003), 106-07, 11- 14; Satō, Kami, hotoke, ōken no chūsei 神・仏・王権の中世, 50-67. 36 Bernard Faure’s arguments about the pantheons of the gods propose an answer to this question. In addition to the explicit hierarchies given in texts in oaths and other medieval texts, Faure also examines implicit pantheons enabling the production of fluid and hybrid identities by relating deities through networks of association. Regardless of formal categories, similarities and differences in specialized power (i.e., function) or symbolic image could be linked together to disentangle the complex relations between the various divinities. Faure supports Satō’s arguments against a Buddhist-kami dichotomy, while questioning the rigidity of his paradigm of mundane/wrathful and supramundane/saving deities. 75 I agree with Faure about the central importance of the interpretive variability offered by the implicit pantheons, but I believe that Satō’s worldly/otherworldly distinction still needs to be recognized as a central and established pantheon of the times. Whether or not the hierarchization of the gods as transcendent and worldly fits every case, it shows how contemporary thinkers were trying to make sense of the various deities, and how controversies were managed within that work. The sources connecting the deities in networks of association often leave the earlier interpretive work unsaid. Therefore, we must search for clues about the production of knowledge about the networks of associations beyond discussions of deities and their relations. The local identity of deities and sacred space was especially difficult to explain through associations. Local manifestations of the same transcendent Buddha naturally shared functional and symbolic traits, making them difficult to differentiate within the frameworks laid out by Faure and Satō. Place -based identities marked individuality, more often than similarity, hampering its incorporation into networks of association. The reference to the place of Itsukushima Shrine in the name of the Itsukushima Deity distinguishes her as the master of her 75 Faure, Gods of Medieval Japan: Volume 2, Protectors and Predators, 335, 42. 37 domain. Faure is right to describe the name of a deity as “the mirage of a god’s individuality,” but it was an important mirage, nonetheless. The Itsukushima Deity was understood by her interrelations with other deities and sacred spaces, and in this manner, was able to transcend the geographic boundaries of its island shrine. Nevertheless, it still needed to be identifiable as individual and spatially fixed. My examination of the Itsukushima Deity’s associations with other local deities is designed to test what happens when we compare the relations between worldly deities instead of those between worldly and otherworldly deities. Honji-suijaku arguments could universalize local deities as manifestations of transcendent Buddhist deities, but this diminished their spatial individuality through abstraction. In other words, religious institutions arguing for the importance of their deities needed interpretive frameworks with the flexibility to relate deities of various sites as similar in certain ways, and different in others. Kuroda Toshio hints at this issue when he describes the ideology of Japan as the realm of the gods, saying “Although such a notion was too abstract to comprise a realistic nation-concept, one can perceive in it an engi-like, mythological understanding that differed from ‘classic’ exo-esotericism, and that incorporated a stress on something peculiarly Japanese.” 76 The universalizing Buddhist logic underlying the realm of the gods could not explain Japan as different. I agree with Kuroda’s recognition of the divergence between Buddhist conceptual abstraction and mythological particularization. However, while Kuroda explains the shift towards myths as “nonrational” (botsuronri teki 没論理的) and “vulgarizing” (hizokuka 卑俗化) Buddhist doctrines, I contend that mythic narratives 76 Kuroda and Rambelli, "The Discourse on the "Land of Kami" (Shinkoku) in Medieval Japan: National Consciousness and International Awareness," 376. 38 (and other visual, performative, and ritual frameworks) connected with Buddhist thought to expand the interpretive potential of the religious episteme. 77 Origin Narratives for Connecting Intellectual and Vernacular Religion I propose that temple and shrine origin narratives mediated between intellectual discourses on the locality and universality of sacred spaces. Origin narratives were told to diverse audiences ranging from low ranking audiences who would assemble before a local preacher or performer, to nobles in the court gathering together to read a beautifully illustrated scroll (emaki 絵巻). 78 As such, origin narratives tend to be examined in terms of popular religion and religious culture rather than the musings of scholarly monks. However, these modern categories obstruct our view of how narratives (and other works of literature, visual culture, and performance) could not only reflect, but moreover, influence intellectual and religious thought. Scholarly monks not only used origin narratives for preaching and proselytizing, but also incorporated them into their studied writings on the gods. The idea of vernacular religion offers a starting place for considering origin narratives as more than just articulations of local forms of religious practice and belief, or conversely, localizations of an implicit orthodox tradition. Leonard Primiano proposed vernacular religion as a theoretical approach for moving beyond value-laden categorical divides between institutional and popular religion by shifting focus to religion as lived by individuals. 79 When he reflected on 77 Ibid., 378-79; Kuroda Toshio 黒田俊雄, "Chūsei no shinkoku shisō: Ko kka ishiki to kokusai kankaku 中世の神国思想：国家意識と国際感覚," in Kuroda Toshio chosakushū 黒田俊雄著作集 (Kyoto: Hō zō kan, 1994), 168. 78 To be clear, this is not to say that the audiences and format of the received audiences was strictly divided by status. Major religious ceremonies often brought together diverse audiences together and this would be an apt time for the telling of origin narratives and image explications (etoki 絵解き) of hanging scrolls depicting origin narratives and other miraculous tales. 79 Leonard Norman Primiano, "Vernacular Religion and the Search for Method in Religious Folklife," Western Folklore 54, no. 1 (1995): 39-44. 39 the reception of vernacular religion almost twenty years after his original article, he was surprised to find scholars using it as an alternative term to pair with institutional religion. His ideas led people to recognize negative connotations preserved in terms like “folk” or “popular” religion and to pay more attention to religion outside of elite institutions and doctrinal formulations. Such developments provide a valuable change in perspective, but they miss Primiano’s larger argument that all religion can be understood to be vernacular. Every institution is made up of people with the creative potential to raise new ideas and practices. 80 While acknowledging the power of established institutions and traditions, he argues that the influences between individuals and their environments are bidirectional. 81 Similarly, religious knowledge needs to be considered in multidirectional relationships with diverse persons, modes of discourse, and cultural circumstances. Religious knowledge likewise needs to be considered in multidirectional relationships with diverse persons, modes of discourse, and cultural circumstances. There remain opportunities for applying vernacular religion as a method to further question preconceived differences between the elite/center/intellectual and the local/peripheral/cultural. Post-colonial scholars have critiqued the coloniality of knowledge as epistemically structured around the center, but such arguments maintain focus on the center and can neglect the potential of the periphery to participate and influence the production of knowledge despite the unequal conditions. Asymmetrical power relations did not always leave those in the periphery powerless, but instead required engagement with different practices and epistemic cultures. 82 Therefore, it is not 80 "Manifestations of the Religious Vernacular: Ambiguity, Power, and Creativity," in Vernacular Religion in Everyday Life : Expressions of Belief, ed. Marion Bowman and Ulo Valk (New York: Routledge, 2014), 383-84. 81 "Vernacular Religion and the Search for Method in Religious Folklife," 44. 82 Leandro Rodriguez Medina use of the term “coloniality of knowledge” works from Aníbal Quijano’s neologism “coloniality” to refer to the forms of culture and knowledge still used today to extort power in the world. Quijano emphasizes the Eurocentered production of knowledge through the paradigms of modernity and rationality as part of the power structure presenting Western society as the center and other countries and cultures as peripheral and other. 40 enough to argue that the study of regional sites like Itsukushima is important because they have been ignored in previous scholarship. Moreover, their study can help to uncover the differing circumstances faced by regional religious institutions and narratives (e.g., origin narratives), as well as their available means for participating in translocal religious, intellectual, and literary spheres. The common view of origin narratives as locally produced and performed by religious officials at the temple or shrine in question encourages concerns over the writer’s unhindered power to fabricate the past to promote devotion and patronage. This led to derogatory images of origin narratives as little more than baseless advertisements in early scholarship. Gorai Shigeru and Sakurai Tokutarō countered these disparaging views of origin narratives by contending that even if origin narratives bend the truth, they still reveal how the producers wanted to represent the sacred site, and how worshippers imagined and experienced it. 83 There remains one notable point that they left unchallenged—that religious institution had sole agency over the production of origin narratives, and consequentially, the identity of the site and its deities. This encourages positioning around the local space of the temple or shrine as the center—its officials as the active powerholders and its worshippers as passive recipients. From this perspective, origin narratives can only serve as objects conveying the views of the institution, but not as things with the power or relevancy to influence ideas. Monks and priests of a site were often involved in the production and performance of origin narratives, but this does not mean that they held control, or that translocal influences were necessarily of secondary importance. Leandro Rodriguez Medina, Centers and Peripheries in Knowledge Production (New York: Routledge, 2014), 2-4; Aníbal Quijano, "Coloniality and Modernity/Rationality," Cultural Studies 21, no. 2-3 (2007): 170, 72-74. 83 Gorai Shigeru 五来重, Jisha engi to denshō bunka 寺社縁起と伝承文化, vol. 4, Gorai Shigeru chosakushū 五来重著作集 (Kyoto: Hō zō kan, 2008); Sakurai Tokutarō 桜井徳太郎, "Engi no ruikei to tenkai 縁起の類型と展開," in Jisha engi 寺社緣起, ed. Sakurai Tokutarō 桜井徳太郎, Hagiwara Tatsuo 萩原龍夫, and Miyata Noboru 宮田登 (Tokyo: Iwanami Shoten, 1975). 41 Unsurprisingly, the translocal contexts of origin narratives becomes more visible when we move beyond the context of a single site and look to the textual and social culture behind their production and use. In a special volume on origin narratives in the Japanese Journal of Religious Studies, Heather Blair and Kawasaki Tsuyoshi raise the untapped potential for more comparative and theoretical scholarship on origin narratives. 84 Nevertheless, despite the abundance of unstudied origin narratives, the diversity of their content, literary styles, and historical contexts makes origin narrative research more amenable to close studies of a single text than comparative or universal analysis. 85 In her monograph on Mt. Kinpusen, Heather Blair examined the origin narrative manuscript history for temples in the Ōmine mountain range to further nuance the socio-spatial power relations articulated and negotiated through the texts. Even if the narrative content linked temples as dominant and subordinate, wider manuscript circulation allowed the texts to serve diverse purposes. 86 Her findings reveal that origin narratives and associations between deities should not be assumed to be solely for the sake of enhancing the site’s image, and were often involved in negotiations over the relations between 84 Blair and Kawasaki, "Editors' Introduction: Engi: Forging Accounts of Sacred Origins," 20. 85 This resembles the situation in the field of Middle English manuscript studies, in which the overwhelming and multifarious nature of larger corpora has led case studies to become the standard methodology. However, as Michael Johnston argues in his study of fourteen and fifteenth century Middle English manuscripts, the difficulty should not overshadow the potential benefits of studies with more comprehensive scope. Michael Johnston, "Copying and Reading The Prick of Conscience in Late Medieval England," Speculum 95, no. 3 (2020): 744, 77-78. 86 Her study joins a growing body of scholarship in Japanese examining the context behind the production of origin narratives for sites in the Nara area. Multiple scholars have discussed how Kōfukuji’s assertions of authority over other sacred sites in Nara (often through head-temple (honji 本寺) and subsidiary temple (matsuji 末寺) relations) or over all of Yamato Province can be examined within the origin narratives of different Nara-based temples and shrines once presented outside of the context of a single sacred site. See Heather Elizabeth Blair, Real and Imagined: The Peak of Gold in Heian Japan (Cambridge, MA: Harvard University Asia Center, 2015), 265-67. Uejima, Nihon chūsei shakai no keisei to ōken 日本中世社会の形成と王権, 528-36, 64-75; Chikamoto Kensuke 近本謙介, "Haimetsu kara no saisei: Nanto ni okeru chūsei no tōrai 廃滅からの再生：南都における中世の到来," Nihon bungaku 49, no. 7 (2000): 29-30, 36-37; Blair, Real and Imagined: The Peak of Gold in Heian Japan, 259-65; Tsuyoshi Kawasaki, "The Invention and Reception of the Mino'odera engi," Japanese Journal of Religious Studies 42, no. 1 (2015): 145-47; Ō hashi Naoyoshi 大橋直義, "Jisha no kūkan to gensetsu: "Jishaken" toshite no Nanto ni oyobu 寺社の空間と言説：「寺社圏」としての南都に及ぶ," in Chūsei jisha no kūkan, tekusuto, gigei: "Jishaken" no pā supekutivu 中世寺社の空間・テクスト・技芸 : 「寺社圈」のパースペクティヴ, ed. Ō hashi Naoyoshi, Kazuhiro Fujimaki, and Yūsuke Takahashi (Tō kyō : Bensei Shuppan, 2014). 42 institutions (religious, political, and social). My study builds on Blair’s work by shifting attention to a provincial site, and expanding the scope of origin narrative circulation beyond a shared region. The use and transmission of origin narratives by central location and elite institutions in Nara unsurprisingly differed from regionally elite sites like Itsukushima Shrine. Origin Narratives and the Culture of Shared Knowledge The comparison of origin narratives and the connections between divinities within shines a new light on the relations between secrecy and shared knowledge in medieval Japan. Previous scholarship has characterized the period as a “culture of secrecy,” based on the centrality of esoteric Buddhism in the medieval episteme. The end of the culture of secrecy in the early modern period is then posited as an epistemological paradigm shift emblematic of the changing times. 87 The death of secrecy might have followed the end of the medieval period, but this does not mean that open forms of knowledge did not coexist with the medieval culture of secrecy. Brian Steininger argued that the expanding number of centers of learning (esp. Buddhist temples) in the thirteenth and fourteenth centuries led to increasingly free textual circulation. Religious and court elites defended their intellectual authority by asserting their exclusive access to information, and trying to hide their inability to prevent others from copying and transmitting that information. 88 Most of the origin narrative manuscripts with colophons recording the copyist and date of writing bear the names of prominent scholarly monks who collected origin narratives 87 William M. Bodiford, "When Secrecy Ends: The Tokugawa Reformation of Tendai Buddhism and its Implications," in The Culture of Secrecy in Japanese Religion, ed. Bernhard Scheid and Mark Teeuwen (London: Routledge, 2006), 311. 88 Brian Steininger, "The Scribal Imaginary in Medieval Japanese Paratexts," The Journal of Japanese studies 45, no. 2 (2019): 264-65. 43 of various temples and shrines. 89 These scholarly monks relied less on individual texts than their comparisons and integration of information from the various origin narratives to provide authoritative knowledge on the gods. 90 The difference between secret and public knowledge was not simply a matter of access to texts and information, but also interpretive authority. As Mark Teeuwen and Bernard Scheid have argued, secret knowledge did not require excluding others from access to information, but rather, provided initiates with inclusionary authority to properly possess, interpret, and use that knowledge. 91 Scholarly monks could claim expertise on the gods, but the thick descriptions about the gods in origin narratives could be interpreted and used to make associations between deities by anyone with access to the information. Therefore, the shared knowledge of origin narratives required different strategies to make one’s voice heard. 89 For example, Kenna (釼阿, 1261-1338) the abbot of Shōmyōji , Zenkai 全海 at Gokurakuji, and Yoshida Kanemigi ( 吉田兼右 1516 －1573) , the head of the Yoshida lineage, all copied the Itsukushima Shrine origin narrative, but this did not mean that they had special access to the texts. For a comparison of the Itsukushima Daimyōjin nikki manuscripts by Kenna and Zenkai see Makino Kazuo 牧野和夫, "Nagatobon "Heike monogatari" to "Itsukushima Daimyō jin nikki": Nagato-bon "Heike monogatari" seiritsu no ichi katei wo "Itsukushima Daimyō jin nikki" shishū ni saguru 長門本「平家物語』と『厳島大明神日記』：長門本『平家物語』生成の一過程を『厳島大明神日記』四周に探る," in Nagatobon Heike monogatari no sōgō kenkyū: Kōchū hen 長門本平家物語の総合研究：校注篇, ed. Asahara Yoshiko 麻原美子 and Nanami Hiroaki 名波弘彰 (Tokyo: Benseisha, 1998). For Yoshida Kanemigi’s copy of the Itsukushima origin narrative, see "Itsukushima shake engi 厳島社家縁起," in Yoshida bunko (Tenri University). 90 While written in the early modern period, Yoshida Kanemigi’s son, Bonshun (梵舜, 1553−1632) describes his copying of the Shoshin kongenshō, an encyclopedic text on the kami purportedly written by Yoshida Kanemigi as “the most important text among our family’s texts,” and “a text nobody outside of our family can see” in his diary entry from 1630, third month, eighteenth day. See Bonshun 梵舜, Shunkyūki 舜旧記, ed. Kamata Jun'ichi 鎌田純一, 8 vols., Shiryō sanshū 史料纂集 (Tokyo: Zoku Gunsho Ruijū Kanseikai, 1970), v. 6, p. 18; Arai Daisuke 新井大祐, "Chūsei kōki ni okeru Yoshida ke no jinja ken kyū to Engishiki ‘Jinmyōchō:’ Bonshun jihitsu Shoshinki wo tsūro toshite 中世後期における吉田家の神社研究と『延喜式』「神名帳」：梵舜自筆『諸神記』を通路として," in Chūsei shinwa to jingi, Shintō sekai 中世神話と神祇・神道世界, ed. Itō Satoshi 伊藤聡 (Tokyo: Chikurinsha, 2011), 472. 91 Bernhard Scheid, "Two Modes of Secrecy in the Nihon shoki Transmission," in The Culture of Secrecy in Japanese Religion, ed. Bernhard Scheid and Mark Teeuwen (London: Routledge, 2006), 284; Mark Teeuwen, "Knowing vs. Owning a Secret: Secrecy in Medieval Japan, as seen through the sokui kanjō Enthronement Unction," ibid., 173. 44 Rather than limiting access, origin narrative producers worked to make the tales as accessible as possible. The easier the information could spread and be used by others, the greater the chance that the deities and sacred spaces described would become better known, and thus, more likely to be used in the origin narratives and discursive writings of other temples and shrines. This served as a marketing strategy to attract pilgrimage, and also as a means of competing for discursive presence. Contending for superiority separates the victor over the others, and consequentially, everybody has to fight alone. Conversely, relational arguments for shared power and prestige strengthen when the members support and reproduce the same arguments. While the chapters of the dissertation discuss different types of relations with the Itsukushima Deity and Itsukushima origin narratives, they can all be said to work by propagating those networked connections over competing discourses. One way to encourage others to copy and spread origin narrative information was to promote sound bites condensing discursive signification into easy to remember forms. Rather than narrative events, lists identifying the Itsukushima Deity as one of the three Benzaiten of Japan, or one of the sisters of the Dragon Princess from the Lotus Sutra, presented the deity as an esteemed sea deity. The concision allowed other texts to copy and spread information about the Itsukushima Deity without having to devote pages to recounting a whole origin narrative. Moreover, the sound bites can function as categorical classifiers, encouraging multiple deities to be described in similar terms, although only a select few might be popularly remembered as the most prominent examples. Similar to the earlier discussion of Furukawa Kōshoken’s comments about Itsukushima as one of the three scenic places of Japan, Itsukushima’s beauty is rarely questioned despite the presence of countless other sites worthy of comparison. Once sound bites successfully become widespread, they can work as a form of “positive censorship (censorship by 45 repletion),” limiting the visibility of alternative arguments by overloading them with repeated mentions of the same sound bite. 92 With thousands of temples and shrines throughout the archipelago, and then countless networks of association between deities, there was just too much information for anyone to fully comprehend. This is not all that different than the situation of modern-day scholars faced with an inordinate amount of potential research objects, and the impossibility of knowing everything. Itsukushima Origin Narrative Manuscripts Itsukushima Shrine origin narratives can only provide a limited glimpse of medieval understandings of the shrine and its deity. It was the interconnected mass of texts and knowledge that solidified ideas about Itsukushima outside of the bounds of individual texts. The extant Itsukushima origin narrative manuscripts still deserve our attention, even if it is just to see the limitations of analyzing a single origin narrative. The Itsukushima Shrine origin narrative told today still shares a narrative base with the medieval versions. Significant differences emerge once one takes a look at the details, but the similarities are enough to allow me to introduce the narrative by summarizing from a modern version. During the reign of Suiko Tennō, a man named Saeki Kuramoto 佐伯鞍職 was fishing one day, and a boat with crimson sails approached from the sea to the west. The stranger asked Kuramoto to give her the island of Miyajima, and then revealed herself to be the Itsukushima Deity. Once Kuramoto accepted, the Itsukushima Deity asked Kuramoto for his assistance in 92 Roland Barthes describes how radio broadcasting created a constant abundance of sounds, and without special means to process or filter the mass of available media, people were likely to consume to that which was easiest to hear and understand. In turn, music that was innovative and subversive became censored just because its lack of popular appeal would result in it being played infrequently. Roland Barthes, "The Grain of the Voice," in Image, Music, Text (London: Fontana Press, 1977), 185. 46 building a shrine. To do so, the Deity advised him to go to the court and ask for their support. When Kuramoto asked how he could convince the court of her divine powers, the Itsukushima Deity responded that she would manifest a star in the sky above the capital as proof. Everything went according to plan, and after Kuramoto’s return, the Itsukushima Deity asked for him to guide her by boat around the island so that she could find the right spot for her shrine to be built. Eventually she found a place and Itsukushima Shrine was born. 93 There are three main medieval variants of the Itsukushima origin narrative and their earliest manuscripts are all from the first half of the fourteenth century: (1) Itsukushima Daimyōjin nikki (Record of the Itsukushima Deity 厳島大明神日記, 1323-1343); (2) the Itsukushima origin narrative within the Shōbōrinzō (Treasury of the Wheel of the True Law, 正法輪蔵, 1317) hagiography of Prince Shōtoku; (3) Itsukushima no honji (The Original Form of the Itsukushima [Deity] 厳島の本地, 1346). We still cannot know when these narratives were first written or the extent to which they were transmitted and transformed through oral recitations. 94 That being said, the time and circumstances of their copying can uncover hints about how the variants circulated and spread over time. For example, the dating of these three texts in the first half of the fourteenth century does not appear to be coincidental. As will be 93 The publication by Itsukushima Shrine, Itsukushima, also includes three other legends about the origins of Itsukushima: (1) as the Ichikishima goddess of Munakata Shrine in Kyushu, who eventually moved to Miyajima and had Itsukushima Shrine built for her; (2) Jinmu Tennō stopping at t he island of Miyajima to wait until the waters calmed; (3) Kūkai visiting Itsukushima while traveling back from China and building Buddhist halls on Mt. Misen. See Itsukushima jinja 厳島神社, Itsukushima 伊都岐島 (Miyajima-chō : Itsukushima Jinja Shamusho, 1976), 8-11, 14; Araki Hiroyuki 荒木博之, ed. Nihon densetsu taikei 日本伝說大系, 17 vols., vol. 10 (Tokyo: Mizuumi Shobō , 1987), 30-31. 94 Tokuda Kazuo cites numerous examples from Muromachi period diaries describing the person hearing a sacred site’s origin narrative or miracle tale from monks and priests at the site, itinerant biwa lute performers, miko, and renga poets. In the case of the Itsukushima origin narrative for example, Zuikei Shūhō’s diary, the Gaun nikkenroku 臥雲日件録, explains how he heard the origin narrative from a blind biwa lute performer named Jōro 城呂 in Kyoto after returning from a pilgrimage to Itsukushima. See Tokuda Kazuo 徳田和夫, Otogizōshi kenkyū お伽草子研究 (Tokyo: Miyai Shoten, 1988), 165-80; Zuikei, Gaun nikkenroku batsuyū 卧雲日件錄抜尤, 13. 47 discussed in more detail later, petitions for rewards after the Mongol invasions led to a surge in the production of new origin narratives and more comparative thinking about the relations between the various deities. This is not to say that all early fourteenth century origin narratives were produced to compete for recognition in defeating the Mongols, but it is fair to describe them as produced within the resultant discursive environment which stressed comparisons of different deities and sacred spaces. 95 There are no extant premodern Itsukushima origin narratives explicitly produced for the shrine, and only one historical document describing the reproduction of an illustrated scroll version of the origin narrative for the shrine beginning in 1542, but the scrolls did not survive to the present. The document describes its translocal production: the painting was begun by a painter called the counselor (sanshō 三省) of Tosa Province and finished by Kano Shōei ( 狩野松栄, 1519-1592), and the explanatory text was written on Mt. Hakusan. 96 Head priest Fusaaki recorded this information when he transferred the origin narrative to his son, Mototsuna 元綱 in 1569, and told him that “it should be stored in secret for generation after generation.” 97 However, the content of the Itsukushima Shrine illustrated origin narrative was based on the Itsukushima 95 The specific context of the Mongol invasions for the Itsukushima Daimyōjin nikki will be discussed in more detail in chapter 1. It is worth noting here that there are some differences in the Shōbōrinzō compared to other biographies of Prince Shōtoku, and between the 1346 version of Itsukushima no honji and later manuscripts that stress the importance of numinous power to protect the realm in these versions. The 1346 Itsukushima no honji alone includes an oracle with the Itsukushima Deity identifies herself as a protector of the realm and the province. Concerning the Shōbōrinzō, Matsumoto Shinsuke discusses how the Rinnōji -bon version incorporates a discussion of Jingū’s invasion of Korea and connects it with a rite performed by Prince Shōtoku at Shittenōji to protect the realm. See "Kaidai: Itsukushima no engi 解題：いつくしまのゑんぎ," 382; Matsumoto Shinsuke 松本真輔, Shōtoku Taishi den to kassentan 聖徳太子伝と合戦譚 (Tokyo: Bensei Shuppan, 2007), 251-52. 96 Kano Shōei painted a n ema with the image of the demon of the Rashōmon gate (Rashōmonki zu 羅生門鬼図) for Itsukushima Shrine on the third day of the first month of 1569, and he probably worked on the Itsukushima Shrine origin narrative during the same visit. He stopped at Itsukushima as well as Shōtenji 承天寺 (in modern-day Fukuyama, Hiroshima Prefecture), while making his way to Kyushu to meet with the daimyo of Bungo province Ōtomo Sōrin ( 大友宗麟, 1530~1586). Tanaka Toshio 田中敏雄, "Kano ha no gajin to Ōtomo Sōrin 狩野派の画人と大友宗麟," Geijutsu: Ō saka geijutsu daigaku kiyō 芸術 : 大阪芸術大学紀要. 15 (1992): 17-18. 97 See doc. 23 in "Nosaka monjo 野坂文書," in Hiroshima kenshi: Kodai chūsei shiryō hen 広島県史. 古代中世資料編, ed. Hiroshima-ken 広島県 (Hiroshima: Hiroshima, 1978), 449. 48 no honji variant, and would have been widely accessible. 98 Illustrated scrolls could serve as symbols of prestige and exclusivity, and as such, worked best when viewed by limited audiences. 99 In contrast, short origin narratives without visual accompaniments were easier to copy, and as a result, proliferate. Visual images and colophons are invaluable subjects for scholarly analysis, but we also need to stress the reproducibility of unadorned texts. The three variants contemporaneously circulated apart from one another, providing us with a window for comparing their influences on later writings about Itsukushima Shrine. Each of the variants provides a different date for the founding of Itsukushima Shrine, and I use this detail to help trace their respective dissemination (table 1). The dates are specific enough to surmise that the information trace back to the respective variants, rather than from later writers changing the dates themselves. 100 Medieval and early modern writers often had enough information about the origins of Itsukushima Shrine to notice multiple explanations for the date 98 The only detail about the origin narrative content is a mention of the place of Mt. Karahiku 迦羅比兵山, and this mountain only appears in the Itsukushima no honji variant. See ibid.; "Kaidai: Itsukushima no engi 解題：いつくしまのゑんぎ," 378. Regarding the accessibility of the Itsukushima no honji variant, there are at least twelve extant copies, with two from the medieval period, and the rest from the early modern period. This is far more than any other variant. For a list of the extant texts, and a comparison of them, see Matsumoto Ryūshin 松本隆信, "Zōt ei Muromachi jidai monogatari genzonbon kanmei mokuroku 増訂室町時代物語類現存本簡明目錄," in Otogizōshi no sekai 御伽草子の世界, ed. Nara Ehon Kokusai Kenkyū Kaigi 奈良絵本国際研究会議 (Tokyo: Sanseidō , 1982), 63; Matsumoto Ryūshin 松本隆信, Chūsei ni okeru honjimono no kenkyū 中世における本地物の研究 (Tokyo: Kyūko Shoin, 1996), 90-103. 99 In the Muromachi period, the shogun Ashikaga Yoshinori was able to borrow illustrated origin narratives and other precious works from temples and shrines, and the scrolls sometimes traveled from the shogunate to other elites in the capital. Karen Brock interprets this as demonstrating Yoshinori’s authority in cultural and religious spheres, rather than as reflective of broader lending practices for temples’ illustrated scroll manuscripts. Karen L. Brock, "The Shogun's ‘Painting Match’," Monumenta Nipponica 50, no. 4 (1995): 458-59. 100 The dates themselves were not meaningless; they often corresponded with the timing of annual rites at Itsukushima Shrine. For example, within the monk Zuikei Shūhō ( 瑞渓周鳳, 1391-1473)’s record of hearing about the origins of Itsukushima Shrine in his diary, the Gaun nikkenroku 臥雲日件録, he mentions a large-scale ceremony on the thirteenth day of the ninth month. Later, in the seventeenth century, the 1692 Itsukushima michishiba no ki 厳島道芝記 gazetteer cites the biography of Prince Shōtoku ( Shōtoku taishi den 聖徳太子伝記) as dating the founding of Itsukushima Shrine to the twelfth day of the eleventh month and uses this date to explain the timing of the enshrinement festival (chinzasai 鎮座祭) in the eleventh month. See Zuikei, Gaun nikkenroku batsuyū 卧雲日件錄抜尤, 13; Kojima Tsunenari 小島常也, Itsukushima michishiba no ki 嚴島道芝記 (Miyajima, Hiroshima: Miyajima-chō , 1971), 25-26. 49 of the shrine’s founding. The compilers were concerned with the accuracy of the origin narrative information they were recording. 101 This resulted in a body of texts with a high degree of consistency with the three variants, despite chances for variations from oral transmissions and mistakes over the centuries. 101 The Shōbōrinzō, for example, highlights a discrepancy in the year of founding as either the fifth year of Tanshō ( 端正, 593) or the ninth year of Kikki ( 吉貴, 602). The difference matters for the organizational structure of the Shōbōrinzō, because it is organized chronologically with one chapter for each of the years of Prince Shōtoku’s life . The compilers do not explicitly argue for one date over the other, but they follow the 602 date by presenting the Itsukushima origin narrative when Shōtoku was thirty -one years old. Additionally, the sixteenth century Itsukushima shake engi 厳島社家縁起 manuscript demonstrates that later copyists continued to pay attention to the dates. Although the Itsukushima shake engi copies the Shōbōrinzō’s Itsukushima origin narrative almost word-for-word, it leaves out the first sentence giving the date of the ninth year of Kikki, and changes the day of the remaining date from the twelfth day of the eleventh month to the thirteenth day, and corrects the age of Prince Shōtoku during the fifth year of Tanshō to twenty -two rather than twenty-one as given in the Shōbōrinzō. Given the copyist’s attention to the date in the Itsukushima shake engi, the altering of the day was likely a choice to follow an alternative source (the date matches the 1346 version of Itsukushima no honji, titled Itsukushima no engi. See "Shōbōrinzō 聖法輪蔵," in Shinshū shiryō shūsei 眞宗史料集成, ed. Ishida Mitsuyuki 石田充之 and Chiba Jō ryū 千葉乘隆 (Kyoto: Dō hō sha, 1974), 487; "Itsukushima shake engi 厳島社家縁起," unpaginated. 50 Table 1: Itsukushima Origin Narrative Variants (italics mark the texts with full origin narratives) Date of Founding of Itsukushima (in text) Text Date (of Text) 9th month, 13th day Itsukushima Daimyōjin nikki 厳島大明神日記 1323-1343 Mineaiki 峯相記 Post 1348 Nagatobon Heike monogatari 長門本『平家物語』 ~15th century 11th month, 12th day Shōbōrinzō 正法輪蔵 1317 Genpei jōsuiki 源平盛衰記 ~15th century Itsukushima Daimyōjin engi shō 厳島大明神縁起抄 * Also significant similarities with the 1346 version of Itsukushima no honji Pre 1602 Honchō shosha ichiran 本朝諸社一覧 1685 Itsukushima yurai いつくしま由来 1779 Geihan tsushi 芸藩通志 1825 12th month, 13th day Itsukushima no honji 厳島の本地 Late Muromachi Shoshin kongenshō 諸神根元抄 Post 1556 Yoshida Kanemigi Itsukushima sha sengū ganmon 吉田重右厳島遷宮願文 1571 Fusaaki oboegaki 房顕覚書 1580 Itsukushima go engi 厳島御縁起 * Variant of Itsukushima no honji Early modern Exceptions 11th month, 11th day Ruiju kigenshō 類聚既験抄 Late 13th century 11th month, 13th day Itsukushima no engi いつくしまのゑんぎ * Variant of Itsukushima no honji 1346 Itsukushima shake engi 厳島社家縁起 *Copies main body of engi from Shōbōrinzō, and adds a summary of the Itsukushima no honji narrative 1516-1573 11th month, 17th day Mantokujibon Shōtoku taishi den 満徳寺本聖徳太子伝 * Variant of Shōbōrinzō 1462 11th and 12th months both given Itsukushima Michishiba no ki 厳島道芝記 1692 Itsukushima kōtaijin gochinza ki 伊都岐島皇太神御鎮座記 1751~1764 51 Abbreviated to "the time of Suiko Tenno" Honchō jinja kō 本朝神社考 1628~1645 A brief examination of the chart of the three variants and their later influences reveals a few important points about consistency and contingency in the Itsukushima Shrine origin narrative. First of all, the variant lineage of the Itsukushima no honji origin narrative (founding date of twelfth month, thirteenth day) is the only one with any source explicitly produced by an official of Itsukushima Shrine. The Fusaaki oboegaki 房顕覚書, produced by the high priest (tanamori 棚守) of Itsukushima Shrine, Fusaaki 房顕, in 1580 provides an overview of the shrine’s history. As such, its contents would have matched the origin narrative possessed by the shrine, and the information conveyed to pilgrims and prominent guests. The origin narrative manuscript used by the Itsukushima shrine priests during the seventeenth century no longer exists, but there are enough details to conclude that it was based on the same Itsukushima no honji variant. 102 Yoshida Kanemigi ( 吉田兼右, 1516-1573), head of the Yoshida household, also included the same date for the shrine’s founding in the dedicatory vow he recited when came in the twelfth month of 1571 to perform the enshrinement ceremony for the Itsukushima Deity following the shrine’s restoration. Kanemigi had connections with Fusaaki tracing back to 1544, and it makes sense that his information about the shrine would have originated with Fusaaki. His relationship with Fusaaki also helps to explain the use of the same date in the section on Itsukushima Shrine in the Shoshin kongenshō 諸神根元抄, a compilation on the various shrines by a member of the Yoshida household. 103 102 Doc. 23 in "Nosaka monjo 野坂文書," 449. 103 The compiler and dating of the Shoshin kongenshō and its better-known variant the Shosha kongenki 諸社根元記 are unclear, but Shimasue Kiyoshi argues that the current form of the Shoshin kongenshō was likely produced by Yoshida Kanemigi after 1556 and prior to the Shosha kongenki. See Shimasue Kiyoshi 島居清, "Shoshin kongenshō, Shoshinki, Shosha kongenki ni tsuite (2): Shoshinki to Shoshin kongenshō 諸神根源抄・諸神記・諸 52 Further discussion of the origin narrative variants would be unnecessary if the officials at Itsukushima Shrine controlled the transmission of texts and information about its origins. But this was not the case. Even the writings of Yoshida Kanemigi hint at the influence of multiple sources on Itsukushima’s origin narrative. The Itsukushima Shrine origin narrative in the Yoshida bunko archive called the Itsukushima shake engi (Origin Narrative of the Itsukushima Shrine Family, 厳島社家縁起) gives Yoshida Kanemigi as the original writer, but the content and dating are closer to that of the Shōbōrinzō than the other Yoshida texts following the Itsukushima no honji variant. The use of the term “shrine family” (shake 社家) in the title further suggests that Kanemigi received the information directly from the shrine, most likely from Fusaaki. 104 From our limited knowledge of the circumstances, it is impossible to confirm exactly how Kanemigi received the origin narrative or why its content differs from the contemporary writings from Fusaaki and Yoshida Kanemigi discussed above. The Itsukushima shake engi begins with an almost exact copy of the Shōbōrinzō, continues with a summary of the Itsukushima no honji narrative and its identifications of the shrine’s deities, and ends with a list of seven secret places on the island. 105 One possibility is that Fusaaki possessed multiple 社根元記に就いて(2) ：諸神記と諸神根源抄," Biburia, Tenri toshokan hō 31 (1965): 56. For their identical sections on Itsukushima Shrine, see "Shoshin kongenshō 諸神根元抄," (National Institute of Japanese Literature: Hizen Shimabara Matsudaira bunko 肥前島原松平文庫), coma 60; "Shosha kongen ki 諸社根元記," in Jingi zensho 神祇全書, ed. Saeki Ariyoshi 佐伯有義 (Tokyo: Kō ten Kō kyūjo, Jingū Hō saikai, 1906), 21. 104 Although the colophon for the Itsukushima shake engi does not explain how Kanemigi acquired the text to copy, colophons from texts written by Kanemigi, as well as some entries in the early modern Shosha kongenki describe how members of the Yoshida family borrowing or copying origin narratives and other information on shrines when they visited and met with members of the shrine family. See Arai, "Chūsei kōki ni okeru Yoshida ke no jinja kenkyū to Engishiki ‘Jinmyōchō: ’ Bonshun jihitsu Shoshinki wo tsūro toshite 中世後期における吉田家の神社研究と『延喜式』「神名帳」：梵舜自筆『諸神記』を通路として," 474; Shimasue Kiyoshi 島居清, "Yoshida bunko no Kanemigi jihitsubon ni tsuite: Jinja jinmyō sais hi bukki nado ni kansuru mono (ni) 吉田文庫の兼右自筆本に就いて：神社神名祭祀服忌等に関するもの（二）," Biburia, Tenri toshokan hō 27 (1964): 49; "Shosha kongen ki 諸社根元記," 5, 320. 105 The text even follows the Shōbōrinzō in beginning the origin narrative with Prince Shōtoku’s age, even thou gh the Itsukushima shake engi extracts the Itsukushima origin narrative from the Prince Shōtoku biography. "Itsukushima shake engi 厳島社家縁起." 53 versions of the origin narrative and shared two versions with Kanemigi. Another is that Kanemigi (or a later editor) gained access to the Shōbōrinzō and Itsukushima no honji narratives separately and later compiled them together. In the end, there is no way to know for certain, and the answer would likely tell us little. Shifting our attention away from individuals involved in the production and transmission of origin narratives to its later usage—the information that continued to be reproduced, transformed, and remembered—illuminates the power of origin narratives to influence religious knowledge regardless of the intentions of those involved in its production and dissemination. The transmission and transformation of origin narratives was less direct and more contingent than conventionally thought. Focus on temple and shrine clergy as the producers and transmitters of their site’s origin narratives have led scholars to develop theories about human networks linking temples and shrines with similar origin narratives. It is easy to be enticed by the logic of such arguments, and moreover, the conclusions about the transmissions do not necessarily reveal the significance of the received information or the involved networks. In the case of Itsukushima Shrine, Kanai Kiyomitsu hypothesized about the development of intertextual connections between the origin narratives of Itsukushima, Mt. Kōya, Zenkōji, and Kumano through mountain ascetics traveling between the sites. 106 Makino Kazuo also suggested that the Itsukushima Daimyōjin nikki origin narrative made its way to the temples of Gokurakuji and Shōmyōji in Kamakura through connections with temple networks centered around Mt. Hiei and Saidaiji. 107 These arguments are possible, but the information could have spread faster and farther through texts in broader circulation. It makes sense to try and explain the transmission of 106 Kanai Kiyomitsu 金井清光, Ippen to Jishū kyōdan 一遍と時宗教団 (Tokyo: Kadokawa Shoten, 1975), 179-80. 107 Makino, "Nagatobon "Heike monogatari" to "Itsukushima Daimyō jin nikki": Nagato-bon "Heike monogatari" seiritsu no ichi katei wo "Itsukushima Daimyō jin nikki" shishū ni saguru 長門本「平家物語』と『厳島大明神日記』：長門本『平家物語』生成の一過程を『厳島大明神日記』四周に探る," 243-46. 54 texts in relation to the known information networks of Mt. Hiei and Saidaiji, but we cannot assume that the origin narratives circulated in a logical or straightforward manner. Methodologically, it is safer to analyze the information we have, the intertextual discursive connections that consistently link the same groups of deities together. Origin narratives are finished products that hide the interpretive work involved in their creation, but we can work to retrace the steps of their producers and readers by following the connections between deities and sacred spaces in relation to contemporary discursive controversies. Conclusion Why did Itsukushima Shrine’s fame survive the Taira? There are two related aspects to consider. The first is that Itsukushima’s image did not endure but was reborn in various forms. The second is that Itsukushima Shrine was well connected. I am not arguing for the agency of the Itsukushima Shrine priests to employ their networks to adapt the shrine’s image with the changing times. Instead, Itsukushima Shrine’s connections were discursive, and as such, enabled and encouraged others to transform Itsukushima’s hybrid identities. However, this is not my conclusion. It is a starting place for investigating how and why others understood the sacred space and deity of Itsukushima as they did. Examining the work involved in interpreting and describing the Itsukushima Deity through her relations with other divinities, I will argue that the Itsukushima Deity gained renowned by gaining a special place within desirable deity networks. The Itsukushima Deity was a manifestation of Dainichi Buddha, Kannon, or Benzaiten; one of the three Munakata goddesses who descended from Amaterasu; and a dragon deity of royal descent. These identities mattered, but they often mattered because they ranked her as one of the two manifestations of Dainichi in 55 Japan (Amaterasu being the other); one of the daughters of Dragon King Sagara (siblings with the Dragon Princess from the Lotus Sutra and other female dragon deities); one of the three Benzaiten (along with Chikubushima and Enoshima); and one of the tutelary deities of an elite historical household (similar to the royal family with Ise’s Amaterasu, the Fujiwara and the Kasuga Deity, or the Minamoto with Hachiman). As we shall see, origin narratives and the writings of scholarly monks inconsistently listed different gods within these networks, and they gave varying reasons for emphasizing the importance of some of the deities over the others. These networks left readers with a simple conclusion: the Itsukushima Deity and the other selected deities are special because they were included over countless others. Furukawa Kōshoken was r ight to question Itsukushima Shrine’s position as one of the three scenic places of Japan, and to note how other places get ignored because of their exclusion from the list. Discussions of sacred space, though, often gave more weight to common knowledge than to reason. Itsukushima was one of the three scenic places once that became common knowledge. Furukawa’s modern rationalism and well-traveled perspective made him skeptical of hearsay. Noting discrepancies between the encyclopedic Wakan sanzai zue ( 和漢三才図会, 1712) and other printed texts, he remarked, “Hearing something one hundred times is not the same as seeing it once. It is hard to know a place’s true nature if you cannot see it with your own eyes.” 108 Furukawa might have had more experience of places than most, but he did not realize the power of familiar phrases. People tend to believe what they have heard before, particularly when discussing places and deities outside the scope of the majority’s travels. 108 Furukawa, "Tōyū zakki 東遊雑記," 445; Herbert E. Plutschow, A Reader in Edo Period Travel (Folkestone, Kent, UK: Global Oriental, 2006), 104. 56 Considered in terms of the production of knowledge, we can go so far as to say that facts need to be accepted and repeated. Bruno Latour argues, “Fact construction is so much a collective process that an isolated person builds only dreams, claims, and feelings, not facts.” 109 Furukawa might be able to make more informed claims about places than most, but that does not matter if the ideas do not reach and convince people to further share them with others. The semiotic play involved in the linking of associations between the gods was not all fun and games; it was hard work that only succeeded when the connections spread and become known. Discussing the process of linking associations might make them sound artificial, but this is a fault of western rationalism. Karen Barad contends that the conventional separation of ontology and epistemology leaves us with two potential, both problematic, approaches towards materiality. It is either “a given or a mere effect of human agency.” 110 While not identical to Barad’s theorization of onto-epistemology, the nondual relationship between the signifier and the signified in the medieval episteme presents knowledge as reflective of being. A more onto- epistemological viewpoint allows us to consider how the use of origin narratives to produce knowledge of the gods and sacred space relates to the ontology of local gods and sacred spaces. The comparison of site-specific deities equated with universal Buddhas revealed logical conflicts. This did not simply result in new interpretations of Buddhist or Shinto thought. Kami existed as local and universal through networks of association that simultaneously mapped commonality and difference. This was not merely a resolution for religious questions, but a 109 Latour, Science in Action: How to Follow Scientists and Engineers through Society, 41. 110 Karen Barad brings together theories of physics and knowledge production to propose a performative model of agential realism in which things come into being through the mapping of differential boundaries. From this perspective, no things exist prior to their intra-actions which allow them to be understood as distinct. The analogical epistemological mode of medieval Japan presumes that being exists prior to their understanding, but the process of knowledge production results in a similar relationality between being and knowledge. Karen Michelle Barad, Meeting the Universe Halfway: Quantum Physics and the Entanglement of Matter and Meaning (Durham, NC: Duke University Press, 2007), 183-85. 57 necessity for interacting with the kami and sacred spaces as individual divinities. From this perspective, the Buddhas and kami of worldly sacred spaces not only made the sacred more accessible, they made it more diverse. The Itsukushima Deity at Itsukushima Shrine, and the Itsukushima deities enshrined elsewhere were both the same and different, and this was not a contradiction. The medieval period might have been the age of the “warring deities,” but its battles were fought by scholarly monks trying to understand the nature of the realm of the gods. This is a rare story where victories were decided through studied arguments and the sharing of information, rather than deceit and abuses of power. Of course, there were also cases where religious officials altered information to gain advantages over rival institutions. The problem is that the framework of competing power-bloc temples and shrines can overshadow the common practices of scholarly monks to faithfully collect, record, and interpret information. When honji-suijaku complicated the mythological hierarchies of the kami, it did not only threaten the authority of the elite temples and shrines. It challenged the individuality of every worldly deity and sacred space. Networks of associations and origin narratives developed together to provide new frameworks for hierarchical and spatial relations. The systems that developed were not always consistent, fair, or unbiased, but they provided enough stability to keep the sacred of/in the world meaningful. The picture of the past drawn by following the traces might not always be salacious, or even sensible, but it deserves to be seen on its own terms—in its own frame. 58 Chapter 1 The Production of Origin Narratives and the Networking of Dragon Sisters On the nineteenth day of the third month of 1314, the holy man Nyoen 如円 told the sovereign, Hanazono Tennō ( 花園天皇, 1297-1348), about an oracle from the Kitano Deity 北野 of the Aoki estate in Chikuzen Province (modern day Fukuoka Prefecture in Kyushu). According to the entry in Hanazono Tennō’s diary, the deity manifested as an injured snake and sp oke through the voice of a female shrine attendant (miko 巫女). The god proclaimed: “The foreigners have already attacked. The gods of Kashii, Hakozaki, and Kōra shrines and I have battled them. The Kashii Deity is half dead. Because of my virtue as the Great Unhindered One, I have notified people by manifesting myself as a snake,” and continued, saying: “If there are more prayers, I will once again depart to subjugate the foreign countries.” 111 People attributed the typhoon winds that sank the Mongol ships in 1281 to the deities protecting Japan. Without any way to know if there would be further attacks, prayers for the subjugation of the Mongols continued into the early fourteenth century. 112 The foreign threat provided opportunities for temples and shrines to gain rewards in exchange for prayers, but it also created a problem—when countless temples and shrines are all trying to assert the efficacy of one’s own deity’s prayers against the Mongols, how can any one site demonstrate its superiority over the others? Within this context, the oracle from the Kitano Deity might sound like a self-serving lie, and in fact, that is how one shrine official interpreted it. Writing on the reverse side of the paper provides the deity’s response to the official’s skepticism, “The miko was possessed again and 111 Hanazono Tennō 花園天皇, Hanazono Tennō shinki 花園天皇宸記, ed. Murata Masashi 村田正志, 3 vols., vol. 1, Shiryō sanshū. Kokiroku hen 史料纂集. 古記錄編 (Tokyo: Zoku Gunsho Ruijū Kanseikai, 1982), 150. Translation based on Thomas Conlan, In Little Need of Divine Intervention: Takezaki Suenaga's Scrolls of the Mongol Invasions of Japan (Ithaca, NY: East Asia Program, Cornell University, 2001), 252-53. 112 Official petitions for prayers from the court and bakufu governments stopped in 1293 and 1311 respectively. Nevertheless, temples and shrines continued to argue for the need for further prayers until around 1321. Kawazoe Shō ji 川添昭二, Chūsei Kyūshū no seiji, bunkashi 中世九州の政治・文化史 (Fukuoka: Kaichō sha, 2003), 59-62. 59 said: ‘Within three days you will suffer [divine] punishment.’ That night [the official] suddenly died. After that, all returned to [a state of] belief.” There was a price to pay for doubting the words of the gods. The intellectual historian Satō Hiroo describes the medieval period (approximately 1200 - 1600) as “the warring deities period” (kamigami no sengoku jidai 神々の戦国時代), in which Amaterasu 天照 no longer safely occupied the position as the foremost deity in Japan, and each of the myriad deities sought to assert preeminence. 113 Temple and shrine origin narratives (jisha engi 寺社縁起) were indispensable resources within this competition. More than just detailing the origins of a sacred space, the narratives date the arrival of the deity at the site, describe the founders and the deities, and tell stories of the miraculous events associated with the site. In short, origin narratives presented the identities of temples and shrines. Furthermore, the genre of origin narrative itself added an air of authority. 114 Audiences expected the information to originate from the temple or shrine itself, and thus to be accurate. Scholars too have assumed that each religious institution produced its own origin narratives, leading some to criticize the narratives as “nothing more than baseless and preposterous stories and images used for the self- promotion of the temple or shrine.” 115 113 Satō, Kami, Hotoke, ōken no chūsei 神・仏・王権の中世 (Kyoto: Hōzōkan, 1998), 313. 114 The Chōkan kanmon 長寛勘文, as a document made to assist the court in determining whether or not the Kumano Deity should be treated as the same deity as Amaterasu, is a valuable reference for seeing what types of texts were seen as credible sources for understanding the deities. Looking at what texts are quoted, we find the Kumano origin narrative Kumano gongen go suijaku engi 熊野権現御垂迹縁起, as well as myths from the Nihon shoki, Sendai kuji hongi 先代旧事本紀 and Kogoshūi 古語拾遺, Nihon shoki commentaries, official histories of the realm (kokushi 国史), and diaries (nikki 日記). Nakara Naritomo 中原業倫, "Chōkan kanmon 長寛勘文," in Gunsho ruijū. Zatsu bu 群書類從. 第 26 輯 (雑部), ed. Hanawa Hokinoichi 塙保己一 (Tokyo: Yagi Shoten kosho shuppanbu 2013). 115 The quoted description of origin narratives refers to a common view of origin narratives within previous scholarship, which Gorai Shigeru argued against in his scholarship. Gorai Shigeru 五来重, Jisha engi to denshō bunka 寺社縁起と伝承文化, vol. 4, Gorai Shigeru chosakushū (Kyoto: Hō zō kan, 2008), 3. For an overview of the previous scholarship on origin narratives in English and the development of origin narratives as a field of study, see 60 It is clear that origin narratives were altered or written anew to connect a temple or shrine’s power to the defeat of the Mongols. 116 However, thinking of these narratives as mere fabrications can lead scholars to overstate the agency of religious institutions in producing the origin narratives of one’s own temple or shrine. If the audience could, like the skeptical shrine attendant, question the accuracy of a narrative, do we not need to ask about the factors that would make an origin narrative convincing? As Heather Blair and Kawasaki Tsuyoshi note, Western scholarship often approaches the study of origin narratives through the close analysis of a single text in relation to the historical and cultural contexts of the space. 117 Within a methodology focused on a particular text or site, it is not surprising that the analysis of origin narratives often highlights how the stories were strategically constructed to meet the contemporary needs of the institution. However, origin narratives were not produced within a vacuum; the writers needed to consider external factors so that audiences would not read the tales as “baseless” and “self-promotional.” This chapter takes a different approach to origin narratives. Examining origin narratives of more than one sacred site raises new perspectives on the production and reception of origin narratives and shared knowledge about the gods. Beginning from a single origin narrative, the Itsukushima Daimyōjin nikki (Record of the Itsukushima Deity 厳島大明神日記), and tracing Heather Blair and Kawasaki Tsuyoshi 川崎剛志, "Editors' Introduction: Engi: Forging Accounts of Sacred Origins," Japanese Journal of Religious Studies 42, no. 1 (2015). 116 Kawazoe Shō ji 川添昭二, Chūsei Kyūshū no seiji, bunkashi 中世九州の政治・文化史 (Fukuoka: Kaichō sha, 2003), 59-60. 117 Blair and Kawasaki, "Editors' Introduction: Engi: Forging Accounts of Sacred Origins," For other scholarship situating origin narratives in historical and religious contexts, see Thal, Sarah, Rearranging the Landscape of the Gods: The Politics of a Pilgrimage Site in Japan, 1573-1912 (Chicago: University of Chicago Press, 2005); Caroline Hirasawa, Hell-Bent for Heaven in Tateyama Mandara : Painting and Religious Practice at a Japanese Mountain (Leiden: Brill, 2013); D. Max Moerman, Localizing Paradise: Kumano Pilgrimage and the Religious Landscape of Premodern Japan (Cambridge, MA: Harvard University Asia Center, 2005); Anna Andreeva, "The Karmic Origins of the Great Bright Miwa Deity: A Transformation of the Sacred Mountain in Premodern Japan," Monumenta Nipponica 65, no. 2 (2010). 61 overlapping discursive elements with other texts, the chapter connects networks between the origin narratives and deities of multiple sites. As we shall see, the texts providing meaning to Itsukushima did not always originate from Itsukushima Shrine. It was not only priests from Itsukushima, but also scholarly monks at temples near the capital and in Kyushu, who articulated and spread ideas about the Itsukushima Deity. More specifically, I focus on the novel association made between Itsukushima and Queen Consort Jingū (Jingū kōgō 神功皇后) in the Itsukushima Daimyōjin nikki. Famous as the leader of an expedition to conquer the Korean kingdoms, and as the mother of the war god Hachiman, Jingū became a symbol used by medieval myths to demonstrate Japan’s ability to defeat foreign enemies. 118 Therefore, many origin narratives asserted connections with Jingū to provide a precedent for their deity’s role in defeating the Mongols, but some were better received than others. Examining how and why the association between the Itsukushima Deity and Queen Consort Jingū became widely accepted, this chapter show that Itsukushima origin narratives did not construct the deity’s identity independently. The Itsukushima Daimyōjin nikki provided the essential information for connecting Itsukushima with Jingū, but it only succeeded because the writings of other religious institutions reproduced, spread, and standardized the association. Even if an origin narrative was written with the intent of promoting specific interpretations of the deity and its space, its fate was decided by a mix of contingency and standardization. In the following pages, I will show how the Itsukushima Daimyōjin nikki encouraged the texts of other religious institutions to incorporate and spread the connection between Itsukushima and Queen Consort Jingū through networks of association. The Itsukushima Daimyōjin nikki succeeded by 118 Abe Yasurō 阿部泰郎, "Hachiman engi to Chūsei Nihongi: "Yurikawa Daijin" no sekai kara 八幡縁起と中世日本紀：「百合若大臣」の世界から," Gendai shisō 20, no. 4 (1992): 78. 62 presenting the Itsukushima Deity as distinctive and, more importantly, as relatable. Rather than claiming superiority, the most effective tactic was to make others want to draw similarities with Itsukushima. Networks for Mediating Connections between the Gods The issue that needs to be addressed is not whether the origin narratives are historically accurate, but how the narratives convinced readers and listeners to accept them. The narratives are undeniably constructed, but that is not to say that the writers could say whatever they wanted. Questions of accuracy rose to the forefront when countless religious institutions sent petitions for rewards after the Mongol invasions. While some temples tried to emphasize the unique efficacy of its own prayers against the Mongols, such assertions lost credibility when numerous sites were making similar claims. For example, the Illustrated Scroll of the Suwa Deity (Suwa Daimyōjin ekotoba 諏訪大明神絵詞, post 1356) claims, “Although it can be said that the way of the sovereign and peace [in the realm] are protected by all of the various deities together, the subjugation of the foreign enemy is the miraculous feat of this shrine.” 119 Suwa priest Enchū ( 円忠, 1295-1364) initiated production of the origin narrative and designed it as an officially authorized history of Suwa, including an introduction by Go-Kōgon Tennō ( 後光厳, 1338-1374) and a colophon by the shogun Ashikaga Takauji ( 足利尊氏 1305-1358). 120 Nevertheless, when Enchū asked Urabe Kanetoyo ( 卜部兼豊, 1305-1376), an official in the Council on Shrine 119 Enchū, 圓忠, "Suwa Daimyō jin ekotoba 諏訪大明神絵詞," in Zoku Gunsho ruijū (Tokyo: Zoku Gunsho Ruijū Kanseikai, 1934), 496. 120 Haruko Wakita, "The Creation of Fabricated Myths in the Medieval Age: An Examination of Shintoshu, Histories (Engi), Noh Plays, and Other Sources," in Writing histories in Japan: texts and their transformations from ancient times through the Meiji era, ed. James C. Baxter and Joshua A. Fogel (Kyoto: International Research Center for Japanese Studies, 2007), 91-92. 63 Affairs (Jingikan 神祇官), to read the work, Kanetoyo responded by referring to a similar episode about another temple’s prayers against the Mongols. He concluded, “Like this, it was not [a deed] limited only to the one shrine of Suwa.” 121 Kanetoyo goes on to explain that multiple temples and shrines should receive rewards, but his statement suggests that rhetoric emphasizing a single site’s conquering of the Mongols might not be convincing to contemporary scholars or the court. A closer look at the interactions between Enchū and Urabe Kanetoyo illuminates the issues involved in the production of a convincing origin narrative. First of all, Enchū did not originally have connections with Kanetoyo. Enchū asked Tōin Kinkata (1291 -1360, 洞院公賢), a court official known for his knowledge and scholarship for his assistance, but Kinkata responded that he was not knowledgeable in the subject and passed the request to Kanetoyo. As a member of the Urabe family, a preeminent family of expertise in the gods and the ancient myths, Kanetoyo was an ideal advisor for the production of an origin narrative. Additionally, Enchū’s writing to Kanetoyo, likely sent through Kinkata as an intermediary, show that he wanted Kanetoyo to confirm whether or not the content was based on correct and authoritative sources. Enchū listed multiple sections from the origin narrative, noted his sources, and concluded with the following request: Concerning the miraculous deeds in the various passages given above, they are recorded in the “True Personal Records of the Monk Ryūben” (Ryūben sō shōshiki 隆弁僧正私記) and the “Petition to the Gods by the Courtier Nakanori” (Nakanori Chōjin saimon 仲範朝臣祭文) and others. [Like this, they are representative] of the religious sphere of the Kanto. However, their origins are unknown. Whether or not [this information] can be seen in the Nihongi or the old records, this should be [examined and] shown. 122 121 Urabe Kanetoyo’s comments are recorded in the diary of Tōin Kink ata, in entries from 1356, 3 rd day of the 8 th month. Tō in Kinkata 洞院公賢, ed. Entairyaku 園太曆, 7 vols., vol. 5, Shiryō sanshū. Kokiroku hen 史料纂集. 古記錄編 (Tokyo: Zoku Gunsho Ruijū Kanseikai, 1970), 239. 122 Ibid, 237-38. 64 Because Enchū’s discursive knowledge came from sources from the Kanto region of eastern Japan, it was difficult for him to know whether or not the content in the Suwa Daimyōjin ekotoba would be accepted by more distant audiences. Both Ryūben 123 and Fujiwara no Nakanori 124 were esteemed intellectuals, but it was still possible that their records were based on regional information that diverged from the knowledge base of readers from the capital. In his reply, Kanetoyo notes that he sent back as separate papers information on the content he found that correlated with entries in the Nihon shoki or the Sendai kuji hongi, and then goes on to list the content which he was not able to verify in the canonical texts on the gods or the historical texts available to him. 125 Most readers would lack Kanetoyo’s expertise, and would only doubt the origin narrative’s accuracy if the content was noticeably questionable. However, the competing claims in origin narratives to justify rewards for defeating the Mongols could help readers to notice inconsistencies between the origin narratives of different temples and shrines. 123 Ryūben was a scholarly monk who served simultaneously as the abbot of Tsurugaoka Hachiman Shrine in Kamakura, and as the chief administrator (chōshi 長吏) of Onjōji. Additio nally, he had close connections with Hōjō Tokiyori, who in turn had connections with Shōmyōji, the temple where the monk Ke nna copied and collected numerous origin narratives, including the Itsukushima Daimyōjin nikki. For a detailed examination of these two monks and their writings, see Nihonmatsu Yasuhiro 二本松康宏, "Suwa engi no henyō : Suwa daiō kara Kō ga Saburō he 諏訪縁起の変容：陬波大王から甲賀三郎へ " in Suwa shinkō no Chūsei: Shinwa, denshō, rekishi 諏訪信仰の中世 : 神話・伝承・歴史, ed. Fukuda Akira 福田晃, Tokuda Kazuo 徳田和夫, and Nihonmatsu Yasuhiro 二本松康宏 (Tokyo: Miyai Shoten, 2015), 127-28. 124 Nakanori was the deputy commissioner of the Bureau of Justice (keibushō 刑部少輔). He appears in the Taiheiki and was known for his scholarship. In a side note in the Taishiden gyokurinshō 太子伝玉林抄, written by Kunkai ( 訓海, 1386-1457), Nakanori is described as “the preeminent man of letters amongst all under heaven.” Makino Kazuo works from several later texts citing writings of Nakanori, such as the Hōkke sōan shō 法華草案抄 by the Nichiren monk Nicchō ( 日朝, 1422-1500) to argue that Nakanori was involved in Tendai, Nichiren, and waka related spheres. Hasegawa Tadashi 長谷川端, ed. Taiheiki 太平記, 4 vols., Shinpen Nihon koten bungaku zenshū (Tokyo: Shō gakkan, 1995), 255; Kunkai 訓海, Hōryūji zō Son'ei-bon Taishi den gyokurin shō 法隆寺藏尊英本太子傳玉林抄, 3 vols., vol. 1 (Tokyo: Yoshikawa Kō bunkan, 1978), 133; Makino Kazuo 牧野和夫, Engyōbon "Heike monogatari" no setsuwa to gakumon 延慶本「平家物語」の說話と学問 (Kyoto: Shibunkaku Shuppan, 2005), 188, 258. 125 Tō in, Entairyaku 園太曆, 238-39. 65 Various factors influence the perceived accuracy of a narrative—familiarity with the narrative or at least certain elements contained within it (i.e. whether or not it is widespread and thus seen as accepted by others); consistency with other texts and discourses; and the authority of those involved in the production, transmission, and later citation of the narrative. Enchū paid close attention to the authority of the producers (collaborating with the sovereign and the shogun), and consistency with the canonical myths, but was less mindful of consistency with other origin narratives. The need for compatibility with other origin narratives led to new strategies for depicting a deity’s power. Rather than constructing arguments of superiority or uniqueness, it was safer to stress similarities with other deities and origin narratives. As Bruno Latour has argued, “By itself a given sentence is neither a fact nor a fiction; it is made so by others, later on.” 126 Within his study of scientific fact-making, Latour demonstrates how controversial issues can become accepted fact through the recognition and writings of other scholars. Scientists might engage alone in practices of experimentation and writing, but it is collective practices, such as citations by later scholars, that provide credibility to the earlier work. Similarly, it is not the accuracy of an origin narrative that makes it true or false, but the usage of the narrative in later texts. References to the origin narratives of other sites and the inclusion of intertextual details provided evidence for the validity of the cited and citing texts. The frequency of citations can grow exponentially once a few notable sources begin to make references to it; when one text references, reproduces, and validates the identity of a god, it increases the likelihood of others reading and writing about that identity. Over time, that which is repeatedly cited becomes widely accepted, and other conflicting discourses are forgotten or condemned as fictional fabrication. 126 Bruno Latour, Science in Action: How to Follow Scientists and Engineers through Society (Cambridge, MA: Harvard University Press, 1987), 25. 66 In order for people to receive an origin narrative as true, two factors came into play: (1) accessibility and (2) consistency. Origin narratives would only be reused if they could be easily referenced, and if there were benefits to quoting them. I will examine these two issues through two types of networks: (1) transmission networks and (2) networks of association. Transmission networks consist of the people, texts (both oral and manuscript), and places which allowed for and stimulated the circulation, collection, and re-writing of origin narratives in temples and shrines spread throughout the Japanese archipelago. As for networks of association, they consist of shared discursive elements that could be used to relate different deities, sacred spaces, or origin narratives. In contrast to conventional views emphasizing the individuality of local deities, it was their commonalities that gave them value within networks of association. From this point of view, the Itsukushima origin narrative was quotable because it was unexceptionable. It was easy for scholarly monks from other religious institutions to find similarities between their deities and the Itsukushima Deity, and thus, to incorporate connections with Itsukushima into their writings. Networks of association pervaded medieval understandings of the gods. Japanese kami deities are often understood through connections with Buddhist deities, and recent scholarship has demonstrated the various types of sacred beings that could be linked through networks of association. 127 In particular, Bernard Faure’s explanation of the pantheons of gods provides a framework for examining the networked relationships between deities. As Faure notes, although 127 Notably, and in a revisionist vein, Mark Teeuwen and Fabio Rambelli argue that the concept of honji suijaku (original forms [of deities] and [their] provisional manifestations 本地垂迹) does not work as precise one-to-one correspondences between kami and buddhas / bodhisattvas, but instead “took the form of complicated networks of associations, establishing links between kami and buddhas, but also between kami and other kami, kami and Yin- Yang deities, buddhas and other buddhas, Wisdom Kings, historical culture heroes both from Japan, China and India, and even demons and witch animals.” Mark Teeuwen and Fabio Rambelli, "Introduction," in Buddhas and Kami in Japan: Honji Suijaku as a Combinatory Paradigm, ed. Mark Teeuwen and Fabio Rambelli (New York: RoutledgeCurzon, 2003), 47. 67 Buddhism presents a strict hierarchy of gods with buddhas and bodhisattvas at the top and worldly deities, like the Japanese kami, at the bottom, 128 deities throughout the pantheon are defined as much by their shared features as they are by the differences which determine their place within the Buddhist pantheon. To make sense of this seeming paradox, Faure utilizes two theoretical approaches, structuralism for the strict, explicit pantheon of deities, and actor-network theory for the implicit pantheons linked together through common characteristics. 129 Understood within Faure’s terms, the networks connecting Itsukushima with other deities and sacred sites make up an implicit pantheon. Instead of beginning from the pantheons of the gods, I work outwards from interrelated deities in a single origin narrative and trace how networks of association developed as other texts reproduced and altered the connections between deities. Scholarly monks made associations with increasingly complex discursive as they read across more texts and worked to make sense of them together. We lose sight of the monks’ interpretive work and intellectual concerns if we only examine the pantheons, and not their origin narrative building blocks. 130 Inspired by Latour’s work, Faure argues that by seeing the gods as “ever-changing nodes within a network constantly in flux, we can more accurately perceive the Japanese gods as the unstable aggregates that they are rather than as static projections of human minds, society, culture, or what have you.” 131 The 128 Satō Hiroo’s examination of kishōmon 起請文 provides clear examples of the hierarchy of deities (both Buddhist and kami), described by Faure. It is worth noting though that Satō argues that the position of the deitie s within the hierarchy is not based on Buddhist or kami designations, but rather on the position of the deities as worldly or otherworldly. While kami are usually of this world, and buddhas often otherworldly, the local instantiations of buddhas (ex. the Great Buddha of Tōdaiji) were separated as worldly deities. Satō Hiroo, "Wrathful Deities and Saving Deities," in Buddhas and Kami in Japan: Honji Suijaku as a Combinatory Paradigm, ed. Mark Teeuwen and Fabio Rambelli (New York: RoutledgeCurzon, 2003). 129 Bernard Faure, Gods of Medieval Japan: Volume 1, The Fluid Pantheon (Honolulu: University of Hawai`i Press, 2016), 1-2, 25-26. 130 Sujung Kim similarly notes the efficacy of Bernard Faure’s methodological approach to Buddhist mythology and the gods of medieval Japan while also arguing for the importance of more detailed case studies which can examine how the systems discussed by Faure operated within certain specific contexts and situations. Kim, Shinra Myojin and Buddhist Networks of the East Asian Mediterranean, 9-10. 131 Faure, Gods of Medieval Japan: Volume 1, The Fluid Pantheon: 10. 68 multidirectional agency allowed by networks creates room for changing and ambiguous identities, where gods are defined in part by the ever-changing web of connections in which they participate. That being said, the work of networks also leads to stability and standardization. Once we turn our attention to endeavors to accumulate information (e.g., from origin narratives and related discursive texts) in one place, regulate the usage of the information within certain textual and discursive forms, and translate disparate or conflicting elements into sensible and useful forms, we can see that the networked associations between deities were doing more than just engaging with one another as shifting streams of interrelations. The networks allowed for flexibility and movement, as well as exclusion, regulation, and standardization of meaning. The distinctions between the deities might be porous, but there were still boundaries. Furthermore, the networks could provide certain actors with power over the production and regulation of those boundaries. 132 Connection rather than Contestation for Affirming Origin Narratives In contrast to the Suwa Daimyōjin ekotoba’s claim for the Suwa Deity’s individual agency in defeating foreign enemies (i.e., the Mongols), other thirteenth to fourteenth century origin narratives began to emphasize the band of deities that accompanied Queen Consort Jingū on her expedition to subjugate the Korean kingdoms. Rather than only mentioning a single deity, 132 Manuel Vásquez distinguishes between flows and networks as methodologies for the study of religion, arguing that while the concept of flows draws attention to the importance of mobility, change, and interrelations in religion, it also risks overemphasizing the pervasiveness and ease of those changes and movements. Instead, Vasquez suggests that network analysis can provide a more balanced perspective on the mechanisms of power in these networks, explaining both how the meaning, mobility, and connectivity of sacred spaces could be made or contested, and also how those same aspects could be controlled and regulated through negating practices such as exclusion or suppression. Manuel Vásquez, "Studying Religion in Motion: A Networks Approach," Method & Theory in the Study of Religion 20, no. 2 (2008): 178-79. 69 incorporating a multitude provided mutual reinforcement between the various texts of other temples and shrines while also demonstrating one’s own deity’s assistance to Jingū. In turn, the repeated listing of the deities strengthened the associations between certain deities as the protectors of the realm, and thus the power of those deities in particular to protect Japan from the Mongols. In the case of Itsukushima, the Itsukushima Daimyōjin nikki never refers to the Mongols or foreign enemies directly, and instead mentions the Itsukushima Deity and Jingū as sisters to provide a basis for other narratives to incorporate Itsukushima into Jingū’s retinue. As I will detail later, the Itsukushima Daimyōjin nikki’s associations worked to incorporate the Itsukushima Deity into the genealogies for Jingū in contemporary origin narratives for Hachiman shrines. It was not necessary to compete for superiority or invent new forms of association; it was enough just to be included. Nevertheless, this was not always easy to accomplish. There was no shortage of medieval origin narratives retelling Queen Consort Jingū’s subjugation of Korean kingdoms, 133 but it was the Hachiman gudōkun (“Teachings on Hachiman for Ignorant Children” 八幡愚童訓, ca. 1293~1318), a contemporary origin narrative for Iwashimizu Hachiman Shrine near Kyoto, that played a key role in standardizing the deities accompanying Jingū. 134 The text was copied and circulated extensively over hundreds of years. Comparing the Hachiman gudōkun with other medieval Hachiman origin narratives, and the over forty Hachiman gudōkun manuscripts, scholars have recognized the Hachiman gudōkun as the 133 For a table comparing twenty different texts including narratives of Jingū’s subjugation of the Korean kingdoms compiled between 1200-1500, see Tada Keiko 多田圭子, "Chūsei ni okeru Jingū kōgō zō no tenkai: Engi kara 'Taiheiki' he 中世における神功皇后像の展開：縁起から『太平記』へ," Kokubun mejiro 31(1991): 194-96. 134 The given production dates for the Hachiman gudōkun of 1293~1318 brings together two different estimates for the dating of the earlier kō 甲 variant. Hagiwara Tatsuo argued that it was made during the reign of Hanazono Tennō (1308-1318), and Shinma Mio suggested that its dates are slightly earlier, between 1293-1300. For Hagiwara’s dating, see his introduction to "Hachiman gudō kun (kō ) 八幡愚童訓（甲)," in 寺社緣起, ed. Sakurai Tokutarō 桜井徳太郎, Hagiwara Tatsuo 萩原龍夫, and Miyata Noboru 宮田登 (Tokyo: Iwanami Shoten, 1975), 168. For Shinma’s dating, see Shinma Mio 新間水緖, Shinbutsu setsuwa to setsuwashū no kenkyū 神仏説話と説話集の研究 (Osaka: Seibundō , 2008), 103. 70 most influential narrative of its time on Hachiman worship, and the same can be said about its influence on the Jingū discourse. 135 Given the Hachiman gudōkun’s wide circulation, its mention of the Itsukushima Deity could potentially influence understandings of Itsukushima more so than any Itsukushima origin narrative. This occurs in the scene prior to Queen Consort Jingū’s departure for Silla, in which the Hachiman gudōkun presents a detailed record of the deities that accompanied her. Because the Hōman Bodhisattva and the Kawakami Deity are the Q ueen Consort’s younger sisters, although Jingū has the body of a woman, her wielding of bow and arrow and armor were the same as everybody else. Suwa, Atsuta, Mishima, Munakata, Itsukushima Deity, and others, together three hundred and seventy-five in total, boarded forty-eight boats at Shika no shima. Each of these three hundred and seventy deities changed their forms so that in every boat there were three hundred and seventy people, each with the same body. Among these [deities], the helmsman was the Shikashima Deity, the general was the Sumiyoshi Deity, and the lieutenant general was the Kōra Deity. 136 The inclusion of details about the three hundred seventy deities in total, as well as all the deities transforming to take the same form, generates an image of deities from across the realm coming together to support Queen Consort Jingū. The passage acknowledges that various deities could be connected with Jingū, but it still elevates certain deities by mentioning their names, and in some cases, their military posts. Additionally, before their departure, the Hachiman gudōkun distinguishes the Itsukushima Deity’s importance as a female dragon deity who, together with the Munakata Deity (of Munakata Shrine in Kyushu) as a female water deity, assists Jingū in a 135 Ono Takashi gives a comprehensive summary of each of the over forty extant manuscripts and categorizes the texts into ten different variations. Ono Takashi 小野尚志, Hachiman gudōkun shohon kenkyū: Ronkō to shiryō 八幡愚童訓諸本研究：論考と資料. (Tokyo: Miyai Shoten, 2001), 7-177; Shinma, Shinbutsu setsuwa to setsuwashū no kenkyū 神仏説話と説話集の研究: 102-03. Emily Simpson compares the Hachiman gudōkun with other Hachiman origin narratives to likewise conclude that the text circulated widely, and served as a model for other Hachiman origin narratives. See Emily Blythe Simpson, "Crafting a Goddess: Divinization, Womanhood and Genre in Narratives of Empress Jingū" (Ph.D. diss, University of California, Santa Barbara, 2019), 165. 136 "Hachiman gudō kun (kō ) 八幡愚童訓（甲)," 175. 71 divination rite to foretell the outcome of the expedition against the Korean kingdoms. 137 Selected as one of the ten deities given by name, Itsukushima became known as part of Jingū’s retinue by the countless readers of the Hachiman gudōkun. Accordingly, all the other deities excluded from mention were at a disadvantage. Origin narratives could make arguments for connections between a deity and Queen Consort Jingū, but preexisting associations in other texts helped make the connections between deities a matter of course. The inclusion of the Itsukushima Deity within the Hacihman gudōkun cannot be linked to the direct influence of a specific Itsukushima origin narrative, but that does not mean that Itsukushima’s success came from luck alone. While the producers of an origin narrative could not control how readers would interpret or use their work, they could employ strategies to lead audiences in certain directions. 138 The scores of references to other deities, sacred spaces, and miraculous events in origin narratives, like the Hachiman gudōkun or the Itsukushima Daimyōjin nikki, created an excess of information which could encourage audiences to search for new interpretations or connections. Ideally, readers would interpret, extract, and reuse associations corresponding with the intentions of the producers of the origin narrative. It did not matter if no one could understand the relevance of every deity referenced. Grasping one association could be enough to then spread the identities of those gods. 137 Six days after the divination with the Itsukushima and Munakata deities, the Hōman Bodhisat tva and the Kawakami Deity appear before Jingū and recite prayers for her success. All of the deities here, aside from the Itsukushima Deity, are powerful deities from Kyushu, and, based on the ease in which associations between deities were made at the time, it is also possible to identify the two pairs of deities, with the Itsukushima Deity being the same as the Hōman Bodhisattva, matching the Itsukushima Daimyōjin nikki, and the Munakata Deity being identical to the Kawakami Deity. Ibid., 171. 138 Henry Jenkins discusses the strategies through which modern day media producers seek to influence the reception of their works in a similar manner. Taking the development of transmedia storytelling across the Matrix movies, comics, video games, and fan action as a case study, Jenkins reveals how both collaborators and audiences were given the freedom to create works within the world of the Matrix. Additionally, the vast numbers of references and allusions within the Matrix films combined with the information from additional transmedia Matrix productions to create complex networks of information that were not necessary to understand the films, but still motivated communities of fans to search for hidden details and new interpretations. Henry Jenkins, Convergence Culture: Where Old and New Media Collide (New York: New York University Press, 2006), 97-101, 22-23. 72 Strategic Networks in the Itsukushima Daimyōjin nikki The Itsukushima Daimyōjin nikki provides a glimpse of the tactics employed to encourage other origin narratives to present the Itsukushima Deity together with Queen Consort Jingū. While we do not know who was involved in the earliest production of the Itsukushima Daimyōjin nikki, writings on the back of one of the two extant manuscripts led Fukushima Kazuhiro to date that copy to the two decades between 1323 and 1343. 139 This manuscript was copied by the monk Zenkai 全海 at Gokurakuji, and the other was made by the monk Kenna ( 釼阿, 1261-1338) at Shōmyōji, both located around the eastern city of Kamakura. 140 These two monks led extensive projects to collect, copy, and circulate religious texts, including numerous temple and shrine origin narratives. 141 In addition to Shōmyōji and Gokurakuji, othe r temples such as Shinpukuji, Daigoji, Enryakuji, and Tōji engaged in similar practices of collecting sacred texts (shōgyō 聖教), and monks at these sites developed extensive networks throughout the 139 Due to the scarcity of paper, texts were often written on the backs of previously used paper. Fukushima Kazuhiro worked from the contents and dates mentioned both on Zenkai’s copy of the Itsukushima Daimyōjin nikki as well as other origin narrative texts copied by him, in order to estimate the dates of the copying of the origin narratives. Fukushima Kaneharu 福島金治, Kanazawa Hōjō-shi to Shōmyōji 金沢北条氏と称名寺 (Tokyo: Yoshikawa Kō bunkan, 1997), 234-36. 140 Makino Kazuo compared the contents of the two manuscripts, concluding that although the content is almost identical, a significant number of examples of divergent wordings suggest that the two manuscripts were based on separate transmissions of the Itsukushima origin narrative. Makino Kazuo 牧野和夫, "Nagato-bon "Heike monogatari" to "Itsukushima Daimyō jin nikki": Nagato-bon "Heike monogatari" seiritsu no ichi katei wo "Itsukushima Daimyō jin nikki" shishū ni saguru 長門本「平家物語』と『厳島大明神日記』：長門本『平家物語』生成の一過程を『厳島大明神日記』四周に探る," in Nagato-bon Heike monogatari no sōgō kenkyū: Kōchū hen 長門本平家物語の総合研究. ：校注篇, ed. Asahara Yoshiko 麻原美子 and Nanami Hiroaki 名波弘彰 (Tokyo: Benseisha, 1998). 141 Nōtomi Jōten estimates that Kenna copied over one thousand texts, although the lack of a signed copyist on many of the documents makes it difficult to calculate an exact number. Nōtomi provides a list of documents signed b y Kenna which continues for eleven pages, and archival research since the publication of his book has continued to uncover additional manuscripts by Kenna. Nō tomi Jō ten 納富常天, Kanazawa Bunko shiryō no kenkyū 金沢文庫資料の研究 (Kyoto: Hō zō kan, 1982), 359-70. 73 realm. 142 More than just enabling texts from as far away as Kyushu to reach Kamakura, 143 the circulation of texts within such networks worked to standardize the narratives and discourses employed within the origin narratives. Bringing a large number of texts together enabled and encouraged previously separate agents (narratives or other texts) to be brought together within shared discursive matrices. 144 As long as the writers and readers of origin narratives were familiar with origin narratives of other temples and shrines being circulated, the individual writers and the other origin narratives both possessed agency in the production of new origin narratives. In this manner, the Itsukushima Daimyōjin nikki could circulate among writers of origin narratives, who could then include the Itsukushima Deity within their discussions of gods associated with Jingū. Working primarily from networks of scholarly monks connected with the copyists of the Itsukushima Daimyōjin nikki (Kenna of Shōmyōji and Zenkai of Gokurakuji), Makin o Kazuo suggested that Kenna received the text through contact with networks connected with the orders at Saidaiji or Tōdaiji Kaidan’in 東大寺戒壇院, while Zenkai received his text separately through connections with the scholarly monks at Mt. Hiei. 145 While this is certainly possible, the Saidaiji 142 Concerning the networks developed by medieval Japanese monks broadly see William E. Deal and Brian Douglas Ruppert, A Cultural History of Japanese Buddhism (Malden, MA: Wiley-Blackwell, 2015), 153-60. Concerning the copying and collecting of texts related to the deities (kami) at the temple Shinpukuji, see Abe Yasurō, "Shintō as Written Representation: The Phases and Shifts of Medieval Shintō Text s," Cahiers d'Extrême- Asie 16 (2006). 143 In one case, a disciple of Kenna named Kiyosane-bō 浄実房 from Dainichiji Temple 大日寺 in Ise Province was sent to Daikakuji 大覚寺 in Kyushu (Dazaifu, in modern day Fukuoka) to collect a text that had been borrowed by a royal prince there. Fukushima, Kaneharu 福島金治, "Kanazawa Shōmyōji to Ise, Chinzei 金沢称名寺と伊勢・鎮西," in Bijutsushi ronsō zōkei to bunka 美術史論叢造形と文化, ed. Shimizu Mazumi 清水眞澄 (Tokyo: Yūzankaku Shuppan, 2000), 265-66. 144 Bruno Latour, Reassembling the Social: An Introduction to Actor-Network-Theory (New York: Oxford University Press, 2005), 228-30. 145 Makino Kazuo 牧野和夫, "Nagatobon "Heike monogatari" to "Itsukushima Daimyō jin nikki": Nagato-bon "Heike monogatari" seiritsu no ichi katei wo "Itsukushima Daimyō jin nikki" shishū ni saguru 長門本「平家物語』と『厳島大明神日記』：長門本『平家物語』生成の一過程を『厳島大明神日記』四周に探る," in Nagatobon Heike monogatari no sōgō kenkyū: Kōchū hen 長門本平家物語の総合研究：校注篇, ed. Asahara Yoshiko 麻原美子 and Nanami Hiroaki 名波弘彰 (Tokyo: Benseisha, 1998), 240-48. 74 network proposed by Makino corresponds with discussions of the transmission of the Shōbōrinzō ( 正法輪蔵, 1317), an account of the life of Prince Shōtoku that includes a di fferent account of the Itsukushima origin narrative. 146 This raises the question—how likely is it that two texts with divergent associations for the same detail, the sisters of the Itsukushima Deity, received their information from networks at the same place during around the same time? Considering that different discourses on Itsukushima could have traveled at different times, or to different people within the same larger network, it is not necessarily problematic. If this is the case though, simply saying that a discourse circulated through such extensive networks might not elucidate much other than the importance of these networks for discursive transmissions. Rather than speculating about the transmission of the Itsukushima origin narratives through temple networks, I propose examining transmission networks and networks of association together. During the medieval period, the Itsukushima Deity was described in multiple ways, but as shown by Tanaka Takako, the most common identification was as the daughter of the Dragon King Sagara and the sister of the eight year old Dragon Princess, famous from the episode of her enlightenment in the “Devadatta” chapter of the Lotus Sutra. 147 It was through this association with the Dragon Princess that the Itsukushima Daimyōjin nikki linked the Itsukushima Deity with 146 The Saidaiji order in fact played an important role in promulgating the Prince Shōtoku cult in western Japan, particularly through networks along the Inland sea. See David Quinter, "Localizing Strategies: Eison and the Shōtoku Taishi Cult," Monumenta Nipponica 69, no. 2 (2014): 194-95. For a more historical discussion of the Saidaij order in relation to water ways in western Japan, see Amino Yoshihiko 網野善彦, Mōko shūrai 蒙古襲来, ed. Inaba Nobumichi 稲葉伸道, 19 vols., vol. 5, Amino Yoshihiko chosakushū 網野善彦著作集 (Tokyo: Iwanami Shoten, 2007), 290-93. 147 Tanaka Takako has argued that the identity of the Itsukushima Deity as a dragon deity and the sister of the Dragon Princess underlay the description of the Itsukushima Deity as Princess Ashibiki in the Itsukushima no honji variant of the Itsukushima origin narrative. While her analysis brings together and compares some of the texts in table 1, her discussion of the texts does not examine how or why the connections between these texts developed. See Tanaka Takako 田中貴子, Gehō to aihō no chūsei 外法と愛法の中世 (Tokyo: Sunagoya Shobō, 19 93), 21-38. For a brief summary of the Dragon Princess episode in the Lotus Sutra and its significance, see Ryuichi Abe, "Revisiting the Dragon Princess: Her Role in Medieval Engi Stories and Their Implications in Reading the Lotus Sutra," Japanese Journal of Religious Studies 42, no. 1 (2015): 32-40. 75 Queen Consort Jingū. Discussions of the familial relationship between Itsukushima and the Dragon Princess were prevalent enough in contemporary texts that it provided a convincing base to help legitimize the new connection with Jingū. Looking beyond origin narratives, descriptions of the Itsukushima Deity as the sister of the Dragon Princess are found in encyclopedic compilations of religious texts, Nihon shoki commentaries, Lotus Sutra jikidan (straight talk 直談) commentaries, nō plays, the Tale of the Heike, Prince Shōtoku narratives, and pilgrimage accounts (table 2). Through such mediums, discourses on the Itsukushima Deity spread far beyond its origin narratives, increasing accessibility and consistency. 76 Table 2: Mentions of Itsukushima as Daughter of Dragon King Sagara Text Date Sisters Gukanshō 愚管抄 1219 Itsukushima; others not listed Shinra Myōjin -ki 新羅明神記 1268-1344 Itsukushima; Shinra Deity (of Onjōji); Dragon Prin cess (from Lotus Sutra) Itsukushima Daimyōjin nikki 厳島大明神日記 1290-1333 Itsukushima; Queen Consort Jingū; Yodohime; Dragon Princess Shōbōrinzō 正法輪蔵 1317 Itsukushima; Chikubushima; Enoshima; Dragon Princess Keiranshuyōshū 溪嵐拾葉集 Pre 1318 Itsukushima; others not listed Jittokushū 寺徳集 1344 Itsukushima; Shinra Deity; Dragon Princess Onjōji denki 園城寺伝記 Post 1343 Itsukushima (subscript: Toyotama-hime); Shinra Deity; Dragon Princess Itsukushima no engi いつくしまのゑんぎ 1346 Princess Ashibiki (Itsukushima Deity)'s younger sister as daughter of Sagara (i.e., Dragon Princess) Kakuichibon Heike monogatari 覚一本平家物語 1371 Itsukushima; others not listed Rokuon'in Saigoku geko-ki 鹿苑院西国下向記 1389 Itsukushima; Kehi Deity; Shinra Deity; Ōmuro Deity Hakuhōshō 白宝抄 14th century Itsukushima; Dragon Princess; Seiryū Gongen (of Daigoji) Nihon shoki shishō 日本書紀私抄 1341-1420 Itsukushima; Toyotama-hime; Tamayori-hime Sanbō -in denbō kechimyaku 三宝院伝法血脈 1336-1573 Itsukushima; Dragon Princess; Seiryū Deity Jimon denki horoku 寺門伝記補録 1394-1424 Itsukushima; Shinra Deity; Dragon Princess Nihon shoki kikigaki 日本書紀聞書 1419 Itsukushima; Shinra Deity; Queen Consort Jingū; Seiryū Gongen; Dragon Princess Itsukushima jinja shuzō kan'ensō narabini jo 厳島神社修造幹線疏井序 ~1420 Itsukushima (as youngest daughter); others not listed Kō ban kuzetsu 弘鑁口説 Pre 1426 Itsukushima (subscript: same in essence as Tenkawa); Seiryū Gongen; Shinra Deity; Dragon Princess Gaun Nikkenroku 臥雲日件録 1447 Itsukushima; Kasatori Deity; Izu Enoshima Deity Jinkeishō 塵荊鈔 1482 Itsukushima; others not listed Kechimyaku shishō 血脈私抄 1487 Itsukushima; Seiryū Gongen; Dragon Princess - similar text to Sanbō-in denbo kechimyaku Nagatobon Heike monogatari 長門本平家物語 ~15th century Itsukushima; Queen Consort Jingū; Yodohime; Dragon Princess Genpei jō suiki 源平盛衰記 ~15th century Itsukushima; others not listed 77 Hakurakuten 白楽天 Mid Muromachi Itsukushima; others not listed. But notably, deities mentioned strongly resemble those in Hachiman gudōkun Nihon shoki sanso 日本書紀纂疏 Mid 15th century Itsukushima; others not listed Hokkekyō jikidanshō 法華経直談抄 1546 Itsukushima, Seiryū Gongen, Dragon Princess Itsukushima - Daimokutate 厳島：題目立 Pre 1573 Itsukushima; others not listed Itsukushima Daimyōjin engi sh ō 厳島大明神縁起抄 Pre 1602 Itsukushima; others not listed Figure 1: Map of sacred sites included in table 2 (size of point reflects number of occurrences) 78 While some texts gave differing accounts of the siblings of the Dragon Princess, they almost always included the Itsukushima Deity. With no one text as the authoritative reference for the Dragon Princess’s family tree, the best way to provide the inclusion of a particular deity with authority was to match the associations with other texts. Even if some texts worked to emphasize a particular deity as the sister of the Dragon Princess, retaining the other sisters often included elsewhere made the passage less suspicious. From this perspective, it makes sense that texts such as the Kōban kuzetsu ( 弘鑁口説, pre 1426), a compilation of oral transmissions by the Daigoji monk Kōban, incl udes not only the Daigoji deity Seiryū Gongen as the Dragon Princess’s sister, but also the Itsukushima Deity, and the Shinra Deity of Onjōji. 148 An earlier Daigoji origin narrative from 937, for example, only mentions Seiryū Gongen as the third sister, but Kōban apparently knew of other transmissions that included the Shinra Deity, 149 and considered it best to include them. 150 At first, the basis for the Itsukushima Deity being the sister of the Dragon Princess might not have been any stronger than the case for the Seiryū Gongen or the Shinra Deity, and in fact, the 937 Daigoji engi dates the identification of the Seiryū Gongen as the daughter of the Dragon King earlier than any extant text for Itsukushima. Later, however, the frequent and consistent mentions of the Itsukushima Deity made it worthy of inclusion. More 148 "Kō ban kuzetsu 弘鑁口説," in Zoku gunsho ruiju. Shakke bu 續群書類從. 第 27 輯上 ( 釈家部), ed. Hanawa Hokinoichi 塙保己一 and Oota Toshiro 太田藤四郎 (Tokyo: Yagi Shoten, 2013), 718. 149 There are at least four texts which give the Shinra deity, Itsukushima Deity, and the Dragon Princess from the Lotus Sutra as the three daughters of the Dragon King Sagara, and all four texts are compilations of Onjōji rela ted discourses. See Kuroda, Satoshi 黒田智, "<Shiryō shōkai> Shinra Myōjin ki < 史料紹介> 新羅明神記," Tōkyō daigaku shiryō hensanjo kenkyū kiyō 11(2001): 87; Suishin 水心, "Jitokushū 寺徳集," in Zoku gunsho ruiju. Shakke bu 續群書類從. 第 28 輯上 ( 釈家部), ed. Hanawa Hokinoichi 塙保己一 and Toshiro Oota 太田藤四郎 (Tokyo: Yagi Shoten, 2013), 18; "Jimon denki horoku ( 寺門伝記補録)," in Dai Nihon bukkyō zensho 大日本仏教全書 (Tokyo: Daihō rinkaku, 2007), 118; "Onjōji denki 園城寺伝記," in Dai Nihon bukkyō zensho 大日本仏教全書 (Tokyo: Daihō rinkaku, 2007), 7. 150 The Daigoji engi in the Dai Nihon Bukkyō zensho does not included a dated colophon, but an earlier manuscript has been dated to 937. "Daigoji engi 醍醐寺縁起," in Dai Nihon bukkyō zensho. Jishi sōsho daiichi 大日本佛教全書 . 寺誌叢書第一 (Tokyo: Daihō rinkaku, 2007), 247. On the 10 th century dating of the engi, see Sarah Alizah Fremerman, "Divine Impersonations: Nyoirin Kannon in Medieval Japan" (Stanford University, 2008), 51. 79 than anything about the Itsukushima Deity herself, it was these references that made and preserved her identity as the sister of the Dragon Princess. This is not to say that these associations were never doubted. The courtier and scholar Ichijō Kanera ( 一条兼良, 1402-1481), in a commentary on the Nihon shoki, the Nihon shoki sanso 日本書紀纂疏, touches on the origin narrative of the Itsukushima Deity, “It is popularly said, ‘The Itsukushima Deity is the daughter of Dragon King Sagara.’ However, the basis for this is still not known.” 151 As Ichijō suggests, there are many religious commentaries and narratives which describe the Itsukushima Deity as the daughter of the dragon king, but there is no definitive text for this characterization. Ichijō was trying to make sense of two sets of associations, one with Itsukushima as the daughter of the Dragon King, and the other with Itsukushima as Ichikishima 市杵島, one of the three daughters of Amaterasu. 152 Bringing these originally separate discourses together raised a problematic question: what is the relationship between the dragon kings of Buddhist cosmology and the genealogy of the Japanese kami deities? The “Genealogy of the Gods” (Jingi keizu 神祇系図) by Ichijō’s contempo rary, the Tendai monk Ryōhen 良遍, identifies the dragon deities of the Nihon shoki myths as the children of Izanami and Izanagi, which positions them before Amaterasu or her descendants. 153 Ryōhen 151 Ichijō Kanera 一條兼良, Nihon shoki sanso 日本書紀纂疏, ed. Kiyohara Nobukata 淸原宣賢 (Tokyo: Seisaku hatsubai Yagi Shoten, 1977), 243. 152 The earliest identification of Itsukushima as Ichikishima appears in the Shaku Nihongi, compiled by Urabe Kanekata between 1274-1301. However, a record from 867 in the Nihon sandai jitsuroku 日本三代実録 mentions Itsukushima and Munakata enshrined together (Itsukushima Munakata otōm e no kami 伊都岐島宗形小専神), suggesting that the identity of Itsukushima as Ichikishima (one of the three Munakata deities) existed since the ninth century. For the Shaku Nihongi see Urabe Kanekata 卜部兼方, Shaku Nihongi 釋日本紀, ed. Onoda Mitsuo 小野田光雄, Shintō taikei. Koten chūshaku hen (Tokyo: Shintō Taikei Hensankai, 1986), 147. Concerning the Nihon sandai jitsuroku see Matsuoka Hisato 松岡久人, Aki Itsukushimasha 安芸厳島社 (Kyoto: Hō zō kan, 1986), 34. 153 While Ryōhen’s “Genealo gy of the Gods” is undated, Ryōhen wrote commentaries on the kami deities and the Nihon shoki myths between 1390 (Taike zasshi 台家雑誌) and 1424 (Shindai no maki shikenmon 神代巻私見聞). Even if the “Genealogy of the Gods” was written somewhat earlier or later, it could still coincide with the time of Ichijō Kanera’s studies. For an overview of Ryōhen’s commentaries, see Abe Yasurō 阿部泰郎, "Ryōhen Nihon 80 avoids the disparate timing by only identifying Itsukushima with the Ichikishima Deity, but it would not have been hard for scholars like Ichijō Kanera to notice the problem. 154 In the end, however, he does not judge either of them to be false. In practice, the circulation and multiplication of the idea mattered more than its origins. We cannot trace the exact origins of the discourse on Itsukushima as the daughter of the dragon king, but looking at how it spread and circulated, we can see how the identity standardized, and thus legitimized, understandings of the Itsukushima Deity. Incorporation and Standardization of the Jingū Association Utilizing shared discursive matrices with the Dragon Princess, the Itsukushima Daimyōjin nikki worked to expand Itsukushima’s network of associations to include Queen Consort Jingū through links with other sacred sites. The sisters of the Dragon Princess worked particularly well as the basis for drawing connections because the Dragon Princess was well known but without any authoritative source for the number or identities of her sisters. Within medieval networks of association, the arguments for the association mattered less than consistency with other texts. It was difficult to raise doubts about the connections between deities unless the associations conflicted with those made in other texts. Therefore, the ideal subject for connecting deities had shoki chūshaku no yōsō: Gakumon no gendan kara ‘monogatari’ toshite no <Nihongi> e 良遍「日本書紀」注釈の様相-- 学問の言談から" 物語" としての< 日本紀> へ," Kokugo to kokubungaku 71, no. 11 (1994). 154 Ryōhen’s “Genealogy of the Gods” identifies the Seiryū Gongen of Daigoji with the Nihon shoki dragon deity Tamayori hime 玉依姫 (the genealogy also includes two other dragon deities from the myths, Hodakami 穂高見 and Toyotama hime 豊玉姫), presumably because of its association with the Dragon Princess. However, this does not become a problem within the genealogy, because the Seiryū Gongen is not identified with any other gods from the myths. Nevertheless, a subscript within the genealogy suggests that the identities of the dragon deities within the kami genealogies was a debated topic. Ryōhen prefaces the section on the dragon deities with an annotation saying, “It is popularly said.” Unfortunately, the subsequent text is illegible, but Ryōhen follows with “I state,” suggesting that his view of the dragon deities differs from some others. See Ryōhen 良遍, "Jingi keizu 神祇系図," in Tendai Shintō 天台神道, ed. Tamura Yoshirō 田村芳朗, Sueki Fumihiko 末木文美士, and Satō Masato 佐藤真人, Shintō taikei. Ronsetsu hen (Tokyo: Shintō Taikei Hensankai, 1990), 604. 81 to balance specificity and imprecision. It had to defined narrowly enough to make the connection clear, while still providing room for explaining differences with other texts. The Itsukushima Daimyōjin nikki itself stresses the importance of the Itsukushima Deity’s associations, as can be seen from the subtitle “The subject of being one in essence (ittai 一体) with Kamado and Hakusan.” 155 Neither Kamado Shrine, located on Mt. Hōman 宝満山 outside of the city of Dazaifu in modern day Fukuoka prefecture, nor Hakusan 白山 in the Hokuriku region in the northwest by the Sea of Japan, are known for having close relations with Itsukushima, but discursive commonalities allowed for the connections to be made. The first four sentences of the text define the Itsukushima Deity through these networks of association. She is the younger sister of the eight-year old daughter of Dragon King Sagara, and the older sister of Queen Consort Jingū and Yodo-hime. On the contemplative peak of Hakusan in the northern realm, the provisional manifestation Zenchō Gongen appears — her virtue full, her will heavenly, and her benefits pervasive. [For the sake of] all people, the traces of [Mt] Hōman come from the western sea and mani fest in the high peaks of Kamado—her power new, her will heavenly, her miraculous efficacy eternal. Her spiritual power illumines the day in truth. Therefore, in order to protect the kings of one hundred generations and spread the teaching of the [Buddhist] Law, [the Itsukushima Deity,] wishing to come closer to the royal palace, approached from Kyushu, and was born [at Itsukushima] during the reign of Suiko Tennō. 156 Without situating these statements within the context of the referenced networks of associations, the passage makes little sense. If the reader understood the connections though, these first few sentences could become more important than the rest of the narrative for providing Itsukushima’s identity. First, the associations with Dragon King Sagara and Mt. Hōman provide a basis for connecting Itsukushima with Queen Consort Jingū by overlapping with Hachiman discourses that linked Mt. Hōman and the daughters of the Dragon King with Jingū. Secondly, 155 Kanazawa Bunko, 金沢文庫, ed. Kanazawa Bunko no chūsei shintō shiryō 金沢文庫の中世神道資料 (Yokohama: Yokohama kenritsu Kanazawa Bunko, 1996), 51-53. 156 Ibid., 52. 82 the associations with Hakusan and Kamado connected Itsukushima with a separate discourse about the gods near the capital who have traditionally served as protectors of the realm. Either one of these associations could have been enough to make a compelling case for Itsukushima’s power to defeat the Mongols but including both increased the chances of others citing Itsukushima within discussions of either discourse. Separate genealogies of dragon deities (e.g., Itsukushima) and Hachiman deities as sisters of the Dragon Princess might seem incongruous, but they instead provided a foundation for Itsukushima’s association with Queen Consort Jingū. Comparing the associations in the Itsukushima Daimyōjin nikki with the Hachiman Usagū gotakusenshū (“Hachiman Usa Shrine Oracle Collection,” hereafter Usa takusenshū, 八幡宇佐宮御託宣集) illuminates the role of Hachiman discourses on Queen Consort Jingū in mediating the connection with Itsukushima. Bringing together origin narratives from various Hachiman sites in Kyushu, the compiler of the Usa takusenshū, Jin’un ( 神吽, 1231-1314), situated the history, myths, and identities of the Usa Hachiman Deity in relation to the other sites of Hachiman deities. Similar to the Itsukushima Daimyōjin nikki, this was done in part by establishing kinship lines between the deities. After recording the origin narrative of a site, Jin’un added diagrams mapping the relations between the different deities as described in that text. For example, one account provides two genealogies of the five daughters of the Dragon King Sagara side by side. The first lists the daughters as: (1) the Dragon Princess: attained Buddhahood at eight years of age (2) the Kamado Deity: Chikuzen Province (3) the Kashii Deity: Chikuzen Province (4) the Kawakami Deity: Hizen Province (5) the Takeo Deity: Higo Province The second continues: (1) Jewel Brocade (Tamananishiki 宝錦): the eight-year old Dragon Princess (2) Hōman Bodhisattva: the Kamado [Deity] 83 (3) Ōtarashi hime 大帯姫 (Queen Consort Jingū): the Kashii [Deity] 157 (4) Toyo-hime (Yodo-hime): the Kawakami [Deity] (5) the Takeo Deity. 158 The first genealogy focuses on Hachiman deities in Kyushu, and the second matches them with deities from the Nihon shoki. Together, they harmonize the genealogies of the dragon sisters and Queen Consort Jingū. While the Usa takusenshū itself does not mention Itsukushima, Queen Consort Jingū and Yodo-hime are the two deities given as the sisters of Itsukushima in the Itsukushima Daimyōjin nikki. Additionally, the Itsukushima Damyōjin nikki presents the Itsukushima Deity as one in essence with the Hōman Deity, who is also included in the Usa takusenhū’s second genealogy. It is impossible to know if the producers of the Itsukushima Daimyōjin nikki directly referenced the Usa takusenshū, but it is clear that they knew about the same grouping of sisters and worked to make the Itsukushima Deity fit into this account without contradictions. A notation written on the back of a later manuscript of the Usa takusenshū shows that the Itsukushima Daimyōjin nikki’s identification of Itsukushima with Jingū was recognized by later readers. Composed on the back of the genealogy chart with the five sisters of the Dragon Princess, the comment reads, “It is said: the Itsukushima Deity and Hōman [Bodhisattva] are one in essence. This can be seen with certainty in the Sutra of the Ocean Dragon Kings. It states there, ‘The girl of the dragon king, the Jewel Brocade Sage, after passing unmeasurable kalpas, 157 This entry on Queen Consort Jingū also includes Hachiman as Queen Consort Jingū’s son, and as one in essence with the Hakozaki Deity, and Ugafukiaezu as Hachiman’s father. The 1680 Ō tani University Library Manuscript (fig. 1) separates Ugafukiaezu from Queen Consort Jingū and Hachiman, but this appears to be a mistake made while copying and trying to make sense of the formatting of the genealogy. In the Nihon shoki and Kojiki, Ugayafukiaezu is the child of Hikohohodemi no mikoto 彦火火出見尊 and Toyotama-hime 豊玉姫, and the father of Jinmu tennō. In the Usa takusenshū, Ugayafukiaezu appears before Jingū, and they make a deal that he will help her if she agrees to marry him after subjugating the Korean kingdoms, returning to Japan, and giving birth to Hachiman. See Shigematsu Akihisa 重松明久, ed. Hachiman Usagū gotakusenshū 八幡宇佐宮御託宣集 (Tokyo: Gendai Shichōsha, 1986), 100 -101. 158 Ibid, 101. 84 became a buddha named Universal World Thus Come One.’” 159 Even though there is no mention of Itsukushima or Hōman within the sutra, the use of the same na me, Jewel Brocade, for the Dragon Princess in the Usa takusenshū lends credibility to the related genealogy in the Usa takusenshū. 160 Once that is accepted, the only step remaining is to show that the Itsukushima Deity is the sister of the Dragon Princess, which was already widely known. Networks for the Protection of the Realm of Japan The discourses concerning Queen Consort Jingū developed and spread widely after the Mongol invasions, and they coexisted with a discourse on the deities protecting the realm through their ability to protect the sovereign. 161 Within the latter however, the distance of Kyushu from the capital became an argument against a site’s ability to protect the court. Strategically choosing both lines of argumentation, scholarly monks used the multiple associations within origin narratives such as the Itsukushima Daimyōjin nikki to create links with discourses concerning both Jingū and the protection of the sovereign. The Itsukushima Daimyōjin nikki presents the deity as both a protector of the realm and a subjugator of foreign enemies through additional networks of association. Returning to the 159 Ibid., 102. This citation from the sutra condenses a larger passage within the Sutra of the Buddha’s Preaching on the Dragon King of the Sea (Ch. Hailong-wang jing; J. Kairyūōkyō 佛説海龍王經 T. 598). The passage begins, “At that time, the dragon king of the sea had a daughter. Her name was Jewel Brocade Free from Impurities. This Jewel Brocade girl, at some time, will become [a buddha] of the unsurpassable, correct, true enlightenment, and the supreme, perfect awakening.” The text later continues, “The Buddha understood the intentions of the various devas, dragon deities, and gandharvas and addressed the monks, ‘This Jewel Brocade girl after passing three hundred unmeasurable kalpas, became a buddha named Universal World Thus Come One, and reached true, correct awakening. T 598, 15:149-150. 160 Ibid., 101. 161 Satō Hiroo describes the basis for the protection of the realm, and Japan as a real m of the gods (shinkoku 神国) in the classical period (kodai) as the protection of the sovereign. Edicts from 869, recorded in the Sandai jitsuroku 三代実録 describe the deities leading all of the other deities of the realm, similar to the Hachiman gudōkun, but it is the relationship with the royal family that is emphasized in these earlier examples, using titles like kōtaijin ( 皇大神, literally, great deity of the sovereign), and gobyō (御廟, the ancestral shrine). See Satō Hiroo 佐藤弘夫, Kami, hotoke, ōken no chūsei 神・仏・王権の中世 (Kyoto: Hōzōkan, 1998), 319 -22. 85 subtitle’s characterization of the Itsukushima Deity as one in essence with Kamado and Hakusan, these associations also emphasize Itsukushima’s role as a protector of the realm. A comment by a writer or later copyist of the Itsukushima Daimyōjin nikki at the end of the manuscript explains the significance of the Hakusan Gongen reference: “I examined and cite the Sannō kōshiki, it says: ‘The fifth [Hie Deity] Marōdo no miya 客人宮 is Eleven-Headed Kannon, and manifests in response as the miraculous deity Hakusan Zenjō.” 162 This passage provides a rare view of how medieval intellectuals interpreted the associations within origin narratives. The commenter found an association with the Hakusan Gongen in the Sannō koshiki 山王講式, a liturgical text for the worship of the Sannō deities of Mt. Hiei, and then used that external information to u nderstand the connection between Hakusan and Itsukushima. Because the Marōdo Deity was also enshrined at Itsukushima Shrine, the connection between Hakusan and the Marōdo Deity strengthened Itsukushima’s connection with Mt. Hiei and its stature as a site for the protection of the realm. Enryakuji Temple and Hie Shrine, the religious sites on Mt. Hiei, share a long history as the protectors of the realm par excellence. Beginning with Saichō ( 最澄, 766-822), the founder of Enryakuji, the Tendai school and its institutions stressed their ability to protect the realm. 163 While the threat of a foreign invasion by the Mongols led to increased attention to Kyushu temples and shrines associated with Queen Consort Jingū, Mt. Hiei ritualists continued to perform rites for the sovereign and for the protection of the realm, maintaining Mt. Hiei’s place 162 The Sannō koshiki text within the Gyosan soshō 魚山叢書 matches the quotation word for word, supporting the writer’s claim of referencing the Sannō koshiki text itself. For the quotation within the Itsukushima Daimyōjin nikki, see Kanazawa Bunko, Kanazawa Bunko no chūsei shintō shiryō 金沢文庫の中世神道資料, 53. For the Sannō koshiki text (containing a colophon dating an earlier copying of the manuscript to 1306), see “Sannō koshiki 山王講式,” in Gyosan soshō 魚山叢書, Tokyo University Historiographical Insitute manuscript, 94. 163 Mikael S. Adolphson, The Gates of Power: Monks, Courtiers, and Warriors in Premodern Japan (Honolulu: University of Hawai`i Press, 2000), 33-34. 86 as a key site ensuring the safety of the realm. 164 Therefore, the Itsukushima Daimyōjin nikki’s connections with the deities of Hie, in turn, helped to similarly present Itsukushima as a protector of the realm. Ideally, texts about the Hie Sannō Deity would help the association to spread by referencing the connection with the Itsukushima Deity in turn, but this was out of the control of the Itsukushima origin narrative. The only thing the Itsukushima Daimyōjin nikki could do was to make a logical connection based on records about the Hiei deities and hope for others to accept and reproduce the association. Even while the Mongol invasions influenced religious practices at Mt. Hiei, religious intellectuals on the mountain continued to base their claims on the established idea of the mutual interdependence of the Royal Law and the Buddhist Law. In the 1284 Tendai bosatsu kai shinzoku ikkanshō (Notes on the Tendai Bodhisattva Precepts and the Unity of the Real and the Conventional 天台菩薩戒真俗一貫抄) by the Enryakuji precept lineage monk Ejin 恵尋, the author details a line of argumentation for the Tendai bodhisattva precepts at Mt. Hiei as “one and the same with way of the governance of the realm” ( 政道一同之戒儀), and thus “Sanmon (Mt. Hiei) as the only one called the site for the protection of the sovereign” ( 山門独名皇帝本命道場事). 165 As Funata Jun’ichi explains, the table of contents alone discloses Ejin’s line of reasoning in connecting the bodhisattva precepts, governance of the realm, and protection from the Mongols. The fourth and fifth sections argue that the people who receive the bodhisattva 164 Through a summary of requests for prayers to subjugate the Mongols by the court and by the Kamakura bakufu, Aida Nirō’s research sho ws that early prayers were mostly ordered by the court or the royal family and sent to powerful temples and and shrines (particularly those within the twenty-two shrine system) near the capital. However, once the bakufu became more involved in ordering requests for prayers, prayers to the kokubunji and ichinomiya temples and shrines in the provinces became more common. Nevertheless, Enryakuji and other elite temples continued to receive orders for prayers. Enryakuji performed rites against the Mongols in 1271; 1274; and 1276. Aida Nirō 相田二郎, Mōko shūrai no kenkyū 蒙古襲来の研究 (Tokyo: Yoshikawa Kō bunkan, 1958), 59-90, see 86-87 for prayers performed at Enryakuji. 165 Funata Jun'ichi 舩田淳一, "Chūsei Eizan no kairitsu fukkō: Rissō Ejin no shisō to kokka -kan wo megutte 中世叡山の戒律復興 : 律層恵尋の思想と国家観をめぐって," Bukkyō daigaku sōgō kenkyūjo kiyō 16(2009): 361, 63. 87 precepts and the lands they stand on cannot be defeated by foreign realms. Then, the sixth and seventh sections assert that the sovereign of the realm is one in essence with the bodhisattva precepts. Finally, the eleventh section focuses on Mt. Hiei as the only site for the protection of the sovereign, and the thirteenth section concludes that the sovereign of Japan shall defeat foreign realms through the power of the precepts. 166 The connection with the Kamado Deity also aligns Itsukushima with the protection of the realm, but in this case, by emphasizing Itsukushima’s movement away from Kyushu. The connection with Kamado risked situating Itsukushima together with all the sacred spaces in Kyushu associated with Queen Consort Jingū as distant from the capital, but the Itsukushima Daimyōjin nikki preemptively provides a counter-argument. The Itsukushima Daimyōjin nikki states, “Therefore, in order to protect the kings of one hundred generations and spread the teaching of the [Buddhist] Law, [the Itsukushima deity] wishing to come closer to the castle of the sovereign (ōjō 王城) approached from Kyushu, and was born [at Itsukushima] during the reign of Suiko Tennō.” 167 The term “castle of the sovereign” refers to the capital, but was rarely used except when discussing the shrines within the twenty-two shrine system, which were all shrines near the capital designated as protectors of the realm. 168 Likewise, Iwashimizu Hachiman origin narratives used this same phrasing to argue that the Hachiman Deity moved from Usa Shrine in Kyushu to Iwashimizu Hachiman Shrine south of Kyoto in order to guard the royal court. For example, the Miyadera enji shō (Notes on Matters of the Shrine Temple 宮寺縁事抄), 166 Ibid., 361. 167 Kanazawa Bunko, 金沢文庫, ed. Kanazawa Bunko no chūsei shintō shiryō 金沢文庫の中世神道資料 (Yokohama: Yokohama kenritsu Kanazawa Bunko, 1996), 52. 168 On the Twenty-two shrine system and the term “ōjō,” see Uejima Susumu 上島享, Nihon chūsei shakai no keisei to ōken 日本中世社会の形成と王権 (Nagoya: Nagoya Daigaku Shuppankai, 2010), 244-51. On the Twenty-two shrine system more generally, see Allan G. Grapard, "Institution, Ritual, and Ideology: The Twenty-Two Shrine- Temple Multiplexes of Heian Japan," History of Religions 27, no. 3 (1988). 88 compiled by the Iwashimizu Hachiman abbot Tanaka Dōsei ( 田中道清, 1169-1206), presents its case in an oracle, “I (Hachiman) moved close to the capital, wanting to protect the castle of the sovereign.” 169 The Itsukushima Daimyōjin nikki never presents Itsukushima as unique in its ability to protect the realm and subjugate foreign enemies. Instead, the text replicates the arguments made by other sites by incorporating Itsukushima through networks of association. Other sites, like Kamado and Munakata shrines in Kyushu, tried but struggled to present their deities as simultaneously protectors of their individual regions, protectors of the realm, and subjugators of foreign enemies. For example, the “Record of the Mt. Kamado Hōman Bodhisattva,” (Kamadosan Hōman daibosatsu ki 竈門山宝満大菩薩記) also copied by Kenna at Shōmyōji, employs an argument of the deity manifesting in two bodies, one being a protector of the realm (i.e., the court), and the second being a subjugator of foreign realms. The text recognized the importance of both discourses, but as a Hachiman site in Kyushu, it struggled to substantiate both roles. “[As for the] Hachiman bodhisattvas, when the deities to protect the various provinces were decided by the [Hachiman] of Ōita Shrine, even though the Hōman bodhisattva received [this designation to protect its province], it was saddened about [not receiving] the title of protector of the realm.” 170 Hōman’s role as a protector of the province in which it is located was clear, but in order to make a claim about protecting the realm that would not be immediately rejected as false, the origin narrative resorts to an argument about the deity’s deep-rooted desire to protect the realm. While arguments about deity’s appearing in multiple forms were neither uncommon nor dubious, the supporting evidence of the deity’s 169 "Miyadera enji shō 宮寺縁事抄," in Shintō taikei. Jinja hen, Iwashimizu 神道大系. 神社編：石清水, ed. Murata Masashi 村田正志 (Tokyo: Shintō Taikei Hensankai, 1988), 48. 170 Kanazawa Bunko 金沢文庫, ed. Kanazawa Bunko no chūsei shintō shiryō 金沢文庫の中世神道資料 (Yokohama: Yokohama kenritsu Kanazawa Bunko, 1996). 89 disappointment at not receiving the role it wanted lacks conviction. Similarly, the Munakata-ki (Record of Munakata 宗像記), also copied by Kenna, introduces the Munakata Deity as “the master of Japan, the child of Amaterasu, the miraculous deity for subjugating the Korean kingdoms, and the number one ancestral shrine in Kyushu.” 171 The text aimed to situate the Munakata deity as a prominent god amongst all of the deities of the realm, but it still provides the caveat that it is only the primary ancestral shrine within the Kyushu region. Being a powerful deity in Kyushu was not sufficient. Kyushu was the center of the military defenses against the Mongols and the worship of Queen Consort Jingū, but the protection of the gods was based on the framework of a unified realm. Assembly with and in Relation to other Networks Placing Itsukushima within the shared discourses about Queen Consort Jingū and the protection of the realm was not enough in itself to raise Itsukushima’s status as a preeminent site for the protection of Japan. The Itsukushima Daimyōjin nikki moved Itsukushima into these discourses, and consequentially, situated Itsukushima as a place that could be incorporated into the origin narratives of other religious institutions participating within the same discourses. In the end, it was the inclusion of Itsukushima within these other texts that stabilized the Itsukushima Deity’s place as one of the main deities protecting the realm from foreign enemies. Looking beyond the Hachiman gudōkun demonstrates that Jingū’s band of deities was systematized and influenced discourses both inside and outside of origin narratives. The Kōra tamadare no miya shinpisho (Secret Writings on Kōra Tamadare Shrine 高良玉垂宮神秘書) is an origin narrative for Kōra Shrine in Kyushu, believed to have been written in the late s ixteenth 171 Ibid. 90 to early seventeenth century, but the structure of the text resembles a reference text about Kōra more so than a tale of its origins. 172 Beginning with a genealogy of the deities of Japan, the text situates Kōra as the husband of Queen Consort Jingū, and thus, focuses much of its attention on Jingū’s subjugation of the Korean kingdoms. This section of the text reads like a condensed version of the Hachiman gudōkun with much of the story removed and only the information on the deities remaining as bulleted points. The synopsis of one Hachiman gudōkun passage about Itsukushima concisely reads, “There were two women, a female water deity and a female dragon deity. The dragon deity was the Itsukushima Deity. The water deity was the Munakata Deity.” The text then moves on to identifying the Shika Deity 志賀 in Echizen, the Kashima Deity 鹿島 in Hitachi, and the Kasuga Deity 春日 in Yamato as one in essence, and the sisters of Queen Consort Jingū as the Hōman bodhisattva and the Kawakami Deity. 173 This structure helps us to discern how the compiler was reading the Hachiman gudōkun. The relationship of the deities with Jingū was of primary importance, and the geographical location of the deities was also valued. Moreover, the last pages of the text present the information about the deities within the Hachiman gudōkun in diagrams, beginning with a genealogy of the deities and ending with a list of the deities who accompanied Jingū to subjugate the Korean kingdoms, that call to mind the genealogical charts in the Usa takusenshū. 174 In this manner, the Kōra tamadare no miya shinpisho extracted information from the Hachiman 172 While the colophon for the text is incomplete and does not provide a date for the production of the text, the available evidence suggests that it was written sometime between the late sixteenth century and the early seventeenth century by the head priest of Kōra Shrine. Araki Hisashi 荒木尚 ed. Kōra tamadarenomiya shinpisho dō shihai 高良玉垂宮神秘書同紙背 (Kurume: Kō ra Taisha, 1972), 182-83, 92-94. 173 Ibid., 17. 174 The only inconsistency with the Hachiman gudōkun is the addition of the Ninety-nine Bodied King, the deity of Fūrō Shrine (Fūrōgū 風浪宮) in modern day Fukuoka Prefecture, to the list of deities in Jingū’s retinue. Ibid., 175, 193-94; "Hachiman gudō kun (kō ) 八幡愚童訓（甲)," in 寺社緣起, ed. Sakurai Tokutarō 桜井徳太郎, Hagiwara Tatsuo 萩原龍夫, and Miyata Noboru 宮田登 (Tokyo: Iwanami Shoten, 1975), 175. 91 gudōkun and organized it into an easier to use format. While this information is used within the Kōra tamadare no miya shinpisho to situate the Kōra Deity within the larger discursive associations of the deities, the organization and inclusion of details about the other deities suggest that the information was valued in itself, and worth being rewritten within a reference guide. The use of origin narratives as sources of information enabled associations with the Itsukushima Deity to be incorporated into the writings of scholarly monks and intellectuals affiliated with various religious institutions. The producers and readers of the texts could unintentionally contribute to the dissemination of Itsukushima’s association with Jingū by copying the list of deities accompanying Queen Consort Jingū in the Hachiman gudōkun or the sisters of the Dragon Princess from the Usa takusenshū. While the Kōra tamadare no miya shinpisho provides a clear view of the Hachiman gudōkun’s influence on later texts, the case of the Itsukushima Daimyōjin nikki is more complicated. The development of Itsukushima within the Queen Consort Jingū discourse cannot be simplified to a chronological sequence from the Usa takusenshū’s recording of the sisters of the Dragon Princess, to the Itsukushima Daimyōjin nikki’s suggestive incorporation of the Itsukushima Deity as one of the sisters of Jingū and the Dragon Princess, and finally to the Hachiman gudōkun’s inclusion of Itsukushima alongside the deities most closely associated with Jingū. These three origin narratives reveal how these intertextual connections developed and spread, but other texts could have played the same roles. With the Itsukushima Daimyōjin nikki, Hachiman gudōkun, and Usa takusenshū, all written at around the same time in the late thirteenth to early fourteenth century, we cannot decisively place any of the texts before the others. The earliest datable example to link Itsukushima with other deities protecting Japan from the Mongols is a 1293 dedicatory vow 92 (ganmon 願文) by the priest Tahōbō 他宝坊 of Ikinomatsubara Shrine in Kyushu. The vow explains that Tahōbō reported to the Bakufu government in Kamakura that the deities were deflecting a curse back towards the Mongols, and in return, an order was made for a shrine to be built at Ikinomatsubara. Notably, the deities mentioned by Tahōbō was not limited to those enshrined at Ikinomatsubara Shrine and almost perfectly matched the deities accompanying Jingū in the Hachiman gudōkun: Suwa, Kashima, 175 Mishima, Itsukushima, Izumo, Hakozaki Hachiman, Sumiyoshi, Kawakami, the eight Dragon Kings, and the Dragon King of the Ocean. 176 While there are slight differences, the retinue remains largely intact, even though the vow was dedicated before the Hachiman gudōkun was finished. While it is possible that the Hachiman gudōkun section was being transmitted before the compilation was completed, it is more likely that the grouping of deities existed in texts prior to both Tahōbō’s dedicatory vow and the Hachiman gudōkun, and that they adopted the existing discourse. 177 The Hachiman gudōkun’s influence cemented the Itsukushima Deity’s place within Jingū’s retinue, but we cannot go so far as to credit it, or any of the other texts discussed, as the first to make the connection. 175 Kashima can be interpreted as another reading for Shikashima 志賀島 in Kyushu. For an example, see the Kōra tamadare no miya shinpisho. Araki Hisashi 荒木尚 ed. Kōra tamadarenomiya shinpisho dō shihai 高良玉垂宮神秘書同紙背 (Kurume: Kō ra Taisha, 1972), 17. 176 Kawazoe Shō ji 川添昭二, ed. Genkō bōrui hennen shiryō: Chūkai ikoku keiko ban'yaku shiryō no kenkyū 元寇防塁編年史料 : 注解異国警固番役史料の研究 (Fukuoka: Fukuoka-shi Kyō iku Iinkai, 1971), 277-78. For a translation of the dedicatory vow, see doc. 64 in Thomas Conlan, In Little Need of Divine Intervention: Takezaki Suenaga's Scrolls of the Mongol Invasions of Japan (Ithaca, NY: East Asia Program, Cornell University, 2001), 251-52. 177 Noting shared discourses between the Usa takusenshū and the Keiran shūyōshū, which were written during the same time period, Murata Shin’ichi concludes that it is most likely that both Jin’un and Kōshū, the compilers of the two texts, were working from shared sources rather than it being the case that the Usa takusenshū referred to the Keiran shūyōshū or vice versa. Murata Shin'ichi 村田真一, Usa Hachiman shinwa gensetsu no kenkyū: "Hachiman Usagū gotakusenshū" o yomu 宇佐八幡神話言說の研究：「八幡宇佐宮御託宣集」を読む (Kyoto: Hō zō kan, 2016), 409-10. 93 Additionally, the existence of four other texts incorporating the same sisters of the Dragon Princess as the Usa takusenshū hinders attempts to directly connect the Itsukushima Daimyōjin nikki with the Usa takusenshū. Any of the texts could have spread the discourse. Moreover, all of these texts were collected at the Kanazawa Bunko archive from the late thirteenth to the first half of the fourteenth centuries, with two of the four signed by Kenna or Zenkai, the same monks who copied the Itsukushima Daimyōjin nikki. 178 The fact that all of these texts contain the same information within similar narratives, and were copied and stored by the same persons, suggests that these texts were being produced and circulated in overlapping networks, wherein any one text could have influenced another’s production, and through which they all eventually reached Shōmyōji and Gokurakuji in Kamakura. Undoubtedly, the texts extant today only reveal a small glimpse into the complex movements of texts that enabled associations to be drawn between various sites and deities. The indeterminant relations between these multiple, discursively interconnected texts prevent us from providing a clear, linear account of the progression of the texts identifying the Itsukushima Deity in relation to Queen Consort Jingū. However, the diffused agency within networks of association, wherein various texts all had the potential to transmit the same discursive connections, is what made it successful. Scholarly monks with extensive libraries and networks with other religious institutions had advantages for producing and transmitting their writings, strategic use of networks of association could lead various other origin narratives to become widely received and influential. 178 These texts are as follows: (1) the “Record of the Mt. Kamado Hōman Bodhisattv a” (“Kamado-san Hōman Daibosatsu ki” 竈門山宝満大菩薩記), copied by Kenna; (2) “Secret Transmission [about the] deity Hachiman” (“Jingi hiden” 神祇秘伝八幡), copied by Zenkai; (3) “Oral Transmission [about the] Hachiman Bodhisattva (version A)” (“Hachiman daibosatsu kuketsu” 八幡大菩薩口決（甲本）); (4) “Oracle of the Great Bodhisattva” (“Dai bosatsu go takusen tō” 大菩薩御託宣等). For the first three texts, see Kanazawa Bunko, Kanazawa Bunko no chūsei shintō shiryō 金沢文庫の中世神道資料. For the fourth text, see doc. 6768 in Kanazawa Bunko, 金沢文庫, "Kanazawa Bunko komonjo 金澤文庫古文書," (Yokohama: Kanazawa Bunko, 1956), 302-03. 94 It is worth taking the look at the Mineaki (Record of Joint Peaks, 峯相記), a fourteenth century Harima Province gazetteer, as a counterexample showing how some associations with Queen Consort Jingū failed to gain traction. The Mineaiki includes an origin narrative of the Keisokuji 鶏足寺 temple that tried to draw connections between its deity and Jingū, but the associations were overly specific and conflicted with details in other origin narratives. The Mineaiki retells the Jingū narrative in order to the Keisokuji affiliated deity, Ōki no mikoto 男貴尊, as her lieutenant general. However, this would become problematic if compared with texts, like the Hachiman gudōkun, that identified the lieutenant general otherwise. The Mineaki account states: Susanoo no mikoto’s first [son], royal prince Ōki no mikoto, manifests as Hakusan Myōri Gongen. Now, when Queen Consort Jingū attacked the three Korean kingdoms, he (Ōki no mikoto) acted as the lieutenant general and made his way to the battlefield. Afterwards, peace was restored, and the Queen Consort returned to the capital. Now, if it is the case that [a deity] participated in the victory over the foreign enemies, [that deity] should be grouped with the various deities of the Middle Kingdom (Japan) that carried out their responsibilities in battle. They should receive the promised recognition. 179 Following a format similar to the Itsukushima Daimyōjin nikki, the Mineaiki utilizes familial relations to associate the local deity Ōki no mikoto with a major god from the Nihon shoki, Susanoo, and with a deity from another province, the Hakusan Deity. 180 However, while the Itsukushima Daimyōjin nikki does not dwell on the associations with the Jingū discourse, the Mineaki emphasizes Ōki no mikoto’s status as lieutenant general and his right to be rewarded. The Mineaiki makes its message clear, but it is the lack of detail in the Itsukushima Daimyōjin nikki that makes the association between Itsukushima and Jingū useful for other texts. The 179 "Mineaiki 峯相記," in Zoku gunsho ruijū 續群書類從. 釈家部, ed. Hanawa Hokinoichi 塙保己一 and Oota Toshiro 太田藤四郎 (Tokyo: Zoku Gunsho Ruijū Kanseikai, 1957), 233. 180 It is worth noting that the Itsukushima Daimyōjin niki also associates the Itsukushima Deity with the Hakusan Gongen, and the two texts likely due so for similar reasons. 95 Mineaiki’s specificity made it harder for recognition of the Ōki no mikoto Deity within Jingū’s entourage to spread. The only other texts which also include this deity within the Jingū narrative are two eighteenth century Harima gazetteers, the Harima kagami (Mirror of Harima, 播磨鑑) and the Harima no kuni fudoki (A Record of the Natural Features of Harima Province 播磨国風土記). 181 In other words, the discourse was sustainable within the region, but not outside of local knowledge. Compared with the Itsukushima Daimyōjin nikki, the Itsukushima narrative’s discursive overlap with the Dragon Princess and Hachiman discourses on Queen Consort Jingū made it more convincing, and more accessible for use in non-local discourses on other deities. Popular Reception of the Itsukushima Origin Narrative Associations The complex intertextual connections likely would not have been of much interest to most readers. Scholarly monks and intellectuals realized the need to read and write origin narratives in relation to one another, but most people would not have had access to the texts to reference, or any desire to compare the gods. However, that does not mean that the networks of association connecting Itsukushima with the Dragon Princess and Queen Consort Jingū were only significant to a small group of religious elites. The repeated mentions of these networks of association in later texts helped the identifications to spread to wider audiences. Pilgrims traveling to Itsukushima Shrine might not have been familiar with the reasons for the Itsukushima Deity’s connections with the Dragon Princess or Jingū, but they could still recognize that the Itsukushima Deity was a dragon deity who could protect threats from overseas. 181 Yamaguchi Makoto 山口真琴, "Harima nashonarizumu to Jingū kōgō densetsu: "Mineai ki" josetsu 播磨ナショナリズムと神功皇后伝説：『峯相記』序説," Problématique 3(2002): 69-75. 96 As Lori Meeks discusses in the case of the Nara convent Hokkeji, origin narratives were often disseminated through their incorporation within origin narrative anthologies and pilgrimage diaries rather than as individual texts. Pilgrims read these diaries and origin narratives like guidebooks, using the texts as resources for learning pilgrimage itineraries, the sights to see, and understandings of a site’s histories and narratives. Repeating the steps traveled by those before them, the reading of diaries and the origin narratives within further promoted the standardization of discourses about sacred site amongst popular audiences. 182 Likewise, the Itsukushima origin narrative text itself was not always what reached worshippers, and the reception of the Itsukushima origin narrative could change depending on the context in which it was experienced. The role of Jingū as a victor over foreign enemies provided opportunities for the associations with her to be applied to new contexts after the threat of the Mongols subsided. Over one hundred years after the Mongol invasions, the band of deities assisting Jingū was still remembered and reconstructed when worries of foreign powers reemerged. For example in the nō play Haku Rakuten ( 白楽天, post 1419) when the Sumiyoshi Deity summons a divine wind to blow the poet Bai Juyi back to China after defeating him in a poetry match, the Sumiyoshi Deity is accompanied by an almost identical band of gods: Ise, Iwashimizu, Kamo, Kasuga, Kashima, Mishima, Suwa, Atsuta, and Itsukushima, who perform a dance before giving Bai Juyi his hasty return. 183 As Susan Blakely Klein has argued, this 182 Lori Rachelle Meeks, Hokkeji and the Reemergence of Female Monastic Orders in Premodern Japan (Honolulu: University of Hawaii Press, 2010), 34-42. The reading of origin narratives within diaries was particularly significant in the temples and shrines in the provinces. Origin narratives provided reasons for worshippers to travel and make offerings to distant sites, and knowledge for understanding the site once they got there. Even if readers did not travel all the way to the site itself, the origin narratives provided an awareness of the major religious institutions spread across the realm. See Nishiguchi, Junko 西口順子, Heian jidai no jiin to minshū 平安時代の寺院と民衆 (Kyoto: Hō zō kan, 2004), 127-28 183 Concerning the dating of Haku rakuten, Susan Blakely Klein mentions that previous scholars (Kume Kunitake, Takano Tatsuyuki, and Amano Fumio) have suggested that the play was commissioned by the shogun Ashikaga Yoshimochi ( 足利義持, 1386-1428) in response to the shogunate’s insecure foreign relations with Chosŏn Korea and Ming China. Susan Blakeley Klein, "Nō as Political Allegory: The Case of Haku Rakuten," in Like Clouds or 97 evocation of the “divine wind” which protected Japan from the Mongol invaders makes the scene into a “metaphorical sea battle,” in the deities triumph over the foreigners through both military and cultural strength. 184 Therefore, it is no surprise that the deities summoned are the same as those who were known for protecting against the Mongols. Haku rakuten recalls the defeat of the Mongols at a time shortly after the 1419 Ōei “invasion,” in which the Chosŏn government sent ships to attack Tsushima and subdue the pirates (wakō 倭寇) causing disturbances in international waters. Just as the precedent of Queen Consort Jingū became central at the time of the Mongol invasions, those deities incorporated into Jingū’s retinue in medieval origin narratives reemerged when concerns of a foreign threat were raised anew. Itsukushima origin narratives, and their networks of association, were incorporated into later narratives about Itsukushima, but the association with Queen Consort Jingū became less prominent once distanced from the context of foreign threats. For example, the Itsukushima Daimyōjin nikki is reproduced almost word for word in the Nagatobon Tale of the Heike 長門本平家物語. This variant of the Tale of the Heike has a regional emphasis on western Japan and was likely used for proselytization in the locale around Itsukushima. 185 The Itsukushima Daimyōjin nikki might have primarily circulated among scholarly monks, but the Nagatobon presented the origin narrative, and its description of the deity as Jingū’s sister, to wider audiences. However, the Nagatobon is the only Heike variant to identify the Itsukushima Deity in relation to Jingū. Instead, the Genpei jōsuiki 源平盛衰記 and Kakuichibon ( 覚一本, 1371) Heike variants both stress the Itsukushima Deity’s identity as the daughter of the Dragon Mists: Studies and Translations of Nō Plays of the Genpei War, ed. Elizabeth Oyler and Michael Geoffrey Watson (Ithaca, New York: East Asia Program, Cornell University, 2013), 402-03, 19-20. 184 Ibid, 405. 185 See chapter four for a more detailed discussion of the Itsukushima Deity, the Taira, and the Nagatobon Tale of Heike. 98 Princess. 186 Even if audiences of the Tale of the Heike did not refer to the work as a pilgrimage guide, they could still become familiar with the Itsukushima Deity’s relationship with the Dragon Princess by hearing the narratives. Beyond the association with the Dragon Princess, the Genpei jōsuiki includes an alternative version of the Itsukushima origin narrative similar to the one within the Shōbōrinzō Prince Shōtoku biography. 187 The Genpei jōsuiki lacks some of the discursive details found in the Shōbōrinzō, such as a list of the other sisters of the Dragon Princess, but it still provided readers with a detailed telling of the origin narrative. Additionally, the use of the Shōbōrinzō and the Genpei jōsuiki outside of the regional contexts of the Nagatobon helped to strengthen the popular image of the Itsukushima Deity as the sister of the Dragon Princess. The Shōbōrinzō was used as a script for etoki 絵解き image explication of illustrated scrolls of the Prince Shōtoku narrative, and the Genpei jōsuiki’s Itsukushima origin narrative was retold in the kyōgen interlude of the nō play “Itsukushima” by Kanze Nagatoshi ( 観世長俊, 1488-1541). In this manner, both the Shōbōrinzō and the Genpei jōsuiki provided opportunities for people to hear about the Itsukushima origin narrative and the dragon princess discourse, without having to read the texts themselves. 188 The nō play focuses on the interplay between the Itsukushima Deity as the daughter of the Dragon King and the episode of the Dragon Princess in the Lotus Sutra. Its content and rhetoric distances it from the concise information found in origin narratives, but the 186 Ichiko Teiji 市古貞次, ed. Genpei jōsuiki 源平盛衰記., Chūsei no bungaku 中世の文学 (Tokyo: Miyai Shoten, 1991), v. 3, p. 14; Heike monogatari 平家物語, vol. 1, Shinpen Nihon koten bungaku zenshū (Tokyo: Shō gakkan, 1994), 177. 187 Makino Kazuo compared the Itsukushima origin narrative passages in the Shōbōrinzō and the Genpei jōsuiki and argued that it is more likely that both texts were based on the same preexisting text, than that the Genpei jōsuiki was adapted from the Shōbōrinzō. That being said, he still acknowledges the possibility that the Genpei jōsuiki was based on the Shōbōrinzō. Makino, Engyōbon "Heike monogatari" no setsuwa to gakumon 延慶本「平家物語」の說話と学問, 333-36. 188 Concerning the use of the Shōbōrinzō for preaching and etoki image explication, see Abe Yasurō 阿部泰郎, "Shōbōri nzō 「正法輪蔵」," Kokubungaku kaishaku to kanshō 54, no. 10 (1989): 95-96. 99 kyōgen interlude fills this gap. Repeating the Genpei jōsuiki’s Itsukushima origin narrative, the interlude provides the audience with a basis for understanding and imagining the Itsukushima Deity. The medieval pilgrimage records for Itsukushima are not extensive, but the sources available often describe the Itsukushima Deity as the daughter of the Dragon King, repeating and reinforcing the same discourse used within the Itsukushima Daimyōjin nikki. The Rokuon’in Saigoku gekō-ki 鹿苑院西国下向記, a travel record of Ashikaga Yoshimitsu’s ( 足利義満, 1358-1408) travels to the western provinces by a travel companion named Mototsuna 元網, describes the visit to Itsukushima: In the same month, on the 11 th day, we went to the [Itsukushima] Shrine. [As for] the one called the Itsukushima Deity, she is the Princess of Dragon King Sagara (subscript: original form is either Eleven Headed Kannon or Bishamon). Her miraculous efficacy is without equal. Therefore, when the Taira Great Minister (Kiyomori) repaired the great pagoda at Mt. Kōya, [he received] an [oracular] message from Kukai and thus piously built this shrine. The corridors of one hundred and eighty ken were built exceptionally. He worshipped the deity and subsequently received the deity in a dream, and then held all of the four oceans under heaven in the palm of his hand. The Honorable Great General [Ashikaga Yoshimitsu]’s pilgrimage as well is outstanding. Usually, in our realm, the siblings of this deity are [said to be] Kehi Shrine in Echizen Province, the Shinra Deity in Ōmi Province Take District, and Ōmuro Shrine in Musashi Province. 189 Beginning with the identification of the Itsukushima Deity as the daughter of the Dragon King, followed with a note on the honji suijaku associations of the deity, and ending with her relatives, the passage is structured around the standard types of associations used to identify and understand the deity. Mototsuna then overlays their pilgrimage with Kiyomori’s worship of Itsukushima in the Tale of the Heike by recalling the origins of Kiyomori’s worship from an 189 Mototsuna 元網, "Rokuon’in Saigoku gekō -ki 鹿苑院西国下向記," in Sankeiki 參詣記, ed. Shinjō Tsunezō 新城常三, Shintō taikei. Bungaku hen (Tokyo: Shintō Taikei Hensankai, 1984), 156. 100 oracle received while at Mt. Kōya. 190 This allows him to associate Yoshimitsu with the success achieved by Kiyomori after his pilgrimage to Itsukushima, while conveniently ignoring the fall of the Heike. While the connections of the Itsukushima Deity with Dragon King Sagara and Kehi also appear in other versions of the Tale of the Heike, the addition here of the Shinra Deity and the Ōmuro Deity shows that Mototsuna’s knowledge of Itsukushima was not reliant on the Heike narrative alone. As discussed earlier, the Shinra deity was often listed as one of the daughters of the Dragon King Sagara and sisters with the Itsukushima Deity. Like this, details from multiple narratives merged together, expanding and disseminating the networks of associations. While not describing the exact association with the Dragon King Sagara, other travelers employed the related association of Itsukushima as the Dragon Palace for writing about one’s pilgrimage. Sometime after the defeat of Go-Daigo in 1333, the Rinzai Zen monk Chūgan Engetsu ( 中巌円月, 1300-1375) stopped at Itsukushima on his way back to the capital from Hakata in Kyushu. Only two poems remain to tell of his experience of Itsukushima, but they reflect the same motifs of the Dragon Princess from the Lotus Sutra and the Dragon Palace. Beautiful hair, dressed in crimson, thin sweeping eyebrows. A renunciant dressed in darkness sails forward with oars of lapis lazuli. I come to visit the inner courts of the Dragon Palace. Sooner or later it is the time of the offering of the jewel and becoming a buddha. My mind travels off to a beautiful place revealing sacred footprints. A mysterious mountain surges forth from the sea. The moon illumines the corridors as the water ebbs and wanes. In the depths of night, who is there in the crystal palace? 191 The first poem combines imagery of Itsukushima and the episode of the Dragon Princess gaining enlightenment in the Lotus Sutra, allowing for Chūgan’s pilgrimage to join that play of 190 Royall Tyler, trans., The Tale of the Heike (New York: Viking, 2012), 146-49. 191 Chūgan Engetsu 中巌円月, "Tōkai ichiōsh ū 東海一漚集," in Gozan bungaku shinshū 五山文學新集, ed. Tamamura Takeji 玉村竹二 (Tokyo: Tō kyō Daigaku Shuppankai, 1970), 328. 101 associations. The term for renunciant (sen 仙) can refer specifically to mountain recluses and Daoist sages, but also to beings with supernatural powers, such as Buddhist practitioners, bodhisattvas, buddhas, and Japanese deities. These multiple meanings provided the potential for the renunciant to refer to the Itsukushima Deity, Dragon Princess, or Chūgan himself, as is suggested by the next sentence’s switch to the first person. 192 From this ambiguity in the subject, the reference to the Dragon Princess’s offering of the jewel can also be taken as Chūgan’s offering to the Itsukushima Deity. In this manner, Chūgan employs the multiple linked meanings in the networks of association between Itsukushima and the Dragon Princess as a lens for portraying his own pilgrimage experience. At the risk of over-interpreting the actions of another traveler to Itsukushima, it is possible that Imagawa Ryōshun ( 今川了俊, 1326-1420)’s 1371 offering to the Itsukushima Deity, two relics dropped into the sea, is similarly based on the Itsukushima Deity’s relation with the Dragon Princess. Because relics are commonly seen as equivalent to wish fulfilling jewels, the relic offering can be read as referring back to the Dragon Princess’s offering to the Buddha, the Itsukushima Deity as a dragon deity living in the Dragon Palace, or even the episode of Queen Consort Jingū dropping two jewels into the sea to defeat the foreign enemy in narratives such as the Hachiman gudōkun. 193 With the entry ending with a description of Itsukushima as 192 The imagery of the beautiful goddess sailing on the seas further calls back to the origin narrative scene of the Itsukushima Deity meeting Saeki Kuramoto at sea. 193 Considering the context of Imagawa Ryōshun’s visit to Itsukushima while on his way to battle in Kyushu, Inada Toshinori suggests that this prayer could be read as a prayer for success in battle more specifically. Inada Takeshi further argues that Imagawa’s writing in the Michiyukiburi was influenced by the Hachiman gudōkun, and cites a passage within the Michiyukiburi referring specifically to the episode of Jingū using the two ebb and wane jewels against the Korean kingdoms from the Hachiman gudōkun. Inada Toshinori 稲田利徳, "Imagawa Ryōshun 'Michiyuki buri' chūshaku (3) 今川了俊「道行きぶり」注釈（3 ）," Okayama daigaku kyōiku gakubu kenkyū shūroku, no. 91 (1992): 7; Kobayakawa Takeshi 小早川健, "'Hachiman gudōkun' to Chūsei kikōbun 『八幡愚童訓』と中世紀行文," Kōbe shiritsu kōgyō kōtō senmon gakkō kenkyū kiyōki 34 (1996): 99; Imagawa Ryō shun 今川了俊, "Michiyuki buri 道行きぶり," in Chūsei nikki kikōshū 中世日記紀行集, ed. Nagasaki Ken 長崎健, Shinpen nihon koten bungaku zenshu 新編日本古典文学全集 (Tokyo: Shōgakkan, 1994), 407. 102 “the dwelling of the master of the palace in the sea,” the association with the Dragon Palace is clear, and although we cannot know Imagawa Ryōshun’s intentions, the networks of association related to the Dragon Palace could have led Ryōshun and future readers to imagine the pilgrimage through these connections. Conclusion The production of the Itsukushima Daimyōjin nikki did not revolutionize the imagination or the prestige of the Itsukushima Deity. The achievement of the Itsukushima Daimyōjin nikki was making the associations with the Dragon Princess and Queen Consort Jingū into a compact, easily usable form. Condensable as a list of siblings, the associations could be included within various texts and performances, expanding the audience and agents for further reproduction of the discourse. Even a century after the Mongol invasions, the association between Itsukushima and Queen Consort Jingū was not forgotten. In the Nihon shoki kan daiichi kikigaki (1419, 日本書紀巻第一聞書), Ryōhen describes an oracle by the Itsukushima Deity, but this tim e with five sisters: (1) the Dragon Princess; (2) the Shinra Deity; (3) Hachiman’s mother, Queen Consort Jingū; (4) the Itsukushima Deity; and (5) Seiryū Gongen. 194 Corresponding with the texts discussed earlier, this grouping includes the most common deities identified as the sisters of the Itsukushima, and adds Jingū. Recorded as a transcript of Ryōhen’s lectures on the Nihon shoki at Jōseji 常施寺 in Kyoto, the Nihon shoki kan daiichi kikigaki hints at the many forms through which this discourse could spread. 195 194 Ryōhen 良遍, "Nihon shoki kan daiichi, daini kikigaki (Tawa bunko zō) 日本書紀巻第一・第二聞書 ( 多和文庫蔵)," in Sonarejō: Murasame hen 磯馴帖：村雨篇, ed. Itō Masayoshi 伊藤正義, Koten kenkyū shiryōshū (Osaka: Izumi Shoin, 2002), 355. 195 Even lectures though, traveled in part through its transcription. A colophon for another manuscript, transmitted to the Hachiman shrine in Mitsugi, Hiroshima, gives a lecturer from Harima Province less than six months after 103 The importance of the Itsukushima Deity’s associations with the Dragon Princess and Queen Consort Jingū only become evident once we stop reading origin narratives as self- contained stories. Origin narratives work as cultic literature because they are directed towards a particular deity or sacred space, and because they are eminently quotable. Umberto Eco argues that a film or book can be made into a cult object when one is able to “break, dislocate, unhinge it so that one can remember only parts of it, irrespective of their original relationship with the whole.” 196 The same can be said about origin narratives. Itsukushima Daimyōjin nikki helped spread the identity of the Itsukushima Deity as a protector from the Mongols by condensing the argument into a succinct and separable association. The Itsukushima Daimyōjin nikki can be read and appreciated as a story, but that could lead us to overlook the most influential aspects of the text. Only a few scholarly monks might have had access to the Itsukushima Daimyōjin nikki, but countless people had opportunities to hear about Itsukushima as the sister of the Dragon Princess and Queen Consort Jingū. Shifting the focus away from Itsukushima allows us to see the significance of networks for the Itsukushima origin narrative. Even if a scholarly monk at Itsukushima came up with the idea of linking Itsukushima with Queen Consort Jingū through the Dragon Princess, this connection was not produced or transmitted through the sole efforts of the individual. Rather, it was the assemblage of different origin narratives and networks of associations that allowed the association with Jingū to be received widely and positively. In particular, the compilation of multiple origin narratives and networks of association within texts like the Usa takusenshū and Ryōh en (1419, 6 th month, 17 th day). While the name of the lecturer is illegible in the manuscript, and it is unknown how it reached Hiroshima Prefecture, it still provides evidence that the lectures disseminated beyond the original audience of the lecture in the capital, and their circles of acquaintances. 196 Umberto Eco, "Casablanca: Cult Movies and Intertextual Collage," in Travels in Hyper Reality: Essays (San Diego: Harcourt Brace & Company, 1986), 198; Henry Jenkins, Convergence Culture: Where Old and New Media Collide (New York: New York University Press, 2006), 97-98. 104 the Hachiman gudōkun played an essential role in bringing together discourses on the many gods of the Japanese archipelago. Likewise, in evaluating the impact of the networks of association involving the Itsukushima Deity, we also need to look beyond the reception by devotees of Itsukushima Shrine. It was in no small part the work needed to rewrite origin narratives after the Mongol invasions, especially the gathering of information about competing deities, which led to standardized understandings of the deities in relation to one another. No one text and no one thinker could construct the identity of the Itsukushima Deity. Innumerable persons, texts, and gods all played a role. 105 Chapter 2 Benzaiten Outside of Itsukushima: Spatial Networks and Deity Multiplication The boundaries of a temple or shrine are clearly labeled. In Japan, shrine torii and temple gates announce the entrance into the sacred, and pathways guide devotees to buildings for enshrined deities, ritual spaces, and traces of miraculous past events. Taking a step back though, the lines between sacred and worldly spaces begins to blur. In the case of gardens within Indian Buddhist monasteries, Gregory Schopen demonstrated that the ideal distance between sacred and profane spaces was “neither too far nor too near.” The outside world was supposed to be visible while kept at arm’s length. 197 It might seem hard to measure such a distance, but that can be a benefit rather than a problem. As will be discussed in the following pages, seeing space with blurred boundaries can lead us to understand space beyond the two-dimensional confines of a map. We tend to think of space as singular and unchanging. 198 With modernity came the idea of the “world as exhibition,” of the world as an external object that could be displayed, studied, and systematized. The turn to postmodernism and cultural studies in the 1980’s challenged the aim of an objective view of the world, moving the scholar’s perspective towards the world from outside and total to inside and confined. 199 Instead, space became the stable ground from which to locate one’s own perspective or situate the cultural context of the object of study. In the study of East Asian religions, debates over the relationship between local or popular religious 197 Schopen, Gregory, "The Buddhist "Monastery" and the Indian Garden: Aesthetics, Assimilations, and the Siting of Monastic Establishments," Journal of the American Oriental Society 126, no. 4 (2006): 504-05. 198 Doreen Massey problematizes the binary distinction between space and time, and the resultant view of space as a-temporal and unchanging. This perspective reinforces views of space as a perfect structural system, wherein each part is adjacent and unconnected with other spaces. Consequently, people fail to see the happenstance changes or contradictions in spatial understandings, and the narrative and discourses made to explain or hide those structural imperfections. Doreen B. Massey, For Space (Thousand Oaks, CA: Sage, 2005), 39. 199 Derek Gregory, Geographical Imaginations (Cambridge, MA: Blackwell, 1994), 34-37, 72-77. 106 communities and more institutional religions has recently led to an influx of scholarship examining a single cultic site. Instead of doctrinal and institutional forms of religious practices and teachings, the site-based study works from the beliefs and practices at a site to reveal the complex ways in which diverse religious elements can come together in lived religious practice. 200 However, the emphasis on the particularities of a single site can risk taking the space as a bounded totality apart from the outside world. 201 In this chapter, I explore the possibilities for sacred space when unrestrained by singularity. I argue for a new approach to sacred space by examining how relationships between multiple sacred spaces and deities constructed the meanings of those spaces, and allowed them to transcend the structural constraints of cartographic space. Maps define spaces through names and lines differentiating adjacent spaces, but the sacred spaces and deities of medieval Japan moved, multiplied, and connected with one another. In this manner, sacred space follows geographer Doreen Massey’s notion of space as “the product of interrelations; as constituted through interactions, from the immensity of the global to the intimately tiny.” 202 Within these terms, the interrelations can be considered as a network with the different constituting elements mediating 200 In this sense, as a methodology, the site-based study can resemble a phenomenology of religious space and faces some of the same issues. The site-based study functions by gathering together knowledge of all the elements of a sacred space (ritual, iconography, myths, and individual worship), and then assembles the pieces to gain a better understanding of the nature of the space and/or Japanese religions. In doing so, however, the space takes the form of a singular object without agency, change, or comparables. Because the researcher must determine which materials concern the site and which do not, she or he must begin from a preconceived notion of the space, despite the intention to let the data speak for itself. On this subject, see Walter Capp’s discussion of Cornelius Petrus Tiele’s phenomenology of religion. Capps, Walter H., Religious Studies: The Making of a Discipline (Minneapolis: Fortress Press, 1995), 117-19. 201 The site-based study avoids some of the issues raised by Robert Campany concerning academic practices reifying religions but raises the risk of reifying the space itself. Concerning the reification of religions in religious studies, see Robert Ford Campany, "On the Very Idea of Religions (In the Modern West and in Early Medieval China)," History of Religions 42, no. 4 (2003): 291-94. Concerning the reification of the local and spatial approaches to religious studies, see Manuel Vásquez, "Studying Religion in Motion: A Networks Approach," Method & Theory in the Study of Religion 20, no. 2 (2008): 167-68. 202 Massey, For Space: 9. 107 the interactions and meanings of one another. Every space simultaneously exists on multiple levels, each based in different connections. Beginning from the space of Itsukushima Shrine, I work outwards through networks of association between local sacred spaces of the Benzaiten deity (Skt. Sarasvatī, Ch. Biancaitian, 弁才天) that linked Itsukushima with other sacred spaces throughout the Japanese archipelago. Unlike conventionally understood space (as singular and unchanging), deities can move, multiply, and manifest in various forms, providing characteristics, identities, or reasonings for the associations between different gods. Relations between spaces and gods provided both opportunities and complications for freeing space from the boundaries drawn on the map. Some cases disregard the fixity of space altogether, such as narratives of sacred mountains flying to new locations. 203 In other cases though, the immobility of space needs to be retained while also allowing for mobility in the meaning of the site and its connections with other sacred spaces. It is this second case that I will be focusing on. The idea of space as immobile empowers local space as unique. However, the problem is that mobile meanings and connections are necessary in order to make the space understandable in comparison to other sites. In short, spaces need to be both the same and different. 203 Origin narratives concerning Kinpusen (Mt. Ōmine as Vulture’s Peak), and of Kumano (as the pure land of Kannon) describe the mountains flying from India and eventually arriving in Japan. In the case of Kumano, the mountain first traveled to China, and then to Kyushu (location of Mt. Hiko 英彦山) and Shikoku before finally arriving at the location which would become Kumano. Itō Satoshi suggests that this movement of spac es from India to Japan applies the logic of honji-suijaku, which normally links deities, to the spaces of India and Japan. In China, Mt. Lu and other mountains (often named Linjiu shan, the translation of Vulture’s Peak (Jp. Ryōju sen 霊鷲山)) similarly were said to have traveled from India to China, either as Vulture’s Peak or as one of its subsidiary peaks. Allan G. Grapard, "Flying Mountains and Walkers of Emptiness: Toward a Definition of Sacred Space in Japanese Religions," History of Religions 21, no. 3 (1982): 218; Itō Satoshi 伊藤聡, Shintō no keisei to chūsei shinwa 神道の形成と中世神話 (Tokyo: Yoshikawa Kō bunkan, 2016), 74-76; James Robson, "Buddhist Sacred Geography," in Early Chinese Religion: Part Two, Period of Division (220-589 AD), ed. John Lagerwey and Lü Pengzhi (Leiden: Brill, 2010), 1370. 108 The coexistence of sameness and difference for a sacred space can be rationalized through the division of the site’s identity into two parts: (1) locality, and (2) function. In theory, every space has its own mapped area, making location a marker of uniqueness. As the name suggests, Itsukushima Shrine is the sacred space of the island of Itsukushima (modern day Miyajima). The function is based on characteristics that clarify the specific miraculous power of the site and its deity. The Itsukushima Deity’s identities as a dragon god, sister of the Dragon Princess, and Benzaiten all reinforced the belief that Itsukushima was a site efficacious for prayers related to the sea (e.g., travel, protection from foreign invaders, rain rites), amongst others like music and performance. Locality and function are compatible in terms of a single sacred space, however, comparing the functions of different sites can call the uniqueness of the sites into question. Corresponding with these two parts, site-based studies have emphasized locality, while studies of Buddho-kami relations and the honji-suijaku (original form and local traces 本地垂迹) system have focused on the functions of the gods. Setting aside spatial characteristics, Bernard Faure demonstrated how the related functions of different gods led to networks of connections between them. 204 And conversely, it is by emphasizing local contexts that site-based studies have been able to separate the gods and religious practices at the site from translocal religious discourses. Both methodologies circumvent the coexistence of locality and function by looking at one or the other, but this chapter examines the two together. One might expect links between sites to focus on the deities, but many medieval texts also made connections between the spaces themselves. It is precisely this discursive disjuncture between sacred spaces and their deities as locally particular and functionally universal that provides a new perspective for understanding 204 Bernard Faure, Gods of Medieval Japan: Volume 1, The Fluid Pantheon (Honolulu: University of Hawai`i Press, 2016), 27. 109 sacred space. Looking to the writings of medieval scholarly monks, we can see how they formulated systems of networked space to make sense of the seeming contradictions arising from spaces and deities being both local and unique, and transcendental and interconnected. Honji-suijaku and the Systematization of Relations As Buddhism spread beyond India, sacred space needed to increasingly detach itself from specific locations, such as those associated with the life story of the historical Śā kyamuni Buddha. As Koichi Shinohara has shown through narratives about sacred spaces in China, mobile objects, such as the begging bowl of the Buddha, relics, and Buddha icons, provided a means of connecting local places with sacred spaces from the broader Buddhist cosmos. 205 The honji-suijaku system similarly enabled places in medieval Japan to be associated with otherworldly Buddhist sacred spaces by abstracting the gods in Japan as the local traces (suijaku) of the original (honji) transcendental Buddhist gods. Following Satō Hiroo’s argument that honji-suijaku’s primary categorical division was not between Buddhist and kami, but saving (transcendental) and wrathful (worldly) deities, we can see how the honji-suijaku system can shift identification from sacred spaces to the gods and their original transcendence. 206 Bernard Faure aptly explains, “Despite its mythological proliferation, the honji suijaku theory was initially a demythologizing attempt in which the gods became increasingly abstract entities.” 207 205 Koichi Shinohara, "The Story of the Buddha's Begging Bowl: Imagining a Biography and Sacred Places," in Pilgrims, Patrons, and Place: Localizing Sanctity in Asian Religions, ed. Phyllis Granoff and Shinohara Koichi (Vancouver: UBC Press, 2003), 90-95. 206 Based on his conclusions, Satō advocated for new methodologies treating Buddhist and kami deities at the same level. While many scholars have acknowledged the importance of his findings, few have incorporated them into one’s own research. By examining different localized manifestations of the Benzaiten deity, this chapter serves as a first step for exploring how the field of Japanese religions can grow from Satō’s work. Hiroo Satō , "Wrathful Deities and Saving Deities," in Buddhas and Kami in Japan: Honji Suijaku as a Combinatory Paradigm, ed. Mark Teeuwen and Fabio Rambelli (New York: RoutledgeCurzon, 2003), 112-14. 207 Bernard Faure, Gods of Medieval Japan: Volume 2, Protectors and Predators (Honolulu: University of Hawai`i Press, 2016), 5. 110 To take Faure’s idea a step further, I propose that the demythologization (abstraction as universals) of the gods itself mediated their mythological proliferation (particularization as local). As local gods or local sacred spaces, they can exist separate from other gods and spaces, but understanding sacred spaces and deities in relationship to one another requires a universal principle to serve as the basis for comparison. Honji-suijaku provides the universalizing logic, and it also blurs the significance of the differences between them. 208 Consequentially, scholarly monks responded through other discourses, such as narrative, to reemphasize differences while still striving to retain the essential unity provided by honji-suijaku. The same unification and removal of differences within the unity can be seen in Zhiyi’s ( 智顗, 538-597) original usage of the terms “original” (hon 本) and “trace” (jaku 迹) to differentiate the first fourteen chapters of the Lotus Sutra as the “trace-teaching” produced by Śākyamuni Buddha, and the latter fourteen chapters as the “original teaching” of the eternal Buddha. Zhiyi used this logic to systemically allow for discrepancies between the teachings of the Buddha in different sutra texts, and to also elevate the teachings of the Lotus Sutra as primary. Within this system, the Lotus Sutra is the original, essential teaching of the eternal buddha, and the other sutras are provisional teachings performed by Śākyamuni to present the teachings in forms suited to the needs of different audiences. 209 Honji-suijaku likewise explains the local deities in Japan (in relation to each other, or those in other countries) by identifying 208 The issue is not unlike the problem of particulars and universals in identity politics, as presented in Linda Zerilli’s reading of Ernesto Laclau’s Emancipation(s). Laclau refutes the ideal of a “pure universality” devoid of any particularity, but instead argues for universals that are constructed through the re-articulation of an always unrealized unifying concept. As Zerilli explains, “the relationship between universal and particular entails not the realization of a shared essence or the final overcoming of all differences but an ongoing and conflict-ridden process of mediation through which antagonistic struggles articulate common social objectives and political strategies. M. G. Zerilli Linda, "This Universalism Which Is Not One," review of Emancipation(s), Ernesto Laclau, Diacritics 28, no. 2 (1998): 9. 209 Mark Teeuwen and Fabio Rambelli, "Introduction," in Buddhas and Kami in Japan: Honji Suijaku as a Combinatory Paradigm, ed. Mark Teeuwen and Fabio Rambelli (New York: RoutledgeCurzon, 2003), 15-16. 111 them as “traces” of the “original” otherworldly buddhas and bodhisattvas, who manifest within the world to lead sentient beings towards salvation. Any differences between worldly deities become explainable as nothing more than deviations from the original form made to better serve the needs of worshippers. Shifting back and forth between explicating universals and particulars led to increasingly complex networks of relations. On the one hand, these shifts prevented the honji-suijaku system from becoming a hegemonic force marginalizing the kami deities at the bottom of the hierarchy of Buddhist gods. On the other hand, they prevented localizing narratives from erasing the translocal networks between gods and sacred spaces. Recent scholarship has rightfully emphasized the flexibility of the associations made through the combinatory logic of honji- suijaku. They were not always one-to-one connections between buddhas and kami; buddhas, kami, holy monks, and other sacred beings could be brought together to provide deities with multiple overlapping levels of identity. 210 But, the focus on the honji-suijaku system as the primary framework for understanding the identity of the gods has limited scholarly consideration of elements, like spatiality, that cannot be abstracted within the system. Scholarly monks instead utilized other frameworks, such as enshrinement, deity genealogies, and various media (e.g., narratives, iconography, maps, ritual) to try and incorporate particularities while still corresponding with existent honji-suijaku associations. When we shift our attention to individual Benzaiten deities, honji-suijaku distinctions normally categorize Benzaiten as the original form, and the local kami deities as worldly manifestations of Benzaiten. Through this logic, each local trace of Benzaiten can take a different form to best suit the needs of regional communities, but the exact meaning of a local 210 Ibid., 47-48. 112 manifestation’s individuality is often left unexplained. Instead, assertions of shared essence and local differences were stated as fact and affirmed through preconceived ideas of local identity. In no small part, the name of the deity acts as an empty signifier of spatial identity. For example, if the Itsukushima Benzaiten and the Tenkawa Benzaiten 天川弁才天 (Tenkawa Shrine, in the Yoshino mountain range, Nara Prefecture) were the same, why would they have different names? Even when the Itsukushima Deity was enshrined at sites other than Itsukushima Shrine, the deity was usually referred to as the Itsukushima Deity or the Itsukushima Benzaiten rather than simply as Benzaiten. 211 Despite local names, the honji-suijaku system enabled the Itsukushima and Tenkawa Benzaiten to also be identified as ontologically the same. 212 The Kōban kuzetsu (“Explications of 211 As one might expect, the Itsukushima Deity was often (but not always) enshrined in the regions near Itsukushima Shrine, where the locale was especially important for the deity and the worshippers. To give two examples in brief: first, Sasai Itsukushima Shrine 佐々井厳島神社 was part of Itsukushima Shrine’s estate lands, as can be seen in an 1199 report to the Itsukushima Shrine administrative office including the village of Sasai as a tax-exempt land whose profits were reserved for funding the sakuhei 朔幣 rite performed on the first day of each month. Second, Kōkai Shrine 光海神社, according to ridgepole inscription tablets (munafuda 棟札) made at the time of the reconstruction of shrine building, included a shrine dedicated to Itsukushima since 1333, and another shrine dedicated to Hachiman since 1316. For an overview of Sasai Itsukushima Shrine and Kōkai Shrin e, see Hiroshima- ken Jinjachō 広島県神社庁, ed. Hiroshima-ken jinjashi 広島県神社誌 (Hiroshima: Hiroshima-ken Jinjachō , 1994). For the 1199 report including Sasai village, see doc. 24 in "Shinshutsu Itsukushima monjo 新出厳島文書," in Hiroshima kenshi: Kodai chūsei shiryō hen 広島県史. 古代中世資料編, ed. Hiroshima-ken 広島県 (Hiroshima: Hiroshima-ken, 1978). Concerning the ridgepole inscription tablets at Kōkai Shrin e, see Suitō Makoto 水藤真, Munafuda no kenkyū 棟札の研究 (Kyoto: Shibunkaku Shuppan, 2005), 65; Kokuritsu Rekishi Minzoku Hakubutsukan 国立歴史民俗博物館, ed. Shaji no kokuhō jūbun kenzōbutsu tō: Munafuda meibun shūsei, Chūgoku Shikoku Kyūshŭ hen 社寺の国宝重文建造物等：棟札銘文集成. 中国四国九州編 (Chiba-ken Sakura-shi: Kokuritsu Rekishi Minzoku Hakubutsukan, 1993), 137. 212 The multiple iconographies of Benzaiten as two-armed, eight-armed, or as Uga Benzaiten (ten-armed, with three snake heads) provided one means of distinguishing between different regional Benzaiten deities, but not all images followed the standard forms of each Benzaiten. For example, a 1528 dated Benzaiten scripture manuscript, “The Sutra of the Buddha Preaching on the Heavenly Woman Benzaiten,” 仏説大弁才天女経 potentially hints at a connection between Itsukushima and Tenkawa. The beginning of the text bears the alternate title, “Ritual invocation of the Itsukushima Benzaiten of Aki” (Aki no Itsukushima dai Benzaiten no saimon 安芸ノイツク嶋大弁才天ノ祭文), appearing to be describing the Itsukushima Deity, but the content of the scripture details an iconography matching that of the Tenkawa Benzaiten in the Tenkawa mandala, with three snake heads, ten arms, and two feet. Itō Satoshi, in an introduction to a transcription of the text, is at a loss for explaining the relation between Itsukushima and Tenkawa in this text, but considering it together with the Kōban kuzetsu entry, we can hypothesize that the depiction of the Itsukushima Deity in the form of the Tenkawa Benzaiten is not coincidental. Itō Satoshi 伊藤聡, 113 Kōban” 弘鑁口説), a compilation of oral transmissions by the Daigoji monk Kōban ( 弘鑁, ?- 1426), identifies the Itsukushima Deity as “same in essence with the Tenkawa [Deity].” 213 The expression “same in essence” (dōtai 同体) frequently appears within explanations of honji- suijaku associations, but aside from the rhetoric, the only explanation for the connection is the statement’s placement within a list of the sisters of the Dragon Princess from the Lotus Sutra. Here they are given as the Seiryū Gongen (the dragon deity of Daigōji), the Itsukushima Deity, and the Shinra Deity of Onjōji. Tenkawa is not on e of the three, but it is still incorporated through the asserted link with Itsukushima. 214 As discussed in the abstract above, the honji- suijaku system, here, connects the gods by removing spatial particularities from the equation. Common sense dictates that “[Benzaiten is] one in essence with [Benzaiten].” While Itsukushima and Tenkawa could be, and were, identified as the same in some cases, explanations of the external enshrinement of local deities outside of the original space enabled deities to retain their spatial identities while also moving outside of that site. One example for Tenkawa is the Washū Hashio Mitsushima Benzaiten engi genki (“Origin narrative and record of the manifestation of Benzaiten at Hashio Mitsuhima,” 和州箸尾満嶋弁財天縁起現記). The text begins by introducing the deity of the site as the Benzaiten of Tenkawa, who is Eleven Headed Kannon. 215 A later section titled Zuimukishō 瑞夢記抄 further details the narrative saying that, originally, Prince Shōtoku enshrined Ugajin 宇賀神 (i.e., Uga-Benzaiten, "Yoshida bunko shozō no Benzai ten kankei gikyō ni tsuite: Sono honkoku to shōkai 吉田文庫所蔵の弁才天関偽経について：その翻刻と紹介," Muromachi 2 (1993): 42-43. 213 Kōban, "Kōban kuzetsu 弘鑁口説," in Zoku gunsho ruijū. Shake bu 續群書類從. 釈家部, ed. Hanawa Hokinoichi 塙保己一 and Ō ta Tō shirō 太田藤四郎 (Tokyo: Zoku Gunsho Ruijū Kanseikai, 1957), 78. 214 See chapter one for a more detailed discussion of the networks of association that developed between deities identified as sisters of the Dragon Princess. 215 Believed to have been copied in 1444. Ichimura Yoshikazu 市村義一, ed. Tenkawa Benzaiten kanjō: Hashio Mitsushima Benzaiten: Zuimukishō 天河弁財天勧請 : 箸尾満嶋弁財天 : 瑞夢記抄 (Kōryō -chō, Nara Prefecture: Hashio mitsushima Benzaiten kenshō kai, 1979), 1 -2. 114 the conventional universal form for Tenkawa Benzaiten) at the site. Later, the text works to present a more specific local identity other than that of Tenkawa. “A celestial woman said, ‘My space should be called Mitsushima. Therefore, the Benzaiten of this space is called the Benzaiten of Mitsushima.” The passage then continues by adding a translocal interpretation of the name Mitsushima. “When one thinks of the true heart of the matter, ‘mitsu’ ( 満) is to provide in full (enman 円満) merit and wisdom universally to the pilgrims of this site, and also, ‘shima’ ( 島) is for the sake of providing salvation to all sentient beings of the island of Japan (Akitsushima 秋津島).” 216 In just a few pages, the origin narrative depicts the Mitsushima Benzaiten through logics of enshrinement, narrative representation, linguistic play, and honji-suijaku abstraction to simultaneously present the space(s) of the deity as Mitsushima, Tenkawa, and Japan. Even in the case of this regional temple in Nara Prefecture (in the area of modern-day Kōryō 広陵), its origin narrative needed to depict the deity as both local and translocal, as a unique god and as one of many manifestations of the Tenkawa Benzaiten. 217 Traversing the Multiple Spaces of Benzaiten 216 Ibid., 9. 217 Nakajima Ayaka provides additional medieval examples of other temples and shrines in Nara Prefecture enshrining the Tenkawa Benzaiten deity to argue that the enshrinement of local Benzaiten deities was an established practice. She mentions the enshrinement of Tenkawa Benzaiten at Jōruriji 浄瑠璃寺 in 1296, Saidaiji 西大寺 in 1365, Yakushiji 薬師寺, and Tōshōdaiji 唐招提寺 in 1582. Additionally, the Kyūchūji origin narrative (Kyūchūji engi, 1330) tells of the sculpting of a new Tenkawa Benzaiten icon and its enshrinement there in 1325. The last scroll of an early modern illustrated scroll manuscript of the Kyūchūji origin narrative visually depicts the Tenkawa Benzaiten icon as following the iconography of the eight-armed Benzaiten. For Jōruriji, see the Nanbokuchō period Jōruriji ruki, "Jōruriji ruki no koto 浄瑠璃寺流記事," in Dai Nihon Bukkyō zensho. Jishi bu 大日本仏教全書.寺誌部, ed. Suzuki Gakujutsu Zaidan 鈴木学術財団 (Tokyo: Kōdans ha, 1972), 338. For Kyūchūji, see "Kyūchūji engi 橋柱寺縁起," in Kizu chōshi: Shiryō hen 木津町史：史料篇 (Kyo ̄to-fu Sō raku-gun Kizu-chō : Kizu-chō , 1984), 802; "Kyūchūji engi emaki 橋柱寺縁起絵巻," (http://mahoroba.lib.nara-wu.ac.jp/y21/kyouchuji_engi/index.html: Daichiji 大智寺). For Nakajima’s discussion of the enshrinement of the Tenkawa Benzaiten, see Nakajima Ayaka 中島彩花, "Chūsei Benzaiten shinkō to zuzō ni kansuru ken kyū: Benzaiten mandara no tenkai ni miru shinbutsu kan 中世弁才天信仰と図像に関する研究 : 弁才天曼荼羅の展開にみる神仏観" (Joshi bijutsu daigaku, 2013), 67. 115 Any Buddhist deity can exist in both worldly and otherworldly forms, but notably in the case of Benzaiten, the Buddhist sutras emphasize her worldly location. In particular, the Golden Light Sutra describes the worldly environs where Benzaiten exists in specific but still widely applicable terms, which led the scripture to become the locus classicus for explaining Benzaiten’s enshrinement at specific temples and shrines. A hymn within the sutra places Benzaiten “either abiding in mountain cliffs deep and dangerous, or in caves on riverbanks, or in trees and groves.” 218 In contrast to other Buddhist deities located within otherworldly pure lands, Benzaiten’s worldly location came to the forefront, and supported the widespread enshrinement of Benzaiten in sacred mountains, islands, and lakes throughout Japan. As Benzaiten became enshrined in sacred mountains, islands, and lakes throughout Japan, the need to explain the multiplicity of Benzaiten beyond one-to-one identifications (such as Itsukushima and Tenkawa) increased. Scholarly monks increasingly used networks of underground tunnels linking geographically separate Benzaiten spaces to represent the relational identities of their sacred spaces. The environs of Benzaiten within the Golden Light Sutra of caves and cliffs provides an effective basis for designating Benzaiten’s true place of enshrinement within underground spaces rather than the shrines on the surface. A whole other sacred world could exist below ground, where Benzaiten spaces could be brought together regardless of their geographic distance above ground. This allowed for easy movement of the same Benzaiten deity between the sites, and thus for the unity of the multiple sites as a unified sacred space. 218 T 665, 16:437a14-a15. Translation based on that of Catherine Ludvik. Catherine Ludvik, Sarasvati: Riverine Goddess of Knowledge, From the Manuscript-Carrying Vina-Player to the Weapon-Wielding Defender of the Dharma (Leiden: Brill, 2007), 211. 116 The useful gap between the spaces above and below ground, and between honji-suijaku abstractions and local particularity, was incorporated into actual maps of sacred spaces. A 16 th century illustrated map of the twenty-one shrines of Mt. Hiei (Sannō nijūissha tōezu 山王二十一社等絵図) (Fig. 1) uses a cave at Chikubushima to connect it with Iwataki Shrine 岩瀧社 within the Hie Shrine complex. Because of Chikubushima’s fame as a prominent Benzaiten site and its nearby location on an island on Lake Biwa, both Chikubushima and Iwataki became identified as the Benzaiten of Hie Shrine. Yet, their separate names and locations naturally created the need to further explain the relationship between the two sites. On this detailed map of the Hie Shrine complex, the buildings are each labeled with the names of the deities enshrined, and this provided the opportunity for elaboration. Next to the shrine for Iwataki is a note, “Iwataki. One in essence with Chikubushima.” This matches with the textual accounts in the Sannō hiki 山王秘記 and the Tenchi jingi shinchi yōki 天地神祇審鎮要記, and then, another passage next to a rock to the left of the shrine uses the logic of underground tunnels to physically connect Iwataki and Chikubushima. 219 “Hole [in the] rocks. A passageway to Chikubushima.” 220 While the honji suijaku system allows for Iwataki and Chikubushima to be “one in essence” as manifestations of Benzaiten, the underground connection provides a more coherent means to understand the relations between manifestations of Benzaiten at different sites. These two sentences allow the reader to imagine Iwataki and Chikubushima as both the same and different, and to use a map to move beyond the limits of mapping. 219 "Sannō hiki 山王秘記," in Zoku Tendaishū zensho: Shintō 續天台宗全書：神道, ed. Tendai Shūten Hensanjo (Tokyo: Shunjūsha, 1987-), 231; Jihen 慈遍, "Tenchi jingi shinchi yōki 天地神祇審鎮要記," in Tendai shintō. jō 天台神道. 上 ed. Sueki Fumihiko 末木文美士, Mizukami Fumiyoshi 水上文義, and Satō Masato 佐藤真人 (Tokyo: Shintō Taikei Hensankai, 1990), 463. 220 Notations in the map transcribed and discussed within Shimizu Minoru 清水実, "Hie Sannō mandala kō: Sa nnō Shintō shiryō kara no yomitoki 日吉山王曼荼羅考：山王神道史料からの読み解き," Mitsui bijutsu bunkashi ronshū 6 (2013): 22-23. 117 Figure 2: Sannō nijūissha tōezu 山王二十一社等絵図, Eizan bunko, 16th century. Circles added by author to show the locations of Iwataki Shrine (on the right) and the rock with the cavern leading to Chikubushima (on the left). Image from Kageyama Haruki, ed., Shintō taikei. Jinja hen, Hie 神道大系. 神社編：日吉, 379-399. The Keiran shūyōshū , an encyclopedic compilation of doctrinal explanations, oral transmissions, and myriad other teachings, includes an illuminating chapter on Benzaiten. 221 221 T 2410, 76:626-628. I will be referring to the Keiran shūyōshū in the Taishō canon unless otherwise noted. The Shintō taikei also includes a transcription of the Keiran shūyōshū working from the same manuscript as the Taishō canon, but it only includes three out of the six sections in the text. See Sueki Fumihiko 末木文美士, ed. Shintō taikei: Ronsetsu hen, Tendai Shintō (ge) 神道大系. ：論說編 , 天台神道. 下 (Tokyo: Shintō Taikei Hensankai, 1993). Two alternate versions of the Benzaiten chapter still exist in the Hanmachi-shi version ( 反町氏本) and the 118 Compiled from 1311-1350 (the Benzaiten chapter was written no later than 1318) 222 by Kō shū ( 光宗, 1276-1350), a monk of the precept lineage (kaike 戒家) at Kurodani on Mt. Hiei, the Keiran shūyōshū provides an expansive perspective on the contemporary religious discourses circulating both within and outside of Mt. Hiei. 223 Incorporating a diverse range of sources, the Keiran shūyōshū lacks a singular or encompassing account of Benzaiten, but instead uses different accounts to make sense of the diverse views of Benzaiten. 224 Sanzen'in En'yūgura version ( 三千院円融蔵). For a discussion of the different versions of the Keiran shūyōshū see Tanaka Takako 田中貴子, "Keiran shūyōshū" no sekai 「渓嵐拾葉集」の世界 (Nagoya: Nagoya Daigaku Shuppankai, 2003).For a transcription of the Hanmachi-shi version see Fujiwara Shigeo 藤原重雄, "Hanmachi Shigeo shi shozō "Keiranshū yō shū" Benzaiten bu (Honkoku, kaidai) 反町茂雄氏所蔵『渓嵐拾葉集』弁才天部（翻刻・解題）," in Kamigami no sugata, katachi o meguru tamenteki kenkyū 神々のすがた・かたちをめぐる多面的研究, ed. Shimane-ken Kodai Bunka Sentā (Matsue-shi: Shimane-ken Kodai Bunka Sentā , 2011). 222 Nanami Hiroaki calculates the date of compilation based on the years of certain monks with whom Kō shū must have interacted with. For example, a certain transmission must have been given by the monk Enkō 円興, who died in 1318. Nanami Hiroaki 名波弘彰, "Nanto bon "Heike monogatari" Tsunemasa Chikubushima mōde to Hiyoshi seijo no miya no biwa hōshi: Hiei san shinkō -ken ni okeru Uga Benzaiten shinkō wo megut te 南都本『平家物語』経正竹生島詣と日吉社聖女宮の琵琶法師 : 叡山信仰圏における宇賀弁財天信仰をめぐって," Bungei gengo kenkyū 11 (1987): 173. 223 Previously, scholars such as Allan Grapard looked at the Keiranshūyōshū primarily in regard to the religious practices and transmissions based around Mt. Hiei, but Tanaka Takako argues that the travels of Kō shū to various temples along with the coming and going of various religious figures to Mt. Hiei allowed Kō shū to collect information from a diverse range of sources. Allan G. Grapard, "Keiranshūyōshū: A Differe nt Perspective On Mt. Hiei in the Medieval Period," in Re-visioning "Kamakura" Buddhism, ed. Richard Karl Payne (Honolulu: University of Hawai'i Press, 1998), 42-78; Tanaka, "Keiran shūyōshū" no sekai 「渓嵐拾葉集」の世界, 165-79. 224 Because the Benzaiten cult rose in prominence at Mt. Hiei before spreading throughout the realm, Kōshū’s production of the Keiran shūyōshū undoubtedly played a significant role in the development of discourses on Benzaiten in the medieval period. Bernard Faure, "The Cultic World of the Blind Monks: Benzaiten, Jūzenji, and Shukujin," Journal of Religion in Japan 2 (2013): 173. 119 In the Keiran shūyōshū, the Benzaiten chapter is divided into two parts, secret teachings (hiketsu 祕決) and origin narratives. 225 The structure of the Benzaiten sections reveals the divergent abstract (doctrinal/cosmological) and particular (origin narratives, local beliefs, and practices) discourses on Benzaiten that Kōshū was compili ng and trying to understand in relation to one another. 226 When discussing doctrine, it is hard to explain the differences between any of the Benzaiten, but when discussing origin narratives, it is hard to explain why there are so many similarities between the supposedly local origins and miraculous deeds of different Benzaiten. To solve these problems, Kōshū maps the most prominent Benzaiten of the realm into networks, wherein the relations can then be used to explain similarities and differences. For example, the section on the Tenkawa origin narratives gives an account of the relationship between three Benzaiten sites: “Tenkawa, Itsukushima, and Chikubushima are three caves mutually penetrating and passing through to one another. The three precious jewels [of the three Benzaiten] are one in body and mutually interpenetrating” (Fig 2). 227 This passage has 225 The Keiran shūyōshū includes accounts of the origin narratives of five of the six Benzaiten, but strangely leaves out an account of the Minoo Benzaiten. Minoo is explained as one in body with the Tenkawa Benzaiten, but especially considering the encyclopedic nature of the Keiran shūyōshū, the absence of a Minoo origin narrative is an uncharacteristic omission. It is possible that Kōshū did not have access to the Minoo origin narrative, or he chose not to include the origin narrative to avoid having to explain the shared narratives in both the Minoo origin narrative and the Sefuriyama origin narrative. However, this latter explanation likely would have been a lesser issue compared with the incompleteness of providing only five of six origin narratives. On the similar elements in the Minoo and Sefuriyama origin narratives, see Yoshida Fukiko 吉田扶希子 Sefuriyama shinkō no genryū: Nishi Nihon chiiki o chūshin to shite 脊振山信仰の源流 : 西日本地域を中心として (Fukuoka: Chūgokushoten, 2014), 58-62; Yabu Motoaki 籔元晶, "Ryūju bosatsu to shugen no yama: Minoo, Atagoyama, Sefuriyama no denshō k ara 竜樹菩薩と修験の山: 箕面・愛宕山・背振山の伝承から," Mikage shigaku ronshū 40 (2015): 4-6. 226 Faure, Gods of Medieval Japan: Volume 2, Protectors and Predators, 201. 227 T2410, 76:625a23-625a24 Figure 3: Three Benzaiten as wish fulfilling jewels. Image from Keiran shūyōshū 渓嵐拾葉集, T2410, 76:625a23-625a24 120 already been cited within Benzaiten scholarship as evidence of connections between the major Benzaiten sacred spaces, and of the presence of a subterranean, invisible realm of connections between the sacred sites. 228 These two points provide the foundation for my ideas here. By situating this passage in relation to other contemporary works, I will demonstrate that these types of connections were not necessarily an innovation by Kōshū, but a far -reaching discourse framing how Benzaiten and her sacred space(s) were imagined. Beginning with the theme of the underground connections, I argue that the tunnels between sites provided the mediation needed to explain connections between spaces while retaining the possibility of local particularity. Then, I will return to the list of the three Benzaiten in the Keiran shūyōshū to argue that the networked connections between sites were not always neutral relations of equivalence but could instead be used as a basis for defining hierarchical relations between different Benzaiten sites. The passage on the three interconnected Benzaiten caves is especially valuable because it succinctly connects abstract Buddhist logic with a concrete spatial logic. Both Bernard Faure and Gregory Smits mention the similarity with Daoist grotto-heavens (Ch. dongtian, Jp. dōten 洞天), but this reveals more than just the influence of Daoism. 229 The grotto-heavens provided a model for how caves could serve as supernatural pathways. The Chinese character dong 洞 literally means cavern, but can also be used interchangeably with the word tong 通 to mean penetrate or communicate. 230 While the Keiran shūyōshū uses the word “cavern” (ana 穴) instead of “grotto- 228 Andrew Mark Watsky, Chikubushima: Deploying the Sacred Arts in Momoyama Japan (Seattle: University of Washington Press, 2004), 55-56; Gregory Smits, "Conduits of Power: What the Origins of Japan's Earthquake Catfish Reveal about Religious Geography," Japan Review 24 (2012): 56-57; Faure, Gods of Medieval Japan: Volume 2, Protectors and Predators, 206. 229 Gods of Medieval Japan: Volume 2, Protectors and Predators, 206; Smits, "Conduits of Power: What the Origins of Japan's Earthquake Catfish Reveal about Religious Geography," 59. 230 Franciscus Verellen, "The Beyond Within: Grotto-Heavens (dongtian) in Taoist Ritual and Cosmology," Cahiers d'Extrême-Asie 8 (1995): 271; Julius Tsai, "In the Steps of Emperors and Immortals: Imperial Mountain Journeys and Daoist Meditation and Ritual" (Ph.D. diss., Stanford University, 2004), 47. 121 heaven,” the phrase “mutually penetrating and passing through to one another” (tagai ni tsutetsu 互通徹) recalls the language of earlier Chinese texts. Moreover, cases from both Chinese and Japanese sources show the caves providing access not only for travel vertically between earth, heavens, and hells, but also horizontally between worldly spaces. 231 The horizontal travel through grotto-heavens provides the necessary mobility for interpreting the ambiguous interrelations between sacred sites. As relational links, the connections could be used to highlight a shared essence (commonality), regional distinctions, or both. To provide another example from the Keiran shūyōshū, Kō shū quotes a writing attributed to the Tendai priest Annen ( 安然, 840-?), which situates the Chikubushima Benzaiten and Enoshima Benzaiten in parallel. The text states: An inscription by Annen says: “Within the lake in Gōshū , there is a sacred island. [It is] because the living body of Benzaiten dwells here [that] the Buddhist Law at Mt. Hiei is able to flourish. Within the sea at Sōshū, there is a sacred island. [It is] because the living body of Benzaiten dwells here [that] the Buddhist Law in Kamakura is able to flourish.” 232 The descriptions of the two Benzaiten are identical aside from the places. The Chikubushima Benzaiten protects Mt. Hiei, and Enoshima Benzaiten protects Kamakura. The emphasis on commonalities fits with the combinative logic of honji-suijaku. Conversely, when we turn to a contemporary Enoshima origin narrative (extant by 1323), 233 the text contains an almost identical 231 Franciscus Verellen cites two examples from Chinese Daoist texts in which a person enters one grotto cave and emerges at another grotto. In both cases, it only takes the person a short amount of time to travel between the two sites, even though they are geographically far from one another, suggesting a supernatural power of travel through the grottos. Verellen, "The Beyond Within: Grotto-Heavens (dongtian) in Taoist Ritual and Cosmology," 269, 71. Also see Miura Kunio 三浦国雄, "Dōten fukuchi shōron 洞天福地小論," Tōhō shūkyō 61 (1983): 8. 232 T2410, 76: 627a9-12. 233 The Enoshima origin narrative (Sōshū Tsumura Enoshima engi 相州津村江之島縁起) manuscript from the Kanazawa Bunko archive contains a colophon dating the copying of the text to 1323. An almost identical manuscript with occasional differences is stored at Enoshima Shrine and was copied post 1413. For an explanation of the dating of the two manuscripts, and side by side transcriptions of the two texts, see Hattori Seidō 服部清道, "'Enoshima engi' kō 『江島縁起』考," Yokohama shōdai ronshū 10, no. 1 (1977): 157-59. 122 passage. The changes that have been made reveal efforts to clarify the relationship between the two Benzaiten sites and distinguish their roles. The origin narrative contrasts the sites: A writing by Annen says, “Within the lake in Gōshū there is a sacred island. I ts name is Chikubushima. The living body of Benzaiten abides at that island. Due to this, Mt. Hiei flourishes, and it serves as a site for the protection of the realm. In the southern sea of Sōshū, the province of my birth, there is a sacred island. It is called Enoshima. The living body of Benzaiten is on this island. Therefore, [it] is certainly the sole source for the flourishing of this province.” 234 Here, Chikubushima assists Mt. Hiei’s role as a temple for the protection of the realm, while Enoshima Benzaiten plays a regional role as the protector of Sagami Province. The text also mentions Annen’s birth in Sagami Province, providing another layer of connection with the province. But, the spaces of the miraculous powers of the two deities are now separated as translocal (Chikubushima) and local (Enoshima). This is not enough for the Enoshima origin narrative though. This passage is followed with a quotation from a secret writing of Annen, which details the interconnected caves where the Chikubushima and Enoshima Benzaiten dwell. The text guides the reader through cave’s entrance and inner chambers, detailing the multiple spaces within. While the descriptions of the two sites are too long to quote in full, the following passage provides a glimpse into the text’s imagination of the space underneath Chikubushima. A secret writing of Annen says, “The island mountain to the west is named woman rock. This, in other words, is [the] female deva [of] the assembly of the womb world. At the southern shore of the island there is a cave. It is called the golden cave. From the middle of the cave, a golden light sometimes shines. This is the reason why it is called the golden cave. Within the cave, its temples are two-fold. In the inner temple as well, there are two caves. In the eastern cave the womb world mandala is enshrined. In the western cave the diamond world mandala is enshrined. Benzaiten is in the middle of them. To [Benzaiten’s] left is Nagarjuna bodhisattva, to [Benzaiten’s] right is the Great King Tokuzen 徳善大王. In front, there is a lapis lazuli platform, and a scroll of the Lotus Sutra is placed on top of the platform. In front of that is a lake. Within the lake there is a 234 Ibid., 147. 123 white dragon. Its height is eight sun 寸. 235 This [dragon] is the dragon king of Munestsu lake 無熱池. This cave is, in other words, Benzaiten’s original jeweled palace.” 236 The passage continues, leading the reader into other connected caves, with one each as the abode for the Dragon King Sagara, a male immortal, a female immortal, and the Kimbila General (Konpira taishō 金毘羅大将). The secret writing then goes on to a similar intricate description of the space of Enoshima, identifying Enoshima as the male deva of the diamond world to pair with Chikubushima as the female deva of the womb world. This type of a mandalic linking of sacred spaces is not uncommon, 237 but the move from symbolic connections to a visualization of the underground spaces of the two sites is significant. As underground ritual spaces and sites of enshrinement, the caves are described as sites of the gods and as sites for religious practice. The secret writing also emphasizes the geographic connection between the two sites through subterranean tunnels. With multiple levels of caves leading to more caves, the passage provides an excess of detail, impressing the reader with a convoluted network of interconnected sacred spaces. “In the space between the two mountains, there is a dragon cavern (ryūketsu 龍穴). This is the cave through which Benzaiten spontaneously flows forth.” 238 Based on the understanding of a sacred space as the space where a deity is enshrined, if a deity travels and dwells at multiple sites, then the sites can be understood as a single sacred space. That is not to say that they are not different, but they are also networked, and potentially unified as one. Gregory Smits, in his study of the catfish underneath Chikubushima, similarly emphasized the underground connections between sacred sites. However, while Smits describes 235 A sun is a unit of measure, with one sun equivalent to approximately 3.03 centimeters in length. Therefore, the eight sun dragon would have a height of 24.24 centimeters, or about 9.54 inches. 236 Hattori, "'Enoshima engi' kō 『江島縁起』考," 147-48. 237 On the mandalization of sacred space see Grapard, "Flying Mountains and Walkers of Emptiness: Toward a Definition of Sacred Space in Japanese Religions," 207-14. 238 Hattori, "'Enoshima engi' kō 『江島縁起』考," 150. 124 the connections between the sites as “resembl[ing] a modern electrical power grid,” 239 I stress the relations between the sites as a network rather than a flow. The deities and their sacred power do not flow freely in all directions, but are linked through select sites and regulated, in part, through narrative and discourse. Manuel Vasquez praises the multivalent power of networks for opening, closing, and hierarchizing relations between spaces. 240 The significance of networked space as a system for contesting and hierarchizing relations between Benzaiten spaces will be discussed later, but it is worth noting here that the connections between Benzaiten sites were not always as unrestricted as they seem. Surveying medieval writings on the sacred spaces of various Benzaiten and dragon deities shows that the motifs of dragon cavern and underground sacred sites spread widely. Dragon caverns connected Atsuta Shrine with a series of caves and sacred spaces spreading out in all directions. 241 Other caves connected individual sacred sites with the dragon palace or lakes in India associated with the dragon king. 242 Unsurprisingly, the Chikubushima origin narrative 239 Smits, "Conduits of Power: What the Origins of Japan's Earthquake Catfish Reveal about Religious Geography," 42. 240 Vásquez, "Studying Religion in Motion: A Networks Approach," 178-79. 241 Even sites less known for associations with Benzaiten or dragon deities utilized the motif of dragon caves to make connections. The “Atsutagū hishaku kenmon” 熱田宮秘釈見聞 maps spaces throughout Japan linked through underground tunnels between mountain caves and bodies of water: Shiratorizuka 白鳥塚 grave tumuli for the Atsuta Deity; Mt. Fuji; Ise Shrine; Kōya’s Oku no in; Mt. Hakusan; Mt. Asama; and Mānasarovara lake in India (Munetsuchi 無熱池). The text follows this list of places with an explanation relating the connected spaces with Kūkai and dragon deities. “Like this, due to the caves passing through to one another, [the lake at Atsuta] is the lake of the gods. Kūkai made performed religious practices there, and the Zennyo Dragon King made a pilgrimage to Atsuta Shrine [through the lake]. Also, the tumulus dragon is one type of animal (i.e., dragon).” An Atsuta liturgy (Atsuta kōshiki 熱田講式) also mentions the nine caves connecting with Atsuta, but instead works from the number nine to provide Buddhist symbolic connections with the lotus thrones of the nine honored ones of the eight petaled hall in the womb world mandala, and then with the symbolic form (samaya, 三摩耶) of Dainichi Buddha. "Atsutagū hishaku kenmon 熱田宮秘釈見聞," in Chūsei Nihongishū 中世日本紀集, ed. Okabe Kaien 岡部快圓, et al., Shinpukuji zenpon sō kan (Kyoto: Rinsen Shoten, 1999), 359; "Atsuta kō shiki 熱田講式," in Chūsei Nihongishū 中世日本紀集, ed. Okabe Kaien 岡部快圓, et al., Shinpukuji zenpon sō kan (Kyoto: Rinsen Shoten, 1999), 366. 242 Examples include Shinsen’en 神泉苑 in the capital, Ishiyamadera 石山寺, Murōji 室生寺, and Gion Shrine 祇園社. On Shinsen’en see Brian O. Ruppert, "Buddhist Rainmaking in Early Japan: The Dragon King and the Ritual Careers of Esoteric Monks," History of Religions 42, no. 2 (2002): 145-46, 59-64; Matsumoto Ikuyo 松本郁代, "Shinsen'en to 'ryūō': Muromachi jidai 'reijō' no isō 神泉苑と「龍王」 : 室町時代「霊場」の位相," Art research 6 (2006). On Ishiyama, see "Ishiyamadera engi 石山寺縁起," in Zoku gunsho ruijū. Shake bu 續群書類從. 釈家部, 125 (Chikubushima engi 竹生島縁起, 1414-1415) also contains a passage about a dragon cavern grotto on the island. In the beginning, En no Gyōja entered a dragon cavern grotto. Passing through roads within this grotto, he entered and exited [the caves], going north and south, back and forth. The cave entranceways are small, narrow, and bending. Within the center there is a lake. Above, there is a large decaying tree. This is the dwelling place of the great dragon. Now, in the southwest area of this island, there is a stupa built by En no Gyōja. 243 While written with less detail than the Enoshima origin narrative example, the account likewise guides the reader into the dragon cavern, and it gives just enough details to complicate the journey within the underground passageways before arriving at the dwelling place of the deity. Even though this motif of the underground tunnels served the purpose of connecting distant places, it was more than just a rhetorical strategy to form associations between distant places and deities. It was part of an imagined other world of the sacred. 244 The Benzaiten temples and ed. Hanawa Hokinoichi 塙保己一 and Oota Toshiro 太田藤四郎 (Tokyo: Zoku Gunsho Ruijū Kanseikai, 1984), 102. On Murōji, see "Ben'ichisan ki 宀一山記," in Zoku gunsho ruijū. Shake bu 續群書類從. 釈家部, ed. Hanawa Hokinoichi 塙保己一 and Oota Toshiro 太田藤四郎 (Tokyo: Zoku Gunsho Ruijū Kanseikai, 1984), 296; Fujimaki Kazuhiro 藤卷和宏, "Hase no ryūketsu to <Nyoi hōju>: Hasedera engi no tenkai, ‘Benic hisan ’ wo meguru gensetsu gun to no kōsa 初瀬の龍穴と < 如意宝珠>：長谷寺縁起の展開・「宀一山」をめぐる言説群との交差," Kokubungaku kenkyū 130 (2000): 103-06, 10. On Gion, see Minamoto Akikane 源顕兼, Kawabata Yoshiaki 川端善明, and Araki Hiroshi 荒木浩, eds., Kojidan, Zoku Kojidan 古事談. 続古事談, Shin Nihon koten bungaku taikei (Tokyo: Iwanami Shoten, 2005), 728; Kawamura Minato 川村湊, Gozu Tennō to Somin Shōrai densetsu: Kesareta ishintachi 牛頭天王と蘇民将来伝說 : 消された異神たち (Tokyo: Sakuhinsha, 2007), 86-87. 243 "Chikubushima engi 竹生島縁起," in Gunsho ruijū. Jingi bu 群書類従. 神祇部, ed. Hanawa Hokinoichi 塙保己一 (Tokyo: Gunsho Ruijū Kankō kai Kantō sha Nai, 1952), 311. 244 If we move forward to the early modern period, we can see that the underground tunnels in the medieval Enoshima origin narrative ended up influencing proselytizing narratives about Chikubushima and Enoshima. In the case of Enoshima, the underground tunnels underneath Enoshima become connected with Mt. Fuji, as seen in the Fuji no hitoana sōshi (Tale of Fuji cave, 富士の人穴草紙). In this story, Nitta no Shirō Tadatsune is sent by the Kamakura shogun, Minamoto no Yoriie, to explore a cave on the side of Mt. Fuji. He ends up meeting the Asama Deity of Mt Fuji, who guides him down the cave and into a tour of the worlds of hell. While most versions of the text end the hell tour simply with the deity returning him to the human world (such as in the 1603 Akagi bunko version), the 1803 Kodani version adds the motif of the Benzaiten cave to have Nitta emerge from the Fuji cave at Enoshima. Similarly, in the sekkyō bushi narrative Sayohime 小夜姫, after the girl Sayohime placates a local serpent deity, the deity transports Sayohime from Lake Biwa, the location of Chikubushima Island, to Sarusawa Pond 猿沢池 in Nara. Nakano Maori 中野真麻理, "Ichijō shūgyokushō shoin waka kō 『一乗拾玉抄』所引和歌攷," Kokubungaku kenkyū shiryōkan kiyō 23 (1997): 177-78; R. Keller Kimbrough, trans., "Sayohime," in Wondrous Brutal Fictions: Eight Buddhist Tales from the Early Japanese Puppet Theater (New York: Columbia University Press, 2013), 187-88. 126 shrines, the lake (Lake Biwa), the decaying tree, and En no Gyōja’s stupa at Chikubushima are then, the external manifestations of the sacred space, but the dwelling place of the deity is below the surface and out of sight. 245 The possibility of adding an invisible second side to the sacred space seen from the surface provided a useful means for making connections through Benzaiten and dragon deities, and also through sacred objects like relics, wish fulfilling jewels, and sutras. Fujimaki Kazuhiro argues that the practice of burying relics and wish fulfilling jewels possessed the power to transform and sacralize the space above. Origin narratives and other records connected individual relics with sacred monks such as Kūkai and Gyōki , and wish fulfilling jewels with Benzaiten and dragon kings, allowing the buried objects to add their own sacred meanings to those of the space, adding new linkages to preexisting networks of association with holy figures, lineages, and other sacred spaces. Moreover, the presence of the sacred objects beneath the surface acted as uncontestable evidence of the site’s sacrality. 246 In this manner, the lack of a visible form of underground space empowered narrative and discourse to imagine the site with connections extending beyond what the eye can see. The functional overlap between dragon deities and Benzaiten as water deities further associated underground locations of the dragon palace under the sea with Benzaiten caves. This connection comes together in the Kuwanomidera origin narrative (Kuwanomidera engi 桑実寺 245 The importance of the invisibility of the space has also been made for the underground sacred space of Daoist grotto heavens in China. For example, Franciscus Verellen described the grotto as “the invisible interior counterpart of the sacred mountain.” It is both part of the mountain and also distinguished in part through the impossibility of seeing it with one’s own eyes. Verellen, "The Beyond Within: Grotto-Heavens (dongtian) in Taoist Ritual and Cosmology," 269. 246 Fujimaki Kazuhiro 藤卷和宏, "Seichi no chika ni ha: Nihon chūsei no hōju, shari shinkō no hitosumi yori 聖地の地下には：日本中世の宝珠・舎利信仰の一隅より," in Seichi to seijin no tōzai: Kigen wa ikani katarareru ka 聖地と聖人の東西 : 起源はいかに語られるか, ed. Fujimaki Kazuhiro 藤卷和宏 (Tokyo: Bensei Shuppan, 2011), 12-13, 29. 127 縁起, 16 th century), which states, “This lake (Lake Biwa) is the pure land of Benzaiten” and “Benzaiten is the transformation body of the eight great dragon kings. At the bottom of Lake [Biwa] is the dragon palace.” 247 Similarly, the Jingi hishō ( 神祇秘抄, Nanbokuchō period) first identifies Benzaiten as the master of the dragon palace, and then as the wish fulfilling jewel given to Śākyamuni by the dragon princess in the Lotus Sutra. 248 This interpretation connecting Benzaiten with both the dragon palace and the dragon princess fits perfectly with Itsukushima’s strong connections with the Dragon Princess. At the end of the Itsukushima shake engi 厳島社家縁起 origin narrative manuscript copied by Yoshida Kanemigi ( 吉田兼右, 1516-1573), Kanemigi added a secret transmission listing seven secret places at Itsukushima. Notably, the first two spaces connect Itsukushima with underground sacred spaces. The first is an underground route connecting the sea in front of Itsukushima with Kannon’s Pure Land called Fudaraku (Skt. Potalaka, Ch. Butuo, 補陀落). The second traces the wish fulfilling jewel of the Dragon Princess from the Lotus Sutra to the Itsukushima Deity, and then to its resting place within a stupa buried on Mt. Misen, the mountain behind Itsukushima Shrine. 249 Incorporating 247 This explanation is given within the narrative by the monk Jōe ( 定恵, 643-666) when he is summoned to interpret a dream for Tenji Tennō ( 天智天皇, 626-671). Jōe prov ides these details about Lake Biwa and Benzaiten as his basis for interpreting the dream as a sign that Benzaiten will summon an icon of the Medicine King Buddha from the treasury of the dragon palace to cure Tenji’s daughter of her sickness. "Kuwanomidera engi 桑実寺縁起," in Zoku gunsho ruijū. Shake bu 續群書類從. 釈家部, ed. Hanawa Hokinoichi 塙保己一 and Oota Toshiro 太田藤四郎 (Tokyo: Zoku Gunsho Ruijū Kanseikai, 1957), 123-24. 248 Bernard Faure mentions the Jingi hishō’s connection between Benzaiten and the dragon palace, but then suggests in the next sentence that all caves and waterfalls could provide access to the dragon palace. While I also stress the prevalence of connections between Benzaiten / dragon deities, caves, and the dragon palace, I propose that the presence of these underground caves and connections works to distinguish sacred caves and waterfalls from those found in profane spaces. Faure, Gods of Medieval Japan: Volume 2, Protectors and Predators, 187; "Jingi hishō 神祇秘抄," in Chūsei Nihongishū 中世日本紀集, ed. Okabe Kaien 岡部快圓, et al., Shinpukuji zenpon sō kan (Kyoto: Rinsen Shoten, 1999), 394-95. 249 “Number 1: Fudaraku is located at the bottom of the sea in front of the torii. Number 2: The [Itsukushima] Deity [received] from the eight year old dragon girl a wish fulfilling jewel, the mother of enlightenment, [and] a lifetime of sacred teachings, and buried them in a glass stupa. This [place where they are buried] is Mt. Misen.” Matsui Teruaki transcribed the first secret place and argues that Fudaraku here is interchangeable with the dragon palace. Using the preexisting idea of the underground sea passageway to the dragon palace helps the text to make a direct route from Itsukushima to Fudaraku as well. Matsui Teruaki 松井輝昭, "Itsukushima jinja no Benzaiten shinkō no seiritsu to 128 both the strategies of underground burial discussed by Fujimaki and underground Benzaiten / dragon palace routes, these two secret places connect Itsukushima Shrine with the nearby spaces of the sea and Mt. Misen, and with the otherworldly spaces of the Fudaraku and the Dragon Palace. In contrast to the earlier origin narrative connecting the far away spaces of Enoshima and Chikubushima, the secret places of Itsukushima extend the sacred space of the shrine to other places within the vicinity. Regardless of distance, the same logics of associations and of underground space work to construct the networked space beyond being a single geographic area marked on the map. To raise one last point about the underground sacred spaces of Benzaiten, it is important to consider these sites as not only imagined spaces, but also potentially as actual sites of religious practice. The inaccessibility of the spaces in some cases could limit the people who would practice at the site, but in some cases, they could provide worshippers with especially close access to the deities. 250 While rare, there are a few records that hint at the presence of pilgrims and ascetic practitioners within the caves. In the case of Enoshima, Lady Nijō (Go -Fukakusa’in no nijō 後深草院二条, daughter of Minamoto no Masatada 源雅忠, 1258–?) recorded that there sono seikaku 厳島神社の弁財天信仰の成立とその性格," Kenritsu Hirshomia Daigaku ningen bunka gaku-bu kiyō, no. 8 (2013): 145; "Itsukushima shake engi 厳島社家縁起," in Yoshida bunko (Tenri University), 10-11. 250 In a separate but conceptually related case, several shrines at Hie Shrine contained underground chambers (geden 下殿) underneath the buildings where rituals and ascetic practices could take place. The space below the shrine provided an accessible space close to the shrines which barred most people from entry. Additionally, Tokuda Kazuo’s analysis of origin narratives and sankei mandara (“pilgrimage mandala,” 参詣曼荼羅) shows how the deity can be situated underneath the buildings, and therefore, the religious practices in those spaces can be seen as corresponding with the narrative and visual depictions of the shrine space. Helen Hardacre, Shinto: A History (New 129 were places to spend the night within the many grottos on the island, and that she met an aged ascetic practitioner within the Senju Grotto (senju no iwaya 千手の岩屋). 251 Additionally, a Muromachi period illustrated scroll (emaki) version of the Enoshima origin narrative depicts En no Gyōja encountering Benzaiten at a grotto cave labeled “golden cavern” ( kinkutsu 金窟) (fig. 4). Besides the visual prominence of the thickly outlined rocks along the water, the naming of the cave suggests that it was a real place that pilgrims could see or visit. In this manner, the religious discourses, narratives, images, and lived experiences could influence one another. A topographical study of the caves within Mt. Hiko in Kyushu revealed remnants of twelfth to thirteenth century ceramics, which were likely used as religious implements. York: Oxford University Press, 2017), 180; Kuroda Ryūji 黒田龍二, "Hie shichisha honden no kōsei: Yukashi ta saijō wo megutte 日吉七社本殿の構成 : 床下祭場をめぐって," Nihon kenchiku gakkai ronbun hōkokushū 日本建築学会論文報告集 317 (1982); Tokuda Kazuo 徳田和夫, Egatari to monogatari 絵語りと物語り (Tokyo: Heibonsha, 1990), 155-56. 251 Nakanoin Masatada no Musume 中院雅忠女, "Towazugatari とはずがたり," in Kenreimon'in ukyō no daibu shū. Towazugatari 建礼門院右京大夫集. とはずがたり, ed. Kubota Jun 久保田淳, Shinpen Nihon koten bungaku zenshū (Tokyo: Shō gakkan, 1999), 430-31; Karen Brazell, trans., The Confessions of Lady Nijo (Garden City, NY: Anchor Books, 1973), 185-86. Figure 4: Enoshima engi emaki 江島縁起絵巻, third scroll, privately owned. Image from Kanagawa Kenritsu Rekishi Hakubutsukan, ed., Kamigami to deau: Kanagawa no Shintō bijutsu 神々と出逢う：神奈川の神道美術, 99. 130 Additionally, comparisons with the 1213 Hikosan ruki 英彦山流記 reveals similarities between the medieval accounts of the caves and the modern-day terrain. 252 The text lists the forty-nine caves within the mountain range and describes the first eight in detail. Two of those eight mention pilgrims secluding themselves within the caves, and four of the eight (the fourth, fifth, sixth, and seventh) include corridors that connect with additional caves, presenting the caves as both unified and combinative grottos. The description of the sixth cave is as follows: Number six: Takasuka cave 鷹栖窟 is a jeweled palace of three ken. The original form [of the deity] is the Medicine King Buddha and the local trace form is the Tiger Celestial Child (Toratendō 虎天童). In front [of the Medicine King Buddha], there are the twelve divine generals, and a large bell of three shaku. As for the pilgrims who seclude themselves [here], they [perform] confessional rites for three hours without any idleness, and do not retrogress. ... Within the corridors, there are three caves. The second Takasuka is the Kimbila General. The third Takasuka is the eight divine children (hachi dai dōji 八大童子). 253 The cave is presented in more descriptive language here than the cases of Enoshima and Chikubushima, where the reader is walked through the subterranean terrain, but the details about the interior ritual space and the tunnel corridors are similar. The differences in rhetoric suggest the usage of the Hikosan ruki as a reference work. The commonalities support the notion that the cavernous landscapes were not a simply a construction made to explain connections between spaces but could be based in the topographical realities of certain sacred spaces. In fact, the topographical survey of Mt. Hiko also confirmed that Takasuka cave’s location between the three butte of Mt. Takanosu 鷹巣山 connects the three, and verified the existence of caves within 252 Yamamoto Yoshitaka 山本義孝, "'Hikosan ruki' ni shirusareta iwaya no sekai 『彦山流記』に記された窟の世界," Sangaku shūgen 29 (2002): 6, 11-12. 253 "Hikosan rūki 彦山流記," in Shugendō shiryōshū 修験道史料集, ed. Gorai Shigeru 五来重 (Tokyo: Meicho Shuppan, 1983), 468. 131 two of the three. 254 Even though the corridors themselves were not found, the presence of the caves makes the connections easy to add. Standardizing and Hierarchizing the Multiple Benzaiten The mapping of interconnected spaces through underground caves provided a network system capable of explaining relations between multiple local sites enshrining the same deity, but placing the deities within the same framework also led to new complications as monks began to compare the Benzaiten of different locales. 255 Doctrinally, all of the Benzaiten were the same, different, and the same but different, and this was not always satisfactory. Some texts added assertions of one Benzaiten being superior to the others, sometimes even situating the statement following an account of multiple Benzaiten as one in essence. Similar to the case with the origin narratives about Queen Consort Jingū (discussed in chapter 1), the inclusion of countless deities heightened the importance of being one of the select deities to be named. All the deities in Japan might participate in the protection of the realm, but it was the named deities who were remembered for it. In the case of Benzaiten, lists of Benzaiten began to appear in the early fourteenth century to name the most prominent Benzaiten and leave the rest to be forgotten. Today, Itsukushima is famous as one of the Three Benzaiten, along with Enoshima, and 254 Yamamoto, "'Hikosan ruki' ni shirusareta iwaya no sekai 『彦山流記』に記された窟の世界," 9-12. 255 Comparisons were not limited to sites enshrining the same deity but could be made through various associations shared between different sacred sites. In the Hikosan ruki, discussed in the previous section, it lists the first cave on the mountain, Hannya cave 般若窟 together with Kongōsan (within Mt. Katsuragi 葛城山, between modern day Osaka and Nara prefectures) and Chikubushima as “the three sacred spaces of Japan” (Nihon mika reichi 日本三ヶ霊地). While Mt. Hiko also enshrines Benzaiten, the shared association connecting the three sites seems to be as sites where En no Gyōja performed religious practices. As Suzuki Masataka argues, the positioning of sac red sites in relation to one another was not limited to places in the same geographic area. While far away from the capital, Mt. Hiko was involved in shared discourses about sacred space as powerful religious institutions near the capital. Suzuki Masataka 鈴木正崇, "Shugen reizan no engi ni kansuru kōsatsu: 'Hikosan ruki' wo yomu 修験霊山の縁起に関する考察 : 『彦山流記』を読む," Girei bunka gakkai kiyō 3 (2016): 55; "Hikosan rūki 彦山流記," 465. 132 Chikubushima. 256 In the medieval period though, lists of Benzaiten were not always uniform. Depending on the text, there were Benzaiten groupings of three, four, and six, and also five and eight by the early modern period (table 3). The frequent inclusion of Itsukushima, Enoshima, and Chikubushima helped to cement their place among the Benzaiten of Japan. The selection and standardization of these Benzaiten was not a complete coincidence; it developed through the work of scholarly monks elaborating the interconnections between Benzaiten. Returning to the first example from the Keiran shūyōshū on the three Benzaiten (Tenkawa; Itsukushima; Chikubushima), there is an irregularity that hints at an intentional positioning of the Benzaiten sites—Enoshima is replaced by Tenkawa. Because this passage is included within an account about Tenkawa, it is plausible that this grouping was intended to emphasize the importance of the Tenkawa Benzaiten. While this Keiran shūyōshū passage is the earliest known source (1318) using the phrase “three Benzaiten,” the grouping of Itsukushima, Chikubushima, and Enoshima appears in the contemporary 1317 Shōbōrinzō and the Hoki naiden ( 簠簋内伝, early Muromachi period). 257 Notably, the 1337 258 Kinpusen himitsu den 金峯 256 Mentions of the three Benzaiten and their identities can be found on the official websites for Itsukushima (Miyajima tourism website), Chikubushima, and Enoshima. Itsukushima: http://www.miyajima.or.jp/event/event_benzaiten.html; Chikubushima: http://www.chikubushima.jp/explore/hondou/ ; and Enoshima: http://enoshimajinja.or.jp/gosaijin/ . All accessed on July 24, 2018. 257 In the Shōbōrinzō, Itsukushima, Chikubushima, and Enoshima are given as the three sisters of the Dragon Princess from the Lotus Sutra, and only Enoshima is specifically described as Benzaiten. The Hoki naiden instead identifies the three deities as the three daughters of a fox king who flew from India to Japan. Bernard Faure and Minobe Shigekatsu both understand this passage as evidence of associations between fox / Dakini and Benzaiten deities. "Shōbōrinzō 聖法輪蔵," in Shinshū shiryō shūsei 眞宗史料集成, ed. Ishida Mitsuyuki 石田充之 and Chiba Jō ryū 千葉乘隆 (Kyoto: Dō hō sha, 1974), 488; "Hoki naiden 簠簋内伝," in Zoku gunsho ruijū. Zatsu bu 續群書類從. 雑部, ed. Hanawa Hokinoichi 塙保己一 and Oota Toshiro 太田藤四郎 (Tokyo: Zoku Gunsho Ruijū Kanseikai, 1984), 399; Faure, Gods of Medieval Japan: Volume 2, Protectors and Predators, 145-46; Shigekatsu Minobe, "The World View of "Genpei Jōsuiki"," Japanese Journal of Religious Studies 9, no. 2/3 (1982): 222-23. 258 While the Kinpusen himitsu den itself does not specify when it was completed, Abe Yasurō dates the work to 1337 based on comparisons with other writings by Monkan from 1337-1339. See Abe Yasurō 阿部泰郎, "Monkan chosaku seikyō no sai hakken: Sanzon gō gyōhō no tekisuto fu chi to sono isō 文観著作聖教の再発見－三尊合行法のテクスト布置とその位相," Hikaku jinbungaku nenpō 6 (2009): 122. 133 山秘密伝 names four Benzaiten, adding Tenkawa to Itsukushima, Chikubushima, and Enoshima. This further supports the idea of Tenkawa intentionally placing itself within the three Benzaiten within the Keiran shūyōshū. 259 The two sources together hint at the need for Tenkawa to add itself as one of representative Benzaiten of the realm, either as an addition or replacement for Enoshima. The grottos passing through to one another utilizes the concept of underground geographical connections, but alone, this system highlighted the connections between sites more than their differences. Therefore, the Tenkawa passage from the Keiran shūyōshū adds other interpretative frameworks to maintain the unity of the Benzaiten sites while also elevating Tenkawa above the others. First, the Keiran shūyōshū gives each of the Benzaiten names to differentiate their original forms: “Yoshino Tenkawa is Jizō Benzaiten 地蔵弁才天. [It is] the number one Benzaiten of Japan. Second, Itsukushima is Myōon Benzaiten 妙音弁才天. Third, Chikubushima is Kannon Benzaiten 観音弁才天.” 260 Myōon Benzai ten for Itsukushima suggests Itsukushima’s common iconography as a goddess of music playing the biwa lute, and Kannon Benzaiten reflects the understanding of Chikubushima Benzaiten as Nyoirin Kannon (Wish fulfilling jewel Kannon, 如意輪観音). 261 In particular, the explicit mention of Tenkawa as “the number one Benzaiten of Japan” places Tenkawa above the other two. 259 Monkan 文観, "Kinpusen himitsu den 金峰山秘密伝," in Nihon daizōkyō 日本大蔵経, ed. Suzuki Gakujutsu Zaidan (Tokyo: Suzuki gakujutsu zaidan, 1976), 9. 260 T2410, 76:625a20-625a23 261 Within a section on the origins of the prospering [of Chikubushima] (kōgyō engi no koto 興行縁起事) within the Benzaiten origin narrative chapter, the Keiran shūyōshū connects the Chikubushima Benzaiten with the Ishiyama Nyoirin Kannon. “It is said, ‘Chikubushima is the one who went out from the great sea and came to Vulture’s [Peak]. Ishiyama Nyoirin Kannon is the one who appeared at the Southern Pure Land and attained enlightenment. Benzaiten is a provisional manifestation of Nyoirin [Kannon].” Here, the Southern Pure Land (nanbō muku 南方無垢) refers to the place where the Dragon Princess attained enlightenment in the Lotus Sutra, thus connecting Ishiyama Nyoirin Kannon as the equivalent of the Dragon Princess with Chikubushima Benzaiten as dragon deities coming from the sea and arriving at the sacred spaces of the Lotus Sutra. For the Keiran shūyōshū, see T2410, 76:625c13-625c-15. For the Lotus Sutra passage, see T262, 9:35c16-35c18, The Lotus Sutra, trans. Burton Watson 134 (New York: Columbia University Press, 1993), 188. Nakajimayaka also uses this passage from the Keiran shūyōshū to demonstrate how Benzaiten could be understood either as an original form for kami deities or as a provisional manifestation (i.e., Nyoirin Kannon). Nakajima, "Chūsei Benzaiten shinkō to zuzō ni kansur u kenkyū: Benzaiten mandara no tenkai ni miru shinbutsu kan 中世弁才天信仰と図像に関する研究 : 弁才天曼荼羅の展開にみる神仏観," 55. 135 Table 3: Benzaiten Groupings Text Date Number of Benzaiten Benzaiten Connections Shōbōrin-zō 正法輪蔵 1317 Not specified Itsukushima, Chikubushima, Enoshima Keiran shūyōshū 渓嵐拾葉集 Pre 1318 3 Chikubushima, Tenkawa, Itsukushima 6 Chikubushima, Tenkawa, Itsukushima, Enoshima, Minoo, Sefuriyama Tenchi jingi shinchi yōki 天地神祇審鎮要記 ~1333 6 Chikubushima, Tenkawa, Itsukushima, Enoshima, Minoo, Kehi Kinpusen himitsu den 金峰山秘密 1337 4 Chikubushima, Tenkawa, Itsukushima, Enoshima Shugen shinaishō 修験指南抄 14th century 4 Only Tenkawa named Hoki naiden 簠簋内伝 Early Muromachi Not specified Itsukushima, Chikubushima, Enoshima Ugajin saimon 宇賀神祭文 1400 Not specified Itsukushima, Chikubushima, Mishima, Enoshima, Suwa, Sumiyoshi, Kashima, Kashii Nihon shoki kikigaki 日本書紀聞書 1419 4 Tenkawa (= Niu), Itsukushima (= Kōya), Chikubushima (= Tsubakimiya), Enoshima (= Wakamiya) *4 Benzaiten are being associated with the four deities of Mt. Kōya ( Kōya shisho myōjin 高野四所明神) Sangoku denki 三国伝記 Between 1407-1446 Not specified Chikubushima, Itsukushima, Enoshima, Minoo Jinkeishō 塵荊鈔 1482 4 Chikubushima, Tenkawa, Itsukushima, Kehi 7 Above 4 Benzaiten plus Enoshima, Minoo, and Kashima Benzaiten nyo koshiki 弁財天女講式 pre 1512 Not specified Chikubushima, Tenkawa, Enoshima Tenkawa Benzaiten engi 天河弁財天縁起 Muromachi Period 4 Chikubushima, Itsukushima, Enoshima, Minoo 8 Above 4 Benzaiten plus Kengosan 剣御山, Yanagisaka 柳坂, Kamiyamadera 神山寺, Yoshino Kumano 吉野熊野 Wakan sanzai zue 和漢三才図会 1712 5 Chikubushima, Itsukushima, Enoshima, Kinkazan, Mt. Fuji Hokuhisenshi 北肥戦誌 1720 6 Only Sefuriyama named 136 Figure 5: Map of Sacred Sites of Benzaiten from Table 3. (Size of points relative to number of times the site appears in the different texts) The Tenkawa source identifying Tenkawa as Jizō Ben zaiten does not provide an explanation itself, but related ideas can be found within sections of the secret teachings of Benzaiten section of the Keiran shūyōshū. From an iconographical framework, the Golden Light Sutra and Mahāvairocana Sutra (Dainichi kyō 大日経) provide iconographical forms for the two-armed Myōon Benzaiten (e.g., Itsu kushima) and the eight-armed Benzaiten (e.g., Chikubushima) respectively. 262 However, this leaves out the combinatory form of Benzaiten as Uga Benzaiten 宇賀弁才天, which was the established form of Tenkawa Benzaiten. Because Uga Benzaiten also played an important role at Mt. Hiei, Kōshū also emphasizes Uga Benzaiten’s importance, presenting the dichotomy of two types of Benzaiten as Myō on Benzaiten and Uga Benzaiten. Within this system, the Keiran shūyōshū identifies Uga Benzaiten 262 Inose Chihiro gives a detailed comparison of the two sutras as iconographical bases for different images of Benzaiten, and also shows how iconographical features of the originally distinct forms came to merge together to create a new combinatory iconography as well. Inose Chihiro 猪瀬千尋, Chūsei ōken no ongaku to girei 中世王権の音楽と儀礼 (Tokyo: Kasama Shoin, 2018), 210-19. 137 as a transformation of Kannon, allowing Uga Benzaiten to merge with Kannon Benzaiten (Chikubushima / 8-armed Benzaiten). 263 Later, though, the explanation moves to a three-part logic. Beginning with an explanation for an identification of the bodhisattvas Kokūzō 虚空蔵 and Jizō 地蔵 as the original forms of Benzaiten, the Keiran shūyōshū then positions Uga Benzaiten as the nondual body overlaying Kokūzō , Jizō , and human beings. Working from a literal translation of the names Kokūzō , storehouse of the empty skies, and Jizō , storehouse of the earth, the text then situates Uga Benzaiten as both heaven and earth through a reading of the “U” ( 宇) as heaven and the “ga” ( 賀) as earth. 264 Thus, Uga Benzaiten manifests in the world of humans, the world between heaven and earth (based on a deconstruction of the word for “human,” ningen 人間 as a combination of “person” 人 and betweenness 間). The Keiran shūyōshū here employs the logic of the Tendai three truths, which works by presenting a dichotomy, and then revealing the non-dual unity of the seemingly opposite elements. 265 From this perspective, it is possible to interpret the name Jizō Benzaiten as possibly referring to Uga Benzaiten and Jizō as the earthly form of Benzaiten which bri ngs worldly benefits to sentient beings. Or, Jizō Benza iten could be meant as the third nondual form which is added to the preexisting two forms of the two-armed and eight-armed Benzaiten. Presented within the triad of the three Benzaiten, the nondual interpretation corresponds more closely, but the lack of direct 263 The Keiran shūyōshū also identifies Uga Benzaiten’s seed character (shuji 種字) as that of the wish fulfilling jewel, which provides another layer of association with Kannon. T2410:76, 620b12-16. 264 The standard meanings of the characters “U” 宇 and “ga” 賀 do not relate to heaven and earth, but Kō shū expresses the authority of the explanation by citing it as a secret teaching (hiketsu 秘訣). T2410:76, 620a27-a28. 265 This section of the Keiran shūyōshū contains multiple examples in which connections with Benzaiten are explained through the threefold logic of the three truths. To give one example, in order to explain the Dragon Princess as a manifestation of Benzaiten, the Keiran shūyōshū fits three forms of the Benzaiten with the three bodies of the Buddha: Nyoirin Kannon is given as the reward body (hōshin 報身), the Dragon Princess from the Lotus Sutra is given as the response (transformation) body (ōshin 應身), and the Lotus Sutra is given as the Dharma body (hōshin 法身). This structure allows for the Dharma body to be stressed as the nondual form, the “gate of both original form and provisional manifestation,” and the “unity of object and cognition.” T2410:76, 622b15-17. 138 explication is part of what provides the Benzaiten names with power. Like this, Kō shū begins from references to the Buddhist scriptures, fills in gaps with secret transmissions, and then employs hidden meanings within the names of the deities to reveal connections between the deities. 266 This play between meanings provides the Keiran shuyōshū with the methodological tools needed to make sense of a multitude of narratives and teachings about the deities, rituals, and doctrine that often conflict and contradict one another. The concurrent sameness and difference within the three truths logic also appears in the passage through the depiction of the three Benzaiten as three interconnected wish fulfilling jewels. Although the wish fulfilling jewels can also be read as three separate jewels, the common iconographic form of the three jewels together could be seen as a single form representative of the three truths. 267 Through a strategy similar to the breakdown of the name Uga Benzaiten, the Keiran shūyōshū’s association of the three Benzaiten as the three wish fulfilling jewels works both metaphorically and linguistically. Matching the numerical counter for three wish fulfilling 266 This interpretation of additional meanings through the breaking down of the individual characters in a name resembles the practice of etymological allegoresis, in which a single character is broken down and then rearranged to reveal another meaning. Susan Blakely Klein details the use of this mode of analysis within commentaries on the Tales of Ise (Ise monogatari 伊勢物語), and explains that the practice developed through the practices of Tendai monks explaining the relationships between the deities of Enryakuji on Mt. Hiei and those of Hie Shrine at the foot of the mountain. Susan B. Klein, "Wild Words and Syncretic Deities: Kyō gen kigo and honji suijaku in Medieval Literary Allegoresis," in Buddhas and Kami in Japan: Honji Suijaku as a Combinatory Paradigm, ed. Mark Teeuwen and Fabio Rambelli (New York: RoutledgeCurzon, 2003), 178-79. 267 The Reikiki 麗気記 for example, uses the simile “like a chicken and egg” (tori no ko no gotoku 如鶏子) to describe the nonduality of the three primordial deities manifesting within the three wish fulfilling jewels. While the phrase, “like a chicken and egg” was used earlier in the Nihon shoki (720) to describe the three primordial deities, the comparison with the three wish fulfilling jewels aligns the comparison with the medieval Buddhist doctrine of the three truths. Similarly, the 1320 Matsuhashi injin kuketsu 松橋流印信口決 describes the combination of the three realms (of the Diamond realm, the Womb realm, and the nondual form of the two mandalas, the Realization realm (Skt. susiddhi, Jp. soshitsuji 蘇悉地) and the various deities as appearing as the three wish fulfilling jewels. On the Reikiki see "Reikiki 麗気記," in Shintō taikei. ronsetsu hen, Shingon shintō 神道大系. 論說編, 真言神道, ed. Wada Shūjō 和多秀乗 (Tokyo: Shintō Taikei Hensankai, 1993), 7; Suzuki Hideyuki 鈴木英之, Chūsei gakusō to Shintō: Ryōyo Shōgei no gakumon to shisō 中世学僧と神道 : 了誉聖冏の学問と思想 (Tokyo: Benseishuppan, 2012), 198-200. For the Nihon shoki passage see Kojima Noriyuki 小島憲之, ed. Nihon shoki 日本書紀, 3 vols., Shinpen Nihon koten bungaku zenshū (Tokyo: Shōgakkan, 1994), 19, vol. 1. For a transcription and explanation of the passage from the Matsuhashi injin kuketsu, see Itō Satoshi 伊縢聡, Chūsei Tenshō Daijin shinkō no kenkyū 中世天照大神信仰の研究 (Kyoto: Hō zō kan, 2011), 251-53. 139 jewels as “san ben” 三弁 with “san ben” as an abbreviation of three Benzaiten (san Benzaiten 三弁才天), the Keiran shūyōshū overlays multiple modes of meaning to simultaneously present the distinct, unified, and non-dual forms of the deities and their sacred spaces. The Kinpusen himitsu den 金峯山秘密伝, by the Shingon monk Monkan ( 文観, 1278- 1357), employs similar strategies to present Tenkawa as the foremost Benzaiten in a list of four Benzaiten. 268 Again, the text distinguishes each of the deities by naming them: Tenkawa is the Benzaiten of the discernment of the [Buddhist] teachings ( 法弁才); Chikubushima is the Benzaiten of the discernment of meaning ( 義弁才); Enoshima is the Benzaiten of the discernment of speech ( 詞弁才); and Itsukushima is the Benzaiten of the discernment of eloquence ( 弁弁才). These names also employ a logic of linguistic play that connects the four Benzai[ten] (shi Benzai 四弁才) with the set of the four [abilities] of discernment (shiben 四弁), the rhetorical abilities that perfectly express the Buddhist teachings. Immediately following this grouping of the Benzaiten as a single set, the text proceeds to discuss the spaces of the sites together, before ending with Tenkawa as superior. Following the structure of the Tendai three truths, it presents the Benzaiten as both the same and different, and distinguishes Tenkawa as the nondual Benzaiten that is both the same and different. The Kinpusen himitsu den states, “These are superlative spaces, unmatched within our realm, the sacred spaces of the rapid attainment [of supranormal powers]. Now, among these [four Benzaiten], Tenkawa Benzaiten is namely the earliest, the original source. [Tenkawa] is the nondual, wondrous body of Kinpusen and Kumano. This is an unconceivably superlative 268 As a compilation of knowledge about the sacred spaces on Mt. Kinpusen, the work is divided into two parts, first bringing together and explaining various teachings about the deities and the space, and then in the second part, applying the information in the first part within an explication of the rituals and religious practices performed on the mountain. 140 space.” 269 On the level of the translocal, the passage brings the representative Benzaiten sites of the realm together to make them comparable, and from there elevates Tenkawa over the other sites. On a more regional level, Tenkawa similarly situates itself on top through an emphasis of its physical location between the mountains of Kumano and Kinpusen. The river at Tenkawa flowing down to Kumano and Kinpusen, acting as both a physical connection between the sites and a symbolic marker of the nondual relationship of the sites. In this manner, doctrinal play over the three truths combines with temporal (Tenkawa as the earliest Benzaiten) and spatial (Tenkawa between Kumano and Kinpusen) positionings to present the shared essence of all Benzaiten sites as the starting place from which to find differences. As short entries within larger compiled works, it is hard to ascertain the motivations for the assertions of Tenkawa as the preeminent Benzaiten, but a few possibilities can be mentioned. Looking at the historical context in which the Kinpusen himitsuden was written, Kawasaki Tsuyoshi situates the importance of the rituals within the text as militaristic subjugation rites to be used by Go-Daigo Tennō ( 後醍醐天皇, 1288-1339) against the Northern Court. This role is clear for such rituals within the Kinpusen himitsuden as the rite of Bishamon who prevails over enemies (shōteki Bishiman hō 勝敵毘沙門法) and the General Jizō rite ( shōgun Jizō hō 将軍地蔵法), and the content of the Tenkawa Benzaiten rite (Tenkawa Benzaiten kudokuten hō 天川弁才天功徳天法), which explicitly presents the power of the rite for military use. 270 For example, the ritual includes the following description of Benzaiten within the contemplation of the ritual space (dōjōkan 道場観), “Raining down with [her] sword, wheel, and bow and arrow, 269 Monkan, "Kinpusen himitsu den 金峰山秘密伝," 9. 270 Kawasaki Tsuyoshi 川崎剛志, Shugendō no Muromachi bunka 修験道の室町文化 (Tokyo: Iwata Shoin, 2011), 242. 141 [Benzaiten] destroys the sworn enemies of the realm and subjugates the obstacles of sentient beings.” 271 From this perspective, it would make sense for Tenkawa to promote the efficacy of the Benzaiten rite by presenting the superiority of Tenkawa over other Benzaiten deities, and over the deities of the other sacred spaces in the Yoshino area (i.e., Kumano and Kinpusen). Fittingly, those are the messages delivered in the above passage on the four Benzaiten. Thinking broadly, the assertion of Tenkawa’s superiority can also be seen as a general response to the growing awareness of the various regional gods and of Japan as a unified realm. Individual temples and shrines were pushed in opposite directions, to proclaim the power of one’s own institution above that of other temples and shrines, but also to present one’s sacred locale as one of many sites working together to protect the realm. While the combinative logic of honji-suijaku assisted with the idea of Japan as a unified “realm of the gods” (shinkoku 神国), it also made it harder to prove the importance of any one sacred space. These opposing viewpoints come together within the Sannō chapter ( Sannō no koto 山王事) of the Yōtenki (“Record of Heavenly Radiance” 耀天記). 272 It begins by tracing the origins of the kami deities from the myths of the Nihon shoki, then explains the proliferation of deities protecting the realm: A long time ago, after this country had been ruled by the Seven Generations of Heavenly kami and the Five Generations of Earthly kami, the august solar kami Amaterasu manifested herself in the two shrines at Ise. Uniting these two shrines she granted her 271 A later 1551 Ten-armed (Tenkawa) Benzaiten rite (Jippi Benzaiten hō 十臂弁財天法) manuscript at Mt. Kōya copies the contemplation of the ritual space section from the Kinpusen himitsu den word for word, except it removes the quoted sentence on subjugating enemies. This suggests that the rite was used later but for specifically non militaristic purposes. It should be also noted that the Tenkawa Benzaiten rite also copied elsewhere, and the undated Zentūji manuscript copies the Kinpusen himitsu den word for word. The Union Catalogue of Early Japanese Books (Nihon kotenseki sōgō mokuroku 日本古典籍総合目録) lists six other manuscripts with two of them datable to the Muromachi period. Monkan, "Kinpusen himitsu den 金峰山秘密伝," 32; Toritani Takefumi 鳥谷武史, "Kōyasan Kongō sanmai'in zō 'Jippi Benzaiten hō' no honkoku to shōkai 高野山金剛三昧院蔵『十臂弁財天法』の翻刻と紹介," Ningen shakai kankyō kenkyū 33 (2017): 54; "Benzaiten hō 弁財天法," (Zentsūji: National Institute of Japanese Literature), 16. 272 While the Yōtenki includes a colophon dating the text to 1223, it is likely that the text was not compiled all at the same time, and the language of the Sannō chapter suggests that it might h ave been added later (possibly in the late Kamakura to early Muromachi period). 142 protection to the hundred rulers and to all people of this land. Then, establishing the kami of Kamo, Kasuga, Matsuo, and Sumiyoshi, she let them take possession of the land at the four corners and protect the residence of the ruler, thus benefiting everyone. As a consequence, some have become accustomed to think and say that “Japan is a land of kami.” 273 So far, the argument makes perfect sense, but the text quickly shifts to arguing for the primary importance of Mt. Hiei as the site where Śākyamuni Buddha manifests himself and protects the Tendai teachings. All the kami deities should be understood as coming from the expedient means of Śākyamuni Buddha. Deliberating over this matter, I, Śākyamuni, ma nifested myself within the realm of Japan as a kami in the form of the Hie Sannō [Deity]. I save sentient beings in both this life and the next, and also, I [came] to protect the foremost, great Buddhist teachings called the perfect school (Tendai). 274 As a space of the local traces of the transcendental Buddhas, Hie Shrine can be seen as the same as any of the other spaces of kami deities protecting the realm, but the key role given to Śākyamuni provides a means of positioning Hie above the rest. In the case of Benzaiten, the desire to describe Tenkawa above the others can be hypothesized as originating from a similar line of thought. Networked Spaces for Mapping Relations in Space The ability to bring together deities from throughout the archipelago required the collection of knowledge about the deities, but then, the symbolic representation of the totality led to unequal representations of the acquired information. The enshrinement of deities from across Japan existed as comprehensive shrines (sōsha 総社) and within rituals summoning the various 273 Translated in Allan G. Grapard, "Linguistic Cubism: A Singularity of Pluralism in the Sannō Cult," Japanese Journal of Religious Studies 14, no. 2/3 (1987): 216; Yōtenki 耀天記. , vol. 2, Zoku Gunsho ruijū. Jingi bu (ge) 続群書類従. 神祇部（下）. (Tokyo: Yagi Shoten, 2013, 1935), 602. 274 Yōtenki 耀天記, 602. 143 deities to protect the realm from at least the tenth century, but later ledgers of the names of the summoned gods (jinmyōcho 神名帳) began to list the names of specific gods rather than simply mentioning the various gods. 275 The jinmyōchō ledgers could contribute to a worldview of a unified realm of Japan, but in the end, the gods selected to constitute this worldview made up only a small percentage of the totality. Therefore, the development of networked connections provided another means for connecting sacred spaces as a unified space of Japan as the realm of the gods, but the relations between the multitude of sacred spaces was not as equivalent as one might expect. The mapping of Benzaiten spaces, or other sacred spaces, could simultaneously present a unified Japan, and certain Benzaiten as better than the multitudes. Maps provide an easy way of representing the structural unity of spatial units, but they can also hide the imperfect selectivity or unequal relations that can occur within space as relational networks. In order to understand medieval Japanese practices of mapping space, scholars also need to move away from cartographic frameworks. Previously, Allan Grapard designated three levels of sacred space, moving from (1) a single “sacred site” where a deity is enshrined, (2) to a “sacred area”, and finally expanding out (3) to a “sacred nation.” 276 These stages of sacred space broaden the roles and relations between sacred spaces, but moving centrifugally outwards from the sacred site, they still rest on the ground of geographically connected places. Grapard’s categorization of sacred space helps to see how sacred space can extend beyond the walls of a temple or shrine, but it remains restricted by the needs of the structural system for geographic adjacent positionings, and the lessening of regional particularities as the map zooms out from the regional to the national. With networked spaces, 275 Uejima Susumu 上島享, Nihon chūsei shakai no keisei to ōken 日本中世社会の形成と王権 (Nagoya: Nagoya Daigaku Shuppankai, 2010), 387-92. 276 Grapard, "Flying Mountains and Walkers of Emptiness: Toward a Definition of Sacred Space in Japanese Religions," 196. 144 the connections might come together to outline a unified sacred realm, but that was not always the case and national unity still might not have been a primary concern. In the case of the Bunkidan 文机談 (likely written between 1272-1283), a compiled text on musical treatises, it connects the space of Benzaiten with the space of Japan, but the linkage is made primarily to praise the biwa lute as an instrument of Buddhism for saving sentient beings. The passage first makes the association between the four bridges and four strings of the biwa lute with a circle encompassing the entire realm. This is then connected with visualization practices based on the transformation of the Sanskrit character “sa” ( 将) into the biwa, and then into Benzaiten. The Bunkidan then takes this even further by having Benzaiten take the form of an island (as the local place of enshrinement) and then from there, the island of Japan. The passage concludes, “Is [Benzaiten] not she who can be relied on to bring salvation to everyone? [Benzaiten] makes the vow, ‘For the sake of all sentient beings who worship and hold me, it shall be the case that they will overflow with treasures and attain benefits.’ It is impossible to explain each and every case of [Benzaiten’s] sympathetic response [to save sentient beings].” 277 More than anything else, the combination of the biwa lute with spatial and Buddhist logic works to present the biwa as bringing together heaven and earth, the worldly and transcendental, as a symbol of Benzaiten’s power to save sentient beings. The Keiran shūyōshū also uses the biwa to make spatial connections, but in contrast to the case in the Bunkidan, maps a localized understanding of Lake Biwa itself as being in the form of a biwa. The lake has the shape of a biwa. Thus, Chikubushima is the string support. Kojima, called the island of the ten raksasa girls, is the plectrum. On the island, there is a palace; [it is] the “hidden moon” sound hole. Shiraishi and Chikubushima are the “half-moon” 277 Ryūen 隆円 and Iwasa Miyoko 岩佐美代子, ed., Bunkidan zenchūshaku 文机談全注釈 (Tokyo: Kasama Shoin, 2007), 33. 145 sound holes. Hochishima is the “distant mountain.” Seta is the “deer” neck. From Uji up until the sea is “shrimp tail” head [of the biwa]. At the very bottom of the lake are four streams. In other words, these are the four strings. It is a symbol of the innate, true nature of things, and the abode of the living body of Benzaiten. 278 The topographic form of the sites on the lake as a biwa provides visible evidence of the place as the sacred site of Benzaiten. Unlike the previous passages which presented underground tunnels and caves as the hidden, invisible dwelling places of Benzaiten, here we have a mapping of the sacred spaces from a bird’s eye view. In this way, the connections are visible, and understandable, as long as you know about the spaces and how to connect them together. Moving our attention to Benzaiten networks mapping the space of Japan shifts the focus to the unity of the realm, but the importance of individual sites still does not leave the picture. The three Benzaiten and four Benzaiten lists in themselves can be seen as symbols of the Benzaiten of a unified realm of Japan, but a late medieval Mt. Fuji origin narrative, the Chōjō Dainichi Nyorai ryaku engi 頂上大日如来略縁記 makes the positioning of Benzaiten sites into an explicit symbol of their role in uniting the realm. 279 Working from an episode in the Kojiki 古事記 of the creation of Japan on top of five divine stakes, the origin narrative locates those five stakes as Itsukushima (in the west), Kinkasan (east), Chikubushima (north), Enoshima (south), and Mt. Fuji (center). 280 The text identifies the deities as manifestations of Amaterasu rather than as Benzaiten to provide a closer connection with the founding of Japan, but the five listed deities matches the five Benzaiten listed in the 1712 Wakan sanzai zue 和漢三才図会. 281 While the Mt. 278 T2410:76, 625c17-c22. 279 The text itself is not dated, but a colophon naming the abbot at Fuji as Sanbō and locating h is position at Murayama provides enough details for Takeya Yukie to date the text to the late Muromachi. A letter by Sanbō in 1602 discusses his post on the mountain itself, showing that he had moved from Murayama by 1602 at the latest. Takeya Yukie 竹谷靱負, Fujisan no saijinron 富士山の祭神論 (Tokyo: Iwata Shoin, 2006), 40. 280 "Chōjō Da inichi Nyorai ryaku engi 頂上大日如来略縁記 ", in Ryakuengi shūsei 略緣起集成, ed. Nakano Takeshi 中野猛 (Tokyo: Benseisha, 1997), 267. 281 Quoted in Nakano, "Ichijō shūgyokushō shoin waka kō 『一乗拾玉抄』所引和歌攷," 178. 146 Fuji origin narrative prioritizes the identification with Amaterasu over Benzaiten, an early modern version of the Enoshima origin narrative includes the same discussion of the five divine stakes, but then immediately follows it with the deities calling themselves Benzaiten. 282 For Enoshima, the identity of the deity as Benzaiten needed to be preserved above all else, but both origin narratives utilize the five divine stakes to present the five deities as the representative deities of Japan. As the four directions plus the center, they are able to support the land of Japan, and can thus become the representative sacred spaces of the realm. The Enoshima origin narrative explicitly makes the case saying, “Because the gods of heaven and earth here built the five stakes as the cornerstones of our realm, it is clear that they are superior to the all the other sacred spaces.” 283 While never fully explained within the text, the positioning of Mt. Fuji as the center can be read as a possible interpretation for positioning Mt. Fuji as the primary sacred space amongst the five stakes. More evidence would be needed for anything more than conjecture, but it is also worth considering the changes from the four Benzaiten (Itsukushima, Chikubushima, Enoshima, and Tenkawa) to the five Benzaiten (Itsukushima, Chikubushima, Enoshima, Kinkasan, Mt. Fuji) in relation to their spatial positionings. Making Kinkasan the eastern Benzaiten shifts the spatial relations so that Chikubushima becomes the north and Mt. Fuji becomes the center. The noting of the directions for each stake within the text shows a spatial awareness, but that is not enough to demonstrate any intentionality. Similar to the argument for Tenkawa’s importance between Kumano and Kinpusen, the center could easily be placed as a position of power. For example, an oral transmission within the monk Ryōhen’s lectures on the Nihon shoki (1419) works from the 282 Mukozaka Takuya 向坂卓也, "Sōshū Enoshima ue no miya engi ni tsuite: Honkoku to shōkai 相州得瑞嶋上之宮縁起について：翻刻と紹介," Kanazawa Bunko kenkyū 319 (2007): 37. 283 Ibid., 38. 147 same discourse of the creation stakes to assert the importance of Mt. Hiei. “Onokorojima 磤馭盧島 is a well known alternative name for this realm. [It is said], ‘On Mt. Hiei, [Onokorojima] is the mantra of the Medicine King Buddha of the Main Hall. That being the case, it is called the pillar at the center of the realm. This is the matter of our mountain.’” 284 There are not enough sources relating the Benzaiten networks with the realm of Japan to fully examine the ways in which local sacred spaces were brought together in relation to a unified realm of Japan. Instead, the variety of interpretations of relations between the Benzaiten sites suggests the need for a more flexible mapping of the sacred spaces located within Japan. The compilation of knowledge about regional sacred spaces and deities influenced understandings of a unified realm of Japan, but the connections were not as simple as one might expect. To give one final example from the Keiran shūyōshū, the Sannō chapter raises the question, “Why is it that various gods came and manifested themselves at Hie Shrine?” Kōshū answered the question by first listing the identities of the twenty-one shrines as kami deities of different regional sacred spaces and then explaining the significance of the gods together at Hie: As for the Sannō Deity, when spoken about in terms of totality, it is sixty -four deities, and when spoken about in terms of difference, it is the seven shrines. This is the protection of the teachings of the Perfect [Tendai] school by the manifestations of all the various great and small deities of our realm. Besides this, the fact that the various great and small deities within the realm of Japan individually and reciprocally constitute the essence and function of the various gods, this is like the platform of the Sannō Deity. 285 From a religio-political perspective, the logic of enshrining kami deities from throughout Japan together at Mt. Hiei makes perfect sense. Rather than relying on the deities of any one region, 284 Ryōhen 良遍, "Nihon shoki kan daiichi, daini kikigaki (Tawa bunko zō) 日本書紀巻第一・第二聞書( 多和文庫蔵)," in Sonarejō: Murasame hen 磯馴帖：村雨篇, ed. Itō Masayoshi 伊藤正義, Koten kenkyū shiryō shū (Osaka: Izumi Shoin, 2002), 344. 285 T2410, 76:519c20, 520a1-520a5 148 gods from throughout Japan could be worshipped together at Mt. Hiei. 286 However, within a text praising the Sannō Deity itself, the u nderstanding of Sannō as an amalgamation of other regional deities can quickly become confusing. To solve the problem, Kōshū again uses the rhetoric of both same and different. Here he utilizes the language of Buddhist scriptures concerning the relationship between fundamental truth and manifest activity to explain how the deities of Japan can both be unified as one, and function as distinct gods serving different purposes. Therefore, the Sannō Deity, as the exemplary manifestation of the unification and di stinction of the various gods of Japan, is interpreted as a sign of Tendai as the perfection of the Buddhist teachings, and the role of all the gods in protecting the Tendai teachings. From this perspective, the multiplicity of the local gods at Hie is not a problem, but rather a means for Hie to transcend beyond its own space to become symbolic of all of Japan, and consequentially, to be posited as the center of the religious world of Japan. The message is simple: Japan is unified as a single, “realm of the gods,” and while each deity is essential to that unity, the Sannō Deity and Mt. H iei are the preeminent god/space of Japan. Fluid Symbols and the Blurred Boundaries of Identity Lastly, I will give one more discussion of the ambiguity in the identification of Benzaiten deities enshrined at particular sites. The previous sections showed how interrelations enabled different explanations of sameness and/or difference, and this section demonstrates the power of blurred identities to allow for multiple and changing interpretations. The hanging scroll of the Four Deities of the Kōyasan Temple Complex ( Kōya shisho myōjin zō 高野四所明神像) depicts 286 This system resembles the practices at comprehensive shrines (sōsha 総社), and the roles of kokubunji temples and ichinomiya shrines to bring together religious institutions throughout the realm into a centralized system for protecting the realm. 149 the four deities enshrined within Amano Shrine at the foot of Mt. Kōya. Scholars had long accepted the standard identification of the four deities as the (1) Mt. Kōya Deity; (2) Niu Deity; (3) Kehi Deity; and (4) Itsukushima Deity, until Sugano Fumi raised doubts about the historical origins of the third and fourth deities as Kehi and Itsukushima (fig. 6). The earliest source labeling the four deities is the Spring and Autumn Annals of Mt. Kōya (Kōya shunju 高野春秋), an early modern compendium covering the history of Mt. Kōya from 816 -1718. 287 An origin narrative in the compendium recounts how the holy man Gyōshō (Gyōshō Shōnin 行勝上人, 1130-1217) enshrined the Kehi and Itsukushima Deities in 1208, but that dates the text to over five hundred years later. Moreover, the year 1208 seems overly early. The earliest historical record mentioning the four deities is from 1234, and moreover, the military rhetoric within the oracle given to Gyōshō resembles that of texts written by Amano Shrine after the 1274 attack by the Mongols. Additionally, the 1839 Kii Province gazetteer Kii zoku fudoki 紀伊続風土記 calls the identification of Kehi and Itsukushima as the third and four deities an “alternative explanation” (isetsu 異説), and instead identifies the third deity as the Aridōshi Deity 蟻通明神. While this record in itself is not any more authoritative than the Spring and Autumn Annals of Mt. Kōya, the naming of the third deity as Aridōshi corresponds with a 1293 document written by Amano Shrine in response to the 1281 Mongol attack. 288 287 The compilation of the Spring and Autumn Annals of Mt. Kōya was completed in 1719 by the monk Kaikei 懐英 (1642-1727). 288 Sugano Fumi 菅野扶美, "'Gyō shō Shōnin' no katararekata t o Amano sha yonjo myōjin 「行勝上人」の語られ方と天野社四所明神," Junreiki kenkyū 7 (2010): 24-30. 150 Figure 6: Kōya shisho myōjin zō 高野四所明神像, Metropolitan Museum of Art, Late 14th century. As is often the case in images of Benzaiten, the depiction shows the fourth deity to be a two-armed Benzaiten holding a biwa lute, and while this fits the iconography of the Itsukushima Benzaiten, others were depicted in the same manner. Sugano Fumi provides compelling Figure 7: Shisho myōjin zō 四社明神像, Saizen’in, 15th century. Image from Ō saka Shiritsu Bijutsukan, ed., Yama no shinbutsu: Yoshino, Kumano, Kōya: tokubetsuten 山の神仏：吉野・熊野・高野 : 特別展, 122. 151 arguments for questioning the historicity of the account, but she leaves one important question unanswered—do the identities of the deities as Kehi and Itsukushima need to be fixed? Like the case of the networked Benzaiten sites, we need to appreciate the power of blurred identities. The Benzaiten enshrined at Amano Shrine can be the same as the Itsukushima Deity and different. Even the iconography of Benzaiten within the hanging scroll of the four deities was not fixed. The fifteenth century version of the scroll at Saizen’in 西禅院 (fig. 7) puts the four deities in military garb, and the two-armed Benzaiten is replaced with the realm protecting eight-armed Benzaiten from the Golden Light Sutra. Likely not coincidentally, the origins of the enshrinement of Itsukushima and Kehi as the third and fourth gods in the Spring and Autumn Annals of Mt. Kōya shares a militaristic coloring: It was not a dream and it was not an illusion. A god dressed in a robe and a cap, and a beautiful lady of the court [suddenly] appeared like this. [The deity] announced, “[Gyōshō] Shōnin, you worship me, and I was pleased. Truly [your reverence] is attentive and without remiss. [As for] the Kehi Deity in Echizen, and the Itsukushima Deity in Aki, they have been my friends since long ago. Now, even though they are each separated, enshrined in two [different] provinces, [I] do not forget our old comradery. [I] seek for it to be as it was previously, [for us] to live together in one place, holding and protecting the esoteric teachings. During times of the subjugation of foreign realms, [they] were the commanders assisting [from my] left and right. Shōnin, that which I command is the enshrinement of the Niu [Deity]. Enshrine these [deities] at a single site, in a line at the Amano Hanazono Shrine. And then, they should be worshipped and given offerings [together] as the four deities. Having spoken, the deity returned to its divine abode. 289 Rather than providing a historical or religious explanation for the enshrinement of the deities together, the passage appeals to the god’s humanity. Any reader could sympathize with the deity’s desire to be reconnected with her old friends. Still, we should not overlook the brief mention of the shared past of the deities in protecting the realm from foreign enemies. The 289 Kaiei 懷英 , Shinkō Kōya shunjū hennen shūroku 新挍高野春秋編年輯錄, ed. Hinonishi Shinjō 日野西眞定 (Tokyo: Meicho Shuppan, 1982), 133. 152 language closely resembles the rhetoric in post Mongol invasion origin narratives about deities who assisted Queen Consort Jingū against the Korean kingdoms. While it cannot be confirmed, it is easy to imagine this explanation of the four deities of Mt. Kōya devel oping around the time of the Mongol invasions to increase the image of Kōya and Amano as sites for the protection of the realm. Fluid identities of Benzaiten deities can also be found in historical records. One case of the enshrinement of the Itsukushima Deity at Iwashimizu Hachiman Shrine further shows how a Benzaiten icon could become Itsukushima. A fourteenth-century map of Iwashimizu from the Kanazawa Bunko archive (Hachimangū zu narabi sharei 八幡宮図并社例) labels one deity as the Shidara Deity 志多羅, a god of disease popular during the Heian period (794-1185), and then in smaller text, also identifies the deity as Itsukushima (Fig. 8). 290 This contrasts with both the earlier Miyadera enji shō 宮寺縁事抄, an encyclopedic compilation of documents on Iwashimizu Hachiman Shrine by abbot Tanaka Dō sei ( 田中道清, 1169-1206), which only identifies the deity as Shidara, 291 and later records of the movement of deities in 1422 to repair the shrines, which only identify the god as Itsukushima. 292 Notably the descriptions of the enshrined Benzaiten icon match in both cases, showing that it is the same deity, but was simply identified differently. 290 Sasaki Yasuyuki 佐々木康之, ed. Mizu: Shinpi no katachi = Prayers to water 水 : 神秘のかたち (Tokyo: Santorī Bijutsukan, 2015), 97, 181-82. 291 "Miyadera enji shō 宮寺縁事抄," in Shintō taikei. Jinja hen, Iwashimizu 神道大系. 神社編：石清水, ed. Murata Masashi 村田正志 (Tokyo: Shintō Taikei Hensankai, 1988), 27. 292 Iwashimizu Hachimangū 石清水八幡宮, ed. Iwashimizu Hachimangū shiryō sōsho 石清水八幡宮史料叢書, 5 vols., vol. 5 (Hachiman-cho: Iwashimizu Hachimangū Shamusho, 1975), 470-71. 153 Figure 8: Hachimangu zu narabi ni sha rei 八幡宮図并社例, Kanazawa Bunko, 14 th Century. Circle added by author to designate the area enlarged below. Image from Sasaki Yasuyuki 佐々木康之, ed., Mizu: Shinpi no katachi = Prayers to water 水 : 神秘のかたち, 97. While the historical records demonstrate that Itsukushima was enshrined at Iwashimizu Hachiman Shrine, the original reasons for the identification remain unclear. The only text to explain the enshrinement of Itsukushima at Iwashimizu is the 1848 Otokoyama kō koroku 男山 154 考古録 compiled by the Iwashimizu Hachiman Shrine worker Fujiwara Naotsugu ( 藤原尚次, 1797-1878). As a comprehensive encyclopedia on Iwashimizu Hachiman Shrine, the work details each of the deities enshrined at the site. In successive sections, the text explains the deity as Shidara, Itsukushima, and Benzaiten, suggesting that it was not that the deity’s identity had been changed but increased. 293 Working from the form of the icon holding a wish fulfilling jewel, Naotsugu connects the Itsukushima Deity with the episode of Queen Consort Jingū retrieving two jewels from the Dragon Palace under the sea and using them to subjugate the Korean kingdoms. This association was further strengthened by the understanding of Itsukushima as a dragon deity, which Naotsugu connected with the white dragon on the head of the icon. 294 As discussed in chapter one, discourses developed connecting Itsukushima, Queen Consort Jingū, and the Dragon Princess from the Lotus Sutra developed after the Mongol invasions, and we see here that the association between Itsukushima and Jingū was not completely forgotten even five hundred years after the production of the post-Mongol invasion origin narratives in the early fourteenth century. Therefore, even if Naotsugu’s explanation cannot be validated with earlier sources, it provides a logical connection between Itsukushima and Iwashimizu and the association between Shidara and Itsukushima. This connection fills in 293 Fujiwara Naotsugu references a 1724 encyclopedia on Iwashimizu, the Iwashimizu jingenshō 石清水尋源鈔 as his source for the earliest mention of Itsukushima at Iwashimizu in a 1339 report (chūshin 注進). The same text from the Iwashimizu map identifying Shidara as Itsukushima can be found almost word-for-word (this text adds Bishamonten as annother honji-suijaku association with Shidara/Itsukushima) in the Iwashimizu Hachimangū masshaki ( 岩清水末社記, dated approximately to the Nanbokuchō (1336 -1392) or early Muromachi period.) See "Iwashimizu jingenshō 石清水尋源鈔," (Yamato bunkakan: National Institute of Japanese Literature), 168; "Iwashimizu Hachimangū masshaki 石清水八幡宮末社記," in Zoku gunsho ruijū. Jingi bu 續群書類從. 釈家部. 神祇部, ed. Hokinoichi 塙保己一 Hanawa and Oota Toshiro 太田藤四郎 (Tokyo: Zoku Gunsho Ruijū Kanseikai, 1957), 118. 294 Fujiwara Naotsugu concludes, “Therefore, concerning the body of the Itsukushima Princess Deity, as was said above, she holds in her hands a jewel and has a white snake on her head. In our realm, the old records call [her] a god of the sea or a dragon deity, and in China as well, there are books that refer to this deity. Therefore, [this Itsukushima Deity] should not be mixed with or mistaken with other deities.” Iwashimizu Hachimangū 石清水八幡宮, ed. Iwashimizu Hachimangū shiryō sōsho 石清水八幡宮史料叢書, 5 vols., vol. 1 (Hachiman-cho: Iwashimizu Hachimangū Shamusho, 1960), 196-97. 155 the gap as the role of the sacred space and its deities change over time. The identification of the Shidara / Benzaiten deity with Itsukushima could have happened for a number of reasons. Many worshippers would not even notice the enshrinement of Itsukushima among the many gods at the site, and others might not have thought to ask about Itsukushima’s role there. In times when people questioned the reasons for its enshrinement, it presents a greater opportunity to construct an explanation than the risk of creating doubt about the legitimacy of the enshrinement. In this regard, monks and texts associated with the site benefited from their assumed authority. Fujiwara Naotsugu did not find any definitive sources on the Itsukushima Shrine at Iwashimizu Hachimangū, but few readers of Otokoyama kō koroku would be in a position to challenge his explanation. Conclusion Proper names and geographical borders could not fix the identities of sacred spaces and their gods, and that was not always a problem for medieval worshippers and scholarly monks. Worshippers at Itsukushima Shrine might not realize the overlap between the Itsukushima Deity and other Benzaiten, but scholarly monks encountered and studied these relations with other sites. In discussing the construction of locality, Arjun Appadurai raises the conflict between the needs of a space to have a unique sense of locality while still having the meaning of the space rely on relations. Appadurai provides a concise explanation of locality as “primarily relational and contextual rather than as scalar or spatial,” and further explains how the relations and contexts are continuously reproduced. 295 Basing locality in its geographic space assumes a false totality and innateness to space. Therefore, the quality of being “local,” of being particular to a 295 Arjun Appadurai, "The Production of Locality," in Counterworks: Managing the Diversity of Knowledge, ed. Richard Fardon (New York: Routledge, 1995), 208. 156 certain neighborhood, relies on there being an idea of an always preexisting neighborhood already distinguished from other neighborhoods. Naturally, the movements of people, the construction of buildings, and new interactions with others (i.e., the contingencies of history) constantly change the neighborhood. As a result, the meaning of the local needs to be able to change while retaining the illusion of fixity. To explain the fluid yet seemingly stable nature of locality, Appadurai suggests a dialectical reproduction wherein the neighborhood both acts as a context in itself (belief in the neighborhood makes it into a shared marker of locality), and produces its own contexts (e.g., the stories, religious practices, and interrelations which give meaning to the local and allow people to participate in it). Additionally, interactions and histories shared between neighborhoods create further levels of relations consisting of both similarities and differences. 296 In other words, it is not simply a matter of how to “map, build, and inhabit” space, 297 but to do so in a way that corresponds with other spaces. Faced with multiple relational conditions, locality rests on a shaky foundation and is always in need of maintenance. 298 Consequently, on the one hand the transmission of information about a sacred space, or the movements of a god, could reproduce or alter networks of space and subsequent conceptions of locality. On the other hand, the systems underlying those movements help to ensure either that the changes will correspond with preexisting relations between spaces, or that actions will be taken to preserve the involved spaces’ locality despite the effects of the changes. These systems of knowledge provided a means to support the fragility of local space and deities, but they also provided means for the control over the understandings of those spaces. In 296 Ibid., 212-15. 297 Thomas Tweed gives these three terms as the processes involved in the production of place as a site of individual or communal dwelling. Thomas A. Tweed, Crossing and Dwelling: A Theory of Religion (Cambridge, MA: Harvard University Press, 2006), 82. 298 Appadurai, "The Production of Locality," 209. 157 describing modern-day economic and metropolitan spaces as networked together as “the space of flows,” Manuel Castells recasts space through the transmission of information rather than geographic contiguity. By moving away from conventional mappings of space, Castell further argues that power over space does not rely on the control of the space itself, but the production of the cultural codes that provide a systemic power structure over the networked spaces. That is to say, the networks are accessible to all, but the system design encourages participants to follow, and thus reproduce, the symbolic and cultural codes put in place by the power-elites. 299 Similarly, unlimited networks of sacred space could be made, but the establishment of certain systems of meaning and comparison led most to follow predetermined types of comparisons and religious practices. In medieval Japan, the power of the systems of honji-suijaku and networked space cannot be ignored. Mappings of the three, four, five Benzaiten etc, show that the mapping were far from firm, but they still worked to fix the definitions of the representative Benzaiten spaces of Japan and their meanings. Network spaces solve the conflicts and doubts that can arise from ambiguity and multiplicity by providing a fluid logic of relations. The networks themselves do not fix the relations as those of sameness or difference, but instead allow explanations of those relations to present a seemingly fixed understanding that can always still be rearticulated. In the beginning of her paper on the cyborg as a subject matter for subverting the constructed boundaries of gender and identity, Donna Haraway presents the essay as “an argument for pleasure in the confusion of boundaries and for responsibility in their construction.” 300 The discussion of networks of sacred space here shares similar aims. Haraway uses metaphors of the cyborg, science, and information 299 Manuel Castells, The Rise of the Network Society (Malden, MA: Blackwell Publishers, 1996), 433-34, 46-47. 300 Donna Jeanne Haraway, "A Manifesto for Cyborgs: Science, Technology, and Socialist Feminism in the 1980s," in The Haraway Reader (New York: Routledge, 2004), 8. Italics in original. 158 systems to shift discourses on gender from the rationales of nature, essence, and objectivity to the polymorphism of flows, system logics, and strategic identities. That is to say, she argues against a single, shared essence or unity in being female, and for blurred boundaries that do not efface race, class, and other identities from gender identity. This is not unlike the relationship between honji-suijaku logic and other medieval religious discourses. Honji-suijaku abstractions allow for associations by ignoring spatial particularities, but the fluidity of the associations and their explanations in origin narratives and other discourses enabled spatial identities to be retained and to grow. In terms of the responsibility for the construction of the system logics for defining sacred space and deities, the sources show the central role of scholarly monks at powerful institutions like Mt. Hiei. But, the focus from the beginning on interrelations distinguishes their work from the boundary constructions in Western discourses of biology and objectivity. The interpretative work of the scholarly monks was often based in a desire to make sense of the complex and ever-growing webs of associations, and the ambiguity of the relations also provided fertile ground for different interpretations of the sites and their equal or hierarchical relations. The networks between Benzaiten sites could draw connections between sacred spaces, but these lines were always empowered in part through their blurriness. 159 Chapter 3 From the Origins of the Shrine to the Original Form of the Deity: Personalizing the Worship of Itsukushima To put it simply, but I hope not too provocatively: if, when hearing about religion, you direct your attention to the far away, the above, the supernatural, the infinite, the distant, the transcendent, the mysterious, the misty, the sublime, the eternal, chances are that you have not even begun to be sensitive to what religious talk tries to involve you in. 301 It is easy to think of sacred spaces and deities as transcendent and divine—as that which is other to the profane world of human existence. Bruno Latour, however, provocatively suggests that we reverse our expectations. When we look outside of the sacred itself to its articulation, religious rhetoric has the power to take what seems distant and make it close. In the case of the buddhas and bodhisattvas, the power of worldly presence is apparent in the place of the historical Buddha Śākyamuni’s birth and preaching in the world of humans, the form of Jizō as a monk, and the name of Kannon 観音 (an abbreviation of Kanzeon 観世音) as “[the one] who observes all the sounds of the world.” Anthropomorphic ideas of the deities help to bring relations between worshippers and deities close, but they also allow for these relationships to be worried about in human terms. Considering the countless prayers addressed to the deities, as can be seen from the stacks of votive tablets (ema 絵馬) hanging at shrines today, their abundance can raise questions about the deity’s ability to respond to all requests with equal urgency and rigor. In other words, why should the deity answer my prayers rather than those of others? These might be unreasonable worries to address towards a deity, but sometimes even the deities seem only human. The relationship between the deity and the devotee requires a balance between closeness and 301 Bruno Latour, "“Thou Shall Not Freeze-Frame,” or, How Not to Misunderstand the Science and Religion Debate," in Science, Religion, and the Human Experience, ed. James D. Proctor (Oxford: Oxford University Press, 2005), 32. 160 distance. People need to feel close enough to a deity to trust that the deity will respond to one’s worship and prayers, and at the same time, there needs to remain enough of a sense of supernatural distinction to reassure worshippers of the deity’s efficacy to fulfill those prayers. The religious literature of medieval Japan elucidates both problems and answers for balancing the distance of the sacred. The problem begins with the Buddhist spatial worldview of the three realms (sangoku shisō 三国思想) and the temporal worldview of the final age of the Buddhist teachings (mappō 末法), which distanced medieval Japan from the space and time of the historical Buddha Śākyamuni in India. According to mappō, there is a continuous decline in the capacity of beings to properly receive and follow the Buddhist teachings after the death (parinirvana) of the Buddha. Such concerns were further compounded by the ideas of Japan’s geographic distance from the Buddhist space of India, and the indirect transmission of Buddhism from India to China, and China to Japan. 302 Rather than responding with despair, many Buddhist thinkers and institutions took advantage of the rhetoric of the sangoku-mappō worldview to promote certain practices as suitable for the time. In particular, the dual nature of the Japanese kami deities as unique to Japan and as worldly manifestations of the Buddhist deities led to the multifaceted positionality of the kami and Japan as both near and far from the sacred space and power of Buddhism. 302 It is worth noting that the worldview of the three realms (sangoku shisō) did not carry a set value judgment. The distance situated Japan as a realm peripheral to India as the central Buddhist realm, and of course, this could be understood in negative terms. On the other hand, Japan’s place as the last to receive the Buddhist teachings could also be interpreted positively, for example, by situating Japan as the place where the Buddhist teachings are preserved, or where the Buddhist teachings blossomed into a form best suited for the people of contemporary times. On sangoku-mappō as a historical consciousness used to interpret Japanese history through Buddhist history, see Mark L. Blum, "The Sangoku-Mappō Co nstruct: Buddhism, Nationalism, and History in Medieval Japan," in Discourse and Ideology in Medieval Japanese Buddhism, ed. Richard Karl Payne and Taigen Daniel Leighton (New York: Routledge, 2006). 161 The kami are often praised in medieval texts as the deities best suited to the needs of people in Japan. The concept of honji-suijaku (original form and local traces 本地垂迹) identifies the kami (and other worldly deities) as earthly forms of the transcendental Buddhist deities taken to accord with the needs of sentient beings. Described as “dimming their radiance and merging with the dust” (wakō dōjin 和光同塵), the kami form softens the blinding brilliance of the Buddhist deities that could overwhelm and distance themselves from those they seek to save. The kami could even be elevated as manifestations of the Buddhist deities distinct to Japan and providing benefits unavailable to the people in India or China. 303 As Uejima Susumu argues, it was necessary to incorporate the kami into a Buddhist framework through honji-suijaku because there were some roles better suited to the kami than the buddhas. Unlike a transcendent buddha, a kami could appear before a worshipper, answer their prayers, and fill them with feelings of devotion and closeness. Even if the Buddhist logic of honji-suijaku can blur the differences between Buddhist and kami deities through abstract arguments of unified essence, 303 The discourse of Japan as the realm of the gods (shinkoku 神国) conceptualized a unified realm of Japan distinguished from other realms by the presence of the kami deities, but this was still far from being a blanket statement asserting the superiority of Japan as a religious or cultural territory. The fourteenth century Shintōshū 神道集 grapples with incongruent ideas of the advantageous positioning of the kami only in Japan and the recognition of there also being local non-Buddhist deities in India and China as well, saying, “Question: Concerning reverence towards the kami deities (myōjin 明神), can we categorically say that this is limited to our realm? Answer: It is explained in the words of a visualization sutra. There are the ideas of India. [But] because there is no such idea [of the kami] in the western kingdoms, how could they be explained? Therefore, a commentary on the visualization sutra says, ‘As for whether or not a king has a child, this is not something that can be secured, even if he calls upon the gods of various places.’ In the Great Treatise (Skt. Mahāprajñāpāramitā -śāstra , Ch. Dazhi du lun, Jp. Dachidoron 大智度論), we see prayers made to the god of a tree resulting in the birth of a child. Also, in China, there are mausoleums for the spirits of the three sovereigns and the five emperors, and the seven spirits and the seven stars. Beginning with this, one often hears of deities great and minor. In our realm, it was the realm of the gods from the beginning, therefore, it began with the pillars of the one hundred and eighty kami and now there are over thirteen thousand seven hundred [deities] all bringing benefits [to the people].” Shintō shū 神道集, ed. Masao 岡見正雄 Okami and Takahashi Kiichi 高橋喜一, Shintō taikei. Bungaku hen 神道大系. 文學編 (Tokyo: Shintō Taikei Hensankai, 1988), 8. 162 scholars need to pay attention to the differences and complementary functions of Buddhism and the kami. 304 While recent scholarship has increasingly emphasized the importance of the human-like characteristics of the kami deities, their humanity raises a heretofore unexamined problem: as the accessibility of the kami increases, worries over competition for the deity’s attention also increase. If anyone can make prayers to any kami, why should the kami answer my prayers? Especially considering the immense number of kami deities and shrines in Japan, it is hard to know which deity would be the most capable and the most receptive to one’s individual needs. On the one hand, there was a need for the deities to inclusively recognize and respond to the needs of all beings. On the other hand, there needed to be means to foster a sense of closeness with a deity that would elevate the connection between the deity and the devotee above others. This chapter examines the role of origin narratives in developing personal feelings of closeness with a deity. Analyzing the plot driven origin narrative subgenre of honji monogatari (original form narratives 本地物語), and the example of Itsukushima no honij 厳島の本地 (Original Form of Itsukushima) in particular, I will demonstrate how the text is structured to bring deities and devotee together. 304 Itō Satoshi also discus ses the significance of kami taking the form of snakes and woman in order to embody the sufferings of worldly beings. Notably, Itō shows that this body of the kami appears not only within narratives and performances, but also within the Ise kanjō consecration ritual 伊勢灌頂, which overlays the human body with that of the kami as snake in order to then link the person with the buddhas as the original form. Uejima Susumu 上島享, "Kami and Buddhism in the Nō Miwa: Rethinking the Study of the Amalgamation of Kami and Buddhas (shinbutsu shūgō) " Japanese Religions 42, no. 1-2 (2017): 36-37; Itō Satoshi 伊藤聡, Shintō no keisei to chūsei shinwa 神道の形成と中世神話 (Tokyo: Yoshikawa Kō bunkan, 2016), 70-73. 163 Even the Gods Have Favorites: Benefits and Anxieties of Personal Connections The case of the monk Myōe ( 明恵, 1173-1232) is often given as an example of medieval concerns over Japan’s distance from the sacred space of India, and the efficacy of the kami deities instead. Myōe wanted to travel to India but decided to stay in J apan after receiving an oracular message from the Kasuga Deity. He regretted being born so far away from the space and time of Śākyamuni’s India and sought to at least make the pilgrimage to see the traces left in the places visited by the Buddha, but the Kasuga Deity herself manifests to advise him against it. Scholars often discuss the deity’s plea to Myōe, but sometimes understate the deity’s emphasis on the personal connection between them. 305 Mark Unno mentions the intimacy of their relationship, and he suggests that their bond could have originated from: (1) Myōe’s worship of Śākyamuni (because the Kasuga Deity is identified as a manifestation of Śākyamuni); (2) Myōe’s familial bond, as a descendant of the Fujiwara family, with Kasuga as their tutelary deity; and (3) the bond of nonduality (i.e., that there is no difference between the mind and the Buddha). 306 All of these elements likely influenced Myōe, but notably, medieval narratives articulating the deity’s request focus on a more personal and affective relationship. The Kasuga gongen genki (“The Miracles of the Kasuga Deity,” 春日権現験記, offered to Kasuga Shrine in 1309), gives a multi-episode account of Myōe’s meetings with the Kasuga 305 Ryuichi Abe has highlighted the familial connections Myōe draws through ritual services to Śākyamuni as father, Buddhalocana as mother, and the Kasuga Deity as surrogate father, and how this closeness could benefit the divine protection he received. However, this journal article only mentions the Kasuga Deity’s oracle to Myōe in passing and does not connect the oracle with these family connections. Hirano Tae connects Myōe’s worship of Buddhalocana and the oracle from the Kasuga Deity within a larger argument on the important roles played by women in Myōe’s life and religious practices. Hirano mentions Myōe’s tears at receiving the oracle and his consequent deepened worship of Kasuga, but she seems to suggest that his response is due in part to the Kasuga Deity’s choice of his uncle’s (Yuasa Munemitsu 湯浅宗光) wife as her medium. Ryuichi Abe, "Mantra, Hinin, and the Feminine: On the Salvational Strategies of Myōe and Eizon," Cahiers d'Extrême-Asie 13, no. 1 (2002): 101, 05; Hirano Tae 平野多恵, Myōe: Waka to Bukkyō no sōkoku 明恵 : 和歌と仏教の相克 (Tokyo: Kasama Shoin, 2011), 94. 306 Mark T. Unno, "The Body of Time and the Discourse of Precepts," in Discourse and Ideology in Medieval Japanese Buddhism, ed. Richard Karl Payne and Taigen Daniel Leighton (New York: Routledge, 2006), 142-44. 164 Deity. Here, the Kasuga Deity implores Myōe to stay so that the people will not lose such a wise monk and teacher, but the deity also mentions that she will always be there to protect him. There is a clear argument that the presence of the kami as manifestations of buddhas and bodhisattvas makes traveling to India unnecessary, but the narrative bases this argument in the close connection that only exists between the two of them. 307 The Kasuga Deity makes her feelings towards him clear: "There is not one of the gods, good monk," she then continued, "who does not protect you. The Sumiyoshi Deity and I attend you particularly. And I, especially, am always with you in the center of your body, so that even if you were across the sea we would not be parted, and I would not personally mind. But when I remember all the people who can be inspired by you to faith, as long as you are in Japan, my happiness at the thought turns to grief that you should mean to undertake so long a pilgrimage. I love all those who have faith in the Buddha's Teaching, and among them I think particularly fondly of three: yourself, Gedatsu-bō (Jō kei), and another in the Capital. But I am not as devoted to the other two as I am to you." 308 The Kasuga Deity acknowledges that all of the deities of Japan protect Myōe but she situates herself as the closest, always being there within him. The deity even refers to her devotion to him over other worshippers with honorifics (kokoro wo kaketatematsuru 心をかけたてまつる), presenting both Myōe and the Kasuga Deity as revering one another. 309 Although it is possible that the rhetoric of the deity’s love for Myōe exceeding all others is an embellishment made to depict Myōe idealistically, it still suggests that the audience would be receptive to such a relationship with a deity. Additionally, other accounts of the oracle often hint at their distinctive 307 The Kasuga Deity gives him multiple means for him to keep her close: visions of her true form and the unity of Kasuga and Vulture Peak, and transmissions of teachings and poems. Translation adapted from Royall Tyler, The Miracles of the Kasuga Deity (New York: Columbia University Press, 1990), 269-78; "Kasuga gongen genki 春日権現験記," in Gunsho ruijū. Jingi bu 群書類従. 神祇部, ed. Hanawa Hokinoichi 塙保己一 (Tokyo: Gunsho Ruijū Kankō kai Kantō sha Nai, 1952), 47-52. 308 Translation from Tyler, The Miracles of the Kasuga Deity, 272-73; "Kasuga gongen genki 春日権現験記," 49. 309 The phrase (kokoro wo kaketatematsuru 心をかけたてまつる) is translated here as devotion and translates more literally to “humbly gives her heart / attention [to him].” Like “devotion,” the phrase can convey a religious or romantic connotation, or simply mean to focus one’s attention on something. 165 closeness by comparing the deity’s feelings towards Myōe to a parent’s love for the eldest son. 310 All worshippers might be loved like children, but only Myōe is loved as the first born. Myōe is in an ideal position to receive the support of the Kasuga Deity. However, Myōe also stood in stark contrast to the average worshipper’s lesser opportunities and capacities for building connections with a deity. This is, in short, the soteriological anxiety that needs to be ameliorated. Origin narratives and proselytizing works promoting pilgrimage and worship needed to do more than just articulate the sacrality of the site and its deities. They needed to convince the audience that the deities not only had the power to answer their prayers, but also that the deities would be receptive to them. While the average worshipper could not compare with Myōe, origin narratives encouraged their audiences to feel a similar sense of belonging with the deity. That sense of closeness would deepen one’s devotion to the deity and the deity’s response. The protection of the gods might be available to all, but it resonated through individual bonds with the deity. There were multiple means for strengthening connections with a deity (e.g., regional proximity, familial tutelage, belonging to a particular religious institution), and from a more abstract perspective, these can all be understood as feelings of closeness based on a sense of 310 In the Shasekishū 沙石集, Umeo Myōe shōnin den 梅尾明恵上人伝, and the nō play “Kasuga ryūjin” 春日龍神 (pre-1465), the Kasuga Deity further specifies that she thinks of Myōe as her first son and Jōkei ( 貞慶, 1155-1213) as her second son. The statement recognizes both Myōe’s and Jōkei’s devotion to the Kasuga Deity but elevates the relationship with Myōe above that of Jōkei. The Kasuga gongen genki and the Kōsanji Myōe shōnin gyōjō 高山寺明恵上人行状 (by Myōe’s discip le Kikai) also relate the Kasuga Deity’s affection towards Myōe to that of a parent’s love, but do not mention Jōkei. Robert E. Morrell, Sand and Pebbles (Shasekishū): The Tales of Mujū Ichien, A Voice for Pluralism in Kamakura Buddhism (Albany: State University of New York Press, 1985), 84; Mujū Ichien 無住一円, Shasekishū 沙石集, ed. Kojima Takayuki 小島孝之, Shinpen Nihon koten bungaku zenshū (Tokyo: Shōgakkan, 2001), 38; "Umeo Myōe shōnin den 梅尾明恵上人伝," in Myōe Shōnin shiryō 明惠上人資料, ed. Kō zanji Tenseki Monjo Sō gō Chō sadan 高山寺典籍文書綜合調查團 (Tokyo: Tō kyō Daigaku Shuppankai, 1971), 287; "Kōsanji Myōe shōnin gyōjō (Ueyama bon) 高山寺明恵上人行状( 上山本)," in Myōe Shōnin shiryō 明惠上人資料, ed. Kō zanji Tenseki Monjo Sō gō Chō sadan 高山寺典籍文書綜合調查團 (Tokyo: Tō kyō Daigaku Shuppankai, 1971), 112; "Kasuga ryūjin 春日龍神," in Yōkyoku hyakuban 謠曲百番, ed. Nishino Haruo 西野春雄, Shin Nihon koten bungaku taikei (Tokyo: Iwanami Shoten, 1998), 510; Royall Tyler, Japanese Nō Dramas (London: Penguin Books, 1992), 148. 166 belonging. Few people possessed the family ties or regional proximity that Myōe had with Kasuga, but what mattered most was the expression of one’s devotion that would allow the deity to appreciate and reciprocate the worshipper’s feelings. This is what the Kasuga Deity was trying to tell Myōe. He ne eded to realize that the Kasuga Deity was more than just a convenient substitute for Śākyamuni. The unfamiliar Śākyamuni had no reason to devote himself to Myōe like the Kasuga Deity. But this relationship still requires Myōe to respond in turn and devote himself to the Kasuga Deity. The Kingyōku yōshū 金玉要集, a fourteenth century compilation of proselytizing literature, reverses the metaphor of parental affection between kami and worshipper to express the Kasuga Deity’s disappointment at hearing Myōe’s desir e to leave her and travel to India. 311 If he cannot appreciate everything that the Kasuga Deity has done for him, why should the deity continue to help him? She says: As a local trace (suijaku 垂迹), the flavor of my original form’s (honji 本地) teachings can be grasped. You and Gedatsubō are like eldest sons to me, and yet [you speak] of throwing me away and crossing the boundary to the faraway Western Heaven (India). You cannot know how sad this makes me. From Central India, where the Thus Come One appeared in this world, to our realm is a road of one hundred thousand ri (approximately 240,000 miles). To cross the great ocean of ten thousand ri and climb through the mountain peaks and forests, it is impassable even for the birds and beasts. With the vastness of the River of Flowing Sand, this is something humankind has never crossed. Even if I help you to reach [India] in accordance to your prayers, it still will not be easy to see the traces of the Buddha’s sermons. 312 The Kasuga Deity details the difficulty of traveling to India and the uncertainty of what he would even find there. But, this is only said after criticizing him for wanting to go to India in the first place. Where else could Myōe find a deity who would appear before him, much less attest her 311 The nō play “Kasuga ryūjin” also expresses the potential for the Kasuga Deity’s displeasure if Myōe were to leave Japan, with an old man (shite) saying, “In fact, I understand he calls you His first-born son, and treats as his second son the Venerable Gedatsu of Kasagi. You two might as well be His own hands or eyes, He protects you so loyally, night and day. How then, could it please Him to learn that you wish to leave Japan and cross the sea to China and India?” (Translation from Royall Tyler). Japanese Nō Dramas, 148-49; "Kasuga ryūjin 春日龍神," 510. 312 "Kingyōku y ōshū (Naikaku bunko zō) 金玉要集( 内閣文庫蔵)," in Sonarejō: Murasame hen 磯馴帖：村雨篇, ed. Itō Masayoshi 伊藤正義 (Osaka: Izumi Shoin, 2002), 209. 167 special bond with him, and implore him to stay? The Kasuga Deity reveals her human-like feelings and suggests that this is something he will not find elsewhere. This chapter argues, in much the same manner, for two interconnected means through which origin narratives could foster personal closeness with a deity: (1) humanizing the deity and encouraging an affective response from the audience; and (2) presenting the audience with devotional practices mirroring events within the narrative. The connections between the practices and the story enable each person to performatively instill one’s feelings into one’s devotion and one’s devotion into the story. Honji monogatari: The Original Form Narrative as Another Form of Origins This chapter focuses on a subgenre of temple and shrine origin narratives (jisha engi 寺社縁起) called honji monogatari (original form narratives 本地物語). Scholars have tended to categorize honji monogatari in accordance with their literary characteristics as Muromachi period narratives (Muromachi jidai monogatari 室町時代物語, often referred to as otogizōshi 御伽草子), but there is no debate that the tales functioned as origin narratives. 313 I situate honji monogatari as origin narratives to highlight the diverse forms that origin narratives can take, and the religious significance of honji monogatari as narratives that could be read for enjoyment, edification about sacred sites, and as objects of devotion themselves. As origin narratives, honji monogatari still recount the origins of a temple or shrine, but they do so through a past-life story of the deity, who is born as a human, undergoes sufferings, and then becomes a deity to save 313 Because of the imprecision of medieval literary genres, many honji monogatari can correspond with works categorized within literary genres such as Muromachi period narratives, lay preaching works (sekkyō 説経), and ko- jōruri 古浄瑠璃 puppet theater. Ō shima Yukio 大島由紀夫, Chūsei shūsho no bungei bunka: Engi, setsuwa, monogatari no enpen 中世衆庶の文芸文化：縁起・說話・物語の演変 (Tokyo: Miyai Shoten, 2014), 443. 168 others from such sufferings. 314 This structure makes honji-monogatari ideally suited for employing the affective powers of narrative to draw audiences into the story and feel for the characters. As such, honji monogatari were aimed towards promoting worship by popular audiences to a greater degree than temple and shrine origin narratives more broadly, which also needed to convey additional information about the site and its deity for diverse audiences of worshippers, scholarly monks, intellectuals, and political supporters. 315 Ōshima Yukio argues that the di fferentiation of honji monogatari from origin narratives heightened after the late sixteenth century, likely as a result of the growing popularity of honji monogatari. Comparing the titles of honji monogatari manuscripts, early fourteenth and fifteenth century versions are often titled as “Origin narrative of ...” (engi 縁起) or “Tale of ...” (monogatari 物語), while post sixteenth century versions become “Original form of ...” (no honji の本地). 316 The growing popularity and awareness of honji monogatari is significant, but variations in the titles of Itsukushima Shrine honji monogatari texts (as Itsukushima no honji or Itsukushima no engi) complicates arguments for a complete divide from origin narratives. 317 In 314 Tokuda Kazuo gives the Shintōshū ( 神道集, produced between 1352-1360) as a key example of honji monogatari from the fourteenth century. The Shintōshū is a compilation of origin narratives of different temples and shrines, with the first half largely composed of more information-centric temple and shrine origin narratives (jisha engi) and the second half consisting mainly of honji monogatari type origin narratives. Additionally, the compilation of the Shintōshū by the Agui school of preachers at Mt. Hiei suggests that the origin narratives were compiled for use in sermons and proselytization. Tokuda Kazuo 徳田和夫, "Honji monogatari no kiso 本地物語の基礎," in Kami to hotoke 神と仏, ed. Konno Tō ru 今野達, Satake Akihiro 佐竹昭広, and Ueda Shizuteru 上田閑照, Iwanami kō za Nihon bungaku to Bukkyō (Tokyo: Iwanami Shoten, 1994), 90-91. 315 For more information on the various uses and audiences of origin narratives, see the previous chapters of the dissertation. Each chapter examines a different medieval account of the origins of Itsukushima Shrine, and the influences on their production, distribution, and reception. 316 Ō shima, Chūsei shūsho no bungei bunka: Engi, setsuwa, monogatari no enpen 中世衆庶の文芸文化：縁起・說話・物語の演変, 445. 317 There are at least four post-1600 manuscripts with titles corresponding to Itsukushima no engi: The 1622 Tenri Library manuscript; the 1657 National Diet Library manuscript; the early modern Ishikawa Tōru owned manuscript; and the undated Itsukushima Shrine manuscript. For a comparison and categorization of nine of the seventeen known versions of Itsukushima no honji, see Matsumoto Ryūshin 松本隆信, Chūsei ni okeru honjimono no kenkyū 中世における本地物の研究 (Tokyo: Kyūko Shoin, 1996), 90-116. 169 one Itsukushima Shrine manuscript (post seventeenth century), a reader appears to have disagreed with the written title of “Itsukushima go engi” 厳島御縁記, and added in red ink the words “daimyōjin” (deity, 大明神) and “honji” (original form, 本地) to make the title “Itsukushima Daimyōjin on honji engi” (Origin narrative of the original form of the Itsukushima Deity, 厳島大明神御本地縁記). 318 This might seem trivial, but it demonstrates the simultaneous ambiguity and significance of the distinction. For at least one person, it was worth putting brush to paper to designate the work as a honij monogatari. Considering the inconsistent titles and the need for clarity, I will refer to all versions of the Itsukushima honji monogatari as Itsukushima no honji. Thankfully, this corresponds with the earliest fully extant manuscript that is the focus of my analysis, the late Muromachi (1392-1573) Itsukushima no honji from Keiō U niversity Library. There is an earlier 1346 illustrated scroll manuscript, but unfortunately, only a few later sections of the work remain. As will be discussed later, comparing the two reveals how the narrative became increasingly attentive to the interests of popular audiences. 319 Despite the evidence for the popularity and growth of honji monogatari, there are few records detailing how honji monogatari were read. This can be explained in part as a result of the lack of distinction between origin narratives and honji monogatari prior to the seventeenth century. Tokuda Kazuo provides ample evidence from Muromachi period courtier diaries that monks and priests recited origin narratives to pilgrims visiting one’s temple or shrine. The 318 Emaki kenkyūshitsu 中世説話・絵巻研究会 Chūsei setsuwa, "Honkoku: Itsukushima go engi (Itsukushima Jinja zō) 翻刻：厳島御縁起（厳島神社蔵）," Komazawa kokubun 17 (1980): 189, 201. 319 The popularity of Itsukushima no honji flourished in the early modern period and there are over fourteen copies in various forms (written manuscript and wood block print, text only and text with images, scrolls and books) approximately dated to post-1700. That being said, it is not insignificant that there are still two manuscripts from the fourteenth to seventeenth centuries, and one more from the Akaki archive dated to either the late Muromachi period or early Early modern period. We can see that honji monogatari were produced and received more widely than the origin narratives copied and collected by scholarly monks and temple repositories. 170 performances often occurred during the time of rites and festivals, when the sites would be crowded with pilgrims eager to lend their ears, and often their money, for the opportunity to make prayers and build a karmic connection (kechien 結縁) with the deity. 320 The brevity of the records, however, provides an incomplete picture. 321 Koida Tomoko’s investigation of the origin narratives mentioned within the Oyudono no ue no nikki お湯殿の上の日記, a record of the lives of the women serving the sovereign from 1477-1826, reveals that they viewed the Seiganji no honji 誓願寺の本地 and Boroboro no sōshi ぼろぼろの草子 honji monogatari. Notably, even though the entree on Seiganji refers to the work as the “illustrated original form (honji) of Seiganji,” Sanjōnishi Sanetaka’s diary ( 三条西実隆, 1455-1537) appears to refer to the same work as the “illustrated origin narrative (engi emaki 縁起絵巻) of Seiganji.” 322 The entrees resemble those for any other origin narrative, as one might expect given the ability of honji monogatari to function as origin narratives. Nevertheless, we should not go as far as to say that 320 There are also cases of blind biwa lute performers (zatō 座頭), miko (female spirit mediums, 巫女), fund raising monks (kanjin hijiri 勧進聖), and linked verse poets (rengashi 連歌師) performing origin narratives, possibly outside of the grounds of the religious institutions. Additionally, monks and priests sometimes traveled to show an illustrated origin narrative to courtiers and elites who had requested to see the scrolls. As Tokuda argues, the performers did not simply wait for pilgrims to come and listen, and they were also involved in creating opportunities for origin narrative recitation. Tokuda Kazuo 徳田和夫, Otogizōshi kenkyū お伽草子研究 (Tokyo: Miyai Shoten, 1988), 165-74. 321 There are a few cases of individuals hearing an origin narrative and then recording details in their diaries, but many entries omit the content of the narrative altogether. For example, the courtier Sanjōn ishi Sanetaka ( 三条西実隆, 1455-1537) recorded the following after visiting the temple Myōtokuji 妙徳寺 in Harima Province on 7/28/1503 (third year of Bunki 文亀): “[At] Ichijōin, in the se cond year of Shōryaku (991), Jie Daishi’s ( 慈恵大師, = Ryōgen 良源, 912~985) disciple Keihō 慶芳 received an oracle during a memorial ceremony, and this was announced to him. ‘The main object of worship is Monjusri. [At] Sanjōin, dur ing [the current sovereign’s] reign, make the object of worship the Medicine King Buddha. Together, these are the two honored ones [of this temple].’” Sanjō nishi Sanetaka 三條西實隆, Sanetaka Kōki 實隆公記, Third ed., 13 vols. (Tokyo: Zoku Gunsho Ruijū Kanseikai, 1979- 1980), v4, 156; Tokuda, Otogizōshi kenkyū お伽草子研究, 176. 322 Koida Tomoko 恋田知子, "Monogatari sōsh i toshite no keisei to juyō: “Oyudono no ue no nikki” wo tsūjite 物語草子としての形成と受容：『お湯殿の上の日記』を通じて," in Jisha engi no bunkagaku 寺社緣起の文化学, ed. Tsutsumi Kunihiko 堤邦彦 and Tokuda Kazuo 徳田和夫 (Tokyo: Shinwasha, 2005), 232, 35, 46, 48. 171 the honji monogatari were considered and received identically to temple and shrine origin narratives during the medieval period. The diaries just do not tell us enough. Looking outside of the context of medieval Japan, the devotional bhakti literature and religious practices in South Asia provide another avenue for considering the intended reception of honji monogatari. Bhakti is often defined as “devotion,” a fittingly wide-ranging word to signify the breadth of bhakti as encompassing various practices through which a worshipper’s acts and experiences are directed to a god. Karen Pechilis works from Charles Hallisey’s centering of devotion around self-involvement in making a god one’s own, to argue for bhakti as participation in a relationship with a god. More specifically, Pechilis presents bhakti as balancing emotion and intellection within one’s actions. While bhakti is sometimes seen in comparison with Protestant Christianity as a form of personal devotion based solely on emotion, the unison of emotion and knowledge grounds individual action in culturally structured and articulated experiences of closeness with the god. 323 The combination of emotion and knowledge likewise works in honji monogatari through literary conventions to encourage their audiences to experience the same feelings as the characters, fill one’s heart with devotion, and consequentially, allow for the deity to come close. 324 The transmission of the Sanskrit epic, the Ramayana, to Japan and its adaptation within setsuwa tales (first included within the 1180 Hōbutsushū 宝物集) influenced the plot and themes within honji monogatari. Nevertheless, we cannot go so far as to say that the Ramayana was used to incorporate bhakti rhetoric into honji monogatari. Fabio Rambelli, building on the 323 Karen Pechilis Prentiss, The Embodiment of Bhakti (New York: Oxford University Press, 1999), 19-24. 324 For example, the Bhā gavata Mā hā tmya (approximately sixteenth century), describes the reading of the bhakti text, the Bhā gavata Purā ṇa, as proper devotional practice and as a knowledge sacrifice (jñānayajña ) that adorns the reader’s heart, and through this embodiment, provides the devotee with a proper form to receive and please the deity. Ibid., 31-33. 172 scholarship of Tamori Masakazu, demonstrates the close overlap between the plot structures of honji monogatari and the Ramayana. As Rambelli notes, the Chinese translated Buddhist texts transmitting the Ramayana to Japan provided either summaries or variants as Buddhist jataka tales, and these condensed forms empowered Japanese writers to localize Ramayana elements without having to comply with literary or religious aspects of the Ramayana in South Asia. For example, both the Ramayana and the Shintōshū 神道集 compilation’s Suwa Shrine honji monogatari, Suwa engi no koto 諏訪縁起の事, involve the wife of the male protagonist being abducted by a demon (or tengu in the case of Suwa) while he is hunting a deer in the forest; the wife refusing to break her faithfulness to her husband; the husband traveling to another world to rescue her; and ending with the deification of both protagonists. 325 The narrative structure from the Ramayana provides a foundation for the Suwa Shrine narrative general enough to support elements related to kami worship at Suwa Shrine and Japanese literary motifs. These narrative similarities alone are insufficient for presenting a direct connection with bhakti devotion, but they can suggest how Ramayana themes could fit with literary conventions and devotional reading practices in both India and Japan. The Ramayana itself did not originate as a bhakti text, but the themes of separated lovers, loyal devotion despite suffering and adversity, heroic reunification, and Rama’s identification as an avatar of Vishṇu, made it easy for bhakti devotional traditions to incorporate and reinterpret the emotionally charged events of the Ramayana. 326 In the case of Japan, the emotional force likewise given to the suffering of 325 Fabio Rambelli, "The Story of Prince Rama in Japan: Sources and Transformations," in Ramayana in Focus: Visual and Performing Arts of Asia, ed. I. Made Bandem and Gauri Parimoo Krishnan (Singapore: Asian Civilisations Museum, 2010), 29-31, 36-37; Tamori Masakazu 田森雅一, "Nihon ni okeru Ramayana no juyō to hen'yō 日本におけるラーマーヤナの受容と変容," in Rā mā yana no uchū: Denshō to minzoku zōkei = Cosmos of Ramayana: Tradition and Ethno-forms ラーマーヤナの宇宙 : 伝承と民族造形, ed. Kaneko Kazushige 金子量重, Sakata Teiji 坂田貞二, and Suzuki Masataka 鈴木正崇 (Tokyo: Shunjūsha, 1998), 198-99, 207-12. 326 John Brockington gives the purā ṇa narratives of the fourth to twelfth centuries and the Bhakti movement in Tamil from the seventh century onwards as two important stages leading the epics of the Ramayana and the 173 separated lovers in Heian court literature and waka poetry made the Ramayana a fitting source for producing devotional narratives about the past lives and deification of Japanese kami deities. Therefore, even if the Buddhist variants of the Ramayana received in Japan did not convey ideas of bhakti devotion, it does not seem to be mere coincidence that honji monogatari began to incorporate plot elements and motifs from the Ramayana. 327 Scholarly monks in medieval Japan could have easily seen the potential in the Ramayana for developing new origin narratives meant to elicit feelings of closeness and devotion. Humanizing the Original Form of the Deities Honji monogatari took advantage of being a form of origin narratives (gaining their authority as a mythic record of a sacred space’s past), while also incorporating narrative strategies for encouraging personal devotion to a sacred space and its deities. Admittedly, the lack of historical records detailing the reception of origin narratives, much less honji monogatari, Mahabharata to develop into bhakti texts. Additionally, Ajay Rao examines later discourses (1250-1600) by Tamil Brahmin intellectuals on soteriological and devotional significances of the Ramayana. John Brockington, "The Epics in the Bhakti Tradition," in The Intimate Other: Love Divine in Indic Religions, ed. Anna S. King and Brockington John (Hyderabad: Orient Longman, 2005), 36-38; Ajay K. Rao, Re-figuring the Rā mā yaṇa as Theology: A History of Reception in Premodern India (New York: Routledge, 2015). 327 Tamori Masakazu gives the Sutra on the Collection of the Six Perfections (Ch. Liuduji jing, Jp. Rokudojū kyō 六度集経, T 152) and the Treatise of the Great Commentary on the Abhidharma (Ch. Da piposha lun, Jp. Dai bibasha ron 大毘婆沙論, T 1545) as two possible sources for the summary of the Ramayana transmitted to Japan. Fabio Rambelli also gives the Vessentara Jataka and Dasharatha Jataka as two examples of Ramayana tales transmitted to Japan. Richard Gombrich argues that these retellings as jataka tales of the past lives of the Buddha were not written “Buddhist answer[s] to the Ramayana” but reconcile conflicts between the values of the Ramayana with those of Theravada Buddhism. Notably, the Dasaratha Jataka positions Rama and Sita as past lives of the Buddha and Yaśodharā and emphasizes emotional restraint at the time of his father’s death in contrast to the Ramayana’s depiction of the depth of Rama’s grief. Tamori, "Nihon ni okeru Ramayana no juyō to hen'yō 日本におけるラーマーヤナの受容と変容," 200-02; Rambelli, "The Story of Prince Rama in Japan: Sources and Transformations," 29; Richard Gombrich, "The Vessantara Jātaka, the Rāmāya ṇa and the Dasaratha Jātaka," Journal of the American Oriental Society 105, no. 3 (1985): 434-36. 174 necessitates a degree of conjecture. Therefore, I use the set format and the narrative explication within the text as guides for analyzing how the work could influence its reading and reception. 328 Honji monogatari borrow the structure of Buddhist jataka tales about the past lives of the Buddha to explain the pre-history of Japanese kami gods. Like jataka tales, honji monogatari separate the narrative and its explication as distinct parts. The first half of jataka tales is the past- life story, and the second is the narrator’s explanation of the identities of the characters as buddhas and bodhisattvas contemporary to the telling of the tale. Honji monogatari then add an additional level of association by identifying each of the characters as a manifestation of a buddha or bodhisattva (as the original form), and as one of the deity’s enshrined at the temple or shrine (as the local trace of the original form). The strong narrative coloring of honji monogatari and jataka tales makes it easy to dismiss the religious significance of these works, but conversely, the separation of narrative and explication enables any narrative to be instilled with religious meaning and affect. This was more than just a way of sugar-coating religious instruction for lay audiences, although we cannot discount the entertainment value. Naomi Appleton argues against the idea that jataka are “simply moral stories purely for the instruction of the laity,” contending that their worldly content does not limit the tales to a single audience or a single level of meaning. 329 Even monastics and scholars could enjoy a good story, and even a good story could be profound or sacred. Similarly, Max Moerman describes Kumano no honji 熊野の本地 as “an epic tale of sex and violence, of conspiracy and sacrifice, of evil women and virtuous mothers,” and he still manages to uncover meaning beyond attracting audiences and 328 Tokuda Kazuo similarly argues for examining the reception of honji monogatari from the form of the narrative presented to audiences, and also examines how the incorporation of popular literary motifs (e.g., child born after prayers to the gods (mōshigo 申し子), exile of a young noble (kishu ryūri 貴種流離), etc.) leads audiences to certain types of interpretations. Tokuda, Otogizōshi kenkyū お伽草子研究, 546. 329 Naomi Appleton, Jataka Stories in Theravada Buddhism: Narrating the Bodhisatta Path (Burlington, VT: Ashgate Publishing, 2010), 11-12. 175 worshippers. Instead, he emphasizes how Kumano no honji presents the deity as a character who has experienced such horrible sufferings, and consequentially will sympathize with the troubles faced by worshippers and listen to their prayers. 330 Moerman’s reading of Kumano no honji accords with the existing scholarship focusing on the nature of the gods in honji monogatari as “suffering gods” (kurushimu kami 苦しむ神). Watsuji Testurō first proposed this neologism to present the anguish as a necessary stepping stone towards rebirth as a kami deity whose human experiences and sentiments can mirror those of worshippers and draw them close. 331 As some honji monogatari narratives explicitly explain, human suffering was crucial for “dimming their brilliance and merging with the dust” (wakō dōjin), and thus enabling people to form karmic connections with them. The honji monogatari of “Mt. Komochiyama in Kōzuke Province” ( Kōzuke no kuni komochiyama no koto 上野国児持山之事) in the Shintōshū compilation explains, The way of the kami, of manifesting traces of the buddhas and bodhisattvas in response [to sentient beings], always begins from connections (en 縁) [with sentient beings]. Therefore, when the various buddhas and bodhisattvas come to our realm, they always borrow a person’s womb, take the body of worldly beings, experience sufferings, and try out wholesome and unwholesome [deeds]. [It is only] after this that they take the body of a deity and confer benefits to the sentient beings of the unwholesome world [of mappō]. 332 Ōshima Yukio provides this passage and two more from the Shintōshū to elucidate the fundamental role of suffering to enable the deities to understand the beings in Japan they came to save. 333 This is clear enough. However, connecting this passage with the larger context of the 330 D. Max Moerman, Localizing Paradise: Kumano Pilgrimage and the Religious Landscape of Premodern Japan (Cambridge, MA: Harvard University Asia Center : Distributed by Harvard University Press, 2005), 216-17. 331 Watsuji Tetsurō 和辻哲郎, Uzumoreta Nihon 埋もれた日本 (Tokyo: Shinchō sha, 1951); Tokuda, Otogizōshi kenkyū お伽草子研究, 131. 332 Shintō shū 神道集, 185. 333 Ō shima, Chūsei shūsho no bungei bunka: Engi, setsuwa, monogatari no enpen 中世衆庶の文芸文化：縁起・說話・物語の演変, 446-47. 176 narrative uncovers further meaning. It is tempting to analyze only the pathos of the narrative, or only the later explication of the gods, but both are needed in order to consider how the explication in the second half guides the audience’s reception of the first half. When we take a closer look at the quoted passage, it is located at the end of the text and followed by two concluding sentences: “[I] took out and cooked a maro 鮃𩵥 fish as a substitute for my child, saving her and making her into a deity. This is what is meant by child substitute (ko no kawari 子ノ代).” 334 This recalls the events in the tale of the protagonist’s stepmother saving her by cooking a maro fish that gave off the odor of a corpse and fooled their adversary into thinking she was dead. Tokunami Miki interprets this as a reversal from the expected wicked behavior of a stepmother into a true mother-daughter relationship. Not only does the stepmother save the daughter in the story, the narrative ends with the powerful words “my child.” 335 This sentiment, though, does not make sense as a conclusion until we consider the relationship between the audience and the narrative/narrator. As a worshipper seeking salvation, the audience can hear the words “my child” as coming from the deity. The position of the protagonist transforms from saved to savior not only to connect the deity with human suffering, but also to connect the audience with the deity as one’s child to be substituted out and saved. If we recall the example of Myōe and the Kasuga Deity, the deity used the same rhetoric of familial love to assure Myōe that she would devote herself to him more than any other worshipper. Although the above quotation can be read by itself as a doctrinal honji-suijaku argument for the role of the 334 Shintō shū 神道集, 185. 335 Tokunami Miki 十九波美希, "’Shintōshū ’ ni okeru zaichisei: Tōgoku bungaku ni zaisuru chienteki, ketsuenteki chūtai ni chakumoku shite 『神道集』における在地性：東国文学に存する地縁的・血縁的紐帯に着目して," Bungakushi kenkyū 49 (2009): 78. 177 kami, placing it within the context of being part of an explication of the narrative enables the deities to be positioned in personal relations with the audience. Honji monogatari do more than just present suffering as a stepping stone towards deification; they present the tale of a suffering god as a stepping stone towards practices establishing a close devotional relationship with that deity. It is always an interdependent relationship relying both on the deity’s blessings and the devotee’s worship. First the narrative encourages an emotional response to the telling of the deity’s past life, and afterwards, the narrator urges the audience to instill those feelings into one’s worship. This could be done through making pilgrimages to the site and making offerings, but this is not a requirement. Reading or hearing the work and feeling an emotional closeness with the deity can, in itself, become a form of worship. In this manner, honji monogatari make the worship of site-specific deities, like the Itsukushima Deity, accessible to all regardless of proximity to the sacred space. Explication and the Reinterpreting of the Narrative and the Deity There remains a crucial question: how can we ascertain if audiences interpreted the work in an individual and personal manner? Medieval pilgrims undoubtedly heard origin narratives recitations at temples and shrines within a group audience. Even readers of a written manuscript often gathered together with one person reciting for the group. Individuals with more learning (due to age or education) might have been tasked to read for the group, and regardless of literacy, the expense of owning narratives and practices of borrowing texts lent themselves to collective reading practices. The beginning of Itsukushima no honji takes the tone of an orally explicated text and introduces the narrative as if answering a question from the audience, “When one 178 inquires about the original form of this deity ...” 336 The group setting helps make the building of a personal connection with the deity into a social performance produced through the interaction between the narrator, the audience, and the deity, and working from shared cultural understandings of the narrative’s affect, and the received explication of devotional responses to the engendered feelings. 337 The explication of the narrative in the second half of honji monogatari suggests how the audience should cultivate those feelings of closeness into a personal connection through worship. Sanō Midori argues that, in the case of reading illustrated Tale of Genji scrolls, the performance of narration/reading produces a localized space-time of the here and now that brings together the worlds of text and the reader. The narration draws the audience into the story, but rather than assuming any simple identification between the reader and protagonist, the reader incorporates one’s external knowledge and expectations to recreate the story-world from a perspective beyond that available to any character. 338 In the case of honji monogatari, as an origin narrative of the past life of a Japanese deity, its reading brings together the time of the mythic past into the reader’s present, and the spaces of the deity’s past with the deity’s current space of enshrinement. Specifically, the division of the narrative and the explication creates two 336 Mori Masato cites similar rhetoric speaking out to the reader from Dōjōji engi emaki to show how the text reflects how the text would have been recited to audiences, and accompanied by explanations of the images (etoki 絵解き). Mori Masato 森正人, Ba no monogatariron 場の物語論 (Tokyo: Wakakusa Shobō , 2012), 382-83. Unless otherwise noted, all translations of Itsukushima no honji are from the late Muromachi period Keiō Library manuscript. "Itsukushima no honji 厳島の本地," in Muromachi jidai monogatari taisei 室町時代物語大成, ed. Yokoyama Shigeru 横山重 and Matsumoto Ryūshin 松本隆信 (Tokyo: Kadokawa Shoten, 1973), 270. 337 In the case of bhakti devotion, Christian Lee Novetzke argues for bhkati as an established emotional state (bhāva ) experienced by the devotee that is performed publicly through a tripartite structure of the speaker or poet, audience, and the gods. In other words, bhakti functions by constructing relationships through the giving and receiving of devotion rather than through internal feeling of devotion alone. Christian Lee Novetzke, "Bhakti and Its Public," International Journal of Hindu Studies 11, no. 3 (2007): 255-57, 61-62. 338 It is for this reason that illustrations of court tales often incorporate court ladies (nyōbō) watching the scene, even though the text never mentions such characters. They provide readers with a body and gaze to view the narrative from within the image while still retaining the external viewpoint of the reader. Sano Midori, "The Narration of Tales, The Narration of Paintings," in Reading the Tale of Genji: Its Picture Scrolls, Texts and Romance, ed. Richard Stanley-Baker, Murakami Fuminobu, and Jeremy Tambling (Leiden: Global Oriental, 2009), 38-39, 48-49. 179 consecutive opportunities for reimagining the narrative and one’s relation to it. First, the narrative brings the reader into affective personal relations with the characters. Afterwards, the explication ties the current space-time of the reader and the sacred space with the mythic past of its founding in the narrative. Even when an audience hears a honji monogatari for the first time, the standardized structure (birth as human—undergoing the travails of human life—deification and the founding of a sacred space in Japan) allows them to anticipate events and bring the narrative into one’s individual lived world. The narrative’s use of popular motifs and familial relationships could further assist readers in abstracting the story and identifying oneself within the story. 339 In other words, readers might not identify as the deity, but they could identify with the sufferings of the pre-deity protagonist, and the feelings of other characters towards her. When the protagonist becomes a deity afterwards, the reader can feel a sense of affinity akin to a family member or a member of the local community. Melissa McCormick uses the example of an unnamed protagonist in a narrative scroll made for Hosokawa Masamoto ( 細川政元, 1466-1507) to argue for an individuated reading of the scroll apart from identifying as the character. She explains, “These personalized readings were not, however, predicated upon a direct correspondence between fictional characters and historical viewers but rather emanated from the ambiguous but dynamic space the scroll provided for envisioning complex and reciprocal bonds between the men in Masamoto’s household.” 340 Well-known motifs, themes, and character relations guided readers towards certain imaginings of the narrative and their relation with it, while the absence or 339 Tokuda, Otogizōshi kenkyū お伽草子研究, 534. 340 Melissa McCormick, Tosa Mitsunobu and the Small Scroll in Medieval Japan (Seattle: University of Washington Press, 2009), 203. 180 inconsequentiality of details provided readers with the freedom to individually mold oneself with the narrative. The explication then added details about the sacred space, deities, and rites of the present- day site to translate the personal reading of the narrative into one specific to that place and deity. The degree of detail in the latter part depends on the narrative, and Itsukushima no honji’s relatively comprehensive account makes it an apt case for analysis. Some cases condense the explanation to a single paragraph, but this is still enough room to identify the characters with the deities of particular sites and encourage readers to shift their feelings towards the characters to their worship of the deity. 341 The honji-suijaku associations identifying each character as a deity enshrined at the site fixes the story to the specific sacred space, and also provides information about the multiple enshrined deities that would be of value to worshippers visiting the site. The listing of the dates of the major annual rites also urges readers to respond to the narrative with a pilgrimage to the site. I want to stress, though, that Itsukushima no honji does not present traveling to Itsukushima as a requirement. To the contrary, the transformation from narrative- based feelings of closeness to explication-based devotion works to transcend the boundary of spatial proximity to a site-specific deity. For visitors hearing the narrative recited at the sacred space itself, the past events could be seen taking place at the site that lays before one’s eyes. For readers far away from the site, the 341 For example, the last paragraph in Hashidate no honji ( 橋立の本地, late Muromachi period) abruptly shifts to explaining the two love-struck main characters as the two deities of the temples of Nariai and Kusenoto on the northern and southern ends of the bridge at Ama no hashidate 天の橋立. The narrator praises their power to save worshippers before lastly urging audiences to reflect on their good fortune at being able to hear the narrative. In this manner, Hashidate no honji similarly ends with a recommendation for the audience to reflect on the tale and its importance for them. "The Origins of Hashidate," in Monsters, Animals, and Other Worlds: A Collection of Short Medieval Japanese Tales, ed. R. Keller Kimbrough and Haruo Shirane (New York: Columbia University Press, 2018), 122-23. 181 reading of the protagonist’s travel to deification can become one’s own virtual pilgrimage to the sacred space. The explication of Itsukushima no honji, at the end of the text, equates the merit of its reading with that of ten pilgrimages to Itsukushima Shrine. It might be easy to take this as a strategy for proselytizers and fund-raising monks to attract audiences and compel them to make offerings or pilgrimages. Economic motivations likely spurned such thinking, but that does not mean that the producers or reciters of the narrative were unconcerned with the religious basis for the practices being promoted. Max Moerman examines similar passages of the religious efficacy of reading in the Kumano no honji and compares them to the sacred materiality of Buddhist scriptures and shrine mandalas (miya mandara 宮曼荼羅). 342 In these cases, the words of the Buddhist teachings are the body of Buddhist teachings, and the painting of the shrine environs is also the body of the deity itself. From another perspective, Charlotte Eubanks and Natalie Gummer argue for performative approaches to sutra reading, copying, and recitation as reproducing the sacred text through its embodiment in readers and listeners, and this approach can be applied to honji monogatari as well. 343 More than just acting as a talismanic body of the deity, the reading of honji monogatari manifests the deity and its sacred space in connection with the worshipper. This difference between the text as being the body of the deity and becoming the body of the deity for the devotee is not unlike the difference between shrine mandalas and pilgrimage 342 Moerman, Localizing Paradise: Kumano Pilgrimage and the Religious Landscape of Premodern Japan, 220-21. 343 Even though honji monogatari are not necessarily the words of the Buddha or the kami discussed in the narrative, the recounting of the god’s deification acts as a similar articulation that manifests the body of the deity to be received by the audience. Charlotte Eubanks, Miracles of Book and Body: Buddhist Textual Culture and Medieval Japan (Berkeley: University of California Press, 2011), 43-54. Natalie D. Gummer, "Listening to the Dharmabhā ṇaka: The Buddhist Preacher in and of the Sūtra of Utmost Golden Radiance," Journal of the American Academy of Religion 80, no. 1 (2012): 137-40. 182 mandalas (sankei mandara 参詣曼荼羅). Explaining the development from shrine mandala to pilgrimage mandala, Matsumoto Kōichi argued that the shrine mandala was seen as equivalent to the actual sacred space, and thus as a form of the deity itself. There was no room in the shrine mandala to include anything other than the shrine or temple buildings, surrounding environs, and possibly iconographic images of the deities, because the space of the deity metonymically served as the body of the deity itself. Afterwards, the images expanded to include miraculous events and worshippers, and developed towards the form of pilgrimage mandalas. As people came to be included within the images, it lessened the distance between the deities and devotees and humanized the space of the mandala. As a sacred space within a profane world, the honji-suijaku colored spatiality of shrine mandalas had already merged worldly and otherworldly spaces, but pilgrimage mandalas shifted the balance in favor of the worldly elements. 344 Building on Matsumoto’s research, Talia Andrei hypothesizes on different modes of viewing shrine and pilgrimage mandala. Andrei emphasizes the use of pilgrimage mandala within etoki image explication and kanjin fundraising practices to suggest that the audience’s hearing about the images was likely less an act of worship itself, but rather a means to virtually visit the site and then worship the deity. 345 Although there are reasons for considering the hearing of etoki image explication or honji mongatari explication as practices generating devotional sentiment, and thus as worship, Andrei’s distinction between the viewing of shrine mandala and pilgrimage mandala illuminates the performative viewing of pilgrimage, and by extension, honji monogatari as well. 344 Matsumoto juxtaposes the space of shrine mandala as the “space of the deities” (kami no kūkan 神の空間) and the space of pilgrimage mandala as the “space of people” (hito no kūkan 人の空間). Matsumoto Kōichi 松本公 —, "Shintō mandara no kosumoroji: Miya mandara no shōchō suru sekaikan ni tsuit e 神道曇茶羅のコスモ口ジー：宮曼荼羅の象徴する世界観について," in Kami to Hotoke no kosumologie 神と仏のコスモロジー, Nihon no Bukkyō (Kyoto: Hō zō kan, 1995), 153-54, 61. 345 Talia Andrei, "Mapping Sacred Spaces: Representations of Pleasure and Worship in Sankei Mandara" (Ph.D. diss., Columbia University, 2016), 40. 183 Illustrated versions of honji monogatari sometimes end with images of the sacred space, such as the 1656 Itsukushima no honji woodblock printed book (fig. 9). However, only a few cases, like the late Muromachi period Tenjin engi emaki 天神縁起絵巻 (fig. 12), depict pilgrims worshipping at the site. In contrast to the above discussion of pilgrimage mandalas, the performative reading of honji monogatari lessens the value of depicting a pilgrimage landscape full of people. Finishing the story and arriving at the image of the unpeopled sacred place can provide the impression of having traveled there through one’s reading. Including worshippers within the image can instead risk presenting the images as illustrations of the honji monogatari text. A picture of pilgrims visiting the site could represent the text’s explanation of the equivalency between the reading of honji monogatari and the act of pilgrimage, and consequentially weaken the feeling of individual closeness. For example, the Tenjin engi emaki ends with an image of monks reading a text, assumed to be the origin narrative itself, to pilgrims and asking for donations. Within the context of the narrative, Tokuda Kazuo reads the image as corresponding with the passages expounding the merits of reading and pilgrimages. 346 Similarly, different versions of Kibune no honji 貴船の本地 end with images of either a pilgrimage scene at the shrine (fig. 10) or a reading of the narrative (fig. 11). In contrast, the absence of people in the Itsukushima no honji book provides readers with the opportunity to reflect on the narrative and explication they read, reimagine themselves traveling to the site depicted in the picture, or worship the image within the book as the body of the deity. The text and images provide the 346 Tokuda Kazuo emphasizes the value of this scene for providing us with a picture of how origin narratives were performed at temples and shrines. He gives Kumano no honji, Kibune no honji, and Tenjin engi emaki as important exceptions visually presenting worshippers at the sites. He likewise reads these images as corresponding with the concluding passages expounding the merits of making pilgrimages to the site. Tokuda, Otogizōshi kenkyū お伽草子研究, 197, note 56. 184 reader with both guidance and freedom in regard to how to best move from the reception of the narrative to the giving of devotion. 185 Figure 9: Itsukushima on honji いつくしま御ほん地, 1656, National Diet Library Database. Figure 10: Kibune no honji きふねの本地, Undated, 3 rd volume, National Diet Library Database 186 Figure 12, Tenjin engi emaki 天神縁起絵巻, Late Muromachi period, Tenri Library. Image from Ishikawa Tō ru 石川透, Kanamitsu Keiko 金光桂子, Koida Tomoko 恋田知子, and Saitō Maori 齋藤真麻理, eds. Nara ehonshū 奈良絵本集, Shin Tenri Toshokan zenpon sō sho 23, 36. Figure 11: Kibune no honji きふねの本地, Early modern, Daitōkyū Memorial Library. Image from Shimazu Tadao 島津忠夫, ed., Monogatari sōshi II 物語草子 II, 362. 187 The performative reading/pilgrimage of honji monogatari as devotional practice culminates in the final sentence: “It is said, ‘The gods increase their power through the reverence of the people.’” 347 The merits of the reading and devotion are no longer something just for the reader, but for the deity as well. That is to say, the relationship between the deity and the devotee is two-sided. The closeness fostered from the honji monogatari narrative make the reader feel a sense of belonging with the deity that deepens one’s devotional practices, and in turn, this empowers the deity to manifest in close geographical and sentimental proximity to the individual worshipper reading and reflecting on the honji monogatari. Summary of Itsukushima no honji The discussion above surveyed the genre of honji monogatari as a whole and the rest of the chapter will support the above arguments through a close reading of Itsukushima no honji. Before beginning with the analysis, it will be helpful to summarize the plot in brief. In the eastern country of Tō jō in India, there was a king named Tō zen. He had one thousand consorts, but he despaired because he still did not have a son. He prayed to the buddhas and deities, and afterwards, one of the consorts became pregnant and gave birth to a son called King Senzai. When King Senzai was thirteen years old, he caught a glimpse of an image of a woman on a precious fan that had been passed down in his family for generations, and instantly 347 Similar passages appear in a number of other medieval texts, such as the 1232 Kamakura bakufu legal code, the Goseibai shikimoku 御成敗式目, the nō play Miwa 三輪, and the Enkyokushū 宴曲集, a pre-1296 collection of banquet songs and prose compiled by the Buddhist monk Myōkū 明空. All three of these cases include a second clause stating the dependence of the people on the benevolence of the gods. In the Enkyokushū for example, “Things like this are said: ‘The godliness of the gods is due to the reverence of the people. The humanity of the people is due to the benevolence of the gods.’” In this manner, we see the interdependent relationship between the gods and worshippers (i.e., the audience of texts like original form narratives). On Miwa and the Goseibai shikimoku see Uejima, "Kami and Buddhism in the Nō Miwa: Rethinkin g the Study of the Amalgamation of Kami and Buddhas (shinbutsu shūgō) " 34. For the passage in the Enkyokushū see Myōkū 明空, "Enkyokushū 宴曲集," in Zoku gunsho ruijū. Yūgi bu 続群書類従. 遊戯部, ed. Hanawa Hokinoichi 塙保己一 and Oota Toshiro 太田藤四郎 (Tokyo: Zoku Gunsho Ruijū Kanseikai, 1957), 84-85. 188 fell in love with her. His longing for her was so great that he became sick, and to make matters worse, he learned that the woman on the fan was the goddess Kichijō ten, who is not of this world. A confidante recommended that he instead turn his attention to Princess Ashibiki, known as the most beautiful person in the world. Hearing this, King Senzai sent his messenger, a five- colored crow, to her with a poem expressing his love. The crow faced a dangerous journey across stormy seas, and with fears of death, prayed to the dragon deities to provide him with a place to take shelter. Thankfully, a dragon came to the surface of the water and acted as an island for him to rest until the weather cleared. At last, the bird delivers his message, the Princess responds after some deliberation, and their correspondences lead Senzai to make his way to Saijō and marry her. He brings her back to his country of Tōjō three years later, but his thousand consorts become jealous of her beauty and devise a scheme to kill her. First, they pretend to be sick and have a report sent to King Tō zen stating that the only medicine for their sickness is in the country of Kiman, and that King Senzai must go and retrieve the medicine himself. Then while King Senzai is away, the consorts hire warriors to kill Princess Ashibiki on Mt. Asaka. Princess Ashibiki was seven months pregnant at the time, but she miraculously gives birth to a son before being killed. Even after her death, her body fills with milk so that her body can care for her child. Afterwards, the prince is raised and protected by the animals and deities of the mountain, until the dead Princess Ashibiki appears in the dreams of the son and King Senzai to tell them what happened. With the information she provides, King Senzai goes into the mountains and recovers their son, who he never knew he had. Next, father and son go a holy man in the country of Kaira, and with his help, are able to bring Princess Ashibiki back to life. They live happily for some time, until King Senzai falls in love with Princess Ashibiki’s younger sister when travelling 189 through her place of Sagara. It is then that Princess Ashibiki decides to leave King Senzai and their son, boards a flying carriage and ends up arriving in Western Japan. It is only at this point that the narrative connects with the story of Saeki Kuramoto found in other Itsukushima origin narrative. Saeki Kuramoto was exiled from Harima Province to Aki Province (the location of Itsukushima Shrine) after shooting a deer beloved by the sovereign. One day, he sees the sails of Princess Ashibiki’s boat, and ends up providing her with offerings and guiding her to the island of Itsukushima. Princess Ashibiki becomes the Itsukushima Deity, and a main hall is built for her on the island. Kuramoto sends a request by means of the five- colored crow for recognition of Itsukushima as part of Aki Province and as a sacred space, which is granted eagerly by the sovereign, Suiko Tennō . Although the narrative ends here, the narrator goes on to explain the honji-suijaku identities of the characters in the story, the dates of the annual rites at Itsukushima, and various connections between the narrative and the space of Itsukushima. Finally, the work concludes by praising the efficacy of reading this honji monogatari as equivalent to that of a pilgrimage and explains that a copy of this story was offered to Itsukushima Shrine by Taira no Kiyomori. Distance Makes the Hearts of Deities and Devotees Grow Closer The story pulls at the reader’s heartstrings. The bright story of true love and longing transforms into an emotional thriller full of jealousy, murder, rebirth, adultery, and deification. Princess Ashibiki is clearly a “suffering deity,” and the ordeals of her marriage could be read by female audiences in relation to their own worries and prayers. 348 Once we take a step back to 348 More specifically, the medieval ie marriage system led to reliance on the husband for economic and social support, and it fostered competitive and fraught relations between the wives of the same man. Additionally, the dangers of childbirth for both the mother and child would have made the story of Princess Ashibiki’s miraculous birth to a son especially relevant. 190 question the central importance of the suffering deity motif, though, we can begin to recognize further layers of meaning. In this section, I will argue for the importance of two related layers: (1) The affective power of the narrative works from themes of geographical distance and separation, and consequentially, presents emotional closeness as an alternative to spatial proximity. (2) The emotional response of the reader is not limited to a female readership focused solely on the deity’s sufferings. The longing of King Senzai and the devotion of Saeki Kuramoto also provide outlets for linking the audience’s feelings and sense of a close personal relationship with the Itsukushima Deity. The cache of India as a Buddhist sacred space with a foreign allure cannot be ignored when considering Itsukushima no honji’s setting in India. However, we must be wary of overemphasizing India’s centrality and Japan’s peripherality, or the movement of Ashibiki from India to Japan as an “Indianization” of Japan’s sacred spaces and deities. This is not to say that such arguments are problematic, but only to suggest that the relationship between India and Japan is worth considering beyond the framework of a periphery appropriating the power of the center. Examples of sacred peaks in India flying to Japan, or even becoming the islands of Japan as a whole, present Japan as equivalent to a trace (suijaku) of the original form (honji) of India, but honji monogatari demonstrate that priority is not always given to India or the original form. 349 There are many honji monogatari, like Itsukushima no honji and Kumano no honji, which involve travel from India, but there are also numerous cases based around travel to 349 Itō Satoshi aptly presents the stories of mountains flying from India to Japan a s an application of the honji suijaku discourse to the spaces of India and Japan, rather than to deities. Fabio Rambelli also argues for honji suijaku discourses as part of broader cultural practices overlaying aspects of Indian culture onto Japanese equivalents. See Itō , Shintō no keisei to chūsei shinwa 神道の形成と中世神話, 75-77; Fabio Rambelli, "The Idea of India (Tenjiku) in Pre-Modern Japan: Issues of Signification and Representation in the Buddhist Translation of Cultures," in Buddhism Across Asia: Networks of Material, Intellectual and Cultural Exchange, ed. Tansen Sen (Singapore: ISEAS–Yusof Ishak Institute, 2014), 262-63. 191 heavens or hells. 350 In both cases, the fixed ending of honji monogatari narratives with enshrinement at a specific sacred space in Japan makes travel into a fruitful mediator for connections between sacred spaces, deities, and worshippers. Seen through a Japan-India connection, the travel of the Itsukushima Deity from India to Japan transports the desirable attributes of Indian culture and sacrality to the space of Itsukushima. Then, the deification process makes the Itsukushima Deity into a kami ideally situated within Japanese space and culture. In addition to this connection, we must also consider how the narrative’s travel encourages personal connections with the deity through affective response. The idea of the kami as local to Japan narrows the cultural distance between the deity and the people, but the performance of the narrative is needed to personalize the relationship to a single deity and worshipper. Itsukushima no honji cannot vocalize the feelings of the deity towards an individual worshipper, because the narrative needs to be able to connect with each member of the audience. Instead, the emotional context of the narrative merges with a shared spatial belonging between the deity and the audience to affect readers and listeners with a similar feeling of connection with the deity. Itsukushima no honji begins with the conventional sangoku positioning of India as the 350 Murakami Manabu emphasizes the frequency of familial connections between gods of different Japanese sacred spaces within the Shintōshū to argue for the use of Shintōshū narratives by religious institutions in the Kanto area in eastern Japan to connect themselves with the sacred spaces and deities near the capital. Based on this interpretation, he suggests that the Shintōshū was likely produced outside of the capital. While possible, the argument rests on an assumption that religious institutions in the provinces would want to connect with those in the center, and fails to consider that sites near the capital might be interested in connecting with sites outside of the capital. Arguing to the contrary, Ariga Natsuki compares the frequent connections between Kasuga (in Nara, not far from the capital in Kyoto) and Kashima (peripheral location to the east, in modern day Ibaraki Prefecture) in Kasuga texts, and the lack of such associations in Kashima texts to argue that texts and discourses of religious groups from the center played a central role in the production of the Shintōshū. Murakami Manabu 村上学, "Shintō shū," in Hotoke to kami 仏と神 ed. Konno Tō ru 今野達, Satake Akihiro 佐竹昭広, and Ueda Shizuteru 上田閑照, Iwanami kō za Nihon bungaku to Bukkyō (Tokyo: Iwanami shoten, 1994), 111-19. Ariga Natsuki 有賀夏紀, "'Shintōshū' ni okeru Kasuga honji setsu no keisei 『神道集』における春日本地説の形成," Denshō bungaku kenkyū, no. 58 (2009): 27. 192 center of the Buddhist world and Japan as a small peripheral realm, but this is only the starting place. The narrative begins from concerns over the separation between India and Japan in order to close the distance later. The narrator begins: Now, when we speak of the one called the Itsukushima Deity, [she] manifested her traces in our country for the salvation of all sentient beings during the reign of Suiko Tennō , on the twenty first day of the sexegenary cycle, on the thirteenth day of the twelfth month of the fifth year of Tanjō (593), in the village of Tō ke, Sasai District, Aki Province, on the San’yō dō, in Japan. When one inquires about the original form of this deity, [the story goes as follows: [Within] the great realm [of India], middle [sized] realms, ten thousand smaller realms, and innumerable [other] realms scattered about like millet were created. Among them, there was a single realm. Its name was Tōjō and its king was called King Tōzen. 351 The center-periphery binary is apparent from the contrast between the first sentence’s situating of Itsukushima’s origins in Japan and the following situating of Itsukushima’s origin narrative within the sangoku Buddhist worldview and a specific kingdom in India. However, the spaces of Itsukushima and India are presented through distinct rhetoric and for different reasons. The first sentence on Japan provides a quick reference for fundamental information about Itsukushima, the date of its founding and its location. Also, the description of the location emphasizes the site’s area-specific locality by listing specific place names beginning from Japan and continuing to the province, district, and town. The place names could provide nearby residents with a sense of local belonging, recognizing their knowledge of details that would be alien to others. For 351 The manuscript also contains an annotation correcting and providing additional information here: “In India, there are sixteen great countries, five hundred middle countries, ten thousand small countries, innumerable countries scattered about like millet, and within them”. The Buddhist worldview described here closely resembles that in Gyō nen’s “Commentary on the Book of Precepts in the Brahma Net Sutra” ( 梵網戒本疏日珠鈔) which in turn, is citing the Sutra of Benevolent Kings (仁王経). Gyō nen writes, “The Sutra of Benevolent Kings says, ‘Sixteen great countries, five hundred middle countries, ten thousand small countries, innumerable countries scattered about like millet.’” ( 仁王經云。十六大國五百中國十千小國。無量粟散國) T. 2247 62.168c. Gyō nen accurately quotes the Sutra of Benevolent Kings, but he adds to it the phrase “innumerable countries scattered about like millet.” For comparison, see The Sutra of Benevolent Kings T. 235 8.832b. As Max Moerman has demonstrated, the portrayal of Japan as one of the “innumerable countries scattered about like millet” was widespread in Japanese literature of the tenth to fourteenth centuries. Moerman, D. Max, "From Sangoku to Bankoku: Japan and India in an Expanding World," in Beyond Sectarianism: New Horizons for Interdisciplinary Studies in Japanese Buddhism (Harvard University, Cambridge, MA 2012). "Itsukushima no honji 厳島の本地," 270-71. 193 those unfamiliar with the places, the names could instead serve as crucial information for beginning a pilgrimage (real or virtual) to Itsukushima. In fact, discrepancies in the district and town names in different manuscripts suggests that reciters or copyists of the narrative were sometimes unfamiliar with this level of specificity about Aki Province. 352 The specificity of Itsukushima’s location is juxtaposed with the vague descriptions of India. The second sentence provides numerical specificity concerning the makeup of the different realms in India, but the narrative only names the locations of the two protagonists, King Senzai in Tōjō (Eastern Castle, 東城) and Princess Ashibiki in Saijō (Western Castle, 西城). 353 These names do not refer to historical places, and instead, the directional markers in the names hint at the distance separating King Senzai and Princess Ashibiki. 354 The images of the Indian 352 For example, Sasai District is sometimes replaced with Saeki District (An’ei version, 1773), and the village of Tōke is sometimes instead given as Tokaki (Meireki version, 1656) Tokake (Matsumoto version), or Torakake, Nariki (An’ei version, 1773), or Kawai (Keiō and Itsu kushima repository versions). Historical changes in place names explains the shift from Sasai to Saeki, because the district of Saeki divided into Satō 佐東 and Sasai in the late Heian period and were then restored as the singular Saeki District in 1664. However, the slightly altered names for the village of Tōke 鳥翔 appear most to be a result of confusion over the reading of the characters or misremembering. On the division and unification of Saeki District, see Nihon kokugo daijiten 日本国語大辞典, (Shogakkan). For a listing of manuscripts with differences in the place names, see Asano Hideo 浅野日出男, Sugimoto Yoshinobu 杉本好伸, and Shimada Daisuke 島田大助, "Chūshaku: "Itsukushima Daimyō jin go engi" sono ichi 注釈：『厳島大明神御縁記』その一," in Kokubungaku nenjibetsu ronbunshū: Chūsei 国文学年次別論文集：中世, ed. Gakujutsu Bunken Fukyūkai 学術文献普及会 (Tokyo: Hō bun Shuppan, 1998), 554, note 4. 353 Tengu no dairi is another example of a Muromachi period narrative that also uses “Tōjō” as a general name for the place of Eastern India. Notably, the following sentences detail the other directions, but simply refer to them as “Western India” (nishi Tenjiku 西天竺) etc., suggesting that “Tōjō” and “Saijō” similarly re fer to a general area rather than a specific city or place. Keller Kimbrough likewise translates “Tōjō” as “eastern India.” See "Tengu no dairi 天狗の内裏," in Otogizōshi お伽草子, ed. Shimazu Hisamoto 島津久基 (Tokyo: Iwanami Shoten, 1936), 257; "The Palace of the Tengu (Tengu no dairi)," in Monsters, Animals, and Other Worlds: A Collection of Short Medieval Japanese Tales, ed. R. Keller Kimbrough and Haruo Shirane (New York: Columbia University Press, 2018), 129. 354 It is important to note that the naming of the spaces as Tōjō and Saijō, and the base narrative of falling in love with the beautiful woman in Saijō, did not originate with Itsukushima no honji but with Konjaku monogatarishū narrative 5.22 “Tale of Prince Senjōnin of Tōjō Mee ting the Woman Ajurō” 東城国皇子善生人、通阿就頭女語. Notably, Kumano no honji works from the same base narrative, but changes the place of Tōjō to Magadha (Jp. makada 摩訶陀), a central country in India for Buddhism. Itsukushima no honji also reveals its producer’s knowledge of specific place names in India by locating a diviner called by King Senzai at the border between Magadha and Vārā ṇasī (Jp. harana 波羅奈). This detail only appears in Itsukushima no honji. On the usage of Magadha in Kumano texts, see Matsumoto, Chūsei ni okeru honjimono no kenkyū 中世における本地物の研究, 21-27. For the Itsukushima no honji passage, see "Itsukushima no honji 厳島の本地," 277. 194 settings in illustrated versions of Itsukushima no honji also highlight the general foreignness of the space. Likely due to a lack of visual information about India, Indian buildings in Muromachi period images were painted like their equivalents in China. In the Itsukushima no honji images, such as the booklet from the Kōjō 高乗 collection at the National Institute of Japanese Literature, King Senzai’s home is presented in a Chinese style with the checked tile floors and curtains (fig. 13). 355 Unfortunately, it is not possible to examine how the environment changes later in the narrative, because the latter half of the manuscript no longer exists. That being said, it is still a useful example for considering how the foreign landscape comes to the forefront when depicting 355 Komatsu Kazuhiko and Fabio Rambelli similarly examine images of the Dragon Palace and India in medieval illustrated narratives to argue that the depicted other-world is presented in Chinese or Korean architectural styles. See Komatsu Kazuhiko 小松和彦, Ikai to Nihonjin: Emonogatari no sōzōryoku 異界と日本人 : 絵物語の想像力. (Tokyo: Kadokawa Shoten, 2003), 80; Rambelli, "The Idea of India (Tenjiku) in Pre-Modern Japan: Issues of Signification and Representation in the Buddhist Translation of Cultures," 277-79. Figure 13: Itsukushima engi 厳島縁起, Kōjō Bunko 高乗文庫. Image from Image from National Institute of Japanese Literature Database of Pre-Modern Japanse Works, coda 7. 195 an idealized court culture. India is presented as a heterotopic other world, and while its foreignness might excite the reader, it is always something far away. The distance between King Senzai in Eastern India and Princess Ashibiki in Western India is similarly imagined as insurmountable. The language of distance is used here to separate lovers rather than countries or religions. 356 Japanese Heian court literature stressed longing as an essential element of romance, and Itsukushima no honji applies the same motif to the scenes in India. 357 When King Senzai first falls in love with the image of the woman Kichijōten on the fan, he finds himself in an impossible situation of loving a goddess out of his world. A confidante mentions Princess Ashibiki’s beauty to him to bring his aspirations down from the heavens. This alignment of Kichijōten and Ashibiki’s beauty and their contrasting planes of existence hints at Ashibiki’s dual identities as a human woman and as the Itsukushima Deity. While relying on a degree of specialized knowledge, Ashibiki (as the Itsukushima Deity) could be identified as Kichijōten through their shared association with the Benzaiten deity. 358 This is a romantic metaphor for the honji-suijaku framework—Ashibiki is more accessible as a local trace (suijaku) because of her human form and experiences, but she is no different in essence from the woman on the fan that he fell in love with at first sight. The scene of falling in love with a painted woman on a fan occurs frequently enough in Muromachi period narratives and folktales that scholars named the motif “court women of 356 The lack of spatial particularities allows for the distance between Tōjō and Saijō to hint at the dis tance between Japan and India. To be clear, the narrative distinguishes Japan from Tōjō and Saijō, but the depictions of the movements between Tōj ō and Saijō resemble the later movement from India to Japan. This blurs the boundaries between the spaces of Japan and India, and more broadly, the worldly and otherworldly. 357 As Hank Glassman remarks in the case of The Tale of Mokuren (Mokuren no sōshi 目連の草子), the narrative combines an idealized foreign past of Buddhism in India with idealized cultural past of the Japanese court. Hank Glassman, "The Tale of Mokuren: A Translation of Mokuren-no-sōshi," Journal of Buddhist Literature 1 (1999): 123-24. 358 I.e., Itsukushima = Benzaiten, Kichijōten = Benzaiten, therefore Itsukushima = Kichijōten. Tanaka Takako 田中貴子, Gehō to aihō no chūsei 外法と愛法の中世 (Tokyo: Sunagoya Shobō, 1993), 23 -25. 196 portraits” (esugata nyōbō 絵姿女房). 359 The court woman on the fan is presented as out of reach, but the message to readers is not to just focus on the worldly gods instead of the lofty buddhas and bodhisattvas. Like any good romance, the reader anticipates the protagonist’s fight to overcome the odds and live happily ever after. 360 The reader follows from the perspective of the male protagonist to experience his longing, despair, hope, and eventually, fulfillment. These same sentiments can connect the audience and one’s devotion with the deity of the tale. The theme of love in separation also provides a clear parallel to the initial relationship between a worshipper and a deity, and the aspiration for coming together. As a love story that ends with tragedies of murder and adultery, the parallels do not continue from beginning to end. This is one possible reason why the explication shifts away from metaphors of love when addressing the audience directly. The beginning of their courtship, at least, provides an opportunity to suggest a romantic intimacy connecting the deity with the devotee. For example, Senzai’s first poem to Ashibiki expresses his feelings as a love that transcends the space and time of one lifetime: Yume ni dani / mada minu saki no / koishiki ha / umarenu saki no / chigiri narikeri 359 Matsumoto, Chūsei ni okeru honjimono no kenkyū 中世における本地物の研究, 107. Additionally, Yasuhara Makoto emphasizes that the episode involving the king falling in love with a woman on a fan does not appear in Kumano no honji, even though the first two sections of both stories are closely related. Yasuhara connects this with the uta-e illustrated cypress fan and the three smaller illustrated cypress fans given as offerings to Itsukushima, as well as multiple records of offering of women of the Taira clan to argue that connections were made between fans and Itsukushima and women. Yasuhara, Makoto 安原眞琴, "Ōgi bunka no ichi danmen: Ōgi denshō to Heike no onna tachi 扇文化の一断面：扇伝承と平家の女たち," in 'Heike Monogatari' no tenshō to saisei 「平家物語」の転生と再生, ed. Kazuaki Komine 小峯和明(Tokyo: Kasama Shoin, 2003), 381-84. 360 For example, the woman on the fan in Kibune no honji 貴船の本地 is similarly revealed to be from another world, but in this case, she is Princess Otohime 乙姫, the daughter of the king of the realm of demons. The male protagonist, Sadahira is told that he will never meet his love, that he has lost perspective on reality, and that he needs to forget about her. Instead he travels to the realm of demons to meet her and although they undergo a number of trials including her death, they eventually marry in the human realm and are enshrined together as the Kibune Deity and Marōdo Deity of Kibune Shrine. "Kibune no honji (Keiō gijuku toshokan zō ko shahon) きふねの本地（慶応義塾図書館蔵古写本）," in Muromachi jidai monogatari taisei 室町時代物語大成, ed. Yokoyama Shigeru 横山重 and Matsumoto Ryūshin 松本隆信 (Tokyo: Kadokawa Shoten, 1976), 48-49. 197 Before seeing you / not yet even in a dream / my dear love for you / I find it to be a bond / from before we had been born. 361 Just as worshippers of the Itsukushima Deity might not have seen the deity or visited the shrine before, that does not diminish the bond between them. Some worshippers might be skeptical about the likelihood of such a strong bond existing despite the distance, and Princess Ashibiki likewise harbors doubts at first. Receiving the letter, she thinks to herself, “This is a letter from King Senzai, [the king] of Jambudvīpa and all under heaven. Even if I [replied], there certainly should not be a bond [between us] in this lifetime.” 362 Calling him the King of Jambudvīpa , the continent where humans live according to Buddhist cosmology, she relates the distance between them to that between the worldly and the sacred. Strictly speaking, Jambudvīpa contains all of India, and Senzai is only king of Tōjō, but th e Buddhist terminology presents India as an other- world of Buddhism rather than the worldly setting of the tale. 363 Fortunately, her attendant convinces her of the propriety to send a reply, and their marriage proves that they did in fact share a bond. It might not be easy to connect with a faraway lover or a deity, but the distance 361 "Itsukushima no honji 厳島の本地," 273. The motif of longing for a lover who cannot be seen even in one’s dreams often appears in romantic waka, and in particular, resembles the language of poem 628 in the Shin chokusen wakashū (New Royal Anthology of Waka, 新勅撰和歌集, 1235): “yume ni dani / mada minu hito no / koishiki ha / sora ni shimeyuu / kokochi koso sure,” “How I long to see you, who appears not even in my dreams. I feel as if the flames in my heart infuse the sky.” Translation from Pat Fister, ed. Ama monzeki jiin no sekai: Kōjotachi no shinkō to gosho bunka 尼門跡寺院の世界 : 皇女たちの信仰と御所文化 / A Hidden Treasure: Treasures of the Japanese Imperial Convents (Tokyo: Sankei Shinbunsha, 2009), 233-34. 362 "Itsukushima no honji 厳島の本地," 273. Notably, the Shiramineji 白峰寺 version of Itsukushima no honji (an illustrated scroll manuscript called “Itsukushima”) simplifies the passage by removing the name Jambudvīpa . It says instead, “I hear that in the realm east of here called Tōjō there is a King Senzai, and [this] is a lett er from him.” "Itsukushima: Honkoku, gendaigo yaku いつくしま：翻刻・現代語訳," in Itsukushima shinkō jiten 厳島信仰事典, ed. Nozaka Motoyoshi 野坂元良 (Tokyo: Ebisu Kōshō Shu ppan, 2002), 457. 363 To provide another example, when the attendant of Princess Ashibiki sees the five-colored crow for the first time, she is reminded of the birds “[you] hear live in the realm of the Buddha: peacocks, phoenixes, kalaviṅka (karyōbin 迦陵頻), and garuḍa (konjichō 金翅鳥).” Like Senzai, the bird was from the Indian realm of Tōjō and not a Pure - land like realm of the Buddha, but the other-worldly imagination of Buddhist sacred space articulates the extraordinary nature of the five-colored crow. "Itsukushima no honji 厳島の本地," 273. 198 does not need to be seen as a hindrance. Separation and longing reveals the special nature of their bond. To provide a more explicit example before moving on, Kibune no honji ends with the protagonists married and enshrined together. This provides the opportunity to compare their love to the proper devotional practices of the reader. The narrator explains, “When this honji [monogatari] is revealed, and you listen to it, the [god] will appear as a kami deity out of love. Love is the beginning of the true way.” 364 We must note that the narrator goes on to promote the deity’s miraculous efficacy for answering prayers related to love. Nevertheless, the nominalization of love within these statements leaves the subject usefully ambiguous. One can interpret the deity manifesting out of love as referring either to the love between the worshipper and the deity, or the desired love of the worshipper for a partner. Either way, we can see the religious significance of the narrative as something performed and felt, and not just as a strategy for ensnaring audiences. Unifying the Travel of the Deity and the Devotee to Itsukushima The close personal bond between the audience and the deity continues after the love between Senzai and Ashibiki comes to an end. The death of their marriage becomes the impetus for Ashibiki’s travel to Japan, and the transition towards a more standard telling of the Itsukushima Shrine origin narrative. Within this discussion of the founding of Itsukushima Shrine, the head priest Saeki Kuramoto 佐伯蔵本 takes the place of King Senzai as a character for the audience to associate with and emulate. The connection between Kuramoto and Ashibiki 364 The last sentences of the explication emphasize the connection between the manifestation of the god and love: “Therefore, when one receives the origins of this deity, love is the proper action for people. It is out of love that the god appears as a kami deity. Have no doubt and believe this.” "Kibune no honji (Keiō gijuku to shokan zō ko shahon) きふねの本地（慶応義塾図書館蔵古写本）," 67-68. 199 manifests when they meet in the Seto Inland Sea near the shrine’s location on the island of Miyajima. Itsukushima no honji similarly encourages the reader to connect one’s individual travel to Itsukushima Shrine, either through pilgrimage or through reading, with the original arrival of Itsukushima Deity and Kuramoto at Miyajima. In this sense, the reader reconceives the space of Itsukushima Shrine and the deity in association with oneself, and formalizes the personal relationship with Itsukushima by reproducing Kuramoto’s worship of the deity. The spatial separation in the first part of the story pulls on the audience’s heart strings, while the latter part’s movements to Itsukushima pulls the audience and Kuramoto close to the deity and her sacred space. Itsukushima no honji emphasizes the worship of the deity as fundamental to the sacred nature of the deity and her space. The text never explicitly says when Ashibiki is deified as the Itsukushima Deity, but Kuramoto’s behavior towards her shows that she is a god. This contrasts with earlier fourteenth century Itsukushima Shrine origin narratives, including the 1346 illustrated scroll version of Itsukushima no honji, which instead highlight Itsukushima’s role as a protector of the sovereign and the realm of Japan. Both the early fourteenth century Itsukushima Daimyōjin nikki and the 1346 illustrated scroll manuscript of Itsukushima no honji call the island Ongashima (Celebratory Island 御賀島) before the deity comes and renames the site Itsukushima. The Itsukushima Daimyōjin nikki explains the name Ongashima as originating from when the first sovereign of Japan, Jinmu Tennō, came to the island and watched celebratory dances. 365 These dances were performed as prayers for the long life of the sovereign, and 365 Kanazawa Bunko, 金沢文庫, ed. Kanazawa Bunko no chūsei shintō shiryō 金沢文庫の中世神道資料 (Yokohama: Yokohama kenritsu Kanazawa Bunko, 1996), 52. 200 therefore, their location at Itsukushima would connect the deity as a protector of the sovereign’s authority and long reign. 366 The 1346 version of Itsukushima no honji also uses the name Ongashima and the image of the Itsukushima Deity as a protector of the realm, but these details are changed in later manuscripts. Notably, the 1346 text omits the connection between the name Ongashima and Jinmu Tennō, simply providing the name without any background information. This allows the text to balance Itsukushima’s protection for both the realm and Aki Province. The Itsukushima Deity introduces herself to Kuramoto in this version saying, “I am the protector of our realm and this province. I am called the Itsukushima Deity. You should spread word to everybody from the king of the realm to the people of the districts of this province.” 367 For comparison, the later Keiō version gives the following introduction, “I am the princess of King Ten’ichi from the country of Saijō in Southern India. My coming to the island of Japan is for the sake of saving all sentient beings.” 368 The Keiō version shifts the object of the deity’s aid to the people of Japan, instead of the sovereign or the province. If the full text of the 1346 narrative was extant to analyze, it might be possible to situate the narrative in an intermediate position between the Itsukushima Daimyōjin nikki’s religious-political presentation of Itsukushima’s importance to the court and the focus on personal connections in later versions of Itsukushima no honji. We cannot say whether this was a result of referencing and copying from earlier accounts of the Itsukushima origin narrative, or intentions to be able to present the 1346 narrative to political and popular 366 For more information on the celebratory dance (onga no mai 御賀の舞) and royal authority, see Toyonaga Satomi 豊永聡美, "Chūsei ōken to bugaku: Onga no bugaku wo chūshin ni 中世王権と舞楽：御賀の舞楽を中心に," in Chūsei no geinō to bungei 中世の芸能と文芸, ed. Kobayashi Kenji 小林健二 (Tokyo: Chikurinsha, 2012). 367 For a transcription of the extant portions of the 1346 Itsukushima no engi and its mention of Ongashima, see "Kaidai: Itsukushima no engi 解題：いつくしまのゑんぎ," in Muromachi jidai monogatari taisei 室町時代物語大成, ed. Yokoyama Shigeru 横山重 and Matsumoto Ryūshin 松本隆信 (Tokyo: Kadokawa Shoten, 1973), 382. 368 "Itsukushima no honji 厳島の本地," 291. 201 audiences alike. It is useful to think of the narrative like a play script that could be followed or changed depending on the audience. From this perspective, it would make sense for the fourteenth century Itsukushima no honji to carry the same religio-political connotations as other contemporary Itsukushima origin narratives, while also taking advantage of the narrative emphasis of the honji monogatari narrative to present the work in a more personal manner for popular audiences. The Keiō version of Itsukushima no honji (and later texts) replaces the name Ongashima with Kuromasu 黒益 to connect the origins of the sacred space with the act of worship, first by Kuramoto, and then by later worshippers. The name can be translated as “to increasingly blacken,” and refers to the soot produced while cooking offerings for a deity. 369 Itsukushima no honji is not the only Itsukushima origin narrative to give weight to Kuramoto’s worship of the deity, but the name change adds further emphasis. The fourteenth century origin narrative Itsukushima Daimyōjin nikki details Kuramoto making an offering of rice to the Itsukushima Deity when they first meet, and it resembles the description of the same episode in Itsukushima no honji. The Itsukushima Deity explains that she is hungry after her long voyage and asks him for food. The request allows the deity to instruct Kuramoto on the proper form for a food offering, telling him how to prepare the rice, and how to place the rice for each of the thirty-three deities enshrined at Itsukushima Shrine. Despite the similarities in the content, there are notable differences. The 1346 scroll presents a succession of events beginning with the food offering, to becoming her follower, to 369 My interpretation of Kuromasu is based on Aoki Kigen’s explanation of the term kuromasu within the tenth century Izumo kuni no miyatsuko no kamuyogoto. Aoki Kigen 青木紀元, "‘Ama no mikahi’ ‘Ama no mikawa’: Izumo kuni no miyatsuko no kamuyogoto no mondai ten 「天乃美賀秘」「天乃瓸和」-- 出雲国造神賀詞の問題点," in Shintō oyobi shintō shi: Nishida Nagao Hakushi tsuitō ronbunshū 神道及び神道史：西田長男博士追悼論文集 (Tokyo: Meicho Fukyūkai, 1987), 124-25. 202 guiding her around the seven shores of the island to help her decide the location of her shrine. 370 The offerings formalize their relationship as deity and priest, and provides a model of offerings for later followers, and then proceed to establish the shrine. The later Keiō Itsukushima no honji’s account of this episode delays the giving of Kuramoto’s offerings to strengthen the link between the offerings and the making of Itsukushima Shrine. In order to maintain the order of events while moving the location of the offerings, the Keiō version results with an odd sequence of washing the rice, traveling to the island, and then making the actually offerings. 371 There would be little reason to complicate the order of events except to align with the changing of the island’s name to Kuromasu, as the sacred island where offerings are made to the deity. Ritual Re-Production of the Deity’s Travels Both Itsukushima Daimyōjin nikki and Itsukushima no honji (both the 1346 and Keiō versions) stress the scene of Kuramoto guiding the Itsukushima Deity around the nearby islands until she chooses which island to make her own. Itsukushima no honji’s larger narrative revolves around travel—between kingdoms in India, to Japan, and finally, to Itsukushima. This larger context assists the reader in imagining one’s own reading or one’s pilgrimage as equivalent to the travel of the deity and Kuramoto to Itsukushima. 370 While the Itsukushima Daimyōjin nikki refers to Kuramoto as her guide (sendatsu 先達), emphasizing his role in showing her the island that becomes her home, the 1346 version of Itsukushima no honji calls Kuramoto her follower (otomobito 御供人), literally, “one who makes offerings.” Kanazawa Bunko 金沢文庫, ed. Kanazawa Bunko no chūsei shintō shiryō 金沢文庫の中世神道資料 (Yokohama: Yokohama kenritsu Kanazawa Bunko, 1996), 52; "Kaidai: Itsukushima no engi 解題：いつくしまのゑんぎ," in Muromachi jidai monogatari taisei 室町時代物語大成, ed. Yokoyama Shigeru 横山重 and Matsumoto Ryūshin 松本隆信 (Tokyo: Kadokawa Shoten, 1973), 381. 371 "Itsukushima no honji 厳島の本地," 291. 203 Along with the change in name from Ongashima to Kuromasu, Itsukushima no honji works to align the arrival at Itsukushima with the spaces traveled by Princess Ashibiki. The scenes prior to the meeting of Kuramoto and the Itsukushima Deity detail both of their travels. Both the deity and Kuramoto travel south to arrive in Aki Province. The deity goes from Mt. Ishizuchi (in modern day Ehime Prefecture) and Kuramoto goes down to the town of Ō take after being exiled from Harima Province (fig. 14). Additionally, the Itsukushima Deity’s first sight of the island of Itsukushima makes her recall her home in India. Then, the narration immediately zooms in from the wide view of India and Itsukushima to a regional perspective incorporating place names that would only be familiar to actual visitors. The narrator sets the scene: At this Kuromasu island, she looked at its appearance and said, ‘When one looks at the appearance of the hilly island of India, it resembles this island.’ In the northern sea, at Michiku (misspelling of Mikasa) shore, she left her traces, and established the main hall along with corridors of one hundred and eight ken. 372 Moving back and forth between comparisons with India and regional place names, Ashibiki’s perspective overlays translocal and regional frameworks for the space of Itskushima. The passage invites the reader to experience the deity’s discovery of Itsukushima, and her naming of 372 Ibid. 204 the island as Itsukushima (literally, awe-inspiring island) after seeing it for the first time and being impressed by its beauty. The devotional performance of their travels to Itsukushima was not limited to the reception of the Itsukushima no honji narrative and was also incorporated within the annual rites of the kangensai (rite of wind and string [instruments] 管弦祭) and the shima meguri (island circumambulation 島廻り). Both rites serve as offerings to the Itsukushima Deity through the ritual reproduction of the original travel, and the accompaniment of prayers and music. The ritualized movement allows for the continued performance and maintenance of Itsukushima’s spatial and sacred identities. Significantly, this is not limited to the individual space of Itsukushima Shrine, and includes other sacred spaces connected with Itsukushima Shrine, both within and outside of the island. In this manner, the honji monogatari narrative and these Figure 14: Map of the Movements of Saeki Kuramoto and the Itsukushima Deity. 205 associated rituals together provide worshippers with opportunities to connect themselves with Itsukushima and feel a sense of belonging with the site and its deity. Today, the kangensai is the most famous ceremony at Itsukushima Shrine, and it takes place annually on the seventeenth day of the sixth month of the lunar calendar. It is a rite wherein the Itsukushima Deity is moved on a palanquin (mikoshi 神輿) to a boat and taken to subsidiary shrines enshrining the Itsukushima Deity while music is played on the boat for the deity’s pleasure. Conventional explanations of the rite trace it back to Taira no Kiyomori transmitting Heian court culture from the capital to Itsukushima. 373 However, this was likely a later explanation designed to highlight the connection between Itsukushima and Kiyomori. Looking to historical documents, other scholars have found texts showing the performance of the rite at the request of patrons in the Muromachi period (1392-1573), and they estimate the dating of the kangensai as an annual rite as pre-1541 and 1561 as the latest possible date. 374 373 Kitani Masamitsu 木谷昌光, "Itsukushima no tokushu shinji nit suite: Kanmiso kenjō shiki, oshima meguri shiki, kangensai 厳島の特殊神事について：神衣献上式･御島廻式･管絃祭," Geibi chihōshi kenkyū 芸備地方史硏究. 207 (1997): 17. 374 Harada Yoshiko describes the kangensai as a rite that has continued since the Heian period, but only provides a Muromachi period list of the annual rites at Itsukushima as historical evidence. Matsui Teruaki instead relies on the dates within historical documents, finding 1561 as the earliest verifiable date for the performance of the kangensai, but an earlier text mentioning a rite on the night of the seventeenth day of the sixth month during the time of the Ōuchi’s reign leads him to date the rit e to 1541 at the latest. Harada Yoshiko 原田佳子, Itsukushima no sairei to geinō no kenkyū 厳島の祭礼と芸能の研究 (Tokyo: Fuyō Shobō Shuppan, 2010), 63-68, 71; Teruaki 松井輝昭 Matsui, "Sengoku daimyō Mōri shi to Itsukushima jinja: Tanamori Fusaaki no katsudo wo tegakari toshite 戦国大名毛利氏と厳島神社：棚守房顕の活動を手掛かりとして," in Mōri Motonari to chiiki shakai 毛利元就と地域社会, ed. Kishida Hiroshi 岸田裕之 (Hiroshima: Chūgoku Shinbunsha, 2007), 144-45. 206 The shima meguri rite is also based around travel to subsidiary shrines of Itsukushima, but specifically the shrines at seven shores encircling the island (fig. 15). The rite can be traced back to 1435, and appears to have been performed primarily at the request of patrons before becoming an annual rite by the late sixteenth century. 375 The set timing of the annual rites enabled the shrine workers and patron to make elaborate preparations and for devotees to schedule pilgrimages with the rites.. The popularity of the shima-meguri rite, in fact, might have led to the building of a new residence for the shrine family on the island of Miyajima, rather than across the shore as was the case previously. A 1435 dated letter from the head priest, Fujiwara no 375 More specifically, Ōchi Tokuko presents historical sources showing that the shima meguri was performed as an annual rite between 1572-1600. Ōchi Tokuko 大知徳子, "Mōri Terumoto shojō to oshima meguri 毛利輝元書状と御嶋廻," Miyajimagaku senta nenpō 宮島学センター年報 1 (2010): 44, 46-47. Figure 15: Itsukushima zue 厳島図絵, Vol 1, 1842, Published by Yonamiyaihei. Names of all the shores are included in the original, but the author added the names of the seven shores of the shima-meguri in English and the route around the island. National Diet Library Database. 207 Chikafuji 藤原親藤, mentions the requests for shima-meguri offerings before going on express his gratitude for a shrine residence being built on the island. 376 The narrator of Itsukushima no honji mentions the kangensai rite alongside the ceremonies of the third and ninth months as one of the three central annual rites at Itsukushima. Providing the date of each rite, it promoted pilgrimages coinciding with the rites, and helped to attract large crowds to come and participate in the rites and the festivities. There are few extant medieval sources to describe the scene of Itsukushima during the time of the kangensai ceremony, but early modern sources present the island filling with pilgrims, merchants, and entertainment. An 1840 edition of the Nihon meisho kyūseki angya sumō (“Sumo Contest of Famous Scenic, Historic, and Pilgrimage Places,” 日本名所旧跡行脚数望) ranks the Itsukushima kangensai as the third best shrine ritual in Japan, behind the Gion’e rite of Gion shrine in Kyoto, and the Tenmansai rite in Osaka. Diary entrees of kangensai pilgrims support this high ranking with descriptions of the market built for the festival as full of pilgrims from various provinces, merchants, and entertainers. 377 The Fusaaki oboegaki 房顕覚書, written by the Itsukushima priest Tanamori Fusaaki ( 棚守房顕, 1495-1590) in 1580, discusses how the preparations in the days leading up to the kangensai brought together a diverse group of people including low ranking courtiers (jige 地下), members of the shogunal guard (hōkōshū 奉公衆), merchants, musicians, ritualists, and members of the shrine family, with up to three hundred 376 Doc. 184 in "Itsukushima Nosaka monjo 厳島野坂文書," in Hiroshima kenshi: Kodai chūsei shiryō hen 広島県史. 古代中世資料編, ed. Hiroshima-ken 広島県 (Hiroshima: Hiroshima, 1978); Matsuoka Hisato 松岡久人, Aki Itsukushimasha 安芸厳島社 (Kyoto: Hō zō kan, 1986), 160-51. 377 Takahashi Shūzō 高橋修三, "Itsukushima no rekishi wo megutte: Ichi, shibai, soshite tomikuji 厳島の歴史をめぐって：市・芝居、そして富籤," in Nihon sankei e no izanai: Matsushima, Amanohashidate, Itsukushima 日本三景への誘い : 松島・天橋立・厳島, ed. Shimao Arata 島尾新 and Hasegawa Seiichi 長谷川成一 (Osaka: Seibundō Shuppan, 2007), 109-11. 208 people called to the island on certain days. 378 The officials at Itsukushima recognized the importance of the event, and the need to prepare for the crowds that would cross the shore on the day of the ceremony. The bustling environment of the shrine at the time of the ceremonies would make for an enjoyable and memorable experience, and this feeling could carry over to individual worship. It was clear that this was no ordinary ceremony, and therefore, worshippers could feel closer to the deity than ever. There are few medieval sources revealing the experience of the kangensai or shima meguri rites, but there are sources describing the masses of people who would travel long distances to come for the annual rites of the third and ninth months, and the commercial markets that would be built for them. 379 For example, the diary of the Shōkokuji abbot Zuikei Shūhō ( 瑞渓周鳳, 1391-1473) records how a blind biwa lute performer (zatō 座頭) described the ninth month rite (after summarizing the Itsukushima origin narrative) to him: There is a ceremony at Itsukushima on the thirteenth day of the ninth month, and year after year [people] from the many provinces make pilgrimages to attend. Moreover, ninety thousand or one hundred thousand ships come together. Therefore, along the corridors [of the shrine] the hustling and bustling of people is incessant, and there is not even a single person taking a relaxed stroll. Certainly, the Deity likes the lively clamor. 380 The location of Itsukushima in the Seto Inland Sea made it a convenient location for travelers, and it is easy to imagine how the shores would fill up with boats as pilgrims gathered for the rites. The blind performer’s presentation of the ceremony suggests just how striking the bustling 378 Matsuoka Hisato 松岡久人, "Itsukushima monzenchō no keisei 厳島門前町の形成," in Seto Naikai chiiki no shakaishiteki kenkyū 瀨戶內海地域の社會史的研究, ed. Uozumi Sō gorō 魚澄惣五郎 (Kyoto: Yanagihara Shoten, 1952), 77; Tanamori Fusaaki 棚守房顕, "Fusaaki oboegaki 房顕覚書," in Hiroshima kenshi: Kodai chūsei shiryō hen 広島県史. 古代中世資料編, ed. Hiroshima-ken 広島県 (Hiroshima: Hiroshima, 1978), 1131. 379 For a detailed historical analysis of the economic significance of the annual rites at Itsukushima, including mentions of merchant pilgrims coming from Hakata, Sakai 堺, and Iyo Province in medieval period documents, see Matsuoka, "Itsukushima monzenchō no keisei 厳島門前町の形成." 380 Zuikei Shūhō 瑞溪周, Gaun nikkenroku batsuyū 卧雲日件錄抜尤, ed. Myō an 惟高妙安 Ikō , Dai Nihon kokiroku (Tokyo: Iwanami Shoten, 1961), 13. 209 atmosphere would be for the pilgrims who gathered together from near and far. Even the Itsukushima Deity herself is given as enjoying the lively environment. Visitors could experience being part of the ceremony by traveling to Itsukushima and joining the throngs of people. 381 In this manner, the experience of pilgrims could merge with the imagined experience of the Itsukushima Deity in traveling the seas to arrive at Itsukushima, and potentially, enjoying the festivities and rites as well. The explication of Itsukushima no honji only mentions the kangensai rite and connects the rite with the narrative. The brevity of the passage opens opportunities for the narrative and the rite to add to one another while still not being restricted by what is said or done in the other. The narrative never makes a direct connection between the travel to Itsukushima and the kangensai rite, and instead identifies the music performed during the rite as “the music played in the country of Tōjō to console King Senzai when he grew lovesick.” 382 The description tells the audience little about the rite, aside from possibly making the music sound foreign. Instead, it reminds the audience to recall the narrative and link one’s personal and emotional response to the tale with the rite itself. It is up to the audience to realize the connections between the travel in the narrative, the rite, and one’s own experiences. These connections would be clear enough, though, for those who are able to hear the tale and watch the rite. It is also possible that sermons or recitations of the Itsukushima no honji narrative performed at Itsukushima before or after the kangensai rite could expand on the connection in greater detail than that written in the manuscripts. 381 The Mōri family and their estates provided signific ant patronage for the kangensai and shima-meguri rites at Itsukushima, which likely served both religious and political purposes. See Matsui, "Sengoku daimyō Mōri shi to Itsukushima jinja: Tanamori Fusaaki no katsudo wo tegakari toshite 戦国大名毛利氏と厳島神社：棚守房顕の活動を手掛かりとして," 143-50; Ōchi, "Mōri Terumoto shojō to oshima meguri 毛利輝元書状と御嶋廻," 45-48. 382 "Itsukushima no honji 厳島の本地," 292. 210 In the case of the shima-meguri ritual, it is never mentioned by name within Itsukushima no honji, but the circling of the island by boat calls to mind Saeki Kuramoto guiding the Itsukushima Deity around possible island locations for her shrine. The 1346 version of Itsukushima no honji mentions the “seven islands” visited by Saeki Kuramoto and the deity, which can be taken as a reference to the “seven shores” visited during the shima-meguri rite. 383 Additionally, the contemporary fourteenth century Itsukushima Daimyōjin nikki uses the term “shima-meguri” to refer to Kuramoto’s act of guiding the Itsukushima Deity to the island of Itsukushima, and the text continues by listing the names of the seven shores. 384 In contrast, the later Keiō version of Itsukushima no honji vaguely says, “[Kuramoto] directed the boat towards Kuromasu island, took a search, and when they arrived at Kuromasu...” 385 Although it is possible that this omission of the seven shores is an unintended result from emphasizing the island as the site of worship, we should keep in mind the benefits for leaving details out. 386 The relations 383 There are at least two different explanations for the term “seven islands:” (1) as the seven shores of the island of Miyajima (Mikasa; Sugiura; Takanosu; Koshiboso; Aonori; Yamashiro; and Mitoko), or (2) as seven islands near Miyajima in the Seto Inland Sea (including Nōmijima 能美島; Etajima 江田島; Kamagarijima 蒲刈島; Iwashijima 岩子島; and Urasaki 浦崎). The second meaning associated with seven islands in the Seto Inland Sea can be found in a journal article by the folklore research group on Hiroshima folktales (Densetsu Hiroshima, minzoku no kai 伝説広島・民族の会), but the article does not provide any sources for this association. The ambiguity of the term “seven islands” to refer to either the “seven shores” of the island of Miyajima, or seven islands in the Seto Inland sea area can be considered in terms of the multiple ways in which the space associated with Itsukushima can be drawn in multiple ways—as limited to the island of Miyajima, extending to include nearby islands in the Seto Inland Sea area, or as seen in chapter two, including Benzaiten sites spread throughout Japan). See "Kaidai: Itsukushima no engi 解題：いつくしまのゑんぎ," 382; Kojima Tsunenari 小島常也, Itsukushima michishiba no ki 嚴島道芝記 (Miyajima, Hiroshima: Miyajima-chō , 1971), 26; "Shosha kongen ki 諸社根元記," in Jingi zensho 神祇全書, ed. Saeki Ariyoshi 佐伯有義 (Tokyo: Kō ten Kō kyūjo, Jingū Hō saikai, 1906), 21; Hiroshima Minzoku no Kai 広島民俗の会, "Ichikishima hime to Itsukushima jinja 市杵島姫と厳島神社," Hiroshima-ken minzoku shiryō 6 (1973): 60. 384 The names of the seven shores in the Itsukushima Daimyōjin nikki differ slightly from later lists of the shores. More specifically, only the Itsukushima Daimyōjin nikki names the shores of Mukau, Ari, Oishima, Yofue, and Komori while the remaining two shores of Aonori (here read as Aokoke 青海苔浦) and Mikasa correspond with later texts. Kanazawa Bunko, Kanazawa Bunko no chūsei shintō shiryō 金沢文庫の中世神道資料, 52. 385 "Itsukushima no honji 厳島の本地," 291. 386 There are two early modern woodblock prints of the origins of Itsukushima (Itsukushima yurai いつくしま由来) from 1778 and 1831 that list the seven shores of the shima-meguri rite. However, this information is included not incorporated into the origin narrative itself. The detailed information in these texts resembles that of early modern Itsukushima gazetteers like Itsukushima michishiba no ki or Itsukushima zue more than the more narrative focused 211 constructed through the performance of circling around the island would not be difficult for visitors to deduce, and shrine officials and preachers would likely be there to provide further clarification. While there is detailed information in early modern woodblock gazetteers on Itsukushima, like the 1842 Itsukushima zue 厳島図絵 and the 1702 Itsukushima michishiba no ki 厳島道芝記, 387 the Shoshin kongenshō 諸神根元抄 Yoshida family anthology of writings on the myriad gods provides us with a late medieval (post-1562) reference work linking the travel of the Itsukushima Deity with the shima-meguri rite. 388 With little more than a paragraph on Itsukushima, the Shoshin kongenshō gives a summary of essential information on Itsukushima including its location, the identities of its deities, and a brief account of the site’s origins. Here, the origin narrative is simplified to Saeki Kuramoto meeting the Itsukushima Deity, making an offering of rice to each of the deities of Itsukushima Shrine, and “encircling the seven islands” to guide the deity to the place on Miyajima where she decides to enshrine herself. 389 Given the condensed form of the information within a larger compilation, it is likely that similar Itsukushima no honji. Senoo Yoshinobu 妹尾好信, "Ryaku engi ‘Itsukushima yurai’ kō: Fuku, eiin, to honkoku 略縁起『いつくしま由来』考：附・影印と翻刻," Itsukushima kenkyū 3 (2007): 14, 17. 387 For example, the Itsukushima michishiba no ki explicitly ties the shima-meguri rite with the movements of the Itsukushima Deity when she first came to the island, saying, “The [rite] called the shima-meguri is linked with [when] the three deities graciously descended to this island, and circled around the shores to look for the land [where they should be] enshrined.” Kojima, Itsukushima michishiba no ki 嚴島道芝記, 117. 388 Shimasue Kiyoshi argues that the Shoshin kongenshō was likely written by Yoshida Kanemigi ( 吉田兼右, 1516- 1573) from previous Yoshida family writings by Yoshida Kaneatsu ( 吉田兼敦, 1368-1408), Kanetomo (吉田兼倶, 1435-1511), and Kanemitsu 吉田兼満 after 1556. Shimasue Kiyoshi 島居清, "Shoshin kongenshō, Shoshinki, Shosha kongenki ni tsuite (2): Shoshinki to Shoshin kongenshō 諸神根源抄・諸神記・諸社根元記に就いて(2) ：諸神記と諸神根源抄," Biburia, Tenri toshokan hō 31 (1965): 60. 389 "Shoshin kongenshō 諸神根元抄," (National Institute of Japanese Literature: Hizen Shimabara Matsudaira bunko 肥前島原松平文庫), coma 60. While this text has not been transcribed, the same entry on Itsukushima can be found in the Shosha kongenki 諸社根元記 (Dating estimated to around 1680 by Shimasue Kiyoshi). See "Shosha kongen ki 諸社根元記," 21. 212 information circulated elsewhere and might have been available to visitors at Itsukushima Shrine. 390 Origin narratives might have been a sole source of information on a sacred site for some distant worshippers and scholarly monks unable to make the pilgrimage to Itsukushima, but pilgrims would have access to all the sights before their eyes, as well as the information explained at the shrine. As gazetteers and other wood block print publications became more accessible, even individuals far away would have had access to other texts to supplement the information in the origin narratives. Still, the audience of Itsukushima no honji had access to a stronger, personal, understanding of the narrative and the rites, bringing together one’s knowledge and experiences of the site. Honji monogatari as Narrative and Knowledge The narratival bent of honji monogatari positions them towards popular audiences, but that does not mean that scholarly monks and intellectuals ignored the tales. The Itsukushima origin narrative manuscript copied by Yoshida Kanemigi ( 吉田兼右, 1516-1573), Itsukushima shake engi 厳島社家縁起 provides a rare glimpse into his intellectual interest in Itsukushima no honji. The text condenses the plot and explanative information from Itsukushima no honji, and instead focuses on its identification of the deities enshrined at Itsukushima Shrine as Buddhist deities, and the locations of the deities on the island. The narrative summary is shortened to the 390 For a later example, the two-page manuscript Itsukushima Benzaiten sha kakitsuke 厳島弁才天社書付 lists essential information about Itsukushima Shrine and includes the names of the seven shores. The text can be dated to post 1739, because the text includes mentions the great torii gate being repaired at the end of that year. "Itsukushima Benzaitensha kakitsuke 厳島弁才天社書付," (National Institute of Japanese Literature: Ogata Tamotsu Collection 大方保). 213 extent that it would leave anyone interested in the content unsatisfied. Given its brevity, the summary is translated here in full. Concerning the origins of Itsukushima: long ago in India, there was a country called Tōjō. [In Tōjō], there was a king whose name was King [Tōzen] (rip in paper removing name from document). His [son] the prince was called King Senzai. (Rip in page removing approximately three characters.) The princess of the great king of the country of Saijō was called Ashibiki. [They] married. Their messenger (intermediary) was a bird. Because the bird [shone] golden in five colors, it is called Five [Colored] Bird. This Princess is the deity of this shrine. King Senzai is the Marōdo D eity. 391 For Yoshida Kanemigi, the Yoshida Shinto head priest, this description would have been enough to relate the narrative to their knowledge of Itsukushima Shrine (i.e., the honji suijaku identification of the deities at Itsukushima and their relations with one another). In addition to this summary of information from Itsukushima no honji, Kanemigi’s Itsukushima shake engi also includes a word-for-word copy of the Itsukushima origin narrative from the Shōbōrinzō Prince Shōtoku tale ( 正法輪蔵, 1317), and a list of seven secret places on the island of Itsukushima recorded from an oral transmission (kuden 口伝). From this perspective, we can see the Itsukushima no honji segment as supplementary information not found in the other Shōbōrinzō based origin narrative. The Itsukushima shake engi follows the summary of the Itsukushima no honji narrative with an equally long list of its honji suijaku associations. He treated the associations between the gods of Itsukushima, the Buddhist deities, and the characters as vital information. Notably, the number of gods mentioned and the descriptions of their locations far exceed that found in any other extant version. The list of honji-suijaku associations even continues after the narrative runs out of characters. In particular, the gods added to the list of associations focus on gods enshrined 391 "Itsukushima shake engi 厳島社家縁起," in Yoshida bunko (Tenri University), unpaginated. 214 on the shores surrounding the island of Itsukushima, including but not limited to those visited during the shima-meguri rite. For example, the list of associations includes the shores of Ōe, Nagahama, Sugiura, Takanosu, Koshiboso, Aonori, and mentions the shores of Yamashiro, Suya and Mitoko within the following discussion of places where certain gods are enshrined. 392 None of the shores are included within the honji suijaku identifications in other Itsukushima no honji manuscripts, but this case provides clear evidence for the connection between the narrative, the gods, and the rites. It is possible that Yoshida Kanemigi received the information found in Itsukushima shake engi directly from Itsukushima Shrine, either through his connections with the priest Tanamori Fusaaki (datable to 1544), or when he came to Itsukushima in 1571 to participate in a rite for the reinstallation of the deities into the shrine. 393 This would correspond with the title of Itsukushima shake engi 厳島社家縁起, which translates to “Origin narrative [of] the shrine family of Itsukushima.” If this is the case, it would help to explain the incorporation of the shores of the shima meguri into the list of honji-suijaku associations. Unlike manuscripts made to be read by people far from Itsukushima Shrine, the Itsukushima Shrine family could have included additional information to strengthen the connections with the related rites. Knowing of Kanemigi’s interest in the narrative for its information, the priests could be inclined not to leave anything out. 392 Ibid. 393 The 1544 record of contact between Yoshida Kanemigi and Tanamori Fusaaki is from when Kanemigi transmitted a Yoshida Shintō rite to Tanamori. See doc. 1564 in "Itsukushima Nosaka monjo 厳島野坂文書." Concerning Kanemigi’s participation in the rite for the moving of the deities, he arrived at Itsukushima on 12/6/1571, performed the rite on 12/27, and involved himself with Tanamori at Itsukushima, priests at other nearby shrines, and the Mōri family from the first to third months of 1572. 215 Conclusion Bruno Latour compares the language of religion to “love-talk” while comparing the language of religious and scientific discourses. Latour’s thinking is based on the case of Christian preaching but also resonates with honji monogatari. In short, he argues for a similar performative meaning in the utterances of love and religion to create closeness and presence. The words “I love you” possess little meaning in themselves, but they are powerful enough to bring tears of joy when said by the right person, in the right way. 394 Similarly, the words of Itsukushima no honji can be read simply as an entertaining story, but when performed in the “right way,” it makes the deity present and draws her close to the reader. We only have the remaining textual and visual traces of the spoken and read honji monogatari, but nevertheless, reading only the written content would be, in the words of Latour, “as idiotic as if a lover, asked to repeat whether she loves her partner or not, simply pushed the ‘play’ button of a tape recording to prove that five years ago she had indeed said ‘I love you, darling.’” 395 Latour makes a compelling case for reversing the conventional understandings of the proximity of religious and scientific language. The content of religious sermons might be that which is transcendent and otherworldly, but the words serve to make the divine close and accessible. Conversely, science examines the world around us, but leads us to understandings based in terms of that which is invisible and abstract. We should not, however, ignore the complementary relationship between proximity and knowledge. We need to read the narrative and its informational explication as the two halves of the same book that they are. The relationship between knowledge and proximity towards a deity 394 Latour, "“Thou Shall Not Freeze-Frame,” or, How Not to Misunderstand the Science and Religion Debate," 29- 35. 395 Ibid., 34. 216 is not unlike the case of a local’s feeling of belonging towards one’s local space. A New Yorker might love New York but is unlikely to wear a t-shirt stamped with those words. Why? Put simply, because those shirts are marketed for tourists. One belongs as a local through feelings of closeness to the space, and knowledge of the space and the behavior of locals in contradistinction to the knowledge and behavior of others. 396 The feeling of closeness towards the Itsukushima Deity created through the hearing of Itsukushima no honji can be understood as the feeling of belonging towards a local god. In other words, Itsukushima no honji allows anyone, even pilgrims or those without the means to ever reach Itsukushima Shrine, to worship the Itsukushima Deity as a local god. The work provides the emotional proximity and the knowledge of how to properly worship the deity, connecting the spaces of the worshipper and the deity through affect and ritual. The exclusivity of local belonging and the personal connection to a god seem to rest on unstable ground once space is removed as a barrier to inclusion. It is not unlike the case of the Kasuga Deity declaring that Myōe is her favorite worshipper, while also stating that she loves all her worshippers. Honji monogatari offer the opportunity for everyone, anyone anywhere, to be the favorite child of the deity. 396 Nadia Lovell defines locality in terms of feelings of belonging, as part of a larger anthropological discussion of the need to separate locality from its assumed basis in a fixed spatial territory. She describes the feeling of belonging as an experience that can link places and communities in the plural, “Locality in this sense becomes multivocal, and belonging itself can be viewed as a multifaceted, multilayered process which mobilises loyalty to different communities simultaneously. Nadia Lovell, "Introduction," in Locality and Belonging, ed. Nadia Lovell (New York: Routledge, 1998), 5. 217 Chapter 4 The Rise and Fall of the Heike and the Surviving Association with Itsukushima Shrine When Private Negoro Tōkichi ( 根来藤吾, 1879-1929) left home to fight in the Russo- Japanese War (1904-1905), he took in the sights along the long trip from his home in Sendai to Ujina 宇品, the port near Hiroshima where he would depart for Manchuria. He saw Mount Fuji and Lake Hamana and stopped in the cities of Tokyo and Kyoto. But, the sight that most impressed him was the island of Miyajima, off the shore of Hiroshima, and its Itsukushima Shrine. Not unlike the records of pilgrims hundreds of years earlier or the twenty-first century reviews of the shrine on Google Maps, Negoro praised the beautiful scene of the shrine and its torii gate appearing to float over the water at low tide. His view of Itsukushima was built on a base that was simultaneously timeless, historical, and contemporary. Itsukushima’s seemingly changeless natural beauty aligned with the idea of Japan as the realm of the gods (shinkoku 神国). Its past as the site of worship for the famous warrior-aristocrat Taira no Kiyomori ( 平清盛, 1118-1181) connected Negoro with a glorified history of Japanese martial prowess. The soldiers departing and returning from war were encouraged to stop at Itsukushima, an ideally imagined sacred place to pray for success in battle and be instilled with militaristic patriotism. 397 However, the problem with this image of the past is that the Taira lost the Genpei War and were left annihilated. Why worship at Itsukushima Shrine when its deity could not even protect its most fervent followers? Why was this pivotal point so easily overlooked? The answers to these questions lie in simplified understandings of the past. Memory, narrative, and place all have the power to streamline the complexities and contingencies of the past to align with our views in in the present. Even the most detailed scholarship must emphasize 397 Naoko Shimazu, "Reading the Diaries of Japanese Conscripts: Forging National Consciousness during the Russo- Japanese War," in Nationalisms in Japan, ed. Naoko Shimazu (New York: Routledge, 2006), 53-56. 218 certain points and leave others unsaid in order not to confound readers with an impenetrable mass of information. Fortunately (and often unfortunately), our minds are skilled at remembering certain points by forgetting everything else. The historian of Hellenistic Greece, Angelos Chaniotis, argues that cultural memory often functions through select keywords that could carry widely recognized allusions and meanings. 398 These keywords were read through a preestablished cultural foundation, but that does not entail that their meanings were fixed or singular. Conversely, keywords are powerful precisely because their fluidity can be masked by the assumed stability of the past. Today, Itsukushima is largely remembered, and consequently studied, as the shrine of the Taira. The centrality of the Genpei War and the Tale of the Heike in Japan’s history and literary canon supported preconceived associations between “Itsukushima” and the “Taira” by dismissing the need to look elsewhere. The Taira’s exceptional worship and patronage of Itsukushima Shrine merits the attention it has received. However, the image of the Itsukushima Deity as primarily the Taira’s tutelary deity is a post-medieval development that has overshadowed other facets of how Itsukushima was identified and worshipped earlier. 399 The primary issue in this chapter, and the dissertation, is not the medieval identities of the Itsukushima Deity per se, but rather, the development of those identities through Itsukushima’s incorporation into external discourses and bodies of knowledge. Through this approach, we can uncover how and why people throughout the Japanese archipelago were thinking about Itsukushima Shrine, and more specifically, the agency of various actors (beyond Itsukushima 398 Angelos Chaniotis, "Travelling Memories in the Hellenistic World," in Wandering Poets in Ancient Greek Culture: Travel, Locality and Pan-Hellenism, ed. Richard Hunter and Ian Rutherford (New York: Cambridge University Press, 2009), 256. 399 It is important to mention that The Tale of the Heike had a profound influence on the image of the Itsukushima Deity as the protector of the Taira over the course of the medieval period. It is likely that the increase in positive associations between Itsukushima and the Taira in the late medieval period relates to the Heike’s growing influence and permeation into medieval Japanese culture over the course of the medieval period. 219 Shrine priests) in constructing multiple identities for the Itsukushima Deity. The Tale of the Heike, as a narrative and discourse explaining the Taira’s rise and fall from power, highlights their devotion to the Itsukushima Deity. But non-Heike texts tell a different story. Other medieval writings about Itsukushima rarely mention the Taira; and why should they? Connections with the past could provide prestige, but the Taira household’s efforts ended in failure. The Tale of the Heike influenced medieval perspectives towards Itsukushima, but it was not alone. Other texts, and even some variants of the Tale of the Heike, distanced Itsukushima from the Taira by ignoring, altering, and reinterpreting the meaning of their relationship. Itsukushima Shrine’s image survived the defeat of its Taira patrons because its identity was not confined to a single, unchanging relationship. 400 Specifically, there were two interrelated issues regarding the identities of the Taira and Itsukushima whose changing meanings influenced how they were understood in relation to one another: (1) whether or not the Itsukushima Deity was the tutelary deity of the Taira, and (2) the primacy of the Genpei Wars for remembering the Taira. In the first case, being the tutelary deity protecting the Taira family would crystallize a unique relationship between them, and complicate attempts to separate Itsukushima’s identity from that of the Taira. As for the second, the passage of time made it easier for the Taira to carry a positive connotation by distancing their identity from the Genpei War. Logically, their 400 Itsukushima Shrine also possessed means of economic support, such as landed estates (shōen 荘園), that did not rely on the recognition of the Itsukushima Deity as an authoritative and efficacious divinity worthy of worship. Tamura Hiroshi has argued that the shrine estates gained during the period of the Taira’s patronage were essential for the shrine’s survival after the Genpei War. However, the estates alone could not provide economic independence, especially when fires and natural disasters resulted in the need for major reconstruction projects. In such cases, the sacred prominence of the shrine would need to be recognized for successful appeals to political institutions or worshippers for financial support. Concerning Itsukushima’s estates, see Tamura Hiroshi 田村裕, "Itsukushima sharyō shōen no keisei to kurashiki ni tsuite 厳島社領荘園の形成と倉敷について," in Naikai chiiki shakai no shiteki kenkyū 內海地域社会の史的研究, ed. Matsuoka Hisato 松岡久人 (Yamaguchi-ken Tokushima- shi: Matsuno Shoten, 1978); Kadoshige Hajime 角重始, "Kamakura, Nanbōkucho ki Itsukushima sha shihai no tokushitsu 鎌倉・南北朝期厳島社支配の特質," in Itsukushima shinkō jiten 厳島信仰事典, ed. Nozaka Motoyoshi 野坂元良 (Tokyo: Ebisu Kōshō S huppan, 2002). On rebuilding efforts, see Matsuoka Hisato 松岡久人, Aki Itsukushimasha 安芸厳島社 (Kyoto: Hō zō kan, 1986), 97-111. 220 association was often overlooked when it would be conceived negatively, and brought up when positive. In the following pages, I will argue that (1) the Itsukushima Deity’s image was distanced from the Taira in order to avoid the stigma of the fallen household, but the degree of separation depended on the time and place. The affiliation with the Taira becomes more visible when one moves forward in time, and to western Japan, away from the capital. Given the regional history of the Taira household and Itsukushima Shrine in Aki Province, it would be particularly beneficial for local discourses to link them—as long as the Taira could be remembered favorably. This became possible by the late fifteenth century (three hundred years after the end of the Genpei War). As will be discussed later, there were people calling themselves descendants of the Taira by 1492 at the latest, a claim that only holds merit if the Taira could be seen as more than just the losing side of the Genpei War. 401 Secondly, (2) the Tale of the Heike was not a unitary discourse homogenizing the identity of the Itsukushima Deity as the tutelary deity of the Taira. The Heike must also be recognized as a compilation of narratives and ideas whose meaning could be conveyed or altered outside of the context of the temporally linear narrative of the Taira’s rise and fall. The Heike was a part of the larger discursive environments in which it was rewritten and retold, and as such, could influence and be influenced by external views towards Itsukushima. While the Heike’s story of the Taira’s rise and fall cannot avoid involving the Itsukushima Deity, the Nagatobon, a Heike variant with a regional focus on western Japan, attempted to balance the need for narrative consistency with the aim of promoting Itsukushima Shrine as a preeminent sacred space of western Japan. In other 401 Suzuki Akira cites an entry from 2/8/1492 in the Inryōken nichiroku 蔭涼軒日録 in which a cultural advisor to the Ashikaga bakufu (dōbōshū 同朋衆) named Etsu’a 越阿 identifies himself as a descendant of Taira no Norimori ( 平教盛, 1128-1185). This is the earliest dated record I have found showing someone linking oneself with the Taira family. Suzuki Akira 鈴木彰, Heike monogatari no tenkai to chūsei shakai 平家物語の展開と中世社会 (Tokyo: Kyūko Shoin, 2006), 683. 221 words, the Heike served as both a linear and self-encompassing narrative, and as an episodic compilation whose meaning was not confined to the context of the rise and fall of the Taira. Therefore, the changes made in the Nagatobon reveal the extent to which it could positively incorporate places in western Japan into the Heike, despite the tale’s geopolitical conclusions favoring people and places to the east. Sources The argument for a distancing of Itsukushima from the Taira is complicated by a lack of sources. This scarcity could be the result of the disconnect between them. However, it could just be a coincidence arising from the loss of texts over time. 402 A broad survey of the extant texts is both a necessity and an opportunity for uncovering medieval understandings of the association between Itsukushima and the Taira. I am working from a hodgepodge of sources written over four hundred years (1200-1600) for diverse reasons. Nevertheless, comparing them reveals a multiplicity of coexisting ideas about Itsukushima Shrine, and how certain discourses came to prominence over others. While memory dismisses undesirable, confusing, and seemingly inconsequential elements in order to preserve a simplified and comprehensible meaning, my approach assembles and re-entangles the multiple networks of ideas influencing Itsukushima’s identity. In particular, this chapter examines how the association between Itsukushima and the Taira family was understood during the medieval period, and especially in western Japan. The scarcity of regional historical sources can complicate attempts to distinguish the viewpoints of 402 Fires and other natural disasters destroyed many premodern sources from Itsukushima Shrine, leaving approximately 510 documents written between 794-1573. Matsui Teruaki 松井輝昭, Itsukushima monjo denrai no kenkyū: Chūsei monjo kanri shiron 厳島文書伝来の研究: 中世文書管理史論 (Tokyo: Yoshikawa Kōbunkan, 2008), 30. 222 regional communities from the better-preserved voices of political and cultural elites around the capital in Kyoto. Nevertheless, this does not grant scholars license to research only the center. Even if we must work with a few missing pieces, we must strive to add nuance to differences across specific space-times. Following the end of the Genpei War in 1185, there would be little reason to memorialize Itsukushima or the Taira. The Minamoto victors wanted to present the past from their perspective, one that would not portray the Taira kindly. The surviving Taira fled in various directions, leaving no community seeking to dwell in the memory of their fleeting glory. As for regional elites and religious institutions in western Japan, many of whom had allied with the Taira, they were receptive to the Minamoto’s offers to join the newly formed Kamakura bakufu government. Even Itsukushima Shrine received patronage from the Kamakura bakufu despite its support for the Taira. 403 In 1189, four years after the end of the Genpei War, a lower fifth ranked Minamoto lord ordered an Itsukushima family priest Saeki Kiyomoto 佐伯清元 to arrange for kagura performances and prayers to be made for the sake of Minamoto no Yoritomo during the military campaign against the Northern Fujiwara. The foremost historian of Itsukushima Shrine, Matsuoka Hisato, explained these prayers as a demonstration of the Minamoto’s dominance over the Taira. 404 However, this answer relies on the oppositional Minamoto-Taira framework of the Genpei War rather than the postwar situation of the Bakufu needing to secure influence in western Japan. 405 403 On the relationships between the Kamakura bakufu and religious institutions in Kyushu like Munakata Shrine, see Munakata shishi. tsūshi hen 宗像市史. 通史編, 2 vols., vol. 2 (Munakata: Munakata-shi, 1999), 318-19. 404 Matsuoka, Aki Itsukushimasha 安芸厳島社, 97. 405 It is outside of the scope of this chapter to discuss such historical developments of Itsukushima Shrine after the Genpei War, but I hope to write about it in more detail in the future. In short, my argument for interpreting the 1189 prayers as primarily a means of garnering Itsukushima’s support for the bakufu is based on the prayers being addressed to Saeki Kiyomoto rather than the head priest and Taira retainer, Saeki Kagehiro 佐伯景弘. In this manner, the bakufu worked to shift power away from Kagahiro and strengthen relations with other members of the 223 One reason the Itsukushima Shrine survived the loss of its Taira patrons is that its identity was not confined to its relationship with the Taira. The previous chapters detailed the other crucial elements of the Itsukushima Deity’s identity and how they developed. Informed with a more comprehensive understanding of the medieval bodies of knowledge on the Itsukushima Deity, we can reflect on our modern assumptions towards Itsukushima Shrine and reconsider the medieval significance of the association between Itsukushima and the Taira. Regional Perspectives on the Tale of the Heike Itsukushima Shrine often needed to be presented differently depending on the audience, and especially in the centuries following the end of the Genpei War. It could not remain associated with the Taira within political spheres. Yet, the Taira’s historical, cultural, and regional associations could retain significance. On the one hand, we must recognize how the Tale of the Heike’s expansive reach across political, social, and regional audiences could hinder attempts to control the image of Itsukushima. On the other hand, we must also recognize that the Tale of the Heike did not exist within a vacuum. Heike compilers and audiences related their preexisting understandings of Itsukushima with the Heike episodes. In order to examine how the Tale of the Heike narrative related to Itsukushima in both regional and translocal contexts, I will weave together a close reading of the regional Nagatobon variant of the Tale of the Heike with a variety of other medieval sources linking Itsukushima and the Taira. Only in viewing these shrine family. To give another example, the bakufu appointed Saeki Tamehiro 佐伯為弘 as a land steward (jitō 地頭) in 1216, further connecting Kagehiro’s relatives with Kamakura. For the record of the prayer request to Kiyotomo, see doc. 22 in "Shinshutsu Itsukushima monjo 新出厳島文書," in Hiroshima kenshi: Kodai chūsei shiryō hen 広島県史. 古代中世資料編, ed. Hiroshima-ken 広島県 (Hiroshima: Hiroshima-ken, 1978), 181-82. Concerning Saeki Tamehiro’s appointment, see Jeffrey P. Mass, "Patterns of Provincial Inheritance in Late Heian Japan," Journal of Japanese Studies 9, no. 1 (1983): 89, n. 78. Also see doc. 3076 in Takeuchi Rizō 竹內理三, ed. Kamakura ibun 鎌倉遺文, 42 vols. (Tokyo: Tokyōdō Shuppan, 1971). 224 together can we see the extent to which Itsukushima’s image could change within and outside of the Heike narrative. The Nagatobon Tale of the Heike provides a rare glimpse of how the Heike narrative and its image of Itsukushima Shrine could be reconceived for audiences in western Japan. Considering the Heike’s attention to medieval geopolitical discourses, it is easy to imagine the variants produced outside of the capital radically decentering the narrative. However, we must temper our expectations. The multiple revisions within Nagatobon episodes including Itsukushima suggest that the compilers were attentive to the site’s portrayal, but it is the subtlety of the changes that is most striking. As we shall see, the changes do not impact the larger narrative and ideological frameworks of the Heike. There are other episodes in the Nagatobon, as well as in the Genpei tōjōroku ( 源平闘諍録, 1337) a variant compiled in eastern Japan, that incorporate regional sources and depart from a center-based worldview. That being said, these additions are often tangential or complimentary to the overall narrative. The potential for altering Itsukushima’s image was limited by its role in episodes central to the Taira’s rise and fall. In order to examine the Nagatobon’s agency for re-presenting the Itsukushima Deity within the Heike, we must begin with the larger question of regionality in Heike variants. The long-lasting focus among Heike scholarship in Japan on contextualizing the production of the variants and the relationships between them has provided crucial background information on the Nagatobon, but it remains understudied. Broader thematic studies of the Heike have instead focused on the canonical Kakuichibon variant ( 覚一本, 1371), the Engyōbon ( 延慶本, copied from 1409-1410, though an earlier colophon dates the text to 1309), or the Genpei jōsuiki 源平盛衰記. The Nagatobon (in its current form) was likely produced in western Japan between the 225 late fourteenth to sixteenth centuries, 406 and based largely on an older Engyōbon manuscript written before the extant 1409-1410 manuscript. 407 Compilers then incorporated narratives and information specific to places in western Japan. 408 Kanai Kiyomitsu even suggested that ascetic monks traveling between Itsukushima Shrine, Mt. Kōya, and Zenkōji were responsible for writing the fifth scroll of the Nagatobon, after analyzing the Itsukushima Shrine origin narrative incorporated into the Nagatobon. 409 Makino Kazuo, however, complicated this argument by demonstrating that the Itsukushima origin narrative is almost identical to the Itsukushima Daimyōjin nikki ( 厳島大明神日記, estimated to have been copied between 1323-1343) in the Kanazawa Bunko archive in Kamakura. The potential for texts like the Itsukushima Daimyōjin nikki to circulate between western Japan and Kamakura in the east, compels us to consider wider 406 Matsuo Ashie and Shimazu Tadao date the Nagatobon to the late fourteenth to fifteenth century based on the Nagatobon’s stylistic similarity with Muromachi period (1336-1573) narratives, and its shared content with 1423 nō play “Morihisa” 盛久. More recently, Ōgawa Eiichi compared the language of the Amidaji manuscript (believed to be the oldest extant manuscript, also known as the Akamagaseki manuscript) with the National Diet Library manuscript (copied between 1688-1711) and concluded that the Amidaji manuscript was likely transcribed between 1519-1602 based on linguistic similarities with the language of Christian texts from the same period. Matsuo Ashie 松尾葦江, Gunki monogatari genron 軍記物語原論 (Tokyo: Kasama Shoi, 2008), 231-32; Shimazu Tadao 島津忠夫, Heike monogatari shiron 平家物語試論 (Tokyo: Kyūko Shoin, 1997), 242; Ōgawa Eiichi 小川栄一, "Nihongo shiryō toshite no Nagatobon Heike monogat ari 日本語史料としての長門本平家物語 " Musashi daigaku jinbun gakkai zasshi 41, no. 3 (2010). 407 Matsuo Ashie describes the Nagatobon and the Engyōbon as sibling tex ts originating from the same “old Engyōbon,” based on the extensive commonalities and shared passages betwee n the two variants. Additionally, the Nagatobon also includes passages also found in the Genpei jōsuiki and Shibu kassen jōbon 四部合戦状本. Matsuo, Gunki monogatari genron 軍記物語原論, 230. 408 Matsuo Ashie and Sunagawa Hiroshi have suggested that the Nagatobon compilers might have learned about the temple and shrine origin narratives from itinerant monks and performers who had heard the stories while traveling. Matsuo gives the monks making the pilgrimage of sacred sites in the sixty-six provinces (reijō roku jū roku bu 霊場六十六部) as possible transmitters of origin narratives to the Nagatobon compilers, because the pilgrimage route includes many of the sites with origin narratives described within the text: Mt. Shosha 書写山, Ō saka Shō Hachiman 大阪正八幡, Nagato Ichinomiya 長門一宮, Kirishima 霧島, and Itsukushima. However, these sites are prominent sacred sites within their respective provinces, and that could be reason enough for their inclusion in both the Nagatobon and the pilgrimage of the sixty-six sacred sites. Sunagawa instead raises the possibility that itinerant biwa hōshi performers traveling along the San’yōdō route through the western provinces and Kyushu could have heard and transmitted the origin narratives incorporated only into the Nagatobon. Their arguments are plausible, but they do not explain why manuscripts of the origin narratives could not have travelled without the aid of itinerant monks or performers. Matsuo, Heike monogatari ronkyū 平家物語論究 (Tokyo: Meiji Shoin, 1985), 244-45; Sunagawa Hiroshi 砂川博, Heike monogatari no keisei to biwa hōshi 平家物語の形成と琵琶法師 (Tokyo: Ō fū, 2001), 418-19. 409 Kanai Kiyomitsu 金井清光, Ippen to Jishū kyōdan 一遍と時宗教団 (Tokyo: Kadokawa Shoten, 1975), 177-80. 226 possibilities for site-specific narrative and information to travel outside of the scope of oral transmissions originating from local persons and institutions. 410 The oldest Nagatobon manuscript (transcription estimated to 1519-1602) belonged to Amidaji 阿弥陀寺 (converted to Akama Shrine 赤間神宮 in 1875), a temple reconstructed as a mortuary site for Antoku Tennō and the fallen Taira in 1191, but this alone pro vides little ground for speculating on Amidaji’s role in the compilation or recitation of the Nagatobon as a different version of the Heike. 411 It is fair to surmise that the Tale of the Heike was recited there, likely for the sake of the fallen Taira, but that is not to say that the Nagatobon was compiled specifically for that purpose. Amidaji’s location near the site of Antoku’s death and the Taira’s final defeat at Dannoura (in modern-day Yamaguchi Prefecture) positioned it as an ideal site for memorial rites for the Taira, and for explicating hanging scroll images of the Illustrated Origin Narrative of Antoku Tennō (Antoku tennō engi-e 安徳天皇縁起絵, sixteenth century) and reciting episodes of the Tale of the Heike to visitors passing through the nearby travel hub at Shimonoseki. Naoko Gunji argues that the explication of the Illustrated Origin Narrative of Antoku Tennō served a ritual purpose to placate the Taira, and suggests that the Nagatobon was likely recited for similar purposes. 412 The earliest record of somebody seeing the Nagatobon is from Hayashi Razan’s ( 林羅山, 1583-1657) Tsurezuregusa notsui ( 徒然草野槌, 1621), which recalls how the Amidaji 410 Makino Kazuo 牧野和夫, "Nagatobon "Heike monogatari" kan go 'Itsukushima shidai no koto' wo meguru ikkō satsu: "Kamado san hōman da i bosatsu ki" wo kaishite 長門本『平家物語』巻五「厳嶋次第之事」をめぐる一考察 : 『竈門山寶満大菩薩記』を介して," Jissen koku bungaku 50 (1996): 78. 411 On the dating of the Amidaji manuscript, see Ōgawa, "Nihongo shiryō toshite no Nagatobon Heike monogatari 日本語史料としての長門本平家物語 ". For an overview of the extant Nagatobon manuscripts (sixty-three in total) and an introduction to the Amidaji manuscript and issues found comparing it with early modern diary records, see Matsuo, Heike monogatari ronkyū 平家物語論究, 198-216. 412 Naoko Gunji, "The Ritual Narration of Mortuary Art: The "Illustrated Story of Emperor Antoku" and Its "Etoki" at Amidaji," Japanese Journal of Religious Studies 40, no. 2 (2013); "Horrified Victors: Spirit Pacification of Heike Losers," in Lovable Losers: The Heike in Action and Memory, ed. Mikael S. Adolphson and Anne Commons (Honolulu: University of Hawai'i Press, 2015). 227 monks showed him the text during his visit in 1602. 413 Considering that awareness of the Nagatobon only began to spread after Razan’s son, Hayashi Gahō ( 林鵞峰, 1618-1680) borrowed and copied the text from Amidaji (approximately between 1664-1670), it is likely that the Nagatobon had not circulated widely outside of western Japan during the medieval period. It is likely that Gahō only knew about the Nagatobon from his father, and he would not have gone through the trouble of acquiring the text from Amidaji if he knew of another source for the text. 414 One colophon shared among a number of early-modern Nagatobon manuscripts reinforces the idea of the variant as a rarely accessible “secret text” (hisho 秘書) given to Amidaji as a memorial offering for Antoku Tennō, but this colophon is notably missing f rom the one manuscript from Amidaji. 415 Of course, the Nagatobon could only become known as a secret variant, and to quote Hayashi Gahō, as “somewhat differ ent from the Heike that is prevalent in the world” once the secret was out. 416 The Nagatobon’s connection with Amidaji provides only a brief glimpse into the Heike’s presence in western Japan and cannot reveal how or why the Nagatobon developed as a regional vairant. 417 Naturally, the best sources for examining the regionality of the Heike are the variants 413 There are earlier mentions of travels to Amidaji in medieval travel texts like Imagawa Ryōshun’s ( 今川了俊, 1326-?) Michiyukiburi 道行ぶり from 1371 and Sōgi’s ( 宗祇, 1421-1502) 1480 Tsukushi michi no ki 筑紫道記. Matsuo, Heike monogatari ronkyū 平家物語論究, 211. 414 Murakami Mitsunori 村上光徳, "Akama Jingū shozō gojūni gō monjo no imi: Nagatobon Heike monogatari kenkyū no isshu ken toshite 赤間神宮所蔵五十二号文書の意味：長間本平家物語研究の一手懸として," Komazawa tandai kokubun 6 (1975): 57-59. 415 The oldest dated manuscript including this colophon is from 1769 (Naikaku Meiwa roku nen bon 内閣明和六年本). Matsuo, Heike monogatari ronkyū 平家物語論究, 200, 12-13. 416 Murakami, "Akama Jingū shozō gojūni gō monjo no imi: Nagatobon Heike monogatari kenkyū no isshu ken toshite 赤間神宮所蔵五十二号文書の意味：長間本平家物語研究の一手懸として," 58. 417 Aside from the Nagatobon variant, there is evidence that the Tale of the Heike (or at least parts of the narrative) spread to western Japan as written text by 1330. The Itsukushima Shrine treasury Tale of the Heike fragment (Itsukushima jinja zō Heike monogatari dankan 厳島神社蔵平家物語断簡) was found on the back of the paper used for the copying of one scroll of the Daijikkyō (Skt. Mahāsa ṃnipata-sūtra, Ch. Daji jing 大集経). This scroll was a part of a larger project to copy the five-fold Mahayana sutras (gobu daijōkyō 五部大乗経) at the Saikonji temple 西金寺 located on modern-day Mukaishima 向島 (previously called Uta no shima 歌島), west of Itsukushima in Hiroshima Prefecture. While only a short amount of the Heike narrative remains, the presence of 228 themselves. David Bialock has demonstrated the complex spatial practices of the Engyōbon and even the more capital-centric Kakuichibon. 418 The power of peripheral spaces and blurred boundaries to challenge centralized spatial systems could have offered opportunities for regional variants to assert the authority of people and places outside of the center. 419 However, the spatial practices of the Nagatobon and Genpei tōjōroku focus on place more than power. Instead, they reflect a more general principle of linking places, people, and the past. The regionality of the Nagatobon and Genpei tōjōroku variants can be thought of as a layer of meaning superimposed over preexisting Heike narratives. 420 As a result, geopolitical ideas related to the shift in power from the Taira to the Minamoto often remained consistent, with regional emphasis being added when it would not conflict with the larger narrative. This made it easier for the Genpei tōjōroku to highlight people and places in eastern Japan, including the victorious Minamoto no Yoritomo, than for the Nagatobon to do the same with western Japan and the vanquished Taira. The increased praise for the Minamoto in the Genpei tōjōroku holes in the paper for binding a book raise the possibility that this fragment was originally part of a larger record of the Tale of the Heike. Considering that the manuscript was written almost completely in phonetic hiragana script and that this fragment was used as sutra copying paper suggests that the Tale of the Heike manuscript was not necessarily a prized possession. It is possible that it was recorded for reference, or copied for the education of children of regional elites. Yokoi Takahashi 横井孝, "Itsukushima jinja zō Heike monogatari dankan: Eiin to ryaku kaidai 厳島神社蔵平家物語断簡：影印と略解題," in Enkyō-bon Heike monogatari kōshō 延慶本平家物語考証, ed. Mizuhara Hajime 水原一 (Tokyo: Shintensha, 1997), 190-91; Mukaishima chōshi. Tsūshi hen 向島町史. 通史編, (Hiroshima-ken Mitsugi-gun Mukaishima-chō : Mukaishima-chō , 2000), 204. 418 David Bialock emphasized the potential for future scholarship on the Nagatobon and Genpei tōjōroku to reconsider issues of space in the Tale of the Heike in his 2007 monograph, but these variants remain understudied. David T. Bialock, Eccentric Spaces, Hidden Histories: Narrative, Ritual, and Royal Authority from the Chronicles of Japan to The Tale of the Heike (Stanford, CA: Stanford University Press, 2007), 2, 6-9. "The Tale of the Heike," in The Cambridge History of Japanese Literature, ed. Haruo Shirane, Tomi Suzuki, and David Barnett Lurie (Cambridge, United Kingdom: Cambridge University Press, 2016), 302. 419 Bertrand Westphal describes space as inherently transgressive to explain how spatial fluidity, mobility, and polysystemic relationality (especially between a center and multiple peripheries) challenges homogenous and hierarchical systems of space (i.e., based around a fixed and unparalleled center). Bertrand Westphal, Geocriticism: Real and Fictional Spaces, trans. Robert T. Tally (New York: Palgrave Macmillan, 2011), 37-50. 420 Yamashita Hiroaki describes the Genpei tōjōroku and the Shibu kassen jōbon as “introducing the perspective of a eastern reciter over the perspective of a Kyoto reciter.” Yamashita Hiroaki 山下宏明, Heike monogatari no seiritsu 平家物語の成立 (Nagoya: Nagoya Daigaku Shuppankai, 1993), 120. 229 sometimes feels unnatural, such as during sections on the Taira’s rise to power, but this is not a conflict with the overall narrative. 421 Moreover, the Genpei tōjōroku goes on to incorporate elite families and sacred spaces in eastern Japan aiding the Minamoto, creating a clear line of connection with the prestige of Yoritomo, the Minamoto, and their historic triumphs in the Genpei War. It makes sense, as Yamashita Hiroaki has argued, that regional elites like the Chiba family could have participated and offered family sources for the compilation of the Genpei tōjōroku in order to enhance their image. 422 The Nagatobon shares the same goals as the Genpei tōjōroku, but its path is complicated by the historical framing of the Taira as the losers of the Genpei War. The Nagatobon cannot use the Taira’s glory as a base for creating connections with people and places in western Japan, like the Genpei tōjōroku does with Minamoto no Yoritomo. Neither can the Nagatobon, as a written record of the Tale of the Heike, lionize the Taira as tragic heroes who challenged the authority of the center. 423 Instead, the compilers of the Nagatobon accentuated regional affinity while 421 Ibid., 117. 422 Yamashita Hiroaki noted a particular emphasis on the Chiba family, and found strong similarities with other texts about the family and their tutelary Chiba Myōken Shrine ( Chiba Myōken sha 千葉妙見社): the Chiba jitsuroku 千葉実録, Chiba denkōki 千葉伝考記, Myōken jitsuroku senshūki 妙見実録千集記, and Sengaku shūshō 千学集抄. However, Minamoto Ken’ichiro describes the Sengaku shūshō as a late medieval text, and the others are difficult to date, therefore it is a possibility that some, or all, of these texts were influenced by the Genpei tōjōroku rather than the other way around. Ibid., 121-23; Minamoto Kenichiro 源健一郎, "Chiba Myōken no hontai, honji setsu: Genpei tōjōroku to Chiba Myoken sha kankei shiryō to no aida 千葉妙見の本体・本地説：源平闘諍録と千葉妙見社関係資料との間," Junreiki kenkyū 3 (2006): 67. 423 Positive views of the Taira’s tragic suffering and rebelliousness against the courtly elites arose in the Meiji period (1868-1912), when scholars began to analyze the Heike as literature (rather than as historical, as was often the case in the medieval period) and tried to diminish anti-imperial interpretations. At the same time, the Taira’s transgressive actions against the court were recognized in new readings of the Heike as a national epic, but this interpretation only matured post World War II when cultural values shifted to democratic ideas of popular rule and anti-elitism. This view can be seen in Yoshikawa Eiji’s ( 吉川英治, 1892-1962) novel Shin Heike monogatari ( 新平家物語, “The New Tale of the Heike”) which began serial publication in 1950, and its 1955 filmic adaptation by director Mizoguchi Kenji ( 溝口健二, 1898-1956). Concerning the reception of the Tale of the Heike from the Meiji to postwar periods, and including a comparison with its premodern reception, see David T. Bialock, "Nation and Epic: The Tale of the Heike as Modern Classic," in Inventing the Classics: Modernity, National Identity, and Japanese Literature, ed. Haruo Shirane and Tomi Suzuki (Stanford, CA: Stanford University Press, 2000). For an analysis of Mizoguchi Kenji’s film on the Shin Heike monogatari, see Hitomi Tonomura, "Kiyomori and His Family in Postwar Japan: Mizoguchi’s Shin Heike monogatari (The New Tale of the Heike)," in Lovable Losers: The Heike 230 distancing these episodes from the Heike’s unifying plot. For example, the Nagatobon includes a miraculous tale of the Kiyomizudera Kannon Deity saving Taira no Morihisa 平盛久 from execution (scroll 20, “Shume Hachirozaemon no jō Morihisa no koto” 主馬八郎左衛門尉盛久事). 424 There are also detailed battle scenes where the Taira triumph, such as Taira no Moritoshi’s ( 平盛俊, ?-1184) victories in the northern provinces (scroll 13, “Hōboku tok oro dokoro kassen no koto, tsuki kagyū no hakari no koto” 北国所々合戦事付火牛計事) and the battle at Ataka harbor (scroll 14, “Ataka no minato no kassen no koto” 安高湊合戦事). 425 These events might not be among the most famed scenes from the Heike, but it is their lack of centrality to the narrative that provides leeway for praising the Taira and the traces of their deeds in western Japan. Despite the central conflict between the Taira and the Minamoto in the Heike narrative, the Nagatobon’s increased separation between individual episodes and the larger narrative helped to present the Taira as simply an elite and storied warrior family. For example, the Nagatobon removes a detail present in the Engyōbon where the two Taira swords named Nukemaru 抜丸 and Kogarasu 小烏 passed down to Yorimori ( 平頼盛, 1131-1186, Kiyomori’s half-brother) and Kiyomori respectively are used to symbolize the rift between the brothers in Action and Memory, ed. Mikael S. Adolphson and Anne Commons (Honolulu: University of Hawai'i Press, 2015). 424 The Morihisa narrative is also told in the 1423 nō play “Morihisa” by Kanze Motomasa ( 観世元雅, ?-1432) and the 1520 Kiyomizudera engi emaki 清水寺縁起絵巻. For more information on these non-Nagatobon accounts, see Naoko Gunji, "Morihisa and the Cult of Kannon at Kiyomizudera," in Like Clouds or Mists: Studies and Translations of Nō Plays of the Genpei War, ed. Elizabeth Oyler and Michael Geoffrey Watson (Ithaca, NY: East Asia Program, Cornell University, 2013); X. Jie Yang, "A Miracle at Morihisa’s Execution: Reading Legends of the Origin of Kiyomizudera," in Lovable Losers: The Heike in Action and Memory, ed. Mikael S. Adolphson and Anne Commons (Honolulu: University of Hawai'i Press, 2015). 425 Matsuo Ashie also notices that the Nagatobon alone names a warrior Miyazaki Tarō 宮崎太郎 in the two battles mentioned in the paragraph and three others, and she raises the possibility that sources from the Miyazaki family might have been used in the production of the Nagatobon. That being said, it is not uncommon for the Heike variants to similarly incorporate the names of different regional warriors. Matsuo, Heike monogatari ronkyū 平家物語論究, 241, 46. 231 during the episode of the Taira’s fleeing of the capital. Instead, the Nagatobon adds praise to the swords and armor passed down within the Taira family in an earlier scene when Kiyomori begins to flourish (scroll 1, “Tadamori shussuru koto” 忠盛卒事) as “treasures of the family” (ie no hōbutsu 家の宝物) and “protectors of the realm” (honchō no mamori 本朝のまもり). 426 While unable to fully remove the swords from the context of the Taira’s rise and fall, it shifts the timing to the Taira’s ascendance, and presents the swords as atemporal symbols of warrior authority to protect the realm. Suzuki Akira has found examples of late medieval and early modern sources (such as the 1611 Jusaiki 寿斎記 by Futatsugi Jusai 二木寿斎) using the swords’ possession by the Nishina 仁科 and Emon 江聞 to identify themselves as descendants of the Taira. Notably, Suzuki’s examples often specify a transmission of the swords from Kiyomori to Shigemori before continuing to the Nishina or Emon, reflecting the influence of the Heike’s less critical characterization of Shigemori. Nevertheless, the self-recognition of warrior families as descendants of the Taira was not limited to the Heike either, and could also incorporate external representations of the Taira and their past. 427 The positive portrayal of Itsukushima in the Nagatobon can likewise be read as distancing Itsukushima’s meaning from the Heike plot. Itsukushima cannot help but remain associated with the Taira, and Kiyomori, in particular, but it can also be represented more generally as a prestigious and efficacious sacred space of western Japan. As will be discussed later, there are some episodes involving Itsukushima and Kiyomori that display Kiyomori in a 426 The description of the Taira swords and armor as the “protectors of the realm” is the same in the Genpei jōsuiki, but “treasure of the family” is unique to the Nagatobon. Suzuki, Heike monogatari no tenkai to chūsei shakai 平家物語の展開と中世社会, 80, 648-49. 427 Notably, not all texts remained consistent with the fundamental identity of the swords with the Taira family. For example, the late fifteenth century Myōjikō 名字考 emphasizes the origins of the swords with their forgers in Hōki 伯耆 and Bizen 備前 provinces. Ibid., 650-54 (outside of Taira context), 57-62 (in relation to descendants of the Taira) 232 surprisingly positive light, but this need not reflect a larger attempt to rehabilitate Kiyomori’s image. Conversely, the favorable depiction of Kiyomori can be read as a means to protect Itsukushima from the negative associations of the Taira leader. 428 Focusing on Itsukushima rather than Kiyomori reveals how the connection between Itsukushima and the Taira could shift depending on the context. There are times when the Nagatobon can deemphasize the relationship and praise Itsukushima on its own merits. There are also times when Itsukushima’s support for the Taira plays a crucial role in prominent episodes of the Heike, limiting the possibilities for changing the image of Itsukushima within the larger narrative. Itsukushima within the Nagatobon but Outside the Heike The Nagatobon’s incorporation of origin narratives of temples and shrines in western Japan is significant, even if we cannot deduce how the compilers obtained the information. The Nagatobon expands the episode of Fujiwara no Naritsune’s ( 藤原成経, 1156-1202) exile to include the origin narratives of a number of sacred sites encountered as he travels west, including: Cape Ashizuri 足摺岬, Kirishima 霧島, Sumiyoshi 住吉, and Ō saka Shō Hachiman 大阪正八幡. His travel provides an appropriate michiyuki 道行 travel scene to tell audiences about nearby places. The origin narratives, as stories of the past, are usually kept separate from the characters and events of the Heike, and therefore, they do not raise conflicts with the Tale of the Heike itself. (As we will see later, the case is more complicated for Itsukushima.) The 428 Matsuo Ashie compared the characterization of Kiyomori in passages unique to the Nagatobon and found that they consistently portrayed him in conventional terms, as imprudent, arrogant, and thick-skinned. She also acknowledges that the Nagatobon’s stylistic similarities with Muromachi period narratives could be responsible in part for producing exaggerated portrayals of the characters that can lessen the psychological and more sympathetic depiction of the characters in other variants such as the Engyōbon. Matsuo, Gunki monogatari genron 軍記物語原論, 234-38. 233 Nagatobon does weave the Ashizuri origin narrative into the Heike by making Naritsune into the listener of the tale, but he mainly serves to articulate how the Heike audience should receive the origin narratives. After hearing the Ashizuri origin narrative, Naritsune says, “Even though I knew about this [place] from a family diary [I read] when I was in the capital, it is truly something that should be seen with one’s own eyes. My heart feels a little more at ease when I recall the famous places I saw [along route] to the place of my exile.” 429 People could read or hear about distant sacred places and their origins, just as was the case for the audiences of this episode from the Nagatobon. But this is not the same as visiting the site oneself. That is to say, the Nagatobon is promoting these sacred places as famous and numinous sites worthy of worship and pilgrimage. The chapter “Tokudaiji Sanesada’s Pilgrimage to Itsukushima” (Tokudaiji Sanesada no Itsukushima mōde 徳大寺厳島詣) provides an example of the Nagatobon minimizing the relationship between Itsukushima and Kiyomori. Overall, the Nagatobon remains consistent with the Engyōbon, with a few passages revised or omitted. The court official Sanesada ( 実定, 1139~1191) is distraught after learning that he had lost out on an appointment to Kiyomori’s son Munemori ( 平宗盛, 1147~1185), and is advised to get in the good graces of Kiyomori by making a pilgrimage to Itsukushima Shrine. The Engyōbon spells out the reason for this, “Th e Great Minister Lay Priest [Kiyomori] especially worships the foremost shrine (ichinomiya 一宮) of Aki Province; [therefore] you should make the pilgrimage to Itsukushima.” 430 In contrast, the 429 Asahara Yoshiko 麻原美子 and Nanami Hiroaki 名波弘彰, eds., Nagato-bon Heike monogatari no sōgō kenkyū: Kōchū hen 長門本平家物語の総合研究：校注篇, 3 vols., vol. 1 (Tokyo: Benseisha, 1998), 284. 430 The Kakuichibon expresses the same idea in different words. “Itsukushima [Shrine] of Aki Province is intensely worshipped by the Taira, therefore, why don’t you go make a pilgrimage and pray at that shrine?” Royall Tyler, trans., The Tale of the Heike (New York: Viking, 2012); Ichiko Teiji 市古貞次, ed. Heike monogatari 平家物語, 2 vols., Shinpen Nihon koten bungaku zenshū (Tokyo: Shō gakkan, 1994), v. 45, p. 160; Kitahara Yasuo 北原保雄 234 Nagatobon removes the first half of the sentence from the Engyōbon , leaving, “You should make a pilgrimage to Itsukushima of Aki Province.” 431 It is impossible to ignore the relationship between Itsukushima and the Taira in this episode, but it is not necessary to emphasize it either. The Nagatobon makes another minor change to the Sanesada episode to emphasize Itsukushima’s appeal apart from the connection with the Taira. When Sanesada arrives at Itsukushima Shrine and is received with a performance by the naishi 内侍 (title for the miko at Itsukushima Shrine) shrine attendants, the Nagatobon adds a single line to that of the Engyōbon: “The naishi call to mind the [Itsukushima] Deity” (Naishi tachi wo ba, daimyōjin to koso omoi tatematsure 内侍たちをば、大明神とこそ思ひたてまつれ). 432 The beauty of the Itsukushima naishi was often described through comparisons with the Itsukushima Deity, and their role as performers of dance and song could further help audiences to see them as manifestations of the Itsukushima Deity herself. Tanaka Takako thus interprets this passage from the Nagatobon as overlaying the image of the Itsukushima Deity onto the naishi, and further suggests that they might have also served as reciters of origin narratives and tales about the Itsukushima Deity. 433 and Ogawa Eiichi 小川栄一, eds., Enkyōbon Heike monogatari. Honbun hen 延慶本平家物語. 本文篇, 2 vols. (Tokyo: Benseisha, 1990), v.1, p. 66. 431 The wording in the Nagatobon and Engyōbon (for th e part present in both) is identical aside from minor differences at the end of the sentence that do not change the meaning. The sentence ends with “aru beshi to zonji sōrō” あるべしと存候 in the Nagatobon, and “aru beku sōrō” 有ベク候 in the Engyōbon. Asahara and Nanami, Nagato-bon Heike monogatari no sōgō kenkyū: Kōchū hen 長門本平家物語の総合研究：校注篇, 92. 432 Ibid., 93. For comparison with the Engyōbon, see Kitahara and 小川栄一, Enkyōbon Heike monogatari. Honbun hen 延慶本平家物語. 本文篇, 66-68. 433 Tanaka Takako 田中貴子, Gehō to aihō no chūsei 外法と愛法の中世 (Tokyo: Sunagoya Shobō, 1993), 40 -41. Tanaka quotes the Takakura-in Itsukushima gokō ki 高倉院厳島御幸記 for an example likening the appearance of the miko with the Itsukushima Deity. The passage describes the miko saying, “People said, ‘One day, [I] saw the image of an exalted noblewoman behind a screen heading towards the shrine hall.’ An abiding and unimaginable fragrance came out from the shrine hall.” See "Takakura-in Itsukushima gokō ki 高倉院厳島御幸記 ", in Gunsho ruijū. Kikō bu 群書類従. 紀行部, ed. Hanawa Hokinoichi 塙保己一 (Tokyo: Gunsho Ruijū Kankō kai Kantō sha Nai, 1952), 416; Koga Michichika, "Takakura-in Itsukushima gokō ki (Account of the Jour ney of the ex-Emperor Takakura to Itsukushima)," in Four Japanese Travel Diaries of the Middle Ages (Ithaca, N.Y.: China-Japan Program, Cornell University, 1981), 40. 235 In this manner, the viewing and hearing of performances and narratives would be further heightened by identifying the performers with the Itsukushima Deity herself. Instead of removing a mention of Kiyomori or the Taira, the Nagatobon shifts attention to other well- known aspects of Itsukushima’s image. The naishi, their performances, and the opportunity to draw close to the deity were reasons enough for audiences to forget the association with Kiyomori and make pilgrimages to Itsukushima. The chapter “Yasuyori’s Two Poems” (Yasuyori nishu uta no koto 康頼二首歌事), likewise, praises Itsukushima as a sacred place without mentioning the relationship with the Taira. In this episode, Taira no Yasuyori 平康頼 writes two poems and cast them into the sea as a call for help while exiled on the island of Kikai-ga-shima 鬼界ヶ島. One of the poems reaches Itsukushima Shrine and eventually makes its way to Kiyomori in the capital. All variants take this moment to briefly describe Itsukushima Shrine, and the Nagatobon adds a few more details. The chapter in the Nagatobon is, again, almost identical to the Engyōbon, except for two additional passages of praise, saying: Looking at the place of the [Itsukushima] Deity, the mountain rises tall behind [the shrine], [it is] the provisional gate of perfect enlightenment and marvelous enlightenment and copies the autumn moon of the fourteenth and fifteenth. And, [one] sees offerings of spring flowers from the heaven of the thirty-three [celestial beings] at the shrine of inner realization and outer manifestation. Truly, [Itsukushima] is the sacred space of Dainichi, and bound alongside the shore of the Shingon mysteries. ... The deity of this shrine made thirty-three great vows. According to the first vow, [the deity] will protect all whose heart aspire for enlightenment. Therefore, when one makes a single pilgrimage, they will be ensured enlightenment in a future birth. There is [also] a vow to fulfill, without fail, each and every wish of all sentient beings. [The monk who found the poem] prayed, “Even if my brother, Lay Priest Yasuyori, dies, I wish for him to see that omen. If he lives, I hope he hears that message. 434 434 Asahara and Nanami, Nagato-bon Heike monogatari no sōgō kenkyū: Kōchū hen 長門本平家物語の総合研究：校注篇, 306-07; Kitahara and Ōgawa, Enkyōbon Heike monogatari. Honbun hen 延慶本平家物語. 本文篇, 193-94. 236 The passage does not provide much new information about Itsukushima Shrine or its relevance for the Tale of the Heike. Rather, it paints Itsukushima with liturgical rhetoric heaped with Buddhist terminology and imagery to impress upon the reader Itsukushima’s sacred nature. In this manner, we can begin to see how compilers of the Nagatobon paid particular attention to Itsukushima as not only the tutelary shrine of the Taira, but also as a sacred place worthy of devotion. The Problematic History between Itsukushima and the Taira Identifying the Itsukushima Deity as the tutelary deity of the Taira raises questions: Why would anybody worship the Itsukushima Deity given its failure to protect its most fervent worshippers? Or conversely, why did the Itsukushima Deity agree to serve the Taira? On the one hand, because ruling authority was seen as determined by the gods, the Taira’s defeat, in itself, supported notions of the Taira as the enemies of the realm (chōteki 朝敵) who needed to be subjugated by Minamoto no Yoritomo as the “great general (dai shōgun 大将軍) protecting the realm. From this perspective, the Itsukushima Deity sided with the “enemies” and against the judgment of the other gods. On the other hand, prayers and offerings to the gods were expected to be reciprocated with divine assistance. Worship provided chances to influence the gods and change the future. 435 Even if people acknowledged that Itsukushima Deity was obligated to help the Taira, they could not deny that the deity had failed. If the counterargument is raised that the Itsukushima Deity supported the Taira in the short-term, within the long-term goal of instigating events that would lead to the restoration of order, then Itsukushima failed to uphold her 435 Concerning the balance between the fatalistic viewpoint of events as determined by the gods and the potential for influencing fate through prayers, offerings, and rites, see Thomas Conlan, State of War: The Violent Order of Fourteenth-Century Japan (Ann Arbor: Center for Japanese Studies, University of Michigan, 2003), 172-76. 237 obligation to the Taira. 436 This fatalistic viewpoint made it easy to affirm the position of the victors, and difficult to defend that of the losers. The two sides of divine protection of the realm and divine obligation to the Taira placed the Itsukushima Deity in a lose-lose situation. The gods were expected to do what was best for the realm, but they were also expected to serve the individual desires of their worshippers. 437 These were, of course, not always aligned. The monk Nichiren ( 日蓮, 1222-1282) raised this issue for arguing that disastrous events during his time, such as the burning of Tsurugaoka Hachiman Shrine in Kamakura, resulted from the deities prioritizing their worshippers over the needs of Buddhism (and consequentially the realm). In his 1280 Kangyō Hachimanshō 諫暁八幡抄, Nichiren uses the defeat of the Taira and the Itsukushima Deity as a prior example for the problem and the resultant problems. He states, “The Itsukushima Deity of Aki Province is the tutelary deity of the Taira. The Ise Shrine [Deity], Hachiman, and other [deities] beat the 436 David Bialock illuminates how Kiyomori was identified as a reincarnation of the Tendai monk Ryōgen ( 良源, 912-985) to reconceptualize Kiyomori’s “evil deeds” (akugyō 悪行) as actions for the protection of Buddhism. Based in medieval discourses of Ryōgen as a demon-subjugating general, Kiyomori and Ryōgen are together characterized as empowered to defeat the enemies of Buddhism through conventionally improper behavior. Supported by the prevalent medieval doctrine of original enlightenment (hongaku 本覚) in which good and evil, enlightenment and ignorance are nondual, even Kiyomori’s actions against Buddhist institutions and his feverish death could be good deeds promoting Buddhism. However, this line of logic was not applied to the Itsukushima Deity. Bialock, Eccentric Spaces, Hidden Histories: Narrative, Ritual, and Royal Authority from the Chronicles of Japan to The Tale of the Heike, 272, 76-78, 302-17. 437 The conflicting responsibilities of tutelary deities also becomes raised as an issue in medieval narratives about the conflict between Prince Shōtoku’s and Monobe Moriya ( 物部守屋, ?-587) over the introduction of Buddhism to Japan. The general narrative gives Shōtoku killing Moriya with an arrow empowe red with the might of the four heavenly generals protecting Buddhism, demonstrating the superiority of the Buddhist deities over the Japanese kami deities supporting Moriya. Notably, the 1317 Shōbōrinzō 聖法輪蔵 Shōtoku taishi den variant complicates Moriya’s motives by describing how Moriya receiving an oracle to destroy Buddhist temples and icons from Māra , the demonic obstructer of Buddhism, disguised as his tutelary deity, the Futsu Deity 符都大明神. The true identity of the deity as Māra protects the Futsu Deity from blame, but the specification of the deity as Moriya’s tutelary deity similarly problematizes the prioritized relationship with the tutelary family. Moriya trusts that his deity will guide him to the best possible outcome, but he is instead led to his death. "Shōbōrin -zō 聖法輪蔵," in Shinshū shiryō shūsei 眞宗史料集成, ed. Ishida Mitsuyuki 石田充之 and Chiba Jō ryū 千葉乘隆 (Kyoto: Dō hō sha, 1974), 460-61. Matsumoto Shinsuke includes citations from the Jingi kōshiki 神祇講式 and the Taiheiki 太平記 similarly describing Moriya being deceived by Māra , showing that it circulated outside of just Prince Shōtoku narratives. Matsumoto Shinsuke 松本真輔, Shōtoku Taishi den to kassentan 聖徳太子伝と合戦譚 (Tokyo: Bensei Shuppan, 2007), 134-35. 238 [Itsukushima] Deity as punishment for inciting the Taira’s arrogance, and the Taira were overthrown shortly afterwards.” 438 Nichiren provides a clear line of thought: the Taira damaged the authority of the sovereign and Buddhism for protecting the realm, and consequentially the clan and their deity were punished and defeated. The Itsukushima Deity is the model example here for the problems of kami worship. He concludes that the Mongol invasions were a response of the deities failing to devote themselves to the Buddhist teachings. From this perspective, Nichiren employed the defeat of the Taira and the Itsukushima Deity to suggest what will happen again if the people and gods continue to vie for power instead of supporting the Buddhist law and the Lotus Sutra. As Nichiren demonstrated, Itsukushima’s relationship with the Taira threatened the Itsukushima Deity’s image as an authoritative divinity. Additionally, its history with the Taira was of no help for aligning itself with the Kamakura Bakufu government. In 1232, the shrine replaced the annual rite of the Thousandfold [Lotus Sutra] Offering established by Kiyomori with the mitobiraki sechie (“rite of opening the shrine hall,” 御戸開節会) rite including prayers for the family of the shogun. 439 The simplest response for Itsukushima Shrine was to exclude the Taira from accounts of its past. For example, a chronology of Itsukushima’s history and head 438 Nichiren 日蓮, Nichiren: Shugo kokkaron, Hōon shō, hoka nijūichi-hen 日蓮 : 守護国家論, 報恩抄, 他二十一篇, ed. Tokoro Shigemoto 戶頃重基 and Takagi Yutaka 高木豊, Genten Nihon bukkyō no shisō 9 (Tokyo: Iwanami Shoten, 1991), 359. Also see Kawazoe Shō ji 川添昭二, "Mōko shūrai shiryō toshite no Nichiren ibun 蒙古襲来史料としての日蓮遺文," Fukujin 15 (2011): 39-41; Yamashita Masaharu 山下正治, Heike monogatari to hōshitachi: Chūsei no Bukkyō bungakuteki tenkai 平家物語と法師たち：中世の仏教文学的展開 (Tokyo: Kasama Shoin, 2007), 229-30. 439 While the practices involved in mitobiraki ceremonies differed depending on the place and the occasion, opening the doors would have allowed for a closer space with the deity in which to give offerings and prayers. For descriptions of two mitobiraki ceremonies, see the entries for the Katori jingū mitobiraki 香取神宮御戸開 and the Nukisaki jinja mitobiraki shinji 貫前神社御戸開神事 in Katō Tomoyasu 加藤友康 et al., eds., Nenjū gyōji daijiten 年中行事大辞典 (Tokyo: Yoshikawa Kōbunkan, 2009), 192, 539. For the record for inaugurating the mitobiraki sechie at Itsukushima, see doc. 111 in "Shinshutsu Itsukushima monjo 新出厳島文書," 354-55. 239 priests first written in 1457 (and last amended after 1554), starts with the founding of Itsukushima Shrine during the fifth year of Suiko Tennō’s reign 推古天皇 (597) and then skips to Retired Sovereign Takakura’s 高倉院 pilgrimage to Itsukushima in 1178. 440 This pilgrimage would not have occurred if not for Kiyomori’s relationship with Itsukushima, but such details are easily left out of a chronology. By the late sixteenth century, a few sources from Itsukushima Shrine begin to present the association with Kiyomori in a positive manner, but this still might have been exceptions to the norm. 441 In 1580, shrine priest Tanamori Fusaaki ( 棚守房顕, 1495- 1590) added Kiyomori’s restoration of the shrine into his record of Itsukushima Shrine, the Fusaaki oboegaki 房顕覚書. 442 Lived practices at Itsukushima Shrine and regional views towards the Taira in western Japan likely lay between the viewpoints of texts from Itsukushima suppressing the connection with Kiyomori and the Tale of the Heike’s emphasis on Itsukushima as the Taira’s protector. As recitations and common knowledge of the Heike increased, it would have become increasingly difficult to forget. By 1573, Kiyomori had been enshrined on the island at the new shrine 440 The dating to 1457 is based on the following sentence in the chronology, “Retired Sovereign Takakura’s pilgrimage to this shrine, 1178. From then until now, it has been 279 years.” The text ends with a list of years where flood tides damaged the shrine, with the first in 1511 and the last in 1554. See doc. 428 in "Nosaka monjo 野坂文書," in Hiroshima kenshi: Kodai chūsei shiryō hen 広島県史. 古代中世資料編, ed. Hiroshima-ken 広島県 (Hiroshima: Hiroshima, 1978), 1101. 441 In Matsui Teruaki’s discussion of the compilation of the Go hanmotsu chō 御判物帖 (from 1680-1681), a collection of historical documents showcasing the shrine’s distinguished past, he notes that the compilers paid little attention to sources mentioning Taira no Kiyomori or the head priest at the time, Saeki no Kagehiro. Matsui Teruaki 松井輝昭, Itsukushima monjo denrai no kenkyū: Chūsei monjo kanri shiron 厳島文書伝来の研究: 中世文書管理史論 (Tokyo: Yoshikawa Kōbunkan, 2008), 183 -84, 214. For two examples of sources about religious practices at Itsukushima positively portraying Kiyomori, see doc. 174 (1571: “Memorandum on the distribution of ritual objects for the reinstallation of [icons] at Itsukushima” Itsukushima jinja sengū reibutsu haitō oboe 厳島神社遷宮礼物配当覚); and doc. 301 (1616: “Dedicatory vow of Jun’an Sōun” Jun’an Sōun ganmon 準庵宗雲願文) in "Daiganji monjo 大願寺文書," in Hiroshima kenshi: Kodai chūsei shiryō hen 広島県史. 古代中世資料編, ed. Hiroshima-ken 広島県 (Hiroshima: Hiroshima, 1978), 1298-99, 405-406. 442 The text contains a colophon stating that it was finished in 1580, but it also includes events occurring in 1581. Tanamori Fusaaki 棚守房顕, "Fusaaki oboegaki 房顕覚書," ibid., 1107. 240 (imayashiro 今社), and a ceremony memorializing him was significant enough to be patronized by the daimyo lord Mōri Terumoto ( 毛利輝元, 1553-1625). 443 Additionally, a closer look at Tanamori’s inclusion of Kiyomori within the Fusaaki oboegaki reveals how he brought discourses from the Tale of the Heike and local legends into discussion with Itsukushima’s past: The descendants of the worshippers of this shrine will flourish and be exultant in their fortunes and glory. Previously, when Lay Priest Kiyomori made a pilgrimage [to Itsukushima], unsure whether it was a dream or reality, he received a long sword with its shaft wrapped in silver. The Deity told him, “With this blade, I shall lead you to rule over the four seas and everything under heaven. However, if you commit evil deeds, my protection will not extend to your descendants.” Additionally, Kiyomori made monthly pilgrimages from Fukuhara, and the Ondo Seto [strait] was dug during that time. 444 The transmission of the sword and the oracle warning him against committing evil deeds became widely known from the Kakuichibon Tale of the Heike’s telling of the origins of Kiyomori’s worship of Itsukushima. 445 This will be discussed more later, but in short, the scene shows the Itsukushima Deity granting him the sword as a symbol of her support for the Taira and their rise to power as the protectors of the realm. Additionally, the mention of the Ondo Seto strait refers to a local legend glorifying Kiyomori for opening the strait and supporting sea trade and travel. According to late early modern travel diaries, Kiyomori began to consider digging a strait to 443 Matsui Teruaki suggests that Kiyomori’s brother, Taira no Yorimori, might have funded the building of the shrine for Kiyomori in 1212 and that earlier rites performed there could have been for the sake of Kiyomori, even though the connection is not made explicit until the record of the 1573 Grand Minister Taira [no Kiyomori] ceremony (Taira shōkoku onmatsuri 平相国御祭). There are records of Yorimori offering private estate funds to Itsukushima for the building of a new shrine (although using the term shingū 新宮 rather than imayashiro 今社), but he does not expand upon his intentions. Matsui Teruaki 松井輝昭, "Itsukushima jinja ni okeru Taira no Kiyomori shinkō no genryū ni tsuite: 'Imayashiro' no seiritsu wo chūshin ni 厳島神社における平清盛信仰の源流について：｢今社｣の成立を中心に," in Gakusei sanka ni yoru sekai isan Miyajima no kasseika: Gakusei ga Miyajima no miryoku o saihakkenshi, sekai ni hasshinsuru, Saishū hōkokusho 学生参加による世界遺産宮島の活性化 : 学生が宮島の魅力を再発見し, 世界に発信する : 最終報告書 (Hiroshima-shi: Kenritsu Hiroshima Daigaku Ningen Bunka Gakubu, 2009), 187, 90-92. 444 Tanamori, "Fusaaki oboegaki 房顕覚書," 1107. 445 Tyler, The Tale of the Heike, 177; Ichiko, Heike monogatari 平家物語, vol. 45, p. 205. The Nagatobon and Engyōbon variants do not attribute the fall of the Taira to Kiyomori’s misdeeds, and instead, the Itsukushima Deity simply tells him that his success will only last for one generation. 241 improve the route of his monthly pilgrimage to Itsukushima, and at that time, the Itsukushima Deity appeared and promised to become his wife if he built the strait in a single day. On that day, the sun began to set before the work was complete, so Kiyomori used a fan to guide the sun backwards and provide additional midday light. The Itsukushima Deity felt tricked by Kiyomori’s tactics, and in her anger, transformed into a giant snake. Unable to escape because of the tide, Kiyomori stopped the snake in her tracks with his mighty glare, until the tide finally waned, and he was able to run across. 446 This legend also draws on a Heike episode in which spirits haunt the Taira after the move of the capital to Fukuhara, and Kiyomori glares back at the nightmarish apparitions until they disappeared, and then applies it to the context of the Seto Inland Sea area by connecting Kiyomori with Itsukushima Shrine and maritime trade (scroll 5, Mokke no sata 物怪之沙汰). 447 The Ondo Seto strait legend is a rare example of a regional and largely orally transmitted legend that can be traced back to the late medieval period. Placed together with episodes from the Heike, Tanamori Fusaaki’s account of Itsukushima Shrine’s history shows how the Tale of the Heike influenced notions of Itsukushima Shrine but still did not become the primary source for the shrine’s identity. Questioning Itsukushima’s Identity as the Taira’s Tutelary Deity Kiyomori treated the Itsukushima Deity as his household’s tutelary deity (ujigami 氏神), but was that enough to make it so? The Itsukushima Deity’s close relationship with the Taira was undeniable, but its position as tutelary deity was more complicated. In fact, the Taira already had Hirano Shrine 平野社 as their tutelary shrine, and it is unclear to what extent Itsukushima 446 Shimomukai Tatsuhiko 下向井龍彦, "Taira no Kiyomori Ondo seto 'Hi maneki' kaisaku densetsu no keisei to shintō 平清盛音戸瀬戸「日招き」開削伝説の形成と浸透," Geibi chihōshi kenkyū 282-283 (2012): 48, 50-51. 447 Ichiko, Heike monogatari 平家物語, v.45, p. 361; Tyler, The Tale of the Heike, 261. 242 replaced Hirano in the practices of the wider Taira household, or in the recognition of court society. 448 During the Taira’s rise to power, Itsukushima’s service as their tutelary deity positioned it alongside the most elite shrines in the realm, such as Ise Shrine as the god of the royal family, and Kasuga Shrine as the god of the Fujiwara. Likewise, the Taira’s worship of Itsukushima both aligned with those of the ruling elites while also asserting their difference. Itsukushima’s location in western Japan and its associations with sea travel shifted the bounds of geopolitical authority, symbolizing the Taira’s heterodox authority and a new political era. 449 After their demise, however, the idea of a familial relationship connecting Itsukushima and the Taira could increase the difficulty of distancing Itsukushima from association with the Taira’s deeds and failures. Significantly, both the Tale of the Heike and other medieval sources often mention a close relationship between Itsukushima and the Taira, but rarely specify it as tutelary. Itsukushima Shrine was not the only religious institution to be criticized for supporting the Taira, because people recognized that the Taira also worshipped at other sites. In particular, the Enryakuji abbot Myōun ( 明雲, 1115-1183) had a close relationship with Kiyomori, and the temple’s position within the competitive religious-political sphere of the center made it a larger target for censure than the faraway Itsukushima Shrine. The intersecting relations connecting the Taira with Itsukushima, Hirano, and Enryakuji come to light in an episode from 1183. At this time, Kiyomori had already died and the Taira were nearing defeat. Recognizing the need for assistance, Taira no Munemori ( 平宗盛, 1147- 1185) offered to make Enryakuji their tutelary temple in exchange for their divine assistance in defeating the Minamoto. The courtier Yoshida Tsunefusa ( 吉田経房, 1142-1200) describes 448 Tanaka Hisao 田中久夫, Ujigami shinkō to sosen saishi 神信仰と祖先祭祀 (Tokyo: Meicho Shuppan, 1991), 353-60. 449 Blair, "Rites and Rule: Kiyomori at Itsukushima and Fukuhara," 12, 18-22. 243 Munemori’s offer to Enryakuji in his diary, the Kitsuki 吉記 (7/12/1183). Notably, he comments that this could displease the Taira’s tutelary deity, who is given as the Hirano Deity rather than the Itsukushima Deity. He writes: I heard it said, “Ten Taira courtiers, headed by the Inner Palace Minister (naidaijin 内大臣) [Taira no Munemori 平宗盛], [plan to] take Hie Shrine as their tutelary shrine and Enryakuji as their tutelary temple. [They] wrote a vow asking for permission to devote themselves [to the Sannō Deity], and it was sent among the monks [of Mt. Hiei].” Was this supposed to be a secret matter? Hearing this letter, it was hard to hold back sorrowful tears. However, [by] forsaking Hirano Shrine for use as [their] tutelary shrine, might the will of the god be dreadful? 450 Tsunefusa would not yet have known how Enryakuji would respond to the appeal, or how things would turn out for the Taira. His tears, though, suggest that he was aware of the poor outlook for the Taira. Even if they succeeded in gaining the aid of the Sannō Deity and the Hiei monks, their current tutelary deity, the Hirano Deity, could respond and bring misfortune. There are few historical sources that explicitly state the Taira’s tutelary deity, but Tanaka Hisao explains that the four deities of Hirano Shrine were the ancestral deity of the descendants of Kanmu Tennō ’s mother, and thus served as the tutelary deities of the Kanmu Taira. 451 Tanaka acknowledges that Kiyomori and his close family members might have intentionally avoided worshipping at Hirano Shrine and instead treated Itsukushima as their tutelary shrine, but there is still little evidence that Itsukushima officially replaced Hirano as their tutelary shrine. Yoshida Tsunefusa still considered Hirano as the tutelary shrine, and there were also Taira members, including Taira no Nobunori ( 平信範, 1113-1187) and Kenshunmon’in ( 建春門院, 1142-1176, older sister of Taira no Tokiko 平時子, Kiyomori’s wife), who continued to worship at Hirano. 452 Kiyomori’s 450 Yoshida Tsunefusa 吉田経房, Kitsuki 吉記, ed. Takahashi Hideki 高橋秀樹, 3 vols., vol. 3, Nihonshi shiryō 日本史史料 (Osaka: Izumi Shoin, 1965), 82. 451 Taira no Kiyomori and his family were members of the Ise Taira, which was a sublineage of the Kanmu Taira. 452 Tanaka, Ujigami shinkō to sosen saishi 神信仰と祖先祭祀, 355-69. Tanaka emphasizes the role of Taira no Nobunori, in overseeing the Hirano festival in 1166 (eighth day of the eleventh month), because this position was 244 contemporaries recognized the Taira’s growing patronage for Itsukushima Shrine, but they would have still been aware of their relations with Hirano Shrine and Enryakuji. Enryakuji refused Munemori’s offer and never became the Taira’s tutelary temple, but they were still attacked for supporting the Taira. The same Nichiren who had faulted Itsukushima for being the tutelary deity of the Taira in 1280 had made a similar argument against Enryakuji and Hie Shrine five years earlier. In his 1275 Shinkoku ōgosho 神国王御書, Nichiren gives Enryakuji as the Taira tutelary deity as part of a polemic against the increasing use of esoteric rites (rather than devotion to the Lotus Sutra) at Enryakuji and Ninnaji. Notably, he misappropriates the context of the Taira’s appeal to Enryakuji, describing it as an apology letter (taijō 怠状) given from Kiyomori and the Taira clan to Mt. Hiei on the occasion of Myōun becoming the Enryakuji abbot in 1167. In this manner, Nichiren connects Myōun’s historical support for Kiyomori with the false context of Enryakuji accepting the Taira’s offer to become their tutelary temple. This premise allows Nichiren to suggest that Enryakuji performed esoteric rites to protect their tutelary family but lacked the power to succeed. Other texts kept closer to the facts, but still used the common awareness of the close relationship between the Taira and Enryakuji to elevate religious institutions that had supported the Minamoto victors above Enryakuji. For example, a 1319 petition by Onjō ji against Enryakuji (Ō mi Onjōji gakuto shukurō ra mōshijō 近江園城寺学徒宿老等申状) gives Kiyomori’s devotion towards Enryakuji as evidence of Onjō ji’s superiority. It juxtaposes the victors and given to a member of the tutelary family when an official from Board of Controllers (benkan 弁官) was not in attendance. The Nihon kiryaku 日本紀略 entry from 1019 (eighth day of the eleventh month) provides a precedent, stating, “[Today is the] spring festival of Hirano Shrine. The controller did not attend. The Hirano festival [thus] took the tutelary person, Taira no Masayoshi 平理義, as the substitute controller (bendai 弁代).” For the passages concerning the Hirano festival, see ibid., 355; Taira Nobunori 平信範, Hyōhanki 兵範記, 5 vols., Zō ho shiryō taisei 増補史料大成 (Kyoto: Rinsen Shoten, 1965), v. 3, p. 124; Nihon kiryaku 日本紀略, Kokushi taikei 國史大系 (Tokyo: Keizai Zasshisha, 1897), 1122. 245 losers of the Genpei War as the Taira and Enryakuji, and the Minamoto and Onjōji respectively, saying, “Daishōkoku [Taira no Kiyomori] was devoted to the Sanmon; his clan was demolished in the Western Seas. The Minister of the Right Minamoto [Yoritomo] was devoted to the Jimon; his clan has prospered in the east. It is clear that the prayers of the Mountain [Enyrakuji] cannot compare to the protective powers of our temple.” 453 Regardless of the Taira’s affiliation with Itsukushima, their relationship with Mt. Hiei was enough for Onjōji to position Enryakuji as inferior to Onjōji . In short, a simple understanding of the Genpei War as the Minamoto’s righteous victory over the Taira opened opportunities for praising the deities that sided with the Minamoto and disparaging those allied with the Taira. 454 The attacks against the gods aligned with the Taira were not limited to Itsukushima, and in fact, it is easier to find sources criticizing Enryakuji. This is not to say that Enryakuji was considered to be more closely associated with the Taira than Itsukushima, but rather, that more religious institutions stood to benefit from contrasting themselves with Enryakuji. The Tale of the Heike and other historical texts retell the episode of Munemori’s appeal to Enryakuji and reveal the types of arguments raised to defend religious institutions who had supported the Taira. The simple answer is that Enryakuji refused the Taira’s offer and instead supported the Minamoto by granting Kiso Yoshinaka ( 木曾義仲, 1154-1184) and his troops safe passage through Mt. Hiei. The fact that texts like the Rokudai shōjiki ( 六代勝事記, ca. 453 Translation from Wakabayashi, Haruko Nishioka, The Seven Tengu Scrolls: Evil and the Rhetoric of Legitimacy in Medieval Japanese Buddhism (Honolulu: University of Hawai`i Press, 2012), 172. 454 There are also examples of temple and shrine origin narratives which align their deities with the Minamoto without mentioning the tutelary sites of the Taira. For two examples, see the Hachiman gudōkun (approximately 1308-1318) and the Utsunomiya daimyōjin daidai kizui no koto (1484), "Hachiman gudō kun (kō ) 八幡愚童訓（甲)," in Jisha engi 寺社緣起, ed. Sakurai Tokutarō 桜井徳太郎, Hagiwara Tatsuo 萩原龍夫, and Miyata Noboru 宮田登 (Tokyo: Iwanami Shoten, 1975), 196; "Utsunomiya daimyōjin daidai kizui no koto 宇都宮大明神代々奇瑞之事," in Gunsho ruijū. Jingi bu 群書類従. 神祇部, ed. Hanawa Hokinoichi 塙保己一 (Tokyo: Gunsho Ruijū Kankō kai Kantō sha Nai, 1952), 308. 246 1222~1224) and the Hōryaku kanki ( 保暦間記, ca. 1346~1370) specify the reasons for their rejection of the Taira suggests that worries remained about being remembered as having supported the Taira earlier. 455 Tellingly, both of these texts and the Heike argue that Enryakuji could not support the Taira after they had committed evil deeds (akugyō 悪行). For example, the Hōryaku kanki explains: The age of the Taira lasted over twenty years, [but] could not bear the weight of their evil deeds. It seems that their karmic fruits had reached their limit. Additionally, Lay Priest [Kiyomori] can be seen as having become a substitute for Māra . Would Amaterasu or Shō Hachiman have protected him? Would even Hiyoshi Sannō, who was especially worshipped by [Kiyomori], and had manifested before him when he had offered the sutras, have protected him? 456 While the deities were expected to respond to the prayers of their devotees, they still had the choice not to leave some wishes unanswered. Although the Sannō Deit y had responded favorably to Kiyomori’s ardent worship previously, the same support for the Taira could not be expected after their evil deeds had accumulated. The Taira’s perceived misdeeds could be used to represent deities and institutions that continued to support the Taira in a negative light. The safest options for the gods was to distance themselves from the Taira. 455 The Rokudai shōjiki and the Hōryaku kanki are medieval historical narratives of unknown authorship, and they cover the years 1168-1221 and 1156-1339 respectively. Michael McCarty describes the Rokudai shōjiki as a historical commentary written with the intention of presenting Go-Toba as responsible for the Jōkyū disturbance in 1221. Regarding the Hōryaku kanki, it is known by Heike scholars because its sections on the Genpei War are heavily influenced by The Tale of the Heike (and particularly the Shibu kassen jōbon variant), and Saeki Shin’ichi further argues that the Hōryaku kanki producers used the Heike to reflect on more recent events. For an introductory explanation and translation of the Rokudai shōjiki, see Michael McCarty, "Divided Loyalties and Shifting Perceptions: The Jōkyū Disturbance and Courtier Warrior Relations in Medieval Japan" (Ph.D. diss., Columbia University, 2013), 39-40, 224-51. For the episode of the Taira’s appeal to Enryakuji, see ibid., 228-29; "Rokudai shōjiki 六代勝事記," in Gunsho ruiju. Teiōbu 群書類從.帝王部, ed. Hanawa Hokinoichi 塙保己一 (Tokyo: Yagi Shoten, 2013), 406-07. On the Hōryaku kanki see, Saeki Shin'ichi 佐伯真一, "'Hōryaku kanki' no rekishi jojutsu 『保暦間記』の歴史叙述," Denshō bungaku kenkyū 46 (1997): 49-53; "'Heike monogatari' to 'Hōryaku kanki': Shibubon, Seisuiki kyōtsu sohon no sōtei 『平家物語』と『保暦間記：四部本・盛衰記共通祖本の想定," Chūsei bungaku 40 (1995). 456 "Hōryaku kanki 保暦間記," in Gunsho ruijū. Zatsu bu 群書類従. 雑部, ed. Hanawa Hokinoichi 塙保己一 (Tokyo: Gunsho Ruijū Kankō kai Kantō sha Nai, 1960), 128. 247 The Tale of the Heike further emphasizes Enryakuji’s rebuke of the Taira by quoting the letters sent by the three parties of Taira no Munemori, Kiso Yoshinaka, and Enryakuji. The narrative sequence moves from Yoshinaka’s appeal to the Taira’s to present both a sympathetic view of the Taira’s position, and a defensive explanation for Mt. Hiei diverging from its prior support for them. They justify their past actions as protecting the relatives of the sovereign. However, once the Taira doomed themselves by committing evil deeds, siding with the Minamoto can be seen as siding with the future protectors of the realm. The Kakuichibon succinctly states, “The Heike are, in the current reign, His Majesty’s commoner relatives, and they therefore command respect from all of us here on the Mountain. That is why we have always, so far, prayed that they, too, should flourish greatly. However, their evil passes all bounds, and the people have turned against them.” Shortly after, the letter concludes, “Their time is coming. Why should only our Mountain (Mt. Hiei) ally itself with the Heike, whose time is past, and oppose the Genji, whose future stretches before them?” 457 Why would Mt. Hiei or Itsukushima continue to support the Taira once the household’s defeat seemed inescapable? An argument could be made for the Taira’s tutelary deity serving them to the end, but would this be anything more than an empty gesture? The Engyōbon variant raises this question in an episode about an oracle from the Itsukushima Deity. The Taira receive an oracular message saying, “the divine wind of Itsukushima will topple the Minamoto family.” The Engyōbon then explains, “Based on the deity having this oracle, [the Taira] rejoiced at there 457 Translation from Tyler, The Tale of the Heike, 376; Ichiko, Heike monogatari 平家物語, v. 46, pp. 54-55. The corresponding passages in the Nagatobon and Engyōbon are similar. The Nagatobon largely matches t he Engyōbon, although it reduces the length of Kiso Yoshinaka’s letter to Mt. Hiei, removing some of the rhetoric of the Taira’s evil deeds. Additionally, the Nagatobon adds two sentences praising the Taira’s support for Mt. Hiei before citing their appeal to Mt. Hiei. It states, “Mt. Hiei has not allied with our enemies. Additionally, our family has never been disloyal to Mt. Hiei.” Asahara and Nanami, Nagato-bon Heike monogatari no sōgō kenkyū: Kōchū hen 長門本平家物語の総合研究：校注篇, v.2, p. 1012. 248 being no doubt that the Taira will flourish and the Minamoto will fall. However, afterwards it was heard that the Itsukushima head priest intended to pray for the Taira, but deep within his heart, he prayed for an end to the war, [and therefore] fabricated the matter [of the oracle]. 458 The Engyōbon passage recognizes the special relationship between the Taira and Itsukushima, but still questions why the deity would continue to defend the Taira as the household’s inevitable demise became clear. Enryakuji’s support for the Minamoto over the Taira in 1183 provided a foundation for distancing themselves from the losing side, but the best that could be said for Itsukushima was that they were not the Taira’s tutelary deity. Shifting Perspectives from the Early to Late Medieval A broad survey of medieval texts relating Itsukushima and the Taira reveals that most avoid calling Itsukushima the tutelary deity. Out of eleven non-Heike texts, only the following three refer to Itsukushima as the Taira’s tutelary deity: Ruiju kigenshō 類聚既験抄 (late thirteenth century); Jinkeishō 塵荊鈔 (approximately 1482~1491); and the “Itsukushima” daimokutate 題目立 performance libretto (pre-1575). 459 (A list of all eleven are given in table 4 below. 460 ) 458 Kitahara and Ōgawa , Enkyōbon Heike monogatari. Honbun hen 延慶本平家物語. 本文篇, v.2, p. 24. 459 "Ruiju kigenshō 類聚既験抄," in Chūsei shōdō shiryōshū 中世唱導資料集, ed. Okabe Kaien 岡部快圓, et al. (Kyoto: Rinsen Shoten, 2000), 246; Ichiko Teiji 市古貞次, ed. Jinkeishō 塵荊鈔, 2 vols. (Tokyo: Koten Bunko, 1984), v.2, p. 260; "Itsukushima 厳島," in Nihon shomin bunka shiryō shūsei: Dengaku, sarugaku 日本庶民文化史料集成：田楽・猿楽 (Tokyo: San'ichi Shobō , 1973), 565. While the earliest fully extant script for the “Itsukushima” daimokutate is from 1733, colophons from a partial manuscript belonging to Niu Shrine indicate that the daimokutate was performed on 1/17/1575, in the eighth month of 1594, and during the Keichō era ( 慶長, 1596- 1615). "Itsukushima (zanketsu bon) 厳島（残欠本）," in Nihon shomin bunka shiryō shūsei: Dengaku, sarugaku 日本庶民文化史料集成：田楽・猿楽 (Tokyo: San'ichi Shobō , 1973), 581-82. On the dating of the Ruiju kigenshō and the Jinkeishō, see Abe Yasurō 阿部泰郎, "Ruiju kigenshō : Kaidai 類聚既験抄：解題," in Chūsei shōdō shiryōshū 中世唱導資料集, ed. Okabe Kaien 岡部快圓, et al. (Kyoto: Rinsen Shoten, 2000), 335; Matsubara Kazuyoshi 松原一義, "Jinkeishō" no kenkyū 「塵荊抄」の硏究 (Tokyo: Ō fū, 2002), 110-13. 460 See Minamoto Akikane 源顕兼, Kawabata Yoshiaki 川端善明, and Araki Hiroshi 荒木浩, eds., Kojidan, Zoku Kojidan 古事談. 続古事談, Shin Nihon koten bungaku taikei (Tokyo: Iwanami Shoten, 2005), 478-79; Kitabatake 249 Table 4: Medieval Texts Associating Itsukushima and the Taira Text Date Kojidan 古事談 1205 Ruiju kigenshō 類聚既験抄 Late 13th century Jinnō shōt ōki 神皇正統記 1339 Hōryaku kanki 保暦間記 1346-1370 Gikeiki 義経記 Early Muromachi Period Mantokujibon Shō toku Taishi den 万徳寺本『聖徳太子伝』 1462 Chūkai risshi monogatari 忠快律師物語 Pre 1470 Jinkeishō 塵荊鈔 Between 1482-1491 Itsukushima no honji 厳島の本地 Late Muromachi Period "Itsukushima" daimokutate 『厳島』題目立 Pre-1575 Fusaaki oboegaki 房顕覚書 1580 While the Ruiju kigenshō gives a single sentence defining the Itsukushima Deity as the Taira tutelary deity, the Jinkeishō and “Itsukushima” daimokutate include the tutelary role alongside the other well-known identities of the deity as the daughter of the Dragon King Sagara, and as a manifestation of the Dainichi Buddha of the womb world mandala. The daimokutate adds a fourth identity as the sea goddess Benzaiten. 461 As were discussed in chapters one and two, Itsukushima’s associations with the Dragon King and Benzaiten became established as crucial identities of the Itsukushima Deity over the medieval period. Listing the association with Chikafusa 北畠親房, "Jinnō shōtōki 神皇正統記," in Gunsho ruiju. Teiōbu 群書類從. 帝王部, ed. Hanawa Hokinoichi 塙保己一 (Tokyo: Yagi Shoten, 2013), 91; Chikafusa Kitabatake, A Chronicle of Gods and Sovereigns: Jinnō shōtōki of Kitabatake Chikafusa, trans. H. Paul Varley (New York: Columbia University Press, 1980), 214; "Hōryaku kanki 保暦間記," 130; Kojima Satoaki 小島惠昭 and Watanabe Nobukazu 渡辺信和, "(Kyōdō kenkyū) Mantokuji-zō 'Shōtoku taishi den' honkoku 〔共同研究〕万徳寺蔵『聖徳太子伝』翻刻," Dōhō Daigaku Bukkyō Bunka Kenkyūjo kiyō 2 (1980): 271; Kajihara Masaaki 梶原正昭, ed. Gikeiki 義経記, Shinpen Nihon koten bungaku zenshū (Tokyo: Shō gakkan, 2000), 205; "Itsukushima no honji 厳島の本地," in Muromachi jidai monogatari taisei 室町時代物語大成, ed. Yokoyama Shigeru 横山重 and Matsumoto Ryūshin 松本隆信 (Tokyo: Kadokawa Shoten, 1973), 294; Tanamori, "Fusaaki oboegaki 房顕覚書," 1107-08. 461 The Ruiju kigenshō includes three separate accounts about Itsukushima. It is only the first that mentions the tutelary relationship with the Taira, and the other two focus on Kukai’s pilgrimage and worship of Itsukushima Shrine. "Ruiju kigenshō 類聚既験抄," 246, 55, 58; Ichiko, Jinkeishō 塵荊鈔, v.2, p. 260; "Itsukushima 厳島," 565. 250 Kiyomori alongside the others presents it as important, but not as the primary or only source for the deity’s identity. Aside from the Ruiju kigenshō, the other pre-fifteenth century texts (Kojidan, Jinnō shōtoki, and Hōryaku kanki) all present the Taira connection in a negative fashion, although to varying degrees. The Kojidan uses the narrative of the origins of Kiyomori’s worship of Itsukushima to argue that even though both the kami Amaterasu and Itsukushima are manifestations of Dainichi Buddha, Itsukushima is a lesser form and therefore appropriate for Kiyomori. 462 Additionally, the Jinnō shōtōki speaks in the neutral tone of a historical genealogy of past sovereigns, but implicitly contrasts the Taira (and Itsukushima by association) as those responsible for bringing chaos to the realm with Retired Sovereign Takakura as the one who worked to restore order to the realm. 463 The later, post-fifteenth century texts primarily use the connection between Itsukushima and the Taira in more positive, but often still implicit, manners. They avoid mentioning the Taira’s defeat, and instead focus on Kiyomori’s restoration of the shrine during his rise in power. In other words, they recount Itsukushima’s past rather than the Taira past. For example, the Itsukushima no honji narrative ends by identifying Kiyomori as the copyist of the tale, “In the second year of Hō gen (1157 保元), on the sixteenth day of the second month, Taira no Kiyomori, due to his trust in Itsukushima shrine, copied this origin narrative (engi 縁起) anew and offered it to the treasury; this is that [work].” 464 The reference to Kiyomori gives the 462 Minamoto, Kawasaki, and Araki, Kojidan, Zoku Kojidan 古事談. 続古事談, 478-79; Murata Shin'ichi 村田真一, ""Hachiman Usagū go takusenshū" no sekai: Usa Hachiman shinwa gensetsu no kenkyū 『八幡宇佐宮御託宣集』の中世：宇佐八幡神話言説の研究" (Bukkyō University, 2015). 463 Kitabatake, "Jinnō shōtōki 神皇正統記," 91; Kitabatake, A Chronicle of Gods and Sovereigns: Jinnō shōtōki of Kitabatake Chikafusa, 214. 464 "Itsukushima no honji 厳島の本地," 294. 251 narrative a past tracing back to the twelfth century, and suggests a parallel between Kiyomori’s reading of the narrative and devotion to Itsukushima with that of later readers. The 1462 Mantokujibon Shōtoku taishi den is notable for adding Kiyomori into a preexisting Itsukushima Shrine origin narrative as a restorer of the shrine. The Mantokujibon is a variant of the 1317 Shōbōrinzō 正法輪蔵 account of the life of Prince Shōtoku that occasionally incorporates new details and comments made by the compiler. The Shōbōrinzō includes an Itsukushima origin narrative in the chapter on Prince Shōtoku’s thirty -first year, and the Mantokujibon is the earliest variant to add a comment on the later decline of the shrine and its revival by Kiyomori. It says, “Without going into detail, after that, [the shrine] fell to ruin. It was restored by the originally named shogun (honsei no shōgun 本姓ノ将軍).” 465 Kiyomori is never mentioned by name, but the reference would be clear to anyone familiar with the Tale of the Heike or the Genpei War. Additionally, the title of “originally named shogun” positively positions Kiyomori and the Taira as the pre-Minamoto warrior protectors of the realm, and away from the title of “enemy of the realm.” 466 The victory of the Minamoto memorialized them as the protectors of the realm, but the Mantokujibon hints at the other side of the story, remembering the good Kiyomori did for the realm and for Itsukushima Shrine. 465 My reading of “honsei no shōgun 本姓の将軍 (“originally named shogun”) diverges from the transcription’s “kishō no shōgun” 木姓ノ将軍 (“shōgun named of wood”) , but considering the context and the similarity of 木 and 本, it is likely that the copyist wrote kishō 木姓 as an abbreviation. Kojima and Watanabe, "(Kyōdō k enkyū) Mantokuji-zō 'Shōtoku taishi den' honkoku 〔共同研究〕万徳寺蔵『聖徳太子伝』翻刻," 271. 466 While the Heike is conventionally read as villainizing the Taira and legitimizing the Minamoto, Hyōdō Hiromi argues for a cyclic restoral of order through the Minamoto and Taira exchanging positions with one another. The Taira’s first defeat of the Minamoto led Go-Shirakawa to promote Kiyomori to protect the realm and restore order, before disorder again required power to shift to the Minamoto. Even after the Genpei War, the Ashikaga associated themselves with the Minamoto and the vanquished Hōjō 北条 with the Taira. Afterwards, Toyotomi Hideyoshi identified himself with the Taira, symbolizing his reign as a break from the prior rule of the Ashikaga bakufu (as descendants of the Minamoto). Hyō dō Hiromi 兵藤裕己, Ō ken to monogatari 王権と物語 (Tokyo: Seikyūsha, 1989), 82-85. 252 Lastly, the “Itsukushima” daimokutate demonstrates how performance and ritual could temporally shift attention to the Taira’s political and religious apex and away from their downfall. Elizabeth Oyler introduced the genre of daimokutate to audiences outside of Japan, describing it as an aural-centric performance in which participants recount episodes from the Tale of the Heike by each giving voice to a different character. Oyler explains that daimokutate had been performed as part of coming-of-age ceremonies at shrines in central Japan, and based on this, it is difficult to say if the “Itsukushima” daimokutate was originally produced to be performed at Itsukushima Shrine or in other areas of western Japan where local audiences might feel an affinity with the shrine or its deity. Looking more broadly to the significance for members of the warrior class, she argues that they could impress upon performers and audiences their association with past warriors, and moreover, they could negate the need to interpret the past events of the Genpei War in relation to the present. 467 This means of representing the Taira and Itsukushima apart from the outcome of the war would have been enticing for worshippers of Itsukushima Shrine and individuals seeking to associate themselves with the Taira. The daimokutate enacts the moment of a Lotus Sutra recitation offering at Itsukushima Shrine, which were resplendently performed at Itsukushima and Fukuhara during the time of the Taira’s power. 468 Rather than using this setting to ignore the Heike’s account of the Taira’s defeat, the performance recalls moments from the Heike and adapts them to the time of the Taira’s successes. The most notable example is the council of the gods scene. Told through an 467 Elizabeth Oyler, "Daimokutate: Ritual Placatory Performance of the Genpei War," Oral Tradition 21, no. 1 (2006): 90, 115. 468 As Oyler suggests, it could refer specifically to the ritual offering of the Heike nōkyō 平家納経 illustrated scrolls sponsored by the Taira family in 1164, but the Taira also sponsored a similar one-thousand-part sutra memorial offering (senbu kyō kuyō 千部経供養) in 1177. Additionally, the overlaying of this specific ritual event with the ambiguity of which historical event it was, supports Oyler’s arguments about how the performance combines multiple spaces, times, and identities. Ibid., 100, 03-09. For a summary of all the sutra memorial offerings sponsored by Kiyomori, see Mukasa Akira 武笠朗, "Taira Kiyomori no shinkō to heishi no zōji, zōbutsu (jō) 平清盛の信仰と平氏の造寺・造仏( 上) " Jissen joshi daigaku bijutsu-shi gaku 13 (1998): 29. 253 oracular dream received by a retainer of Minamoto no Masayori ( 源雅頼, 1127-1190), the episode tells of a meeting between the deities wherein the sword of the three regalia (symbolizing authority to protect the realm) is transferred from Kiyomori to the Minamoto. Most Heike variants present the Itsukushima Deity within this scene as powerless against the will of the most senior deities. In contrast, the daimokutate recontextualizes the dream in order to show the Itsukushima Deity triumphing over the tutelary deities of the prestigious Fujiwara and Minamoto families. The Itsukushima Deity in the daimokutate says: Now, Kiyomori, listen well: I am at an assembly of the gods. Amaterasu asks that her clan be given control of the world. The Kasuga Deity asks that its clan be given control of the world. And Hachiman asks that the Minamoto be given control of the world. Among all these requests of the gods, I declare that at this moment the control of the world shall go to Kiyomori. By giving you this sword to pacify the realm, I insure that there will be no cause for concern in the world. 469 The daimokutate manages to reverse the meaning of the scene by changing the timing from when Kiyomori lost the sword to the time when he received it. The succinct conclusion that Kiyomori need not worry limits the timeframe to the moment of the rite. This is undoubtedly, in part, a strategy for overlooking the later outcome for the Taira, but the temporal structure of the performance prevents it from feeling deceptive. The daimokutate further praises the Itsukushima Deity for her martial power. As Oyler notes, Kiyomori connects his sponsoring of the rite with the deity’s assistance in ensuring his victories in the Hōgen and Heiji uprisings. 470 Munemori also calls attention to the Itsukushima Deity’s military strength by describing her iconographic form: “Benzaiten of the Matrix-store realm, avatar of Dainichi and third daughter of the Dragon King Sagara, manifests herself. In her six arms she holds the six virtues. Grasping a bow and carrying a halberd, she protects heaven 469 Translation from Oyler, "Daimokutate: Ritual Placatory Performance of the Genpei War," 111. In Japanese, see "Itsukushima 厳島," 568. 470 Oyler, "Daimokutate: Ritual Placatory Performance of the Genpei War," 103. 254 and earth. The Itsukushima Deity is the reason for all our blessings.” 471 The Itsukushima Deity is conventionally presented as the two-armed Benzaiten holding the biwa lute, but the daimokutate’s change to a six-armed Benzaiten arms her for battle. 472 In this manner, the performative and ritual doubling of spaces and times re-presents the Itsukushima Deity as a powerful god of war divorced from the fact that the Taira (and the Itsukushima Deity) lost the Genpei War. The daimokutate provides a thorough reframing of Itsukushima’s relationship with the Taira, but the end result is similar to the other late medieval texts discussed above. They remember the Taira for their historical accomplishments and not their failures. The passage of time made it easier to prioritize the Taira’s historical significance as an elite warrior household and to disassociate them from their end in the Genpei War and the Heike. Reframing Itsukushima within the Nagatobon Tale of the Heike The Nagatobon resembles other contemporary medieval texts in its efforts to separate Itsukushima from responsibility for the fall of the Taira. However, some of the established Heike episodes cannot be told without addressing the relationship between Itsukushima and the Taira. It is central to both the episodes of Kiyomori rebuilding the great stupa of Mt. Koya and receiving an oracular message to worship Itsukushima (scroll 5, Itsukushima shidai no koto 厳島次第事), 473 and of the council of the gods deciding to transfer authority to protect the realm from the 471 Translation from ibid., 104. In Japanese, see "Itsukushima 厳島," 565. 472 The Kinpusen himitsu den ( 金峯山秘密伝, 1337) describes the iconography of the six-armed Benzaiten. Although it presents the weapons as metaphors for Benzaiten’s salvific power to “subjugate the three obstacles and four demonic forces” that can obstruct the Buddhist path, and to lead people to enlightenment like “an arrow of wisdom,” the same text also includes a Benzaiten rite calling on Benzaiten to defeat enemies on the battlefield. See Monkan 文観, "Kinpusen himitsu den 金峰山秘密伝," in Nihon daizōkyō 日本大蔵経, ed. Suzuki Gakujutsu Zaidan (Tokyo: Suzuki gakujutsu zaidan, 1976), 12, 32. 473 This episode will be referred to as the “Rebuilding of the Great Stupa” following the chapter title in the Kakuichibon (scroll 3, Daitō konryū 大塔建立). 255 Taira to the Minamoto (scroll 9, Minamoto chūnagon seishi no yume no koto 源中納言青侍夢事). The first recounts the Itsukushima Deity vowing to support the Taira and their rise to power. The second shows the Itsukushima Deity’s inability to protect the Taira from losing power. The central role of these episodes within the Heike narrative limits the possibilities for altering Itsukushima’s image while maintaining consistency with preexisting variants. Consequentially, the Nagatobon compilers decided to increase focus on the motivations of the Itsukushima Deity and Kiyomori and present more sympathetic images of them as heroic losers. 474 Dream of the Council of the Deities Minobe Shigekatsu draws a parallel between the Heike chapter (found in most variants) on the Minamoto retainer’s oracular dream of the council of the deities and agreements between the gods in Jien’s Gukanshō 愚管抄 (completed around 1219-1220). According to the Gukanshō, the Ise and Kashima deities met at the time of the founding of the royal rule of Japan to decide that their tutelary households would hold royal and courtly authority respectively, and later, Hachiman and Kasuga consulted together after the Hogen and Heiji disturbances to appoint the Minamoto as the protectors of the royal family (i.e., martial ruling authority). 475 In this light, 474 This resembles the theme of hōgan biiki (“sympathy for the lieutenant,” 判官贔屓), which idealizes tragic defeat within court culture. It is most discussed in relation to Yoshitsune 義経 (the lieutenant) and especially his portrayal in the Gikeiki, Ichiko Teiji that the theme existed during the medieval period, although nobody used the phrase hōgan biiki until the early-modern period. The cases concerning Kiyomori here are unusual, because most portrayals of the Taira as lovable losers focused on other members of the household. See Teiji 市古貞次 Ichiko, Chūsei shōsetsu to sono shūhen 中世小說とその周辺 (Tokyo: Tō kyō Daigaku Shuppankai, 1981), 76.; Adolphson and Commons, "Blurring the Lines: Repositioning the Heike," 9-11. 475 The Kashima and Kasuga deities both represent the tutelary deity of the Fujiwara household. The Fujiwara ancestor Nakatomi no Kamatari ( 中臣鎌足, ca. 614-669) lived in Hitachi province and worshipped the Kashima Deity. Additionally, the Kashima deities Ame no koyane and Himegami were enshrined as two of the four deities at Kasuga Shrine. For the Gukanshō passage, see Delmer Myers Brown and Ichiro Ishida, The Future and the Past: A Translation and Study of the Gukansho, an Interpretative History of Japan Written in 1219 (Berkeley: University of California Press, 1979), 228; 慈円 Jien, Gukanshō 愚管抄, ed. Okami Masao 岡見正雄 and Akamatsu Toshihide 赤松俊秀, Nihon koten bungaku taikei (Tokyo: Iwanami Shoten, 1967), 347. Concerning the Fujiwara’s relationship 256 Kiyomori’s rise to martial authority as the hired sword of Go-Shirakawa can be seen as not only unusual, but also as a deviation from the planned turn of events agreed upon by the gods. 476 Minobe’s view of consistent lineages of political authority decided by the gods challenges attempts to legitimize the Taira’s albeit short-lasted reign. While the council of the deities episode in the Heike variants necessarily legitimizes the transfer of power from the Taira to the Minamoto, it does not always show the decisions of the gods to correspond with a predetermined progression of history. The Heike in general, and especially the Nagatobon, show how the gods vied to change the planned course of events to empower their devotees. It is for this reason that the Nagatobon’s account of the council of the deities is the only moment in any Heike variant to identify the Itsukushima Deity as the Taira’s tutelary deity (ujigami 氏神). The Kakuichibon and Engyōbon instead call the deity an “ally” (kataudo 方人) of Kiyomori. 477 As discussed above, there was little prestige to be gained from being the tutelary deity of a fallen household. However, the unquestionable support of each deity for its tutelary family provides a foundation for defending Itsukushima against arguments for the with the Kashima and Kasuga deities, see Allan G. Grapard, The Protocol of the Gods: A Study of the Kasuga Cult in Japanese History (Berkeley: University of California Press, 1992), 23-24. 476 Kujō Kanezane’s diary Gyokuyō 玉葉 includes two entries which similarly explain events as determined by the deities. In the first case (6/18/1179), Kanezane records a rumor that Kiyomori had received an oracular dream from the Kamo Deity showing him receiving treasures from the god, and this was correlated with Kiyomori receiving inheritance after the death of Fujiwara no Motozane ( 藤原基実, 1143-1166). In the second (4/5/1191), Kanezane hears that Minamoto no Yoritomo’s daughter will become a royal consort, and he remarks, “Important events like this are simply the results of plans carried out by the Ise, Hachiman, and Kasuga deities. Could they possibly have resulted from acts of human will?” (Translation based on Delmer Brown and Ichirō Ishida.) O n the Gyokuyō entries, see Delmer Myers Brown and Ichiro Ishida, The Future and the Past: A Translation and Study of the Gukansho, an Interpretative History of Japan Written in 1219, 427; Sasaki Hachirō 佐々木八郎, Heike monogatari hyōkō 平家物語評講, 2 vols. (Tokyo: Meiji Shoin, 1963), v.1, 594-95; Kujō Kanezane 九條兼實, Kujō-ke-bon Gyokuyō 九条家本玉葉, 14 vols., Zushoryō sō kan (Tokyo: Kunaichō Shoryō bu, 1994-2013), v.6, p. 215; v. 13, p. 11. For Minobe’s arguments on the Tale of the Heike and the Gukanshō, see Minobe Shigekatsu, "The World View of "Genpei Jōsuiki"," Japanese Journal of Religious Studies 9, no. 2/3 (1982): 216-17. This account appears similarly in the Kakuichibon. Tyler, Royall, The Tale of the Heike (New York: Viking, 2012), 262-63. 477 Asahara and Nanami, Nagato-bon Heike monogatari no sōgō kenkyū: Kōchū hen 長門本平家物語の総合研究：校注篇, 644; Kitahara and Ōgawa, Enkyōbon Heike monogatari. Honbun hen 延慶本平家物語. 本文篇, 425; Ichiko, Heike monogatari 平家物語, v. 45, p. 364. 257 illegitimacy of its support for the Taira. The episode must end with the Itsukushima Deity trying and failing to protect the Taira, but the Nagatobon can still try to valorize that attempt. The Nagatobon cannot avoid placing Itsukushima on the losing side here, but it can at least show her to be on par with her opponents. Both the Engyōbon an d the Nagatobon list the three main deities in the episode as Hachiman, Kasuga, and Itsukushima, but only the Nagatobon highlights the roles of the three as tutelary deities. Additionally, while the Engyōbon describes the Hachiman Deity as “the one who allowed the sword to be entrusted to the Manager of the Right Military Guard Minamoto no Yoritomo,” the Nagatobon manages to highlight Kiyomori’s perspective within the description of Hachiman. The Nagatobon identifies Hachiman as the one who “took back the sword bestowed to Lay Priest Kiyomori to give it to Yoritomo.” 478 The episode includes a passage questioning the Kasuga Deity’s plea for the Fujiwara to gain the sword after the Taira, and in doing so, shifts the discourse away from the authority of tutelary deities to that of warrior families. Both the Kakuichibon and Engyōbon suggest that a warr ior household like the Taira or Minamoto should hold the responsibility to protect the realm. 479 The Nagatobon avoids an answer and only asks, “After that (the era of the Minamoto), should the Fujiwara have a turn to possess [the sword]?” 480 The question provides an opportunity to assert 478 Asahara and Nanami, Nagato-bon Heike monogatari no sōgō kenkyū: Kōchū hen 長門本平家物語の総合研究：校注篇, 644. 479 Notably, the Engyōbon presents the Taira and the Minamoto as the two households tha t should exchange authority as the “great general” (dai shōgun 大将軍) and pacify the realm during the final age of the Buddhist teachings. It states, “However, during the final age, the descendants of both the Minamoto and the Taira serve [as the great shogun]; how could the Fujiwara emerge as the great shogun? Could an individual prince restore peace to the realm as the great shogun?” The Kakuichibon raises doubts about the appropriateness of the Fujiwara rather than affirming the propriety of the Minamoto or Taira, saying, “Did he (Kasuga Deity) mean that once the Heike are finished, and then in their turn the Genji, Lord Kamatari’s successors, the sons of the regental house, should assume military command of the realm?” Kitahara and Ogawa, Enkyōbon Heike monogatari. Honbun hen 延慶本平家物語. 本文篇, 426. Translation of Kakuichibon from Tyler, The Tale of the Heike; Ichiko, Heike monogatari 平家物語, v. 45, p. 346. 480 Asahara and Nanami, Nagato-bon Heike monogatari no sōgō kenkyū: Kōchū hen 長門本平家物語の総合研究：校注篇, 647. 258 the legitimacy of the Taira as a warrior household, but any answer would legitimize the rise of the Minamoto and challenge the Itsukushima Deity’s defense of the Taira as their tutelary deity. The council of the deities episode forced Heike compilers to face the conflicting interests and agency of the gods. Presenting the shift in power as restoring order provides an intelligible explanation while strengthening the Minamoto’s authority. The Nagatobon takes the opposite approach in order to defend Itsukushima within the episode. Illuminating the contestations for power among tutelary deities reveals that the Itsukushima Deity acted the same as the most powerful gods of the realm and that she made a worthy opponent. It is easy for the Itsukushima Deity to appear weak given the council of the gods’ conclusion—she is thrown out of the meeting, unable to stop the council from taking the sword away from Kiyomori. In the Hōryaku kanki, Kiyomori is surprised to hear that the Itsukushima Deity did not have the power to protect him. It gives Kiyomori’s response upon hearing an oracle that the council of the gods had decided to transfer the Taira’s authority as martial protectors of the realm to the Minamoto, saying, “Lay Priest [Kiyomori] was taken aback. Even the hand spear (teboko 手鉾) provided to him through the incomparable [power] of this shrine, was lost, nowhere to be found.” 481 If Kiyomori’s authority was gained through the shrine’s “incomparable [power],” his loss of power calls for unfavorable comparisons with the other gods of the realm. Most Heike variants present the Itsukushima Deity as powerless before the powerful Kasuga and Hachiman deities, and instead focus on their decision for the Minamoto family to replace the Taira as the martial protectors of the realm. For example, the Kakuichibon recounts: Another time one of Lord Minamoto no Masayori’s young housemen had a frightening dream. He found himself in what he took to be the Bureau of Shrines. A large gathering of senior officials, formally dressed, was engaged in some sort of debate, at the end of which they expelled from their company one whose seat had been the most junior among 481 "Hōryak u kanki 保暦間記," 131. 259 them. This figure appeared to be a Heike ally. “Who is this gentleman?” the dreamer asked an old man. “The divinity of Itsukushima,” the old man replied. Next an imposing elder, seated in the place of honor, announced, “The Sword of Command, bestowed some time ago on the Taira house, we now award to Yoritomo, in exile in the province of Izu.” Another, similarly imposing elder (Kasuga Deity) seated next to him interjected, “But please let it pass thereafter to my descendants.” 482 The Kakuichibon, as well as the Engyōbon, condense the episode to the essential parts: The Itsukushima Deity is removed from the council and the rest of the gods, led by the Hachiman and Kasuga deities, decide to award the sword to Yoritomo. Beginning with the Itsukushima Deity, as the lowest ranking deity, being removed from the council, the scene prevents the Itsukushima Deity from having any agency. The Itsukushima Deity’s brief presence only serves to show that she remained on the side of the Taira but failed to protect them. The Engyōbon and Nagatobon begin similarly, suggesting that they are working from a similar base, but the Nagatobon expands it to add a more favorable image of Itsukushima. 483 The Nagatobon also presents Hachiman and Kasuga defeating Itsukushima, but it expands the scene to show Itsukushima bravely fighting back. She is given a voice to oppose the move to take the 482 Translation from Tyler, The Tale of the Heike, 262-63; Ichiko, Heike monogatari 平家物語, 362-63. 483 Besides the central issue of the Itsukushima Deity’s position within the council of the gods, the episode includes a brief discussion of the deity’s gender. The topic is raised across the variants, but the Nagatobon once again provides a more detailed description. The Heike asks why the Itsukushima Deity appears in the council as a man dressed in ceremonial court dress even though the deity is conventionally known to be the third daughter of the Dragon King. The Engyōbon simply speculates that the idea of Itsukushima as female might be a mistake ( higagoto ni ya 僻事ニヤ), while the Kakuichibon and Nagatobon explain that a deity can take multiple forms. The point of note in the Nagatobon is that it uses the idea of the Itsukushima and Kehi deities as the provisional manifestations of the dual mandalas (womb realm and diamond realm respectively) to argue that they could take different forms within the world of humans. Moreover, the Nagatobon further explains that the Kehi Deity is enshrined at the Marōdo subsidiary shrine 客人の宮 within Itsukushima Shrine, and that the Itsukushima Deity is, in turn, enshrined as the deity of the offing (oki no gozen おきの御前) at Kehi Shrine. In this manner, the Nagatobon employs specific knowledge of the shrine structures at Itsukushima and Kehi to connect the non-dual mandalic spaces of the deities with their worldly and geographically distant shrines. Although the enshrinement of the Kehi Deity within the Marōdo Shrine cannot be ver ified with other sources, the mention of the Marōdo Shrine is evidence enough of external knowledge about Itsukushima Shrine. Kitahara and Ogawa, Enkyōbon Heike monogatari. Honbun hen 延慶本平家物語. 本文篇, v. 1, p. 425; Ichiko, Heike monogatari 平家物語, v. 45, p. 364. Asahara and Nanami, Nagato- bon Heike monogatari no sōgō kenkyū: Kōchū hen 長門本平家物語の総合研究：校注篇, v.1, p. 647. 260 sword away from Kiyomori. After the Hachiman Deity raises the proposal to transfer the sword to Yoritomo, the Itsukushima Deity responds: “Concerning Lay Priest Kiyomori’s recent holding of the sword, has he committed any wrongdoing? Please let him [continue] to have it a little longer.” After she had spoken, a prominent nobleman (Hachiman) became greatly angered, and said, “The one called Lay Priest Kiyomori has undermined the authority of the court and he is the enemy of the Buddhist Law and the Royal Law (ōbō buppō 王法仏法). Why should he continue to be trusted with it? I cannot imagine there being anyone present who would call oneself an ally of Lay Priest Kiyomori. Is there such a person in attendance?” The person in the farthest seat, wearing red robes, a sword, and a baton, 484 said “I am,” and walked towards the front. When [Hachiman responded] saying, “Remove that person,” two people approached [the Itsukushima Deity] together grabbing her left and right arms, pulled her to a far-off gate, and pushed her out. Disheartened, [the Itsukushima Deity] let them remove her. 485 The Nagatobon demonstrates the Itsukushima Deity’s grit and sadness in fighting in vain to keep the sword in Kiyomori’s stewardship. The Itsukushima Deity does not allow her low-ranking position to stop her from opposing the rest of the gods. Moreover, she does not hesitate to defend Kiyomori even after Hachiman lists his crimes and insinuates a warning against allying oneself with the Taira. Crimes against the authority of the sovereign and Buddhism were valid reasons for breaking ties with the Taira (as seen in the earlier discussion of Enryakuji), but this is not an option for Itsukushima within the standard structure of the scene. Therefore, if the Itsukushima Deity must support the Taira despite their deeds, her best option is to do so as a virtuous tutelary deity. The end is the same—she is removed from the room against her will, the Taira lose the war, and the Minamoto gain power through the establishment of the Kamakura Bakufu. For the Itsukushima Deity though, it is significant that the Nagatobon manages to still portray her as 484 Matching the ceremonial dress for members of the court (sokutai 束帯). 485 Asahara and Nanami, Nagato-bon Heike monogatari no sōgō kenkyū: Kōchū hen 長門本平家物語の総合研究：校注篇, 643-44. 261 acting properly and heroically. The underdog might not always win, but that does not make the fight any less commendable. The Origins of Kiyomori’s Worship of Itsukushima The Tale of the Heike provides an origin narrative for Kiyomori’s worship of Itsukushima Shrine, and in doing so, centralizes the importance of Itsukushima’s connection with the Taira. The episode presents Itsukushima favorably as long as the context remains within the temporal moment of the Taira’s rise to greatness, but the Heike connects the scene with foreshadowing of the Taira’s demise. In order to address the conflicting interests of explaining the rise and fall of the Taira and also positively depicting Itsukushima Shrine, most variants of the Heike have the Itsukushima Deity include conditions for her continued protection of the Taira. Specifically, while the Kakuichibon gives the well-known condition that the Taira’s fortunes will fade if he commits evil deeds, the Engyōbon and Nagatobon instead state that the Taira’s power will only last one generation. Both provide an explanation for the fall of the Taira, but the evil deeds condition holds the Taira fully responsible for the consequences of their actions. 486 Further comparison of the Nagatobon and the Engyōbon reveals that even with a preexisting safeguard for distancing the Itsukushima Deity’s protection from the Taira’s downfall, the Nagatobon takes additional steps to improve the connotations of the association between Itsukushima and Kiyomori. 486 Yamashita Hiroaki argues that the emphasis on Kiyomori’s evil deeds as the primary reason for the fall of the Taira was a later development in the writing of the Tale of the Heike. He gives the Nanto-bon as the earliest variant changing details in the narrative to explain events in terms of Kiyomori’s evil deeds, and therefore, suggests that the evil deeds discourse likely did not become prominent until the late Muromachi period, or the mid Muromachi period at the earliest. Yamashita, Heike monogatari no seiritsu 平家物語の成立, 275. 262 The Itsukushima Deity’s rules for granting the Taira success and protection vary depending on the text, but most Heike variants justify the Itsukushima’s failure to support the Taira to the very end. Notably, the earliest and pre-Heike account of Kiyomori’s oracle to worship the Itsukushima Deity in the Kojidan (1205, 古事談) gives no conditions, and it also does not go so far as to say that the Taira will gain royal authority. Instead, the oracle only tells Kiyomori that he will rise to the rank of grand minister (daijō daijin 太政大臣). More than just leaving out support for the limits of Itsukushima’s connection with the Taira, the Kojidan contrasts Itsukushima with Amaterasu as a lesser form of Dainichi Buddha. When the manifestation of Kūkai comes to tell Kiyomori to restore Itsukushima Shrine, Kūkai begins by saying, “The Dainichi buddhas of Japan are the Ise Deity (Amaterasu) and Itsukushima of Aki [Province]. The Ise Deity is extremely wondrous and profound (amari yūgen nari あまり幽玄也). Fittingly, you are the provincial governor [of Aki]. You should quickly go serve the Itsukushima [Deity].” 487 Murata Shin’ichi explains that the Kojidan’s hierarchization of Itsukushima and Kiyomori below Ise distinguishes Ise as the Dainichi Buddha of the royal family and Itsukushima as the lesser Dainichi Buddha for Kiyomori. 488 This separation of Ise from Itsukushima provides a political religious argument for maintaining the central authority of the sovereign and Ise Shrine, regardless of Kiyomori’s rise in court rank. From this perspective, it is the birth of Kiyomori’s grandson, the future Antoku Tennō, that challenges the boundaries between royal and martial authority, and between the center and liminal spaces, that led to chaos. The Tale of the Heike cannot avoid confronting this line of thought, but the different variants and 487 Minamoto, Kawasaki, and Araki, Kojidan, Zoku Kojidan 古事談. 続古事談, 478-79. 488 Murata, ""Hachiman Usagū go takusenshū" no sekai: Usa Hachiman shinwa gensetsu no kenkyū 『八幡宇佐宮御託宣集』の中世：宇佐八幡神話言説の研究." 263 especially the Nagatobon, still aim to present Itsukushima as more than just a regional deity with quixotic aspirations. The Heike traces the origins of Kiyomori’s worship to the events following his restoration of the stupa of Kongōbuji temple on Mt. K ōya in 1156. Aft er Kiyomori worships at the completed pagoda, the Itsukushima Deity appears before him as a manifestation of the great monk of Mt. Koya, Kūkai ( 空海, 774-835), and gives him an oracular message that he will achieve great fortune if he worships and restores Itsukushima Shrine. The auspicious message is then tempered with the caveat that their fortunes will not last. The Nagatobon shows Kiyomori as distraught upon hearing the bad news but unwilling to give up. In this manner, the Nagatobon presents Kiyomori as a sympathetic hero—he knows his fate but cannot help but strive for a better outcome: When [Taira no Sadayoshi 平貞能] told this to the Governor of Aki (Kiyomori), [Kiyomori] hearing that it would [only] last one lifetime, felt alone and without anyone to rely on. One lifetime is like a dream; he wanted his descendants to succeed him and prosper. In order to pray for this, he secluded himself within the temple, drew mandalas, and they were enshrined in the Golden Hall. [Kiyomori] had Jō myō 常明, a painter whose practice and wisdom are such that even the Retired Sovereign employs him, produce the Western Mandala (Diamond World mandala). As for the Eastern Mandala (Womb World mandala), Kiyomori stooped over and did it in his own hand and even drew blood from his own head to paint the Jeweled Crown (of Dainichi Buddha), the nine honored ones, and the eight-petal [hall]. 489 For comparison, the Engyōbon gives Kiyomori a similar but less emotive response: When [Taira] no Iesada 平家貞 490 told this to the Governor of Aki (Kiyomori), Kiyomori responded, “So, we can expect for it to only be for one generation. My heart is restless 489 Asahara and Nanami, Nagato-bon Heike monogatari no sōgō kenkyū: Kōchū hen 長門本平家物語の総合研究：校注篇, 395. Underline added by author. 490 The person accompanying Kiyomori at Mt. Kōya here (Taira no Iesada) differs from the Nagatobon (Taira no Sadayoshi, Iesada’s son), and there is little consistency among the variants. In the Kakuichibon and the Yashirobon 屋代本, Kiyomori is alone, and in the Shibu kassen-bon 四部合戦本, Taira no Morikuni ( 平盛国, 1113-1186) is given. Ichiko, Heike monogatari 平家物語, 204; Asahara Yoshiko 麻原美子, Haruta Akira 春田宣, and Matsuo Ashie 松尾葦江, eds., Heike monogatari: Yashirobon, Kōyabon taishō 平家物語 : 屋代本・高野本対照, 3 vols., vol. 1 (Tokyo: Shintensha, 1990), 250; Takayama Toshihiro 高山利弘, ed. Kundoku Shibu kassenjō-bon Heike monogatari 訓読四部合戦状本平家物語 (Tokyo: Yūseidō , 1995), 90. 264 [knowing] that it cannot continue to my descendants. I will perform prayers here, on this mountain, for [our] enlightenment in future rebirths. If only I can cultivate good roots [of merit]!” Saying this, he drew the Eastern Mandala and the Western Mandala. As for the Eastern Mandala, Kiyomori had Jōmyō do it. As for the Western Mandala, Kiyomori did it in his own hand, and drew blood from his own head to paint the nine honored ones of the eight petal [hall], and built a mandala hall to dedicate them to. 491 The differences are few, but reveal a shift from the Engyōbon to the Nagatobon in characterizing Kiyomori’s view towards the oracle from the Itsukushima Deity. 492 First of all, the Nagatobon increases the emphasis on Kiyomori’s emotional response to the prophetic message. While the Engyōbon passage acknowledges that he was not happy to hear that his fortune would not extend to future generations, the Nagatobon emphasizes the futility and loneliness within his sadness. Moreover, Kiyomori’s emotions are a result of his feelings towards his family. He is portrayed as a clan leader who cannot bear the thought of gaining great fortune and being unable to share it with his descendants. This characterization of Kiyomori allows for a reading of his evil deeds that is both critical and sympathetic. Kiyomori might have brought chaos to the world, but who would not give up the world for one’s family? Secondly, the Nagatobon explains Kiyomori’s motive in painting the mandalas as an attempt to overcome his oracular future and carry his successes over to his descendants. He recognizes that such a feat is beyond his means, but he cannot help but try. Likewise, it is beyond the means of the Nagatobon to change Kiyomori’s downfall, but it can at least depict Kiyomori and the Itsukushima Deity as heroic losers who realize that they cannot win but fight anyways. Conversely, the Engyōbon gives Ki yomori the more realistic aspiration for his religious practices 491 Kitahara and Ogawa, Enkyōbon Heike monogatari. Honbun hen 延慶本平家物語. 本文篇, 338. 492 A minor difference in the description of the mandalas suggests that the Nagatobon passage was based on the Engyōbon and corrected the Engyōbon’s mistak en reversal of the Western and Eastern mandalas. The Western and Eastern mandala refer to the Diamond World and Womb World mandalas respectively, named in accordance with the direction of the walls they are hung in the ritual space. Kiyomori should be painting the Eastern (Womb World) mandala, because Kiyomori paints the nine honored ones of the eight petal [hall] located at the center of the Womb World mandala. Tellingly, this is the case in the Nagatobon. Elizabeth Ten Grotenhuis, Japanese Mandalas: Representations of Sacred Geography (Honolulu: University of Hawai'i Press, 1999), 37, 59-61. 265 to contribute towards the future salvation of himself and his family. Working from the Buddhist framework of karmic merit, Kiyomori is still held at fault for the demise of the Taira, although that responsibility is based on deeds already committed in this lifetime and those prior, rather than those committed later in The Tale of the Heike. The Nagatobon’s explanation of the origins of Kiyomori’s worship of Itsukushima portrays Kiyomori in an unusually positive light, but it would be a mistake to apply this image of Kiyomori to the Nagatobon as a whole. Rather, it makes more sense to read this sympathetic image of Kiyomori as a necessary change for maintaining a positive image of the Itsukushima Deity within this central scene connecting Itsukushima with Kiyomori. The Nagatobon incorporates a unique episode about Kiyomori preceding the origin narrative of Kiyomori’s worship of the shrine, “The Placard Raised at Nishi Hashijō” (Book 5, Nishi hashijō fuda ga tataruru koto 西八条被立札事). Matsuo Ashie describes this scene as stereotypically critical of Kiyomori, but considered with the following restoration of the great stupa episode, it balances out the uncommonly admirable portrayal of Kiyomori when he receives the oracle from the Itsukushima Deity. 493 Furthermore, it provides another example of Kiyomori as a selfless leader, stressing the importance of his fortunes for all of the Taira in the past, present, and future. In this episode, two placards are placed on gates of Kiyomori’s residence at Nishi Hachijō denouncing his arrogance. It begins, “When one looks at the world, it is like the cur rent of the water. The arrogance of people resembles the bubbles on the surface of the water.” 494 Naturally, Kiyomori is upset, and he summons all of the poets and scholars in Kyoto, 1,336 in total, to write a vow (kishōmon 起請文) at Kitano Tenjin Shrine 北野天神 in order to see whose 493 Matsuo, Gunki monogatari genron 軍記物語原論, 235. 494 Asahara and Nanami, Nagato-bon Heike monogatari no sōgō kenkyū: Kōchū hen 長門本平家物語の総合研究：校注篇, v.1, p. 389. 266 handwriting matches that of the placard. Matsuo is not wrong to describe the scene as portraying Kiyomori as indignant, prone to overreaction, and unaccepting of criticism.” However, a closer look at the language of the placard reveals a potential line of defense for Kiyomori as well. The writing on the gate contends that Kiyomori’s rise to power shames his ancestors, and it contrasts his behavior with that of his father Tadamori ( 忠盛, 1096-1153), whose foresight saved him from an assassination attempt made after he was appointed as the governor of Tajima Province. This leads up to the rebuke, “Thinking only of his one desire, how could he possibly know the grief of his descendants? Grand Minister Lay Priest [Kiyomori], how unjust he is!” 495 Situated immediately prior to the episode where Kiyomori laments after hearing that his fortunes will not carry on to his descendants, the Nagatobon calls into question the veracity of the public censure against him. The example of the attack on Tadamori will also make the audience recall the court nobles’ improper behavior towards Tadamori. The nobles were upset that a warrior would gain the position of provincial governor and planned to correct this seeming injustice by taking his life. In Kiyomori’s case, he similarly rose to an unheard position for someone of his class, and people responded by attacking his character. Their criticisms were not unfounded, but they were not faultless either. We can see Kiyomori acting in the opposite manner of what people see—as a selfless person striving to bring respect and prestige to his ancestors and descendants—and yet the people mercilessly attack him for acting outside of what they deem proper. This scene, together with the restoration of the great stupa episode, balance a sympathetic image of Kiyomori with his standard negative traits. The goal here is not to rewrite the character of Kiyomori, but to present a less critical view towards his association with the Itsukushima Deity. 495 Ibid., 390. 267 The Genpei jōsuiki provides a counterexample where the restoration of the stupa scene retains an unfavorable portrayal of Kiyomori and removes the conditions protecting Itsukushima from the Taira’s later deeds. Instead, the Genpei jōsuiki depicts the Itsukushima Deity trying to reform Kiyomori as a proper ruler. When Kiyomori receives the oracular messages from the Itsukushima Deity, he hears that the deity will unquestionably protect both himself and future generations. After Kiyomori restores the shrine, an Itsukushima naishi shrine attendant voices the deity’s message, “Because it was [her] divine promise, the deity shall provide protection to your descendants. This is truly remarkable.” However, the narrator then provides the audience with a further explanation of the deity’s intentions: From long ago when Lay Priest [Kiyomori] was a layman to now, when he has taken the tonsure, he has never been remiss in his devotion. Therefore, from [his] descendants and brothers to the Grand Minister General [Kiyomori] himself, they were all satiated with their royal favor over provinces, districts, and estates. It was the Deity’s plan, that as a result, the Lay Priest’s heart [would] be calmed, and the Retired Sovereign [Takakura] too prayed for [the Deity] to pacify [Kiyomori]. People say that it is for this reason that the Retired Sovereign made the royal pilgrimage to Itsukushima prior to going to the two shrines of Kamo and Hachiman. 496 The Genpei jōsuiki situates this episode on Kiyomori’s worship of Itsukushima following Retired Sovereign Takakura’s pilgrimage to Itsukushima (rather than following the successful birth of Antoku Tennō in the Nagatobon), emphasizing the shared goals of the Itsukushima Deity and Takakura. The earlier pilgrimage scene also includes him praying to the Itsukushima Deity, where he specifically prays to “calm Kiyomori’s rebellious heart” (nyūdō no muhon no kokoro mo yawaragi 入道の謀叛の心も和ぎ). 497 These added details justify their support for Kiyomori as a strategy to prevent him from bringing further chaos to the realm. This allows the Genpei jōsuiki to condemn Kiyomori’s actions without holding Itsukushima or Retired Sovereign 496 Ichiko Teiji 市古貞次, ed. Genpei jōsuiki 源平盛衰記, v.3, p. 13. 497 Ibid., 11. 268 Takakura equally responsible. Nevertheless, the end result is still that the Itsukushima Deity failed to complete its task. The Itsukushima Deity lacks the clear excuse given by the conditional statements of the other variants. The Genpei jōsuiki aligns the relationship between Itsukushima and the Taira with the larger Heike narrative, but it is not a perfect answer. It clarifies the good intentions of the Itsukushima Deity, but it does not allow the deity to fulfill her aim of aligning Kiyomori’s intentions with the good of the realm. No matter how the story is retold for the sake of the Itsukushima Deity, it remains a scenario of making the best out of a bad situation. The Origins of Itsukushima Shrine in the Tale of the Heike The Nagatobon and Genpei jōsuiki are the only variants to include a separate origin narrative of Itsukushima Shrine. The location of the origin narrative immediately after the restoration of the great stupa and the beginnings of Kiyomori’s worship of the shrine might tempt readers to interpret the origin narrative in relation to the larger Heike narrative. Considering the difficulty of separating the image of Itsukushima from Kiyomori within the great stupa episode though, the origin narrative has more to offer to the Itsukushima Deity if it can remain outside of the context of the Heike. As a tangent from the Heike narrative, the Itsukushima origin narrative reveals the power and history of Itsukushima that existed prior to any connection with the Taira. The origin narratives in the Nagatobon and Genpei jōsuiki present Itsukushima in strikingly different manners. Itsukushima’s origins in the Nagatobon can appear more regional or transgressive against the authority of the center. Conversely, the Genpei jōsuiki removes those elements presenting Itsukushima in conflict with the sovereign and the capital. The contrast could be read as reflecting a regional view of sacred and political power in the Nagatobon, but 269 the differences between the variants could be merely coincidence. The origin narratives in both texts were incorporated from distinct textual traditions, with the one in the Nagatobon being largely identical to the Itsukushima Daimyōjin nikki text from Kanazawa Bunko, and the one in the Genpei jōsuiki corresponding with the Shōbōrinzō account of the life of Prince Shōtoku. 498 As such, the differences between the Nagatobon and the Genpei jōsuiki existed before either text was compiled. There is a chance that the compilers intentionally chose a specific account of the Itsukushima origin narrative, but it is just as likely that they only knew about or had access to a single version. That being said, it is still worth summarizing the origin narrative in the Nagatobon and its differences with the Genpei jōsuiki. This introduction to the Itsukushima origin narratives will clarify how the same narrative could be molded in different manners, and provide a basis for analyzing the subtle differences between the Nagatobon and the earlier Itsukushima Daimyōjin nikki. The origin narrative follows the story of Saeki Kuramoto’s 佐伯蔵本 exile, in which he encounters the Itsukushima Deity, guides her to Itsukushima (modern-day Miyajima), and becomes the first head priest of Itsukushima Shrine. The account of Kuramoto’s exile first appears in the eighth century Nihon shoki, which provided it a legitimate and verifiable basis for the origins of Itsukushima Shrine and of the Saeki family as the shrine priests. The Nihon shoki entry focuses on the sovereign and the queen consort, who become enamored with the nightly cries of a deer, but soon learn that the deer had been killed by Kuramoto. As punishment, the royal family exiles him, and he moves to Aki Province where his descendants became local 498 Makino Kazuo compares the Shōbōrinzō and Genpei jōsuiki passages line by line and argues that it is most likely that both texts were based on an earlier Itsukushima origin narrative. He also acknowledges the possibility that the Genpei jōsuiki passages were based on the Shōbōrinzō itself, but lists other texts referenced by both Prince Shōtoku texts and the Genpei jōsuiki as additional support for the possibility that the Shōbōrinzō and Genpei jōsuiki were working from some of the same texts. Makino Kazuo 牧野和夫, Engyōbon "Heike monogatari" no setsuwa to gakumon 延慶本「平家物語」の說話と学問 (Kyoto: Shibunkaku Shuppan, 2005), 333-36. 270 elites. 499 The brevity of the Nihon shoki entry provided opportunities for connecting the origins of the Saeki family with the origins of Itsukushima Shrine, but it made it difficult to avoid the oppositional history between the Saeki and the royal family. Nevertheless, the Shōbōrinzō and Genpei jōsuiki ignore the Nihon shoki episode and begin the Itsukushima origin narrative with Saeki Kuramoto encountering the Itsukushima Deity at sea. 500 This removal helps to align the Itsukushima origin narrative with its larger context in the Shōbōrinzō, in which Prince Shōtoku performs rites to please the gods (hōraku 法楽) at temples and shrines throughout Japan. In this manner, Shōtoku links the sacred spaces of Japan as a unified realm of the gods ( shingoku) supported by the royal family and the court. The Nagatobon instead defends Saeki Kuramoto’s killing of the deer and uses it as a stimulus to bring him to the Itsukushima Deity. The text explains Kuramoto’s killing of the deer by making it a response to an edict from the sovereign to see the deer. Then, when he delivers the deer carcass to the sovereign, the members of the court deduce from the deer’s golden and iridescent hide that the deer was a manifestation of a deity. Consequentially, Kuramoto is exiled for hunting a divine animal, and not for disobeying the sovereign. 501 From another perspective, 499 The Nihon shoki also includes a separate story in which the Saeki family is traced back to prisoners brought back by Yamato Takeru from his eastern expedition and offered to Ise Shrine. However, due to their rough behavior, they end up being sent away to the provinces of Harima, Sanuki, Iyo, Aki, and Awa. Kojima Noriyuki 小島憲之, ed. Nihon shoki 日本書紀, 3 vols., Shinpen Nihon koten bungaku zenshū (Tokyo: Shōgakkan, 1994), v.1, pp. 388 -89 (Yamato Takeru episode), v. 2, pp. 52-54 (deer episode); W. G. Aston, Nihongi: Chronicles of Japan From the Earliest Times to A.D. 697, 2 vols., vol. 1 (London: Pub. for the Society by K. Paul, Trench, Trübner & col., limited, 1896), 211-12 (Yamato Takeru episode), 89-90 (deer episode); Hatsukaichi chōshi: Tsūshi hen 廿日市町史. 通史編, 2 vols., vol. 1 (Hatsukaichi-chō : Hatsukaichi-chō , 1988), 280-81. 500 The Shōbōrinzō identifies Kuramoto as already being a regional elite as the provincial governor of Aki Province, instead of being a recent exile still traveling to find a home. The Genpei jōsuiki, however, does not mention him being the provincial governor. Following Makino Kazuo’s argument about the Shōbōrinzō and the Genpei jōsuiki both being based on an earlier text, it is more likely that the Shōbōrinzō compilers (or other Itsukushima origin narrative compilers) added Kuramoto’s governor position, than the Genpei jōsuiki compilers removing the detail. "Shōbōrin -zō 聖法輪蔵," 487. 501 Kanazawa Bunko 金沢文庫, ed. Kanazawa Bunko no chūsei shintō shiryō 金沢文庫の中世神道資料 (Yokohama: Yokohama kenritsu Kanazawa Bunko, 1996), 52; Asahara and Nanami, Nagato-bon Heike monogatari no sōgō kenkyū: Kōchū hen 長門本平家物語の総合研究：校注篇, 396-97. 271 though, the killing of the deer can be seen as an appropriation of royal hunting practices. 502 Notably, The Nagatobon and Itsukushima Daimyōjin nikki change the location of the deer from Toga fields (Togano 菟餓野) to Inami fields (Inamino 印南野). Both are locations in Harima Province (modern-day Hyōgo Prefecture) with associations as hunting grounds, but only the Inami fields were specified as a “prohibited [hunting] field” (kin’ya 禁野) reserved for the royal family. 503 After Saeki Kuramoto meets the Itsukushima Deity and guides her to her new home at Itsukushima, issues of royal authority and exile emerge again with the need for Kuramoto to appeal to the court for support to build the shrine. The Itsukushima Deity commands Kuramoto to petition the court to offer shrine estates for funding the building of Itsukushima Shrine, but Kuramoto realizes that obeying the deity would be in defiance of the court’s order of exile. Rather than try to hide the conflict, the Itsukushima Deity tells Kuramoto to side with her, and that she will help convince the sovereign to accept Kuramoto’s petition. The deity explains: It was my plan for you to come here. The golden deer who manifested at Inami plain was, in fact, me. I had you come here in order to act as your wet nurse. Quickly, proceed to the capital and pass a memorial to the throne. At that time, I will become a supernatural bird, carry ten to twenty thousand sasaki branches in my beak, and place them on top of the ceremonial hall (shishinden 紫宸殿). [Then,] when three stars, great stars shine three rays of light on the royal palace, [they] will be surprised and undoubtedly commend land to the shrine. 504 502 The Soga monogatari shows Yoritomo’s grandiose hunting expeditions reconstructing royal hunting practices into performances asserting the authority of the Kamakura bakufu. Additionally, Okada Seishi has argued that deer were used in Manyōshū poems and fudōki gazetteer narratives to symbolically connect the deer’s mating call with the sovereign assuring the fertility of the crops and protecting the realm. See David T. Bialock, Eccentric Spaces, Hidden Histories: Narrative, Ritual, and Royal Authority from the Chronicles of Japan to the Tale of the Heike (Stanford, CA: Stanford University Press, 2007), 113; Lisa Grumbach, "Sacrifice and Salvation in Medieval Japan: Hunting and Meat in Religious Practice at Suwa Jinja" (Ph.D., Stanford University, 2005), 38-39. 503 The Nihon shoki includes an episode in which two sons of Chūai Tennō, Kagosakaō 麑坂王 and Oshikumaō 忍熊王, perform a prayer hunt (ukeigari 祈狩) in which the bounty of the hunt would divine the outcome of their revolt against Queen Consort Jingū. Kojima, Nihon shoki 日本書紀, v. 1, pp. 436-37; Aston, Nihongi: Chronicles of Japan From the Earliest Times to A.D. 697, 1, 237. 504 Asahara and Nanami, Nagato-bon Heike monogatari no sōgō kenkyū: Kōchū hen 長門本平家物語の総合研究：校注篇, 399. 272 As the Itsukushima Deity predicted, the members of the court expressed doubts at first, but they changed their minds after seeing the three stars shine upon them. The deity’s miraculous deeds illuminate the potential to be gained from the court supporting Itsukushima Shrine, but this is displayed in a surprisingly confrontational manner. The delivery of sasaki branches to the capital recalls the practice of gōso 強訴 demonstrations in which religious institutions would protest against the court by bringing their deity to the capital in a palanquin (mikoshi 神輿) or sasaki branches. The Shōbōrinzō and Genpei jōsuiki also include the scene of the Kuramoto petitioning the court, but they leave out any mention of his exile. In the Shōbōrinzō, Kuramoto just states that he is incapable of succeeding in the court and asks if the deity might be able to demonstrate her miraculous powers. 505 In the end, Kuramoto arrives at the court, and with the Itsukushima Deity’s show of power, successfully gains the court’s support. It is certainly possible to see regional appeal in the Itsukushima Deity’s show of power against the center, and to see similarities with Kiyomori’s worship of Itsukushima as a geopolitical othering of himself in contrast to the elites of court. It would be worth considering the transgressive potential of this episode for the Nagatobon if the Nagatobon consistently promoted spatial frameworks presenting western Japan and the center in conflict with one another. However, I have not found this to be the case. In order to follow the larger tendency of the Nagatobon to connect places in western Japan with events from the past, the Itsukushima origin narrative does not need to be linked to Kiyomori or the Heike narrative. 505 The Shōbōrinzō concludes the scene of the Itsukushima Deity’s miraculous deeds with Prince Shōtoku explaining the identity of the Itsukushima Deity to Suiko Tennō, and recommending her to offer estates to Itsukushima. "Shōbōrin -zō 聖法輪蔵," 488. 273 Comparison of the Nagatobon Itsukushima origin narrative with the Itsukushima Daimyōjin nikki reveals that the Nagatobon compilers made minor changes, and none that support linking the narrative with the Heike. Makino Kazuo notes four main differences between the two texts. First, (1) in the introductory section on the identity of the deity, the passages connecting Itsukushima with Hakusan and Kamado are removed, and (2) the age of Queen Consort Jingū is changed from her being the older sister of the Itsukushima Deity to the younger. (3) In the conclusion, the Itsukushima Daimyōjin nikki simply says that all under heaven worshipped [Itsukushima], while the Nagatobon adds a more specific example. “[The Itsukushima Deity] is a deity of awesome miraculous power and was even worshipped by the retired sovereign (Go-Shirakawa).” Lastly, (4) an annotation by an earlier copyist quoting the Sanmon kōshiki 山門講式 to explain the connection with the Hakusan Deity through the honji- suijaku association of the Marōdo Deity 客者 at Hie Shrine (and Itsukushima) and Hakusan as manifestations of Eleven-headed Kannon in the Itsukushima Daimyōjin nikki is left out from the Nagatobon. 506 Three of the four changes (1, 2, and 4) relate to connections with other gods and sovereigns which were vital for the Itsukushima Daimyōjin nikki, but unnecessary (and potentially confusing) in the later context of the Nagatobon. 507 In contrast, the one addition of information (3), kept the meaning of the passage the same but adds Go-Shirakawa, one of the most powerful members of the royal family during the time of Kiyomori’s rise to power, to its praise for Itsukushima. While the connection with Go-Shirakawa hints at the connection with the Heike narrative, Go-Shirakawa is a rather innocuous choice for acknowledging Itsukushima’s 506 Makino, "Nagatobon "Heike monogatari" kan go 'Itsukushima shidai no koto' wo meguru ikkō s atsu: "Kamado san hōman dai bosatsu ki" wo kaishite 長門本『平家物語』巻五「厳嶋次第之事」をめぐる一考察 : 『竈門山寶満大菩薩記』を介して," 78. 507 For more information on the connections between these gods in the Itsukushima Daimyōjin nikki, see chapter one. 274 significance in the late twelfth century. There were historical records showing Go-Shirakawa making pilgrimages to Itsukushima and Go-Shirakawa’s mixed support and antagonism with Kiyomori complicates simple attempts to use the association with Go-Shirakawa to link Itsukushima with the Taira. Additionally, minor changes in wording hint at the Nagatobon compilers trying to prevent the Itsukushima origin narrative from conflicting with the Heike narrative. These alterations are few and short in length, making it easy to read too much into the differences. That being said, the scarcity of changes suggests that the revisions were not made carelessly. Most notably, one sentences removes the world “our realm” (waga chō 我朝) 508 and another removes “royal law” from the common pairing of “the royal law and the Buddhist Law” (ōbō buppō 王法仏法). Together, their removal lessens the emphasis on the Itsukushima Deity’s role as a protector of the realm in the Itsukushima Daimyōjin nikki, which could remind readers of Kiyomori’s transgressive acts. While some of the alterations made by the Nagatobon help to make the passages clearer, the removal of “Royal law” from “the Royal law and the Buddhist Law” makes the sentence more confusing. The Nagatobon says, “Like this, [Itsukushima] is a deity who brings together and protects the Buddhist Law.” 509 The verb “bring together” (kanete 508 Itsukushima Daimyōjin nikki: It is [the deity’s vow] that anyone who makes one pilgrimage [to Itsukushima] will not fall into a bad path (of rebirth) for a long time. The proof of this is that [the Itsukushima Deity] revealed a tinge of the knowledge she held to Kōbō Daishi (Kūkai). [His] tr ansmission of the esoteric teachings to our realm was this deity’s vow. Ichido sankei no tomogara wa, nagaku akudō ni ochisu to on chikai ari. Sono shōko wa Kōbō Daish ni on chi’in wo mochite sono iro wo arawasu. Waga chō ni misshū no wataru koto wa, kono kami no gokugan nari. 一度参詣ノ輩ハ永ク悪道ニヲチスト御誓有其証拠ハ弘法大師ニ御知印ヲモチテ其色ヲ顕ス我朝ニ密宗ノ渡事ハ此ノ神ノ御願也 Kanazawa Bunko, Kanazawa Bunko no chūsei shintō shiryō 金沢文庫の中世神道資料. Nagatobon: It is [the deity’s vow] that anyone who makes one pilgrimage [to Itsukushima] will not descend to a bad path (of rebirth) for a long time. The proof of this is that Kō bō Da ishi’s (Kūkai) transmission of the esoteric teachings was this deity’s vow. Ichido sankei no tomogara wa, nagaku akudō ni dassezu to chikai ari. Sono shōko Kōbō Daishi ni misshū no wataru koto wa, kono kami no gokugan nari 一度参詣の輩は、なかくあく道に堕せすと、ちかひあり。その証拠弘法大師に密宗のわたる事は、此神の御願なり。Asahara and Nanami, Nagato- bon Heike monogatari no sōgō kenkyū: Kōchū hen 長門本平家物語の総合研究：校注篇, 400. 509 Nagatobon: “Like this, [Itsukushima] is a deity who brings together and protects the Buddhist Law.” 275 兼ねて) is only fitting when both the Buddhist Law and the Royal Law are paired together. This is an easy change to make though, for the compilers to lessen emphasis on royal authority and instead present Itsukushima Shrine as a sacred place separate from the political contestations in the Tale of the Heike or the ritual needs of the center. Like the other origin narratives of sacred spaces in western Japan added to the Nagatobon, the Itsukushima origin narrative is important because it does not tell the Heike narrative. The origin narratives allow the Nagatobon to connect prominent sacred spaces with the Heike and its historic events, without having to delve into the complicated roles of the site within the Heike. The Nagatobon works to highlight Itsukushima’s importance throughout the narrative, but that is not always a simple task. The Itsukushima origin narrative, though, required little more than being copied into the text to tell of Itsukushima’s ancient origins. Conclusion The modern-day image of Itsukushima Shrine emerged after the end of World War II. While the image of Itsukushima during the Russo-Japanese War was built around nationalistic conceptions of a Japanese martial culture, the conditions of Japan’s defeat in World War II and the subsequent occupation, led the image of Itsukushima to shift from war to peace. Hiroshima tourism grouped the Hiroshima Peace Memorial (Genbaku Dome) with Itsukushima to first call visitors to remember everything that was lost to the destruction of war, and to then see Itsukushima Shrine as a standing symbol of Japan’s cultural roots in the “golden age” of Heian court culture. 510 Itsukushima Daimyōjin nikki: “Like this, [Itsukushima] is a deity who brings together and protects the Buddhist Law and the Royal Law.” 510 Zwigenberg, "The Atomic City: Military Tourism and Urban Identity in Postwar Hiroshima," 628-29; Yoshida, Bui, and Lee, "Does Tourism Illuminate the Darkness of Hiroshima and Nagasaki?," 337-38. 276 Despite the wartime portrayal of Kiyomori as a famed warrior, the postwar period reconstructed the image of Kiyomori as the transmitter of court culture and performing arts to Itsukushima. In a 2001 article for The Unesco Courier, Hiroshima-born novelist Takenishi Hiroko ( 竹西寛子, 1929-) describes the annual kangensai rite at Itsukushima Shrine in relation to the character of Kiyomori, “This festival is an uncanny blend of boldness and grace, nonchalance and rigour. At its heart stands the image of a shogun at the height of his power, faced with a fading destiny and the yearning to perpetuate the golden age of the Heian.” 511 Takenishi instills the complex simplicity of Itsukushima and Kiyomori’s reconceptualization into one concise passage. The image of Kiyomori cannot be completely divorced from his image in the Tale of the Heike, as the leader of the Taira family who leads them to their sudden rise and fall from power. Nevertheless, she does not allow the Taira’s famous downfall to conflict with positive portrayals of Itsukushima as a sacred and traditional space. 512 Takenishi weaves together the images of Kiyomori as a fervent religious devotee, a warrior ruler and catalyst of the Genpei War (1180-1185), and a preserver of court tradition, but we can also see the inconsistencies in her language. In short, the passage anachronistically presents Kiyomori as seeing his own time as a golden age. Articulating Kiyomori’s intentions in terms of his postwar image as a transmitter of culture, results in a disjointed image of the Taira fighting to preserve the Heian period. This is all the more striking given the conventional reading of the Tale of the Heike as blaming Kiyomori’s 511 As a publication for Unesco, Takenishi situated her writing for a broad international audience, and her 1978 novel Kangensai 管弦祭 gives a more personal and localized discussion of Itsukushima and Hiroshima after the atomic bombings. In Kangensai, Takenishi does not emphasize or defend Kiyomori, and only mentions him once, saying “In regards to the origins of the [kangensai] festival, it is said that it was a performance of string and wind music on a boat began by the provincial governor of Aki Taira no Kiyomori, but aside from that, little is known about it.” Takenishi, "Shinto and Shoguns: A Return to Itsukushima," 43. Takenishi, Kangensai 管絃祭, 192-93. 512 Historical sources show that Kiyomori was an active participant in court culture and used his religious worship at Itsukushima as part of his actions to assert his ruling authority. Blair, "Rites and Rule: Kiyomori at Itsukushima and Fukuhara."; Adolphson and Commons, "Blurring the Lines: Repositioning the Heike," 3-4. 277 arrogance, rash decisions, and evil deeds for leading to the Genpei War and the fall of the Taira. While never explicit, Takinishi’s pairing of easily applicable descriptors of Kiyomori as “bold” and “rigorous” with “grace” and “nonchalance” balances out his perceived flaws, defending Kiyomori and Itsukushima from the types of critiques found in the Tale of the Heike. These modern examples present Itsukushima as a site of memory crystallizing the shrine’s connection with Kiyomori as a symbol of Japan’s military, courtly, and religious heritage. However, the same cannot be said for the meaning of Itsukushima in the medieval period (approximately 1200-1600). In order to reexamine the meaning of Itsukushima Shrine, we must distance ourselves from its significance in the present as the tutelary shrine of the Taira. Pierre Nora describes sites of memory (lieux de mémoire) as “fundamentally remains, the ultimate embodiments of a memorial consciousness that has barely survived in a historical age that calls out for memory because it has abandoned it.” 513 As such, the establishment of a site of memory requires fading recollection and a will to remember. Monica Juneja critiques Nora’s studies on sites of memory for presupposing an ahistorical national-consciousness and state institutions promoting the need for preserving unifying memories and overlooking the impact of French colonialism on cultural memory. 514 We must be careful not to fall into the same pitfalls. We must therefore ask: How did the association between Itsukushima and the Taira develop? Who would desire to promote memories of the Taira’s devotion to Itsukushima? When would such intentions begin to arise? 513 Pierre Nora situates memory and history as oppositional. Remembrance keep the past alive, changing, and connected to the people. History reconstructs and represents the past apart from the present, as something which no longer exists. Nora, "Between Memory and History: Les lieux de mémoire," 8, 12. 514 Juneja, "Architectural Memory between Representation and Practice: Rethinking Pierre Nora’s Les lieux de mémoire," 15. 278 Itsukushima Shrine is commonly known by modern-day visitors as a site of beauty, and as the long-ago shrine of Taira no Kiyomori. But this is not the only way to understand the identity of the Itsukushima Deity. In order to uncover Itsukushima’s continued importance after the fall of the Taira, we must reexamine how medieval discourses discussed Itsukushima as associated or apart from the Taira. For the post-Genpei War associations between Itsukushima with the Taira, silence was often preferable to explanation. Kiyomori’s devotion of the Taira was never fully erased from memory, but over time, enough was forgotten to allow texts to highlight the Taira’s stature as an elite warrior family without immediately calling to mind an image of the Taira as the losing side of the Genpei War. This simplified memory of the Taira could be used to promote Itsukushima Shine, and other places and people in western Japan. Nevertheless, there often remained more risks than benefits in noting the connection between Itsukushima and the Taira. Itsukushima Shrine’s other associations as a dragon deity (specifically as the sister of the Dragon Princess from the Lotus Sutra) and as the Buddhist-Hindu goddess Benzaiten became more than capable of preserving Itsukushima’s identity as a preeminent deity of the sea and of western Japan. The relationship between Itsukushima and the Taira in the Nagatobon variant of the Tale of the Heike likewise benefits from what is not said, just as much as from what is. The Nagatobon does not use its regional, western-Japan based voice to reconstruct the geopolitical framework of the Heike or to reveal a hidden, pro-Taira counternarrative. Instead, it strives to follow standard regional spatial practices of associating places with the past, while confronted with the Heike narrative’s positioning of the Taira and their allies in western Japan on the wrong side of history. As a result, the regional emphasis of the Nagatobon is easiest to find in scenes peripheral to the central storyline of the Heike. It is the tangents, the asides, the seemingly 279 meaningless details that provided the compilers with opportunities to incorporate new perspectives into the Heike. The Tale of the Heike, the origin narrative of Itsukushima Shrine, and countless other narratives and discourses interconnected to form discursive networks influencing conceptions of the identity of Itsukushima. Even the positive portrayals of the Taira as warrior elites often reflected the influence of the Heike, ignoring the ending and focusing on other aspects of the narrative. Therefore, we cannot ascertain a cohesive view of Itsukushima from any one body of sources, whether that be the Heike or the writings of officials at Itsukushima Shrine. The connections and significances of the writings often only become clear once they are examined together and we are able to link them together. Reading the Nagatobon as a bounded text could lead us to interpret the Itsukushima origin narrative and all the mentions of Itsukushima in relation to the larger Heike narrative. Relating the Nagatobon with other contemporary sources about Itsukushima and the Taira, we can begin to recognize the importance of reading Heike episodes about Itsukushima outside of the context of the Heike narrative. Distance and forgetting were a vital part of drawing connections and memories. 280 Conclusion The tales told by common folk are simply astonishing to hear. People of refinement never tell tales of the strange and marvellous. Nevertheless, this does not mean one should necessarily disbelieve the stories of the miraculous powers of the gods and buddhas, or legends of their manifesting in earthly form. It is foolish to be credulous of all the tall tales people tell about such things, but there is no point in doubting everything you hear either. As a rule, you should accept such stories at face value, neither believing everything nor ridiculing it all as nonsense. 515 Modern-day scholars are not the only ones who can read temple and shrine origin narratives and question their veracity. Yoshida Kenkō ( 吉田兼好, ca. 1283~1352) had his doubts, but he still refrained from deeming them lies. He recognized that most stories, not just those about the gods, are often exaggerated. As he put it, “Stories that get passed around are for the most part lies, no doubt because the truth is so boring.” 516 If we dismiss everything that could be false, we could be left with nothing. Kenkō’s words are still worth listening to today. Sch olars strive to master their research, but it is impossible to fully escape uncertainty. Even the individuals working with enviably comprehensive sources must recognize that the extant records are incomplete and may include inaccuracies. A hermeneutics of suspicion might help protect us from being misled, but we would benefit from considering other avenues of interpretation as well. Rather than presuming that origin narratives are baseless fabrications, would we not be better off engaging with the tales as potentially true? When we read medieval texts, should we not consider the interpretive practices of medieval individuals like Kenkō? 515 Translation by Meredith McKinney. See Kenko Yoshida, "Essays in Idleness," in Essays in Idleness and Hojoki (London: Penguin Classics, 2013), 57-58; Yoshida Kenkō 吉田兼好, "Tsurezuregusa 徒然草," in Hōjōki, Tsurezuregusa, Shōbō genzō zuimonki, Tannishō 方丈記・徒然草・正法眼蔵随聞記・歎異抄, ed. Kanda Hideo 神田秀夫, Nagazumi Yasuaki 永積安明, and Yasuraoka Kō saku 安良岡康作, Shinpen Nihon koten bungaku zenshū (Tokyo: Shō gakkan, 1995), 139-40. 516 Yoshida, "Essays in Idleness," 57; "Tsurezuregusa 徒然草," 139. 281 The ubiquity of postmodern critique in the humanities has made it easy to question our sources and look for hidden meanings. However, this is not the only way to approach a source. Bruno Latour and Rita Felski have raised issues with the normative position of critique and the resultant lack of alternative approaches. 517 This is not to say that critique has lost its usefulness or validity, but to suggest that it should be one tool within a larger repertoire. To borrow Felski’s phrasing, if our default approach is “looking behind the text—for its hidden causes, determining conditions, and noxious motives,” than we are likely to miss what is in front of the text—“what it unfurls, calls forth, makes possible.” 518 This is a particularly pertinent issue for religious studies, because critique can risk prioritizing historical contexts over the concerns and aspirations of religious thinkers and practitioners. Kimberley Patton warns, “To discount as politically coded ‘pretexts’ the theological, cultic, or philosophical motivations offered by participants themselves in religious phenomena, movements, or controversies, is arrogantly to disenfranchise those we purport to understand.” 519 Modern focus on the social construction of origin narratives has kept us from asking about the questions, concerns, and practices of the people who wrote, read, and retold them. The established method of analyzing a single origin narrative text and situating it within the broader context of the site’s history has led to two common modes of interpretation. They either reveal an origin narrative’s social construction (e.g., influenced by a religious institution’s political or economic needs), or read the narratives as windows into the medieval imaginaire of the sacred place. The former stresses the presumed agency of a religious institution’s clergy to fabricate a 517 Bruno Latour, "Why Has Critique Run out of Steam? From Matters of Fact to Matters of Concern," Critical Inquiry 30, no. 2 (2004): 225-26; Rita Felski, The Limits of Critique (Chicago: University of Chicago Press, 2015), 8-10. 518 Felski, The Limits of Critique, 12. 519 Kimberley C. Patton, "Juggling Torches: Why We Still Need Comparative Religion," in A Magic Still Dwells: Comparative Religion in the Postmodern Age, ed. Kimberley C. Patton and Ray Benjamin C. (Berkeley, CA: University of California Press, 2000), 168. 282 mythological history which will compel audiences to act in the interests of the temple or shrine. The latter argues that regardless of the motivations for its production, an origin narrative could guide collective understandings and experiences of a sacred place and its divinities. Both provide the institution’s clergy as the primary actors in the construction of the site’s origin narratives and identity. Audiences are presumed to naively accept whatever the narratives say. Similarly, origin narrative texts are presented as containers for the clergy’s fabricated mythologies, rather than as things that could be transformed and circulated outside the control of the original producers. Considering the agency of texts and people outside of Itsukushima Shrine shifts our focus away from the original context behind the construction of an Itsukushima origin narrative to how later persons and texts engaged with them. We can begin to see how scholarly monks and worshippers valued origin narratives as sources of knowledge on the local deities once we expand the context of analysis beyond the history of a single cultic site. Postmodern critique’s emphasis on difference over similarity and sociocultural contextualization over comparison similarly encourages two views towards medieval Shinto: as the appropriation of kami worship by Buddhism or defined by its spatial specificity (i.e., locality). Both cases prioritize the study of kami worship in situ over more abstract intellectual concerns. Categorizing Shinto as a part of Buddhism suggests that Buddhism provided the intellectual framework, and that the kami’s importance was bound to local practices and culture. To be clear, I agree with the critiques against the conventional idea of Shinto as the indigenous religion of Japan which has continued from ancient times to the present. This is an important argument, especially considering its postwar development in response to prior nationalistic 283 essentializations of Shinto. 520 Ironically though, the desire to avoid essentializing Shinto has now led us to the dangerous precipice of generalizing kami worship as “local” and “heterogeneous.” When we only turn our attention to particularity, we naturally overlook the shared, translocal bases connecting different religious institutions, deities, and beliefs. There was not a single unified discourse on the kami deities, but knowledge of the deities was far more integrated than one would expect. The deities needed to be comparable in terms of both similarities and differences to be intelligible, and therefore, the gods needed to be more than just local. As Wendy Doniger argues, “Any discussion of difference must begin from an assumption of sameness.” 521 There needs to be some common ground in order to make the particularities meaningful within broader contexts. No matter how closely the Itsukushima Deity was associated with the place of Itsukushima Shrine, its identity could not be an island separate from the other deities of the realm. As the previous chapters demonstrated, the Itsukushima Deity’s commonalities with other divinities and sacred spaces were central for articulating her identities and significance. Considering the analogical logic of the medieval Japanese episteme, we can also reverse Doniger’s phrasing: any discussion of sameness must begin from an assumption of difference. The preconceived spatial particularity of enshrined kami and Buddhist deities helped clarify how honji-suijaku logic could identify worldly deities as manifestations of otherworldly Buddhas without presenting all deities as identical. It is easy to overlook the importance of similarities when you set out to find differences (and vice-versa), but both are always involved. 520 Kimberley Patton similarly contextualizes the development of postmodernism in relation to postwar responses to fascist and genocidal ideologies. Ibid., 167. On the historiography of Shinto essentialism, see Gaétan Rappo, "Essentialism in Early Shinto Studies," in Exploring Shinto, ed. Michael Pye (Bristol, CT: Equinox, 2020). 521 Wendy Doniger, "Post-Modern and -Colonial -Structural Comparisons," in A Magic Still Dwells: Comparative Religion in the Postmodern Age, ed. Kimberley C. Patton and Ray Benjamin C. (Berkeley, CA: University of California Press, 2000), 65. 284 Instead of trying to resolve the complexities of sacred locality myself, I investigated how medieval origin narratives and scholarly monks worked to make sense of how local particularities and translocal commonalities relate. Bruno Latour presents actor-network theory as a means to help enable the contemporary actants to guide the organization and perspective of our studies. We can illuminate how knowledge about the Itsukushima Deity developed by tracing how texts and ideas about the deity changed over time, analyzing those transformations to uncover the influences of differing interpretations and controversies, and finally, examining how they worked to resolve those issues. 522 This methodology is apt for the study of kami worship and origin narratives because many sources provide brief descriptions of the gods without explaining how or why the included details were meaningful. Their portrayals of the kami appear simple until we compare multiple texts and consider how and why certain details changed or stayed the same. The varied connections between different deities might appear random or meaningless from our points of view, but medieval scholarly monks valued them as logical and informative information about the gods. It can be easy to overlook issues that were important in the past if we rely on established categories and conceptual frameworks established in the present. The translocal nature of local deities is one topic which has been understudied, despite being a substantial issue for medieval scholarly monks. When the monks read and wrote origin narratives, they encountered a problem: how to explain the nature of local deities (including both kami and enshrined Buddhist deities) as spatially particular and transcendentally universal. This is a nonissue when the deities are discussed individually and described as local, but 522 Bruno Latour, Science in Action: How to Follow Scientists and Engineers through Society (Cambridge, MA: Harvard University Press, 1987), 15; Reassembling the Social: An Introduction to Actor-Network-Theory (New York: Oxford University Press, 2005), 23. 285 complications arose once monks compared origin narratives and tried to understand their contents and incongruities within a unified body of knowledge. They did not avoid the complications by solely emphasizing their locality or shared identity. Instead, they found other avenues for simplifying and clarifying the relations between the divinities, while still maintaining opportunities for articulating both individual and shared identities. Some origin narratives emphasized certain identities or networked connections over others, but their efforts to remain consistent with the broader body of knowledge on the gods helped balance simple identities and complex interrelations. Scholarly monks’ intellectual labor to make sense of the translocal natures of local deities and origin narratives brought together elements of religious, literary, and historical significance. Worldly influences (political, economic, social) did not corrupt or take priority over origin narratives’ importance as religious and literary works. Instead, they came together to inspire new perspectives and ways of thinking. For example, the first chapter emphasizes the centrality of the Mongol invasions for stimulating new ideas about how to promote certain deities over the rest. On one level, this was a worldly opportunity for temples and shrines to distinguish themselves by asserting the superior power of their deities to defeat the Mongols. This context is crucial, but it does not give us license to assume that their arguments deviated from the accepted intellectual practices of scholarly monks. To the contrary, I stress that their recognition of the difficulties involved in convincingly arguing for a particular deity’s agency in defeating the Mongols led to new ways of thinking about the gods. Knowledge about the kami expanded precisely because scholarly monks needed to keep their arguments within the realm of reason. They needed to be able to assert the particular importance of their site’s deities without it conflicting with the numerous other origin narratives similarly arguing for the prominent role of their deities for 286 defeating the Mongols. As one solution, origin narratives like the Itsukushima Daimyōjin nikki presented their deity as one member of the group of deities associated with Queen Consort Jingū as renowned for subjugating foreign enemies. As long as the producers were able to reference other origin narratives to make sure that the associations between deities were consistent, they were able to proclaim the power of their deity without conflicting with the similar statements made about the other deities in the group. The fourteenth century discussions of the Itsukushima Deity as Benzaiten in chapter two similarly, although less explicitly, hint at competitive comparisons of the different local Benzaiten. This is especially prominent in the writings focused around the elite sites of Mt. Hiei and Tenkawa. Scholarly monks at these sites could not disregard the existence of multiple Benzaiten with the same ontological essence, but they were able to employ connections between Benzaiten sacred spaces to present select manifestations of Benzaiten as superior. Listing Itsukushima, Chikubushima, and Enoshima together as the three Benzaiten implicitly suggests that they are the preeminent Benzaiten within Japan. Although varying groupings of Benzaiten deities emerged to add others to the three above, the selected deities were remarkably consistent. Comparing accounts of different Benzaiten raised the need for closer consideration of the meaning of their individual/shared identities, leading origin narratives and scholarly writings to explain how their sacred spaces were connected through underground caverns. The descriptions of the sacred geography within the narratives provided a logic of connected yet different spaces to complement honji-suijaku’s system of connections based in commonalities. Competitive concerns between Benzaiten did not lead institutional needs to take priority over well-grounded religious thinking. 287 The first two chapters provide an overview of the intellectual concerns and solutions that emerged in origin narratives relating multiple local deities and sacred spaces. The latter two chapters then examine how narratives promoted the Itsukushima Deity’s individuality while recognizing the basis of its identities in interrelations. The third chapter raises concerns over how to choose which deity to worship when faced with numerous similar options. Why make a pilgrimage to Itsukushima Shrine if there are temples and shrines that might be equally or better capable of answering one’s prayers? Notably, the Itsukushima no honji narrative of the past life of the Itsukushima Deity, does not resolve the issue by depicting the deity’s uniqueness, but by encouraging new connections between the deity and individual readers. Its story about the deity’s past human sufferings—competition for a husband’s affection between consorts, fears of dangerous childbirth—was not novel and it was not supposed to be. 523 They resembled the honji monogatari narratives of other temples and shrines, but the familiarity of the topoi and the portrayed sufferings helped strengthen the audience’s emotional response. The narrative then overlay the audience’s emotional connections with devotional practices of pilgrimage and prayers to strengthen their individual bond with the deity. This provides a foundation for trusting the Itsukushima Deity to answer their prayers, because the worshipper and the deity are now separated from the others through their personal connection. Lastly, the fourth chapter confronts worries over potentially negative distinguishing traits. The easiest way to distinguish the Itsukushima Deity was to refer to its historical connection with the Taira household, but the deity could be blamed for failing to prevent the Taira’s famous defeat in the Genpei War. The Nagatobon variant of the Tale of the Heike is 523 Max Moerman shows how Onsenji origin narratives similarly employed conventional themes rather than original stories to strengthen its impact. D. Max Moerman, "The Buddha and the Bathwater: Defilement and Enlightenment in the Onsenji engi," Japanese Journal of Religious Studies 42, no. 1 (2015): 74. 288 special because it attempts to temper the relationship between the Itsukushima Deity and the Taira from preexisting Heike narratives. It takes tangents to tell the Itsukushima Shrine origin narrative and to describe pilgrimages to the shrine to emphasize Itsukushima’s more general individual significance as a renowned sacred space of western Japan. As the literary influence of the Heike grew over the course of the medieval period, it became easier to perceive of Taira as an elite and famed household, and this shift is reflected in the larger body of medieval texts associating Itsukushima with the Taira. That being said, it is notable that the Nagatobon compilers felt the need to defend against potential criticisms of the Itsukushima Deity in relation to the Taira, and that they did not respond by manipulating the discussion of Itsukushima already within the Heike. Neither the Itsukushima no honji nor the Nagatobon Tale of the Heike stress uniqueness over connections, but they do not ignore individuality either. They took seriously the matrices of texts, discourses, and associations that provided meaning and the authority of precedence. The compilers of the Nagatobon and Itsukushima Shrine origin narratives did not construct the identities of the Itsukushima Deity through creative invention, but scholarly logic and references to supporting evidence. Topics for Future Study: Connections through Difference Every methodology highlights certain aspects while underemphasizing others. My translocal approach to the site-based study is no different. Following how texts from temples and shrines far from Itsukushima Shrine incorporated the Itsukushima Deity and information from its origin narratives, led me away from regional contexts and particularizing characteristics. While outside the scope of the dissertation, it is worth taking a moment here to consider how further 289 discussion of these more particular aspects could help to bridge local and translocal associations and significances. Notions of difference and distinction tend to separate the one from the many, but they can still lead to connections between those perceived as similarly exceptional. Itsukushima’s distinctive scenery attracted more than just pilgrims, but also written and visual representations of the site/sight. As a result, writers and painters sometimes chose to include Itsukushima Shrine’s image within broader accounts of the temples and shrines of the realm. This in turn could be interpreted as grouping Itsukushima with the other included temples and shrines, and as more prominent than those left unmentioned. Both the Ippen hijiri-e (“Illustrated Life of the Holy Man Ippen,” 1299, 一遍聖絵) and the Yūzū nenbutsu engi ( 融通念仏縁起, 1314) reference too many local sacred spaces or deities to portray in detail within the scrolls. The compilers had to select which to include in text and images, and therefore, the painting of Itsukushima over renowned temples and shrines could help to promote Itsukushima Shrine’s image and fame. 524 We expect distinctions like fame to separate objects, but they could also serve as a commonality for connecting sacred spaces known for their uniqueness. Neither the Ippen hijiri-e nor the Yūzū nenbutsu engi describe Itsukushima in depth, but its inclusion alone was enough to highlight the importance of the shrine and its deity. Their paintings of the shrine’s memorable architecture, with its corridors raised over the sea, provided additional evidence and emphasis of the site’s exceptionality. The Ippen hijiri-e describes Ippen 524 Abe Yasurō argues for the need for the compilers of th e Ippen hijiri-e to selectively choose which sacred places to include within the work, and also provides a chart listing the sacred sites whose origin narratives are discussed within the Ippen hijiri-e. They include: Zenkōji, Sugō cavern, Tennōji, Mt. Kōya , Mt. Kumano, Inabadō, Jimokuji, Kayazushuku, Sumiyoshi, the grave of Prince Shōtoku, Taimadera, Iwashimizu Hachiman S hrine, Kyōshinji, Mt. Shosha, Hantaji, Mishima, Ninomiya (of Awaji province), and Shizuki Tenjin Shrine. Abe Yasurō 阿部泰郎, "Junreiki toshite no 'Ippen hijiri-e': 'Eden' ni okeru engi no isō 巡礼記としての『一遍聖絵』：『聖絵』における縁起の位相," Junreiki kenkyū 6 (2009). 290 making pilgrimages to Itsukushima Shrine and numerous other temples and shrines throughout the archipelago. Some of these episodes include comprehensive accounts of the site’s origin narrative, but there is only a single sentence recounting his visit to Itsukushima. 525 The painting of the scene commands the reader’s attention instead. The image begins with Ippen’s boat traveling through the torii gate towards Itsukushima before transitioning to the scene of Ippen watching four naishi dancing. (fig. 16). 526 The painting highlights the shrine’s well-known characteristics: its corridors and torii gate above the seas, the mountainous landscape behind the shrine, and the naishi dancers. It is possible that Ippen’s visit to Itsukushima Shrine would have been left out altogether if not for the renown of the site’s beauty and performances. 525 The sentence recounting Ippen’s visit to Itsukushima Shrine says, “Additionally, during the fall of the same year, during Ippen’s pilgrimage to Itsukushima of Aki Province, the naishi shrine attendants, among others, honored him and performed provisional ceremonies and maiden dances.” The image depicts both Ippen’s arrival, with his boat traveling through the torii in front of the shrine. Shō kai 聖戒 and En'i 円伊, Ippen Shōnin eden 一遍聖人絵伝, ed. Komatsu Shigemi 小松茂美, Murashige Yasushi 村重寧, and Furuya Minoru 古谷稔, Nihon emaki taisei 日本絵卷大成 (Tokyo: Chūō Kō ronsha, 1978), 270-73, 388. 526 Abe contrasts the historically inaccurate representations of the architecture at Itsukushima Shrine and Taimadera 当麻寺, with the more consistent iconographical images of Kumano, Mt. Kōya, and Iwashimizu Hachiman Shrine. More specifically, the Ippen hijiri-e’s square representation of Itsukushima Shrine with the stage in the middle diverges from the contemporary depiction of Itsukushima in scroll 5 of Honshōji’s hanging scrolls of the life of Prince Shōtoku). Abe, "Junreiki toshite no 'Ippen hijiri-e': 'Eden' ni okeru engi no isō 巡礼記としての『一遍聖絵』：『聖絵』における縁起の位相," 34-35. Figure 16: Ippen Shōnin eden 一遍上人絵伝, scroll 10, Shōjōkōji 清浄光寺 1299. This was published as three pages in the Nihon emaki taisei and put together as a single image here by the author. The water of the sea begins on the previous page (270), but is not included here. Images from Nihon emaki taisei 27, edited by Komatsu Shigemi 小松茂美, Murashige Yasushi 村重寧 and Furuya Minoru 古谷稔, 271-273. 291 The Yūzū nenbutsu engi says even less about Itsukushima than the Ippen hijiri-e. It merely includes the Itsukushima Deity’s name within a list of the gods who signed the holy man Ryōnin’s ( 良忍, 1073-1132) ledger (kanjinchō 勧進帳) as protectors of the Yūzū nenbutsu school. The illustrated scene shows Bishamonten holding the ledger and surrounded by the myriad deities that signed it—beginning with Buddhist devas, dragon kings, and continuing to kami of prominent shrines. The included deities represent the multitude of deities in Japan, but similar to the case of the Ippen hijiri-e, the compilers needed to be selective about which deities to include. For example, the Seiryōji version (produced mostly between 1414-1415) of the Yūzu nenbutsu engi emaki lists the names of thirty-six kami deities, but visually includes only eleven of them. Depicting the deities as their shrine buildings, the image depicts: Itsukushima, [Iwashimizu] Hachiman, Kumano, Sumiyoshi, Kasuga, Ise, Kamo, Inari, Gion, Kitano, and Hie. 527 Aside from Itsukushima, the sites were all centrally located elite institutions within the 527 Both the textual and visual selections of deities sometimes differ between productions of the Yūzu nenbutsu engi emaki. In contrast to the Seiryōj i version, the earlier Chicago Museum version (14 th century) lists 39 deities, and the later Tokuyūji version (estimated production between 1710-1713) provides a more comprehensive list of 131 deities. Within the images of the shrines, the Tokuyūji scroll removes Itsukushima, Kamo, Gion, and Sumiyoshi, and instead draws the following eleven sites: Ise, Iwashimizu Hachimangū, Kasuga, Kumano, Yoshida, [Kitano] Tenjingū, Inari, Hie, Mt. Fuji, Mt. Tateyama, and Mt. Hakusan. As Abe Mika suggests, the addition of Fuji, Tateyama, and Hakusan, likely reflects the contemporary development of the three mountains as important spaces for Yūzu nenbutsu practice in Eastern Japan. For transcripts and photographs of the Seiryōji, Chicago, and Tokuyūji ve rsions of the kanjin chō ledger scene respectively, see Yūzū nenbutsu engi emaki 融通念仏縁起絵巻, Nihon emaki zenshū 日本絵巻全集 (Tokyo: Tō hō Shoin, 1929), 15-17 (images), 3-5 (transcription); Komatsu Shigemi 小松茂美 and Kanzaki Mitsuharu 神崎充晴, eds., Yūzū nenbutsu engi 融通念仏縁起, Zoku Nihon emaki taisei (Tokyo: Chūō Kō ronsha, 1983), 35-39 (images), 141-42 (transcription); Abe Mika 阿部美香, "[Shiryō] To kuyūji shozō 'Yūzu dainenbutsu engi' 'Yūzu dainenbutsu kamegane engi:’ Kaidai to eiin, honkoku 〔資料〕徳融寺所蔵『融通大念仏縁起』『融通大念仏亀鐘縁起』：解題と影印・翻刻," Gakuen 925 (2017): 175-77 (images), 216-17 (transcription). For the analysis of the Tokuyūji scrolls, see ibid., 154, 56. For introductory information on the Seiryōji scrolls and the other extant scrolls of the Yūzu nenbutsu engi emaki, see Tashiro Shō kō 田代尚光, Yūzū nenbutsu engi no kenkyū 融通念仏縁起之研究, Rev. ed. (Tokyo: Meicho Shuppan, 1976), 107-248. 292 twenty-two shrine system. The distinct image of Itsukushima Shrine’s corridors above the sea might have influenced the decision to include its deity together with these elite shrines near the capital. Notably, while most versions represent Itsukushima Shrine through a small section of the shrine’s corridors, a later woodblock print version from 1691 provides a wider view of the shrine’s architecture (fig. 17). Apart from questions of original intent, the produced images and texts present Itsukushima Shrine as equivalent to the most prestigious shrines in Japan, and implicitly, as superior to the various deities left unmentioned. 528 The difficulties of copying and traveling with paintings limited the circulation of visual images of temples and shrines. Itsukushima Shrine was fortunate to have its shrine depicted in scrolls that were reproduced and shown to public audiences more than most. The Yūzū nenbutsu 528 The Shōbōrinzō’s account of the Itsukushima origin narrative also places the Itsukushima Deity within a list of the prominent kami deities of the realm. It follows the origin narrative with an explanation of Shōtok u introducing the performance of Buddhist ceremonies for the sake of the kami and unifying the deities of Japan through his worship of the thirty-eight deities (including the Itsukushima Deity). "Shōbōrinzō 聖法輪蔵," in Shinshū shiryō shūsei 眞宗史料集成, ed. Ishida Mitsuyuki 石田充之 and Chiba Jō ryū 千葉乘隆 (Kyoto: Dō hō sha, 1974), 489-91; Abe Yasurō 阿部泰郎, "Shōbōrinzō 「正法輪蔵」," Kokubungaku kaishaku to kanshō 54, no. 10 (1989): 97. Figure 17: Itsukushima Shrine on top half of right page. Yūzū dainenbutsu hon engi 融通大念仏本縁起, 1691, Ukai bunko 鵜飼文庫. Grey background cropped by author. Image from National Institute of Japanese Literature, coda 44. 293 engi was copied multiple times during the Muromachi period, often during kanjin fundraising campaigns and ritual events to provide devotees with opportunities to participate in its production and develop karmic connections (kechien 結縁) with the Yūzu nenbutsu school. 529 The monk Ryōc hin even led the production of woodblocks for printing copies of the illustrated scroll for his 1391 kanjin campaign, aiming to spread the Yūzū nenbutsu engi emaki to every province of Japan. 530 In addition to the Yūzū nenbutsu engi, some hanging scrolls of the life of Prince Shōtoku, such as the Kamakura period scrolls at Honshōji 本證寺, give a surprisingly large image of Itsukushima Shrine and the origin narrative episode of the Itsukushima Deity meeting head priest Saeki Kuramoto (fig. 18, top section). 531 These scrolls would have been used together with the Shōbōrinzō to explain the life of Prince Shōtoku to audiences, and we can infer that the Itsukushima origin narrative and Shōtoku’s worship of the various kami was recited at times. Frequent and widespread mentions of Itsukushima Shrine supported its image as a special sacred space of Japan, even if the texts and images only presented the site as beautiful. 529 Although it is impossible to verify the accessibility of the reproduced scrolls, the existence of multiple copies across wide geographical areas could have made the Yūzū nenbutsu engi emaki visible to far greater audiences than most illustrated scrolls (e.g., Ippen hijiri-e), which were commonly limited to elite audiences in the capital. 530 Takagishi argues that Ashikaga Yoshimitsu was closely involved in the production of the 1391 printed version of the Yūzū nenbutsu engi emaki, and later Ashikaga shoguns likewise patronized the production of new versions. Yoshimitsu’s grandfather, Takauji had experience with the printing of a Buddhist scripture (the Mokuren kyūbo kyō, ch. Mulian jiu mu jing, 目連救母経) in 1340. Additionally, the project of disseminating the scrolls across the realm fit with Yoshimitsu’s construction of temples and stupas in every province to create a network of shogunal authority extending outwards from the capital. Concerning the spread of the painted Yūzū nenbutsu engi scrolls, Ikumi Kaminishi argues that a diverse range of professional and amateur painters were involved in the copying of the scrolls and suggests that this might reflect their widespread transmission. See Takagishi Akira 髙岸輝 and Heather Blair, trans., "The Reproduction of Engi and Memorial Offerings: Multiple Generations of the Ashikaga Shoguns and the Yūzū nenbutsu engi emaki," Japanese Journal of Religious Studies 42, no. 1 (2015): 165-66; Ikumi Kaminishi, Explaining Pictures: Buddhist Propaganda and Etoki Storytelling in Japan (Honolulu: University of Hawai`i Press, 2006), 86. 531 I am indebted to Abe Yasurō f or introducing this image to me and providing me with a booklet with images of the Honshōji hanging scrolls. 294 Figure 18: Shōtoku taishi eden 聖徳太子絵伝), scroll 5, Honshōji, 1322 - 1323. Image from Amano Shinji 天野信治, ed., Honshōji shozō Shōtoku taishi eden kenkyū shiryō 本證寺所蔵「聖徳太子絵伝」研究資料, unpaginated. 295 Itsukushima Shrine’s regional status as the ichinomiya of Aki Province similarly afforded opportunities for including the Itsukushima Deity as one of the representative deities of the realm. Political authorities from the center made offerings and pilgrimages to regional temples and shrines like Itsukushima to assert their ruling authority over the realm. Itsukushima’s regional status and convenient maritime location also encouraged pilgrims, merchants, and other travelers to stop at the shrine before continuing their journey. Additionally, itinerant monks called the sages of the sixty-six regions (rokujūroku bu hijiri 六十六部聖) traveled to bury and make offerings of a sutra copy (usually the Lotus Sutra) at one temple or shrine in each of the sixty-six provinces. 532 Itsukushima Shrine might not have been the primary destination or site of significance in any of these cases, but they still supported Itsukushima’s position as one of the prominent sacred spaces of Japan and Aki Province. Regarding Itsukushima’s status within the province of Aki, Itsukushima Shrine held a position of distinction compared to other temples and shrines, while also having numerous connections with them. The Itsukushima Deity was enshrined at many temples and shrines in the region, and their connections with Itsukushima were noted in origin narratives, 533 ridgepole 532 The earliest record listing the places of each province (and the earliest record including Itsukushima Shrine) is the late sixteenth century document from the Yose household (Yose ie monjo 余瀬家文書) called the “Notes on the places of the sutra offerings in the sixty-six regions” (Rokujūroku bu hōnō fudasho oboegaki 六十六部奉納札所覚書). Historical records of sixty-six region sutra offerings exist from 1231 (Kitakidera 黄滝寺 of Iga province); 1280 (Kiyosumiyama 清澄山 of Awa province); fourteenth century (Ōyamadera 大山寺 of Sagami Province and Chōrakuji of Shimotsuke province). See Seki Hideo 関秀夫, Kyōzuka no shosō to sono tenkai 經塚の諸相とその展開 (Tokyo: Yūzankaku, 1990), 535-37; Yunoue Takashi 湯之上隆, "Chūsei Bukkyō to chihō shakai : Rokujūroku bu hijiri wo tegakari toshite 中世仏教と地方社会：六十六部聖を手がかりとして," Nanakuma shigaku 3 (2002): 5. 533 The Tōunji 洞雲寺 temple origin narrative is a rare example of a regional site with a medieval origin narrative, the Kinkō washō gyōjōki ( 金岡和尚行状記, 1547), explaining the decision of the Itsukushima Deity to be enshrined at the temple. I plan to discuss this text more at a future date, but it is notable that the origin narrative employs the logic of a local deity being converted to Buddhism by the temple founder, even though the Itsukushima Deity was already an enshrined deity recognized as a provisional manifestation of a transcendent Buddhist deity (usually Dainichi or Kannon). More generally, the origin narrative of Susa Shrine (Susa jinja engi 須佐神社縁起, 1469) and the Aki Province Kinkameyama Fukuōji origin narrative copy ( Aki no kuni Kinkameyama Fukuōji engi sha 安芸国金亀山福王寺縁起写, 1460) both mention the Itsukushima Deity as enshrined within the premises of the temple 296 inscription tablets (munafuda 棟札) 534 and shared ritual practices. 535 In this manner, Itsukushima Shrine existed as a singular place and institution, but also as networks of sacred places throughout the province. Unfortunately, the extant medieval texts discussing the role of the Itsukushima Deity at other local religious institutions are scarce, but they are enough to highlight the need for further study of interconnected regional spaces and deities in the provinces. Tōunji 洞雲寺, located just east of Itsukushima Shrine in the city of Hatsukaichi 廿日市, is a rare example of a regional temple whose origin narratives, clergy, and rituals were all connected with Itsukushima. According to the temple’s origin narrative, the Kinkō washō gyōjōki complexes. Although undatable, there also orally transmitted tales about the islands where the Itsukushima Deity visited and left her traces before finally deciding to move to Miyajima. Many of these narratives incorporate elements of the Itsukushima origin narrative and the ritual of the annual ceremony of the kangensai. See doc. 22 of "Fukuōji monjo 福王寺文書," in Hiroshima kenshi: Kodai chūsei shiryō hen 広島県史. 古代中世資料編, ed. Hiroshima-ken 広島県 (Hiroshima: Hiroshima), 147; Hatsukaichi chōshi: Tsūshi hen 廿日市町史. 通史編, 2 vols., vol. 1 (Hatsukaichi-chō : Hatsukaichi-chō , 1988), 468-69; "Susa jinja engi 須佐神社縁起," in Setouchi jisha engishū 瀬戶内寺社緣起集, ed. Wada Shigeki 和田茂樹, Tomohisa Takefumi 友久武文, and Takemoto Hiroo 竹本宏夫 (Hiroshima-shi: Hiroshima Chūsei Bungei Kenkyūkai, 1967), 194; Mukoda Hiroshi 向田裕始, "Seto nai no Ichikishima hime densetsu to kangensai 瀬戸内の市杵島姫伝説と管弦祭," in Itsukushima shinkō jiten 厳島信仰事典, ed. Nozaka Motoyoshi 野坂元良, Shinbutsu Shinkō Jiten Shirī zu (Tokyo: Ebisu Kōshō Shuppan, 2002), 214- 21. 534 Ridgepole inscription tablet providing evidence of the enshrinement of the Itsukushima Deity at Kōkai Shrine 光海神社 by 1332 and Nishiyamadera 西山寺 by 1687. Kokuritsu Rekishi Minzoku Hakubutsukan 国立歴史民俗博物館, ed. Shaji no kokuhō jūbun kenzōbutsu tō: Munafuda meibun shūsei, Chūgoku Shikoku Kyūshŭ hen 社寺の国宝重文建造物等：棟札銘文集成. 中国四国九州編 (Chiba-ken Sakura-shi: Kokuritsu Rekishi Minzoku Hakubutsukan, 1993), 112, 37. 535 There are at least two medieval cases of a recitation of the [Lotus Sutra] in one thousand parts (senbu kyō dokushō 千部経読誦), being performed at Itsukushima Shrine and another site (Sakurao [castle] 桜尾 and Itsukushima (pre 1565) and Tōunji and Itsukushi ma, 16 th century). While later, early modern gazetteers describe the Itsukushima annual rite of the kangensai 管弦祭 also being performed at Seiganji Temple 誓願寺 (in modern-day Hiroshima City), as well as Nohatayama 野畑山 and Daizenji Temple 大善寺 in Mihara. Notably, the rites are called the Nohatayama Itsukushima Shrine rite and the Daizenji Itsukushima Shrine rite respectively. See Hatsukaichi chōshi: Shiryō hen 廿日市町史：資料編, 5 vols., vol. 1 (Hatsukaichi-chō : Hatsukaichi-chō , 1975), 590, 695; Hatsukaichi chōshi: Tsūshi hen 廿日市町史. 通史編, 1, 468-69; "Itsukushima zue 厳島図絵," in Chūgoku no maki 中国の卷, ed. Hase Akihisa 長谷章久 and Ikeda Yasaburō 池田弥三郎, Nihon meisho fūzoku zue (Tokyo: Kadokawa Shoten, 1980), 307; "Mihara shibashi 三原志橋," in Mihara shishi 三原市史 (Hiroshima- ken Mihara-shi: Mihara Shiyakusho, 1970), 26-27, 62-63, 154; Nakamichi Gō ichi 中道豪一, "Kyū Hiroshima-shi ni okeru Itsukushima kangensai ni matsuwaru sairei gyōji ni tsuite: Kindai ni okeru Takachōchin, hifuri, otomon -bune no yōsō to kinsairei gyōji no hassei 旧広島市域における嚴島管絃祭にまつわる祭礼行事について : 近代における高ちょうちん・火振り・御供船の様相と新祭礼行事の発生," Hiroshima shūdai ronshū 59, no. 2 (2019). 297 (“Record of the Actions of Master Kinkō,” 金岡和尚行状記, 1547), the Itsukushima head priest Fujiwara no Norichika 藤原教親 made offerings to support the founding of the temple following an oracle received from the Itsukushima Deity. 536 A record of endowment from Tōunji also shows that Norichika’s son Munechika 宗親 made a land offering to Tōunji in 1493, six years after the building of the temple according to the 1825 gazetteer, the Geihan tsūshi 芸藩通志. 537 It is only natural that Itsukushima Shrine’s clergy developed more direct connections with nearby temples and shrines than was possible when trying to link the Itsukushima Deity with divinities from more distant sites. Nevertheless, the potential to use ritual performances and institutional interactions to link Itsukushima and Tōunji did not remove the importance of or igin narratives for connecting their deities and sacred spaces. The producers of the Tōunji origin narrative ( Kinkō washō gyōjōki) faced the difficult task of demonstrating the site’s individual authority while still acknowledging the central role of Itsukushima in its founding, and its enshrinement of the Itsukushima Deity. In order to emphasize the role of Kinkai in founding the temple without challenging the Itsukushima Deity’s choice to come to Tōunji, the Kinkō washō gyōjōki presents Kinkō converting the Itsukushima Deity to Buddhism. The theme of a monk converting local deities to Buddhism is common for temple origin narratives (and especially in cases of snake deities), but usually the local deity is presented as dangerous unless placated. 538 This case regarding the Itsukushima Deity is unusual 536 Hatsukaichi chōshi: Tsūshi hen 廿日市町史. 通史編, 1, 453, 69. 537 Ibid., 454-55; Hatsukaichi chōshi: Shiryō hen 廿日市町史：資料編, 1, 565. 538 For example, in the Enoshima origin narrative (Ensohima engi 江島縁起, Enoshima Shrine’s manuscript version) Benzaiten manifests to subdue the violent nature of a five headed dragon who had started eating children. The Sayohime さよひめ (also called Chikubushima no honji 竹生島の本地, earliest extant version estimated to late Muromachi period) account of the origins of Chikubushima similarly tells of a giant snake who was given a young girl as a sacrifice each year until the girl Sayohime recited the Lotus Sutra to the snake. Hattori Seidō 服部清道, "'Enoshima engi' kō 『江島縁起』考," Yokohama shōdai ronshū 10, no. 1 (1977): 126-27; R. Keller Kimbrough, 298 considering that the Itsukushima Deity had been peacefully enshrined at Miyajima for centuries prior to the founding of Tōunji. In order to make Kink ai’s explication of the Buddhist teachings to the Itsukushima Deity more sensible, the text emphasizes the god’s snake body and only identifies the Itsukushima Deity’s associations with other kami rather than Buddhist deities. 539 For most readers, though, this story presented a simple way to understand Tōunji’s autonomy , as well as its connection with Itsukushima. 540 The Importance of Simple Identities and Connections The above discussion clarifies how my arguments about the translocal connections between deities and sacred sites are compatible with viewpoints stressing the local identity of sacred places. It might be fair to say that local and translocal identities are two sides of the same coin, but we still need to consider both sides. Notions of Itsukushima Shrine as a local and scenic sacred site were crucial, in part, because they offered simple ways to understand Itsukushima’s identity. While scholarly monks worked to look past the facade of local uniqueness and make sense of Itsukushima’s multiple identities and complex interrelations with other deities and sacred places, their efforts often led to questions rather than clarity. The underlying details might trans., "Sayohime," in Wondrous Brutal Fictions: Eight Buddhist Tales from the Early Japanese Puppet Theater (New York: Columbia University Press, 2013), 183-84. 539 The Itsukushima Deity describes her identity to Kinkai, “I am the deity of this shrine. At first, I was a girl of the Dragon Palace in the sea, and now, as a pretense, I was born in this realm of Japan as the child of Amaterasu and raised by Susanoo. Divining the land, I came to this island. Now, hearing your wondrous teachings has illumined the nature of my mind. I request that you come to my shrine hall and confer [to me] the pure precepts and your lineage. Then, I will immediately remove [this] snake-body, attain enlightenment, and serve as a deity protecting the Buddhist teachings and the realm [of Japan].” Hatsukaichi chōshi: Tsūshi hen 廿日市町史. 通史編, 1, 468-69. 540 The Intoku taiheiki ( 陰徳太平記, completed in 1695 and printed in 1712) briefly mentions the Tōunj i origin narrative, but mainly focuses on the pure spring water at Tōunj i that was received as an offering from the Itsukushima Deity. The 1833 Itsukushima zue gazetteer identifies the place on Miyajima where Kinkō meditated and encountered the Itsukushima Deity, and describes how the water that emerged from the ground at that time likewise became famous as “Kinkō water.” In this ma nner, the Tōunji origin narrative influenced the places of both Tōunji and Itsukushima. Kagawa Masanori 香川正矩, Intoku taiheiki: Shōtoku ni-nen hanpon 陰徳太平記 : 正徳二年板本, ed. Yonehara Masayoshi 米原正義, 6 vols. (Tokyo: Tō yō Shoin, 1980), v. 2, p. 209; "Itsukushima zue 厳島図絵," 140. 299 be complicated, but most people in medieval and present-day Japan did not have reason to know or care about such minutiae. My examinations of the complex translocal relations between deities can conversely be read as an argument for the importance of simplicity. Although I argue that the networks of interrelations were crucial for the thinking of scholarly monks, these networks needed to be represented in clear and concise terms in order to be meaningful for others. Origin narratives did not only provide an intellectual means for understanding a deity or sacred space’s individuality. They also conveyed the networks and identities of the deities in intelligible terms for broader audiences. In this manner, origin narratives show how monastics were able to transmit their intellectual work regarding local sacred entities to diverse bodies of pilgrims, lay practitioners, and court officials. It makes sense that many medieval writings on local deities emphasize local identity when we consider that scholarly monks reformulated their data in order to make it more coherent. Locality simplifies the meaning of a god by highlighting one distinguishing characteristic, thus reducing its comparability. To be clear, I am not saying that local identity or site-specific practices or communities were less significant or sophisticated than the networks of association connecting deities. Rather, I am suggesting that medieval texts emphasizing the regional identities of the gods are not necessarily evidence that kami worship’s defining characteristics are locality and particularism. 541 Origin narratives could present relational meanings as peripheral or parallel to portrayals of the deities as local in order to avoid confusing 541 Heather Blair made a related point by presenting the cult on Mt. Kinpusen as “localizing religion” rather than local religion. Although medieval people emphasized the power of the site and its deity the Zaō Gongen through its local uniqueness, Blair argues that Zaō’s localization coexisted with doctrines and rituals that developed external to Kinpusen. Heather Elizabeth Blair, Real and Imagined: The Peak of Gold in Heian Japan (Cambridge, MA: Harvard University Asia Center, 2015), 60-61. 300 audiences. The disjuncture between local individuality and networked commonality was easily missed as long as the two lines of reasoning never crossed paths. I hope that further consideration of tranlocal interconnections will provide new avenues for examining the production of local knowledge, emplaced practices, and regional communities. I expect that considering the potential tensions between the local and the translocal will expand the significance of local belonging. There is a need for balance between simplicity and complexity, particularity and commonality, stability and fluidity. Simplify too much and the discussion becomes meaningless. Add too much complexity and nobody will be able to engage or learn from the discussion. Medieval temple and shrine origin narratives were most effective when they were based on extensive knowledge, yet succinctly defined the deities and their sacred places within broad discursive contexts. In other words, the scholarly monks who produced origin narratives made comparisons and abstractions to make vast and sometimes contradictory bodies of information about the gods comprehensible. All scholarship (in medieval Japan and modern-day academia alike) requires a degree of comparison in order to show readers its significance. As Jonathan Z. Smith argues, “Comparison provides the means by which we ‘re-vision’ phenomena as our data in order to solve our theoretical problems.” 542 From this perspective, eschewing comparison would be equivalent to lacking any methodological or conceptual framework. Of course, we need to take care not to oversimplify nuances, but this not an issue that can always be solved by describing things in ever-increasing detail. 543 The facts will always be more complicated than our 542 Jonathan Z. Smith, Drudgery Divine: On the Comparison of Early Christianities and the Religions of Late Antiquity (Chicago: University of Chicago Press, 1990), 52. 543 Kieran Healy argues that nuance can impede productive theoretical thinking when it becomes a scholarly aim in itself (i.e., arguing that a subject is more complicated than how it is commonly understood). Arguments for increased complexity implicitly disparage theory and abstraction as imprecise or misleading. This can result in inattention to the theoretical significance of a case study, and uncritical reliance on established conceptual categories like culture and power. Kieran Healy, "Fuck Nuance," Sociological Theory 35, no. 2 (2017): 119, 22-23. 301 arguments, but that is not always a bad thing. If anything, some scholarship might benefit from loosening the particularity of spatial (e.g., local) and temporal (e.g., historical) context in order to expand its connections with readers in the present. If we set aside historical particularities for a moment, temple and shrine origin narratives produced by scholarly monks resemble the site-based studies of scholars in the present. Scholarly monks are, after all, both scholars and monks. In both cases, the scholars researched a particular site, but their discussions expand as they encounter broader questions and related contexts. From this perspective, it makes perfect sense to treat origin narratives within the realm of knowledge rather than fiction. Considering origin narratives as myths unconcerned with historical accuracy diminishes the intellectual standards of the scholarly monks involved in their production and transmission. Yes, origin narratives producers often endeavored to promote the site by stressing its miraculous efficacy, political centrality, and historical prestige. Sometimes they even invented stories and tried to deceive their audiences. But baseless fabrication was the exception, not the norm. It can be enlightening, and enjoyable, to read about the relatable issues faced by scholarly monks— overwhelmed by the amount of texts to read, confused by inconsistencies in the sources, and unable to trace the information cited by others. There is a lot we can learn once we step back from our modern categories and frameworks and follow our sources forward instead. Why study Itsukushima Shrine during the medieval period? The best answer for this question is that contemporary persons considered it an important subject to know. Itsukushima is a site that has been understudied because it falls outside of the standard discussions of the political, religious, and cultural histories of medieval Japan. The same could be said about many other understudied temples and shrines that remain understudied. This can produce a vicious cycle where we continue to select sites to study based on their recognized importance, and 302 consequentially, continue to ignore other lesser-known sites. I hope that my study can help serve as a launch pad for discussing how following connections between subjects can shine attention on understudied places, texts, people, or even methodological approaches. A place might seem inconsequential or a narrative might sound absurd, but these are not sufficient reasons to raise doubts. Following Yoshida Kenkō’s advice, we should take the perspectives and concerns of the time seriously. We need to tread a middle ground between trusting everything to be true and dismissing everything as nonsense. This is both a complicated matter and a simple statement, but that is not a problem. It is a path forward towards new ideas and new perspectives. 303 Abbreviations T. Taishō Shinshū Daizōkyō 大正新脩大藏經, 100 vols. Edited by Takakusu Junjirō 高楠順次郎, Watanabe Kaigyoku 渡辺海旭, et al., Tokyo: Taishō issaikyō kankōka i, 1924– 1932. Bibliography Abe Mika 阿部美香. "[Shiryō] Tokuyūji shozō ‘Yūzu dainenbutsu engi’ ‘Yūzu dainenbutsu kamegane engi:’ Kaidai to eiin, honkoku 〔資料〕徳融寺所蔵『融通大念仏縁起』『融通大念仏亀鐘縁起』：解題と影印・翻刻." Gakuen 925 (2017): 151-226. Abe, Ryuichi. "Mantra, Hinin, and the Feminine: On the Salvational Strategies of Myōe and Eizon." Cahiers d'Extrême-Asie 13, no. 1 (2002): 101-25. ———. "Revisiting the Dragon Princess: Her Role in Medieval Engi Stories and Their Implications in Reading the Lotus Sutra." Japanese Journal of Religious Studies 42, no. 1 (2015): 27- 70. Abe Yasurō 阿部泰郎. "Hachiman engi to Chūsei Nihongi: "Yurikawa Daijin" no sekai kara 八幡縁起と中世日本紀：「百合若大臣」の世界から." Gendai shisō 20, no. 4 (1992): p76-87. ———. "Junreiki toshite no 'Ippen hijiri-e': 'Eden' ni okeru engi no isō 巡礼記としての『一遍聖絵』：『聖絵』における縁起の位相." Junreiki kenkyū 6 (2009): 33-46. ———. "Monkan chosaku seikyō no sai hakken: Sanzon gō gyōhō no tekisuto fuchi to sono isō 文観著作聖教の再発見－三尊合行法のテクスト布置とその位相." Hikaku jinbungaku nenpō 6 (2009): 117-32. ———. "Ruiju kigenshō : Kaidai 類聚既験抄：解題." In Chūsei shōdō shiryōshū 中世唱導資料集, edited by Okabe Kaien 岡部快圓, Matsuno Yō ichi 松野陽一, Abe Yasurō 阿部泰郎 and Yamasaki Makoto 山崎誠, 324-40. Kyoto: Rinsen Shoten, 2000. ———. "Ryōhen Nihon shoki chūshaku no yōsō: Gakumon no gendan kara ‘monoga tari’ toshite no <Nihongi> e 良遍「日本書紀」注釈の様相：学問の言談から" 物語" としての< 日本紀> へ." Kokugo to kokubungaku 71, no. 11 (1994): 67-78. ———. "Shintō as Written Representation: The Phases and Shifts of Medieval Shintō Texts." Cahiers d'Extrême-Asie 16 (2006): 91-119. ———. "Shōb ōrinzō 「正法輪蔵」." Kokubungaku kaishaku to kanshō 54, no. 10 (1989): 92-98. 304 Adolphson, Mikael S. "Fukuhara." In Lovable Losers: The Heike in Action and Memory, edited by Mikael S. Adolphson and Anne Commons, 17-38. Honolulu: University of Hawai'i Press, 2015. ———. The Gates of Power: Monks, Courtiers, and Warriors in Premodern Japan. Honolulu: University of Hawai`i Press, 2000. Adolphson, Mikael S., and Anne Commons. "Blurring the Lines: Repositioning the Heike." In Lovable Losers: The Heike in Action and Memory, edited by Mikael S. Adolphson and Anne Commons, 1-14. Honolulu: University of Hawai'i Press, 2015. Adolphson, Mikael S., Edward Kamens, and Stacie Matsumoto, eds. Heian Japan, Centers and Peripheries. Honolulu: University of Hawai'i Press, 2007. Aida Shintarō 相田二郎. Mōko shūrai no kenkyū 蒙古襲来の硏究. Zō hoban ed. Tokyo: Yoshikawa Kō bunkan, 1982. Amino Yoshihiko 網野善彦. Mōko shūrai 蒙古襲来. Amino Yoshihiko chosakushū 網野善彦著作集. Edited by Inaba Nobumichi 稲葉伸道. 19 vols. Vol. 5, Tokyo: Iwanami Shoten, 2007. Andreeva, Anna. Assembling Shinto: Buddhist Approaches to Kami Worship in Medieval Japan. Cambridge, MA: Harvard University Asia Center, 2017. Andrei, Talia. "Mapping Sacred Spaces: Representations of Pleasure and Worship in Sankei Mandara." Ph.D. diss., Columbia University, 2016. Aoki Kigen 青木紀元. "'Ama no mikahi' 'Ama no mikawa': Izumo kuni no miyatsuko no kamuyogoto no mondai ten 「天乃美賀秘」「天乃瓸和」-- 出雲国造神賀詞の問題点." In Shintō oyobi shintōshi: Nishida Nagao Hakushi tsuitō ronbunshū 神道及び神道史：西田長男博士追悼論文集, 117-32. Tokyo: Meicho Fukyūkai, 1987. Appadurai, Arjun. "The Production of Locality." In Counterworks: Managing the Diversity of Knowledge, edited by Richard Fardon, 208-29. New York: Routledge, 1995. Appleton, Naomi. Jataka Stories in Theravada Buddhism: Narrating the Bodhisatta Path. Burlington, VT: Ashgate Publishing, 2010. Arai Daisuke 新井大祐. "Chūsei kōki ni okeru Yoshida ke no jinja kenkyū to Engishiki ‘Jinmyōchō:’ Bonshun jihitsu Shoshinki wo tsūro toshite 中世後期における吉田家の神社研究と『延喜式』「神名帳」：梵舜自筆『諸神記』を通路として." In Chūsei shinwa to jingi, Shintō sekai 中世神話と神祇・神道世界, edited by Itō Satoshi 伊藤聡, 464-94. Tokyo: Chikurinsha, 2011. Araki Hiroyuki 荒木博之, ed. Nihon densetsu taikei 日本伝說大系. 17 vols. Vol. 10. Tokyo: Mizuumi Shobō , 1987. 305 Araki Hisashi 荒木尚 ed. Kōra tamadarenomiya shinpisho dō shihai 高良玉垂宮神秘書同紙背. Kurume: Kō ra Taisha, 1972. Ariga Natsuki 有賀夏紀. "'Shintōshū' ni okeru Kasuga honji setsu no keisei 『神道集』における春日本地説の形成." Denshō bungaku kenkyū, no. 58 (2009): 26-35. Asahara Yoshiko 麻原美子, and Nanami Hiroaki 名波弘彰, eds. Nagato-bon Heike monogatari no sōgō kenkyū: Kōchū hen 長門本平家物語の総合研究：校注篇. 3 vols. Vol. 1. Tokyo: Benseisha, 1998. Asahara Yoshiko 麻原美子, Haruta Akira 春田宣, and Matsuo Ashie 松尾葦江, eds. Heike monogatari: Yashirobon, Kōyabon taishō 平家物語 : 屋代本・高野本対照. 3 vols. Vol. 1. Tokyo: Shintensha, 1990. Asano Hideo 浅野日出男, Sugimoto Yoshinobu 杉本好伸, and Shimada Daisuke 島田大助. "Chūshaku: "Itsukushima Daimyō jin go engi" sono ichi 注釈：『厳島大明神御縁記』その一." In Kokubungaku nenjibetsu ronbunshū: Chūsei 国文学年次別論文集：中世, edited by Gakujutsu Bunken Fukyūkai 学術文献普及会, 552-64. Tokyo: Hō bun Shuppan, 1998. Aston, W. G. Nihongi: Chronicles of Japan From the Earliest Times to A.D. 697. 2 vols. Vol. 1, London: Pub. for the Society by K. Paul, Trench, Trübner & col., limited, 1896. "Atsuta kō shiki 熱田講式." In Chūsei Nihongishū 中世日本紀集, edited by Okabe Kaien 岡部快圓, Matsuno Yō ichi 松野陽一, Abe Yasurō 阿部泰郎 and Yamasaki Makoto 山崎誠. Shinpukuji zenpon sō kan, 363-72. Kyoto: Rinsen Shoten, 1999. "Atsutagū hishaku kenmon 熱田宮秘釈見聞." In Chūsei Nihongishū 中世日本紀集, edited by Okabe Kaien 岡部快圓, Matsuno Yō ichi 松野陽一, Abe Yasurō 阿部泰郎 and Yamasaki Makoto 山崎誠. Shinpukuji zenpon sō kan, 357-59. Kyoto: Rinsen Shoten, 1999. Barad, Karen Michelle. Meeting the Universe Halfway: Quantum Physics and the Entanglement of Matter and Meaning. Durham, NC: Duke University Press, 2007. Barthes, Roland. "The Grain of the Voice." Translated by Stephen Heath. In Image, Music, Text, 179-89. London: Fontana Press, 1977. "Ben'ichisan ki 宀一山記." In Zoku gunsho ruijū. Shake bu 續群書類從.釈家部, edited by Hanawa Hokinoichi 塙保己一 and Oota Toshiro 太田藤四郎, Vol. 27 (ge), 296-98. Tokyo: Zoku Gunsho Ruijū Kanseikai, 1984. "Benzaiten hō 弁財天法." Zentsūji: National Institute of Japanese Literature. 306 Bialock, David T. Eccentric Spaces, Hidden Histories: Narrative, Ritual, and Royal Authority from the Chronicles of Japan to The Tale of the Heike. Stanford, CA: Stanford University Press, 2007. ———. "Nation and Epic: The Tale of the Heike as Modern Classic." In Inventing the Classics: Modernity, National Identity, and Japanese Literature, edited by Haruo Shirane and Tomi Suzuki, 151-78. Stanford, CA: Stanford University Press, 2000. ———. "The Tale of the Heike." In The Cambridge History of Japanese Literature, edited by Haruo Shirane, Tomi Suzuki and David Barnett Lurie, 295-305. Cambridge, United Kingdom: Cambridge University Press, 2016. Blair, Heather. Real and Imagined: The Peak of Gold in Heian Japan. Cambridge, MA: Harvard University Asia Center, 2015. ———. "Rites and Rule: Kiyomori at Itsukushima and Fukuhara." Harvard Journal of Asiatic Studies 73, no. 1 (2013): 1-42. Blair, Heather, and Kawasaki Tsuyoshi 川崎剛志. "Editors' Introduction: Engi: Forging Accounts of Sacred Origins." Japanese Journal of Religious Studies 42, no. 1 (2015): 1- 26. Blum, Mark L. "The Sangoku-Mappō Construct: Buddhism, Nationalism, and History in Medieval Japan." In Discourse and Ideology in Medieval Japanese Buddhism, edited by Richard Karl Payne and Taigen Daniel Leighton, 31-51. New York: Routledge, 2006. Bodiford, William M. "When Secrecy Ends: The Tokugawa Reformation of Tendai Buddhism and its Implications." In The Culture of Secrecy in Japanese Religion, edited by Bernhard Scheid and Mark Teeuwen, 309-30. London: Routledge, 2006. Bonshun 梵舜. Shunkyūki 舜旧記. Shiryō sanshū 史料纂集. Edited by Kamata Jun'ichi 鎌田純一. 8 vols. Tokyo: Zoku Gunsho Ruijū Kanseikai, 1970. Brock, Karen L. "The Shogun's ‘Painting Match’." Monumenta Nipponica 50, no. 4 (1995): 433- 84. Brockington, John. "The Epics in the Bhakti Tradition." In The Intimate Other: Love Divine in Indic Religions, edited by Anna S. King and Brockington John, 31-53. Hyderabad: Orient Longman, 2005. Cali, Joseph, John Dougill, and Geoff Ciotti, eds. Shinto Shrines: A Guide to the Sacred Sites of Japan's Ancient Religion. Honolulu: University of Hawaiʻi Press, 2013. Campany, Robert Ford. "On the Very Idea of Religions (In the Modern West and in Early Medieval China)." History of Religions 42, no. 4 (2003): 287-319. Castells, Manuel. The Rise of the Network Society. Malden, MA: Blackwell Publishers, 1996. 307 Chaniotis, Angelos. "Travelling Memories in the Hellenistic World." In Wandering Poets in Ancient Greek Culture: Travel, Locality and Pan-Hellenism, edited by Richard Hunter and Ian Rutherford, 249-69. New York: Cambridge University Press, 2009. Chikamoto,Kensuke 近本謙介. "Haimetsu kara no saisei: Nanto ni okeru chūsei no tōrai 廃滅からの再生：南都における中世の到来." Nihon bungaku 49, no. 7 (2000 2000): 29-38. "Chikubushima engi 竹生島縁起." In Gunsho ruijū. Jingi bu 群書類従.神祇部, edited by Hanawa Hokinoichi 塙保己一, Vol. 2, 310-14. Tokyo: Gunsho Ruijū Kankō kai Kantō sha Nai, 1952. "Chōjō Dainichi Nyorai ryaku engi 頂上大日如来略縁記 ". In Ryakuengi shūsei 略緣起集成, edited by Nakano Takeshi 中野猛, 6 vols. Vol. 3, 267-70. Tokyo: Benseisha, 1997. Chūgan Engetsu 中巌円月. "Tōkai ichiōshū 東海一漚集." In Gozan bungaku shinshū 五山文學新集, edited by Tamamura Takeji 玉村竹二, 8 vols. Vol. 4, 315-578. Tokyo: Tō kyō Daigaku Shuppankai, 1970. Chūsei setsuwa, emaki kenkyūshitsu 中世説話・絵巻研究会. "Honkoku: Itsukushima go engi (Itsukushima Jinja zō) 翻刻：厳島御縁起（厳島神社蔵）." Komazawa kokubun 17 (1980): 189-207. Conlan, Thomas. In Little Need of Divine Intervention: Takezaki Suenaga's Scrolls of the Mongol Invasions of Japan. Ithaca, NY: East Asia Program, Cornell University, 2001. ———. State of War: The Violent Order of Fourteenth-Century Japan. Ann Arbor: Center for Japanese Studies, University of Michigan, 2003. "Daiganji monjo 大願寺文書." In Hiroshima kenshi: Kodai chūsei shiryō hen 広島県史. 古代中世資料編, edited by Hiroshima-ken 広島県, 5 vols. Vol. 3, 1155-464. Hiroshima: Hiroshima, 1978. Daigoji engi 醍醐寺縁起." In Dai Nihon bukkyō zensho. Jishi sōsho daiichi 大日本佛教全書 . 寺誌叢書第一. Vol. 117, 246-50. Tokyo: Daihō rinkaku, 2007. Deal, William E., and Brian Douglas Ruppert. A Cultural History of Japanese Buddhism. Malden, MA: Wiley-Blackwell, 2015. De Vito, Christian G., and Anne Gerritsen. "Micro-Spatial Histories of Labour: Towards a New Global History." In Micro-Spatial Histories of Global Labour, edited by Christian G. De Vito and Anne Gerritsen, 1-28. Cham, Switzerland: Palgrave Macmillan, 2018. Doniger, Wendy. "Post-Modern and -Colonial -Structural Comparisons." In A Magic Still Dwells: Comparative Religion in the Postmodern Age, edited by Kimberley C. Patton and Ray Benjamin C. Berkeley, CA: University of California Press, 2000. 308 Eco, Umberto. "Casablanca: Cult Movies and Intertextual Collage." In Travels in Hyper Reality: Essays, 197-212. San Diego: Harcourt Brace & Company, 1986. Enchū 圓忠. "Suwa Daimyō jin ekotoba 諏訪大明神絵詞." In Zoku Gunsho ruijū. Vol. 3 (ge), 494-539. Tokyo: Zoku Gunsho Ruijū Kanseikai, 1934. Eubanks, Charlotte. Miracles of Book and Body: Buddhist Textual Culture and Medieval Japan. Berkeley: University of California Press, 2011. Faure, Bernard. "The Cultic World of the Blind Monks: Benzaiten, Jūzenji, and Shukujin." Journal of Religion in Japan 2 (2013): 171-94. ———. Gods of Medieval Japan: Volume 1, The Fluid Pantheon. Honolulu: University of Hawai`i Press, 2016. ———. Gods of Medieval Japan: Volume 2, Protectors and Predators. Honolulu: University of Hawai`i Press, 2016. Felski, Rita. The Limits of Critique. Chicago: University of Chicago Press, 2015. Fister, Pat, ed. Ama monzeki jiin no sekai: Kōjotachi no shinkō to gosho bunka 尼門跡寺院の世界 : 皇女たちの信仰と御所文化 / A Hidden Treasure: Treasures of the Japanese Imperial Convents. Tokyo: Sankei Shinbunsha, 2009. Fremerman, Sarah Alizah. "Divine Impersonations: Nyoirin Kannon in Medieval Japan." Ph.D. diss, Stanford University, 2008. Fujimaki Kazuhiro 藤卷和宏. "Hase no ryūketsu to <Nyoi hōju>: Hasedera engi no tenkai, ‘Benichisan ’ wo meguru gensetsu gun to no kōsa 初瀬の龍穴と < 如意宝珠> ：長谷寺縁起の展開・「宀一山」をめぐる言説群との交差." Kokubungaku kenkyū 130 (2000): 102-12. ———. "Seichi no chika ni ha: Nihon chūsei no hōju, shari shinkō no hitosumi yori 聖地の地下には：日本中世の宝珠・舎利信仰の一隅より." In Seichi to seijin no tōzai: Kigen wa ikani katarareru ka 聖地と聖人の東西 : 起源はいかに語られるか, edited by Fujimaki Kazuhiro 藤卷和宏, 12-32. Tokyo: Bensei Shuppan, 2011. Fujiwara Shigeo 藤原重雄 "Hanmachi Shigeo shi shozō "Keiranshūyōshū " Benzaiten bu (Honkoku, kaidai) 反町茂雄氏所蔵『渓嵐拾葉集』弁才天部（翻刻・解題）." In Kamigami no sugata, katachi o meguru tamenteki kenkyū 神々のすがた・かたちをめぐる多面的研究, edited by Shimane-ken Kodai Bunka Sentā , 14-17. Matsue-shi: Shimane-ken Kodai Bunka Sentā , 2011. 309 "Fukuōji monjo 福王寺文書." In Hiroshima kenshi: Kodai chūsei shiryō hen 広島県史. 古代中世資料編, edited by Hiroshima-ken 広島県, 5 vols. Vol. 4, 135-48. Hiroshima: Hiroshima. Fukushima Kaneharu 福島金治. Kanazawa Hōjō-shi to Shōmyōji 金沢北条氏と称名寺. Tokyo: Yoshikawa Kō bunkan, 1997. ———. "Kanazawa Shōmyōji to Ise, Chinzei 金沢称名寺と伊勢・鎮西." In Bijutsushi ronsō zōkei to bunka 美術史論叢造形と文化, edited by Shimizu Mazumi 清水眞澄. 256-73. Tokyo: Yūzankaku Shuppan, 2000. Funata Jun'ichi 舩田淳一. "Chūsei Eizan no kairitsu fukkō: Rissō Ejin no shisō to kokka -kan wo megutte 中世叡山の戒律復興 : 律層恵尋の思想と国家観をめぐって." Bukkyō daigaku sōgō kenkyūjo kiyō 16 (2009): 359-80. Furukawa Koshōken 古川古松軒. "Saiyū zakki 西遊雑記." In Nihon shomin seikatsu shiryō shūsei 日本庶民生活史料集成, edited by Miyamoto Tsuneichi 宮本常一, Haraguchi Torao 原口虎雄 and Higa Shunchō 比嘉春潮, 31 vols. Vol. 2, 329-96. Tokyo: San'ichi Shobō , 1969. ———. "Tōyū zakki 東遊雑記." In Nihon shomin seikatsu shiryō shūsei 日本庶民生活史料集成, edited by Miyamoto Tsuneichi 宮本常一, Haraguchi Torao 原口虎雄 and Higa Shunchō 比嘉春潮, 31 vols. Vol. 3, 439-594. Tokyo: San'ichi Shobō , 1969. Glassman, Hank "The Tale of Mokuren: A Translation of Mokuren-no-sōsh i." Journal of Buddhist Literature 1 (1999): 120-61. "Go hanmotsu chō 御判物帖." In Hiroshima kenshi: Kodai chūsei shiryō hen 広島県史. 古代中世資料編, edited by Hiroshima-ken 広島県, 5 vols. Vol. 3. Hiroshima: Hiroshima, 1978. Gombrich, Richard. "The Vessantara Jātaka, the Rāmāya ṇa and the Dasaratha Jātaka." Journal of the American Oriental Society 105, no. 3 (1985): 427-37. Gorai Shigeru 五来重. Jisha engi to denshō bunka 寺社縁起と伝承文化. Gorai Shigeru chosakushū 五来重著作集. Vol. 4, Kyoto: Hō zō kan, 2008. Grapard, Allan G. "Flying Mountains and Walkers of Emptiness: Toward a Definition of Sacred Space in Japanese Religions." History of Religions 21, no. 3 (1982): 195-221. ———. "Institution, Ritual, and Ideology: The Twenty-Two Shrine-Temple Multiplexes of Heian Japan." History of Religions 27, no. 3 (1988): 246-69. ———. "Keiranshūyōshū: A Different Perspective On Mt. Hiei in the Medieval Period." In Re- visioning "Kamakura" Buddhism, edited by Richard Karl Payne, 55-70. Honolulu: University of Hawai'i Press, 1998. 310 ———. "Linguistic Cubism: A Singularity of Pluralism in the Sannō Cult." Japanese Journal of Religious Studies 14, no. 2/3 (1987): 211-34. ———. The Protocol of the Gods: A Study of the Kasuga Cult in Japanese History. Berkeley: University of California Press, 1992. Gregory, Derek. Geographical Imaginations. Cambridge, MA: Blackwell, 1994. Grumbach, Lisa. "Sacrifice and Salvation in Medieval Japan: Hunting and Meat in Religious Practice at Suwa Jinja." Ph.D., Stanford University, 2005. Gummer, Natalie D. "Listening to the Dharmabhā ṇaka: The Buddhist Preacher in and of the Sūtra of Utmost Golden Radiance." Journal of the American Academy of Religion 80, no. 1 (2012): 137-60. Gunji, Naoko. "Horrified Victors: Spirit Pacification of Heike Losers." In Lovable Losers: The Heike in Action and Memory, edited by Mikael S. Adolphson and Anne Commons, 166- 81. Honolulu: University of Hawai'i Press, 2015. ———. "Morihisa and the Cult of Kannon at Kiyomizudera." In Like Clouds or Mists: Studies and Translations of Nō Plays of the Genpei War, edited by Elizabeth Oyler and Michael Geoffrey Watson, 309-28. Ithaca, NY: East Asia Program, Cornell University, 2013. ———. "The Ritual Narration of Mortuary Art: The "Illustrated Story of Emperor Antoku" and Its "Etoki" at Amidaji." Japanese Journal of Religious Studies 40, no. 2 (2013): 203-45. "Hachiman gudō kun (kō ) 八幡愚童訓（甲)." In Jisha engi 寺社緣起, edited by Sakurai Tokutarō 桜井徳太郎, Hagiwara Tatsuo 萩原龍夫 and Miyata Noboru 宮田登, 169-206. Tokyo: Iwanami Shoten, 1975. "Hachiman gudō kun (otsu) 八幡愚童訓 ( 乙)." In Jisha engi 寺社緣起, edited by Sakurai Tokutarō 桜井徳太郎, Hagiwara Tatsuo 萩原龍夫 and Miyata Noboru 宮田登, 207-73. Tokyo: Iwanami Shoten, 1975. Hanazono Tennō 花園天皇. Hanazono Tennō shinki 花園天皇宸記. Shiryō sanshū. Kokiroku hen 史料纂集. 古記錄編 edited by Murata Masashi 村田正志 3 vols. Vol. 1, Tokyo: Zoku Gunsho Ruijū Kanseikai, 1982. Harada Yoshiko 原田佳子. Itsukushima no sairei to geinō no kenkyū 厳島の祭礼と芸能の研究. Tokyo: Fuyō Shobō Shuppan, 2010. Haraway, Donna Jeanne. "A Manifesto for Cyborgs: Science, Technology, and Socialist Feminism in the 1980s." In The Haraway Reader, 7-46. New York: Routledge, 2004. Hardacre, Helen. Shinto: A History. New York: Oxford University Press, 2017. 311 Hasegawa Seiichi 長谷川成一. "Nihon sankei no seiritsu to meisho kan no tenkai 日本三景の成立と名所観の展開." In Nihon sankei e no izanai: Matsushima, Amanohashidate, Itsukushima 日本三景への誘い : 松島・天橋立・厳島, edited by Shimao Arata 島尾新 and Hasegawa Seiichi 長谷川成一, 23-42. Osaka: Seibundō Shuppan, 2007. Hasegawa Tadashi 長谷川端, ed. Taiheiki 太平記. 4 vols., Shinpen Nihon koten bungaku zenshū, vol. 54-57. Tokyo: Shō gakkan, 1995. Hashimoto Akihiko 橋本章彦. "Engi gaku he no shō tai 縁起学への招待." In Yūraku to shinkō no bunkagaku 遊楽と信仰の文化学, edited by Tsutsumi Kunihiko 堤邦彦 and Tokuda Kazuo 徳田和夫, 7-26. Tokyo: Shinwasha, 2010. Hatsukaichi chōshi: Shiryō hen 廿日市町史：資料編. 5 vols. Hatsukaichi-chō : Hatsukaichi- chō , 1975. Hatsukaichi chōshi: Tsūshi hen 廿日市町史. 通史編. 2 vols. Hatsukaichi-chō : Hatsukaichi-chō , 1988. Hattori Seidō 服部清道. "'Enoshima engi' kō 『江島縁起』考." Yokohama shōdai ronshū 10, no. 1 (1977): 122-75. Healy, Kieran. "Fuck Nuance." Sociological Theory 35, no. 2 (2017): 118-27. "Hikosan rūki 彦山流記." In Shugendō shiryōshū 修験道史料集, edited by Gorai Shigeru 五来重, 2 vols. Vol. 2, 463-73. Tokyo: Meicho Shuppan, 1983. Hirano Tae 平野多恵. Myōe: Waka to Bukkyō no sōkoku 明恵 : 和歌と仏教の相克. Tokyo: Kasama Shoin, 2011. Hiroshima-ken Jinjachō 広島県神社庁, ed. Hiroshima-ken jinjashi 広島県神社誌. Hiroshima: Hiroshima-ken Jinjachō , 1994. Hiroshima kenshi: Chūsei (Tsūshi hen) 広島県史.中世（通史編）. 7 vols. Vol. 2, Hiroshima: Hiroshima-ken, 1984. Hiroshima minzoku no kai 広島民俗の会. "Ichikishima hime to Itsukushima jinja 市杵島姫と厳島神社." Hiroshima-ken minzoku shiryō 6 (1973): 49-71. "Hoki naiden 簠簋内伝." In Zoku gunsho ruijū. Zatsu bu 續群書類從. 雑部, edited by Hanawa Hokinoichi 塙保己一 and Oota Toshiro 太田藤四郎, Vol. 31 (jo), 374-414. Tokyo: Zoku Gunsho Ruijū Kanseikai, 1984. Holt, John. Spirits of the Place: Buddhism and Lao Religious Culture. Honolulu: University of Hawai'i Press, 2009. 312 Hori, Kyotsu. "The Mongol Invasions and the Kamakura Bakufu." Ph.D. diss., Columbia University, 1967. "Hōrya ku kanki 保暦間記." In Gunsho ruijū. Zatsu bu 續群書類從. 雑部, edited by Hanawa Hokinoichi 塙保己一, Vol. 2, 120-79. Tokyo: Gunsho Ruijū Kankō kai Kantō sha Nai, 1960. Hyō dō Hiromi 兵藤裕己. Ō ken to monogatari 王権と物語. Tokyo: Seikyūsha, 1989. Ichijō Kanera 一條兼良. Nihon shoki sanso 日本書紀纂疏, edited by Kiyohara Nobukata 淸原宣賢. Tokyo: Seisaku hatsubai Yagi Shoten, 1977. Ichiko Teiji 市古貞次. Chūsei shōsetsu to sono shūhen 中世小說とその周辺. Tokyo: Tō kyō Daigaku Shuppankai, 1981. ———, ed. Genpei jōsuiki 源平盛衰記. 4 vols., Chūsei no bungaku 中世の文学. Tokyo: Miyai Shoten, 1991. ———, ed. Heike monogatari 平家物語. 2 vols., Shinpen Nihon koten bungaku zenshū, vol. 45-46. Tokyo: Shō gakkan, 1994. ———, ed. Jinkeishō 塵荊鈔. 2 vols. Tokyo: Koten Bunko, 1984. Ichimura Yoshikazu 市村義一, ed. Tenkawa Benzaiten kanjō: Hashio Mitsushima Benzaiten: Zuimukishō 天河弁財天勧請 : 箸尾満嶋弁財天 : 瑞夢記抄. Kōryō -chō, Nara Prefecture: Hashio mitsushima Benzaiten kenshō kai, 1979. Ihara Kesao 井原今朝男. "Chūsei no kokuga jisha taisei to minshū tōgō girei 中世の国街寺社体制と民衆統合儀礼." In Chūsei ichinomiyasei no rekishiteki tenkai. ge (sōgō kenkyū hen) 中世一宮制の歴史的展開. 下( 総合研究編), edited by Ichinomiya kenkyūkai, 209- 58. Tokyo: Iwata Shoin, 2004. Imagawa Ryō shun 今川了俊. "Michiyuki buri 道行きぶり." In Chūsei nikki kikōshū 中世日記紀行集, edited by Nagasaki Ken 長崎健. Shinpen nihon koten bungaku zenshu 新編日本古典文学全集, 389-426. Tokyo: Shōgakkan, 1994. Inaba Nobumichi 稲葉伸道. "Chūsei no kokka to jisha: Ōchō to bakufu no jisha seisaku 中世の国家と寺社：王朝と幕府の寺社政策." Nenpō chūsei shi kenkyū 28 (2003): 15-28. Inada Toshinori 稲田利徳. "Imagawa Ryōshu n 'Michiyuki buri' chūshaku (3) 今川了俊「道行きぶり」注釈（3）." Okayama daigaku kyōiku gakubu kenkyū shūroku, no. 91 (1992): 1-16. Inose Chihiro 猪瀬千尋. Chūsei ōken no ongaku to girei 中世王権の音楽と儀礼. Tokyo: Kasama Shoin, 2018. 313 "Ishiyamadera engi 石山寺縁起." In Zoku gunsho ruijū. Shake bu 續群書類從.釈家部, edited by Hanawa Hokinoichi 塙保己一 and Oota Toshiro 太田藤四郎, Vol. 28 (jo), 95-118. Tokyo: Zoku Gunsho Ruijū Kanseikai, 1984. Itō Kunihiko 伊藤邦彦. "Kamakura bakufu 'ikoku kōfuku' kitō to Ichinomiya: Shugo seido to no kankei wo chūshin ni 鎌倉幕府「異国降伏」祈禱と一宮：守護制度との関係を中心に." In Chūsei ichinomiyasei no rekishiteki tenkai. ge (sōgō kenkyū hen) 中世一宮制の歴史的展開. 下( 総合研究編), edited by Ichinomiya kenkyūkai, 259-89. Tokyo: Iwata Shoin, 2004. Itō Satoshi 伊藤聡. Chūsei Tenshō Daijin shinkō no kenkyū 中世天照大神信仰の研究. Kyoto: Hō zō kan, 2011. ———. Shintō no keisei to chūsei shinwa 神道の形成と中世神話. Tokyo: Yoshikawa Kō bunkan, 2016. ———. "Yoshida bunko shozō no Benzaiten kankei gikyō ni tsuite: Sono honkoku to shōkai 吉田文庫所蔵の弁才天関偽経について：その翻刻と紹介." Muromachi 2 (1993): 40-46. "Itsukushima (zanketsu bon) 厳島（残欠本）." In Nihon shomin bunka shiryō shūsei: Dengaku, sarugaku 日本庶民文化史料集成：田楽・猿楽, 16 vols. Vol. 2, 581-83. Tokyo: San'ichi Shobō , 1973. "Itsukushima Benzaitensha kakitsuke 厳島弁才天社書付." National Institute of Japanese Literature: Ogata Tamotsu Collection 大方保. Itsukushima kenkyū: Hiroshima daigaku sekai isan Itsukushima naikai no rekishi to bunka purojekuto kenkyū senta kenkyū seika hōkokusho 厳島研究 : 広島大学世界遺産・厳島- 内海の歴史と文化プロジェクト研究センター研究成果報告書. Hiroshima: Sekai isan Itsukushima naikai no rekishi to bunka purojekuto kenkyū senta, 2005-. "Itsukushima no honji 厳島の本地." In Muromachi jidai monogatari taisei 室町時代物語大成, edited by Yokoyama Shigeru 横山重 and Matsumoto Ryūshin 松本隆信, 13 vols. Vol. 2, 270-96. Tokyo: Kadokawa Shoten, 1973. "Itsukushima Nosaka monjo 厳島野坂文書." In Hiroshima kenshi: Kodai chūsei shiryō hen 広島県史. 古代中世資料編, edited by Hiroshima-ken 広島県, 5 vols. Vol. 2. Hiroshima: Hiroshima, 1978. "Itsukushima shake engi 厳島社家縁起." In Yoshida bunko: Tenri University. "Itsukushima zue 厳島図絵." In Chūgoku no maki 中国の卷, edited by Hase Akihisa 長谷章久 and Ikeda Yasaburō 池田弥三郎. Nihon meisho fūzoku zue, 166-411. Tokyo: Kadokawa Shoten, 1980. 314 "Itsukushima 厳島." In Nihon shomin bunka shiryō shūsei: Dengaku, sarugaku 日本庶民文化史料集成：田楽・猿楽, 16 vols. Vol. 2, 565-68. Tokyo: San'ichi Shobō , 1973. "Itsukushima: Honkoku, gendaigo yaku いつくしま：翻刻・現代語訳." In Itsukushima shinkō jiten 厳島信仰事典, edited by Nozaka Motoyoshi 野坂元良, 452-77. Tokyo: Ebisu Kōshō Shuppan, 2002. Itsukushima jinja 厳島神社. Itsukushima 伊都岐島. Miyajima-chō : Itsukushima Jinja Shamusho, 1976. "Iwashimizu Hachimangū masshaki 石清水八幡宮末社記." In Zoku gunsho ruijū. Jingi bu 續群書類從.釈家部.神祇部, edited by Hanawa Hokinoichi 塙保己一 and Oota Toshiro 太田藤四郎, Vol. 2 (jō ), 115-23. Tokyo: Zoku Gunsho Ruijū Kanseikai, 1957. Iwashimizu Hachimangū 石清水八幡宮, ed. Iwashimizu Hachimangū shiryō sōsho 石清水八幡宮史料叢書. 5 vols. Hachiman-cho: Iwashimizu Hachimangū Shamusho, 1960-1975. "Iwashimizu jingenshō 石清水尋源鈔." Yamato bunkakan: National Institute of Japanese Literature. Jenkins, Henry. Convergence Culture: Where Old and New Media Collide. New York: New York University Press, 2006. Jien 慈円. Gukanshō 愚管抄. Nihon koten bungaku taikei 86. Edited by Okami Masao 岡見正雄 and Akamatsu Toshihide 赤松俊秀. Tokyo: Iwanami Shoten, 1967. Jihen 慈遍. "Tenchi jingi shinchi yōki 天地神祇審鎮要記." In Tendai shintō. jō 天台神道. 上 edited by Sueki Fumihiko 末木文美士, Mizukami Fumiyoshi 水上文義 and Satō Masato 佐藤真人, 403-74. Tokyo: Shintō Taikei Hensankai, 1990. "Jimon denki horoku 寺門伝記補録." In Dai Nihon bukkyō zensho 大日本仏教全書. Vol. 127, 115-454. Tokyo: Daihō rinkaku, 2007. "Jingi hishō 神祇秘抄." In Chūsei Nihongishū 中世日本紀集, edited by Okabe Kaien 岡部快圓, Matsuno Yō ichi 松野陽一, Abe Yasurō 阿部泰郎 and Yamasaki Makoto 山崎誠. Shinpukuji zenpon sō kan, 375-401. Kyoto: Rinsen Shoten, 1999. Johnston, Michael. "Copying and Reading The Prick of Conscience in Late Medieval England." Speculum 95, no. 3 (2020): 742-801. Jōkaku 上覚. "Waka iroha 和歌色葉." In Nihon kagaku taikei 日本歌学大系, edited by Sasaki Nobutsuna 佐佐木信綱, 10 vols. Vol. 3. Tokyo: Kazama Shobō , 1974. 315 "Jō ruriji ruki no koto 浄瑠璃寺流記事." In Dai Nihon Bukkyō zensho. Jishi bu 大日本仏教全書.寺誌部, edited by Suzuki Gakujutsu Zaidan 鈴木学術財団, 336-40. Tokyo: Kōdansha, 1972. Josephson, Jason Ananda. The Invention of Religion in Japan. Chicago: University of Chicago Press, 2012. Juneja, Monica. "Architectural Memory between Representation and Practice: Rethinking Pierre Nora’s Les lieux de mémoire." In Memory, History and Colonialism: Engaging with Pierre Nora in Colonial and Postcolonial Contexts, edited by Indra Sengupta, 11-36. London: German Historical Institute, 2009. Kadoshige Hajime 角重始. "Kamakura, Nanbōkucho ki Itsukushima sha shihai no tokushitsu 鎌倉・南北朝期厳島社支配の特質." In Itsukushima shinkō jiten 厳島信仰事典, edited by Motoyoshi 野坂元良 Nozaka, 256-77. Tokyo: Ebisu Kōshō Shuppan, 2002. Kagawa Masanori 香川正矩. Intoku taiheiki : Shōtoku ni-nen hanpon 陰徳太平記 : 正徳二年板本. Edited by Yonehara Masayoshi 米原正義. 6 vols. Tokyo: Tō yō Shoin, 1980. "Kaidai: Itsukushima no engi 解題：いつくしまのゑんぎ." In Muromachi jidai monogatari taisei 室町時代物語大成, edited by Yokoyama Shigeru 横山重 and Matsumoto Ryūshin 松本隆信, 13 vols. Vol. 2, 377-85. Tokyo: Kadokawa Shoten, 1973. Kaiei 懷英 Shinkō Kōya shunjū hennen shūroku 新挍高野春秋編年輯錄. Edited by Hinonishi Shinjō 日野西眞定. Tokyo: Meicho Shuppan, 1982. Kaizu Ichirō 海津一朗. Chūsei no henkaku to tokusei: Shinryō kōgyōhō no kenkyū 中世の変革と徳政 : 神領興行法の研究. Tokyo: Yoshikawa Kō bunkan, 1994. ———. "Ikoku kōfuku kitō taisei to shokoku ichinomiya kōgyō 異国降伏祈祷体制と諸国一宮興行." In Chūsei ichinomiyasei no rekishiteki tenkai. ge (sōgō kenkyū hen) 中世一宮制の歴史的展開. 下( 総合研究編), edited by Ichinomiya kenkyūkai, 291-325. Tokyo: Iwata Shoin, 2004. Kajihara Masaaki 梶原正昭, ed. Gikeiki 義経記, Shinpen Nihon koten bungaku zenshū, vol. 62. Tokyo: Shō gakkan, 2000. Kaminishi, Ikumi. Explaining Pictures: Buddhist Propaganda and Etoki Storytelling in Japan. Honolulu: University of Hawai`i Press, 2006. Kanai Kiyomitsu 金井清光. Ippen to Jishū kyōdan 一遍と時宗教団. Tokyo: Kadokawa Shoten, 1975. Kanazawa Bunko 金沢文庫, ed. Kanazawa Bunko no chūsei shintō shiryō 金沢文庫の中世神道資料. Yokohama: Yokohama kenritsu Kanazawa Bunko, 1996. 316 ———. ed. "Kanazawa Bunko komonjo 金澤文庫古文書." Yokohama: Kanazawa Bunko, 1956. Kanze Nagatoshi 観世長俊. "Itsukushima 厳島." In Kōchū yōkyoku sōsho 校註謡曲叢書, edited by Haga Yaichi 芳賀矢一 and Sasaki Nobutsunma 佐佐木信綱. 3 vols. Vol. 1, 160-64. Tokyo: Hakubunkan, 1914. "Kasuga gongen genki 春日権現験記." In Gunsho ruijū. Jingi bu 群書類従.神祇部, edited by Hanawa Hokinoichi 塙保己一, Vol. 2, 1-57. Tokyo: Gunsho Ruijū Kankō kai Kantō sha Nai, 1952. "Kasuga ryūjin 春日龍神." In Yōkyoku hyakuban 謠曲百番, edited by Nishino Haruo 西野春雄. Shin Nihon koten bungaku taikei, 508-13. Tokyo: Iwanami Shoten, 1998. Kawamura Minato 川村湊. Gozu Tennō to Somin Shōrai densetsu: Kesareta ishintachi 牛頭天王と蘇民将来伝說 : 消された異神たち. Tokyo: Sakuhinsha, 2007. Kawasaki Tsuyoshi 川崎剛志. "The Invention and Reception of the Mino'odera engi." Japanese Journal of Religious Studies 42, no. 1 (2015): 133-55. ———. Shugendō no Muromachi bunka 修験道の室町文化. Tokyo: Iwata Shoin, 2011. Kawazoe Shō ji 川添昭二. Chūsei Kyūshū no seiji bunkashi 中世九州の政治・文化史. Fukuoka: Kaichō sha, 2003. ———. ed. Genkō bōrui hennen shiryō: Chūkai ikoku keiko ban'yaku shiryō no kenkyū 元寇防塁編年史料 : 注解異国警固番役史料の研究. Fukuoka: Fukuoka-shi Kyō iku Iinkai, 1971. ———. "Mōko shūrai s hiryō toshite no Nichiren ibun 蒙古襲来史料としての日蓮遺文." Fukujin 15 (2011): 74-105. Kenritsu Hiroshima daigaku miyajimagaku sentā 県立広島大学宮島学センター, ed. Miyajimagaku 宮島学. Hiroshima: Keisuisha, 2014. "Kibune no honji (Keiō gijuku toshokan zō ko shahon) きふねの本地（慶応義塾図書館蔵古写本）." In Muromachi jidai monogatari taisei 室町時代物語大成, edited by Yokoyama Shigeru 横山重 and Matsumoto Ryūshin 松本隆信, 13 vols. Vol. 4, 47-68. Tokyo: Kadokawa Shoten, 1976. Kim, Sujung. Shinra Myojin and Buddhist Networks of the East Asian Mediterranean. Honolulu: University of Hawai`i Press, 2019. Kimbrough, R. Keller, trans. "Sayohime." In Wondrous Brutal Fictions: Eight Buddhist Tales from the Early Japanese Puppet Theater, 161-90. New York: Columbia University Press, 2013. 317 "Kingyōku yōshū (Naikaku bunko zō) 金玉要集( 内閣文庫蔵)." In Sonarejō: Murasame hen 磯馴帖：村雨篇, edited by Itō Masayoshi 伊藤正義, 2 vols. Vol. 2, 137-235. Osaka: Izumi Shoin, 2002. Kinoshita Fumitaka 樹下文隆 "Chūsei ni okeru Itsukushima shinkō to Hokkekyō: Nō 'Itsukushima' tanjō no haikei 中世における厳島信仰と法華経：能「厳島」誕生の背景." Itsukushima kenkyū: Hiroshima Daigaku sekai isan Itsukushima uchiumi no rekishi to bunka purojekuto kenkyū sentā kenkyū seika hōkokusho 4 (2008): 60-65. Kitabatake Chikafusa 北畠親房. A Chronicle of Gods and Sovereigns: Jinnō shōtōki of Kitabatake Chikafusa. Translated by H. Paul Varley. New York: Columbia University Press, 1980. ———. "Jinnō shōt ōki 神皇正統記." In Gunsho ruiju. Teiōbu 群書類從.帝王部, edited by Hanawa Hokinoichi 塙保己一, Vol. 3, 1-127. Tokyo: Yagi Shoten, 2013. Kitahara Yasuo 北原保雄, and Ogawa Eiichi 小川栄一, eds. Enkyō-bon Heike monogatari. Honbun hen 延慶本平家物語. 本文篇. 2 vols. Tokyo: Benseisha, 1990. Kitani Masamitsu 木谷昌光. "Itsukushima no tokushu shinji nit suite: Kanmiso kenjō shiki, oshima meguri shiki, kangensai 厳島の特殊神事について：神衣献上式･御島廻式･管絃祭." Geibi chihōshi kenkyū 207 (1997): 4-23. Klein, Susan Blakeley. "Esotericism in Noh Commentaries and Plays: Konparu Zenchiku's Meishuku shū and Kakitsubata." In The Culture of Secrecy in Japanese Religion, edited by Bernhard Scheid and Mark Teeuwen, 229-54. London: Routledge, 2006. ———. "Nō as Political Allegory: The Case of Haku Rakuten." In Like Clouds or Mists: Studies and Translations of Nō Plays of the Genpei War, edited by Elizabeth Oyler and Michael Geoffrey Watson, 399-442. Ithaca, New York: East Asia Program, Cornell University, 2013. ———. "Wild Words and Syncretic Deities: Kyō gen kigo and honji suijaku in Medieval Literary Allegoresis." In Buddhas and Kami in Japan: Honji Suijaku as a Combinatory Paradigm, edited by Mark Teeuwen and Fabio Rambelli, 177-203. New York: RoutledgeCurzon, 2003. Kōban 弘鑁. "Kōban kuzetsu 弘鑁口説." In Zoku gunsho ruijū. Shake bu 續群書類從. 釈家部, edited by Hanawa Hokinoichi 塙保己一 and Ō ta Tō shirō 太田藤四郎, Vol. 27 (jō ), 62- 83. Tokyo: Zoku Gunsho Ruijū Kanseikai, 1957. Kobayakawa Takeshi 小早川健. "'Hachiman gudōkun' to Chūsei kikōbun 『八幡愚童訓』と中世紀行文." Kōbe shiritsu kōgyō kōtō senmon gakkō kenkyū kiyōki 34 (1996): 97-102. 318 Koida Tomoko 恋田知子. "Monogatari sōshi toshite no keisei to juyō: “Oyudono no ue no nikki” wo tsūjite 物語草子としての形成と受容：『お湯殿の上の日記』を通じて." In Jisha engi no bunkagaku 寺社緣起の文化学, edited by Tsutsumi Kunihiko 堤邦彦 and Tokuda Kazuo 徳田和夫, 229-48. Tokyo: Shinwasha, 2005. Kojima Noriyuki 小島憲之, ed. Nihon shoki 日本書紀. 3 vols., Shinpen Nihon koten bungaku zenshū, vol. 2-4. Tokyo: Shōgakkan, 1994. Kojima Satoaki 小島惠昭, and Watanabe Nobukazu 渡辺信和. "(Kyōdō kenkyū) Mantokuji -zō 'Shōtoku taishi den' honkoku 〔共同研究〕万徳寺蔵『聖徳太子伝』翻刻." Dōhō Daigaku Bukkyō Bunka Kenkyūjo kiyō 2 (1980): 123-341. Kojima Tsunenari 小島常也. Itsukushima michishiba no ki 嚴島道芝記. Miyajima, Hiroshima: Miyajima-chō , 1971. Kokuritsu Rekishi Minzoku Hakubutsukan 国立歴史民俗博物館, ed. Shaji no kokuhō jūbun kenzōbutsu tō: Munafuda meibun shūsei, Chūgoku Shikoku Kyūshŭ hen 社寺の国宝重文建造物等：棟札銘文集成. 中国四国九州編. Chiba-ken Sakura-shi: Kokuritsu Rekishi Minzoku Hakubutsukan, 1993. Komatsu Kazuhiko 小松和彦. Ikai to Nihonjin: Emonogatari no sōzōryoku 異界と日本人 : 絵物語の想像力. Tokyo: Kadokawa Shoten, 2003. Komatsu Shigemi 小松茂美, and Kanzaki Mitsuharu 神崎充晴, eds. Yūzū nenbutsu engi 融通念仏縁起, Zoku Nihon emaki taisei, vol. 11. Tokyo: Chūō Kō ronsha, 1983. Kondō Yoshihiro 近藤喜博. Nihon no kami: Shintō shigaku no tame ni 日本の神：神道史学のために. Tokyo: Ō fūsha, 1968. "Kōsanji Myōe shōnin gyōjō (Ueyama bon) 高山寺明恵上人行状( 上山本)." In Myōe Shōnin shiryō 明惠上人資料, edited by Kō zanji Tenseki Monjo Sō gō Chō sadan 高山寺典籍文書綜合調查團, 3 vols. Vol. 1, 109-45. Tokyo: Tō kyō Daigaku Shuppankai, 1971. Kujō Kanezane 九條兼實. Kujō-ke-bon Gyokuyō 九条家本玉葉. Zushoryō sō kan. 14 vols. Tokyo: Kunaichō Shoryō bu, 1994-2013. "Kumano-san ryakki 熊野山略記." In Kumano 熊野, edited by Takikawa Masajirō 瀧川政次郎, 416-47. Tokyo: Chihō shi kenkyūjo, 1957. Kunkai 訓海. Hōryūji zō Son'ei-bon Taishi den gyokurin shō 法隆寺藏尊英本太子傳玉林抄. 3 vols. Vol. 1, Tokyo: Yoshikawa Kō bunkan, 1978. Kuroda Akiko 黒田彰子, ed. Shinsen utamakura nayose 新撰歌枕名寄. 2 vols. Vol. 2, Koten bunko 古典文庫, vol. 513-514. Tokyo: Koten Bunko, 1989. 319 Kuroda Ryūji 黒田龍二. "Hie shichisha honden no kōsei: Yu kashita saijō wo megutte 日吉七社本殿の構成 : 床下祭場をめぐって." Nihon kenchiku gakkai ronbun hōkokushū 317 (1982): 148-54. Kuroda Satoshi 黒田智. "<Shiryō shōkai> Shinra Myōjin ki < 史料紹介> 新羅明神記." Tōkyō daigaku shiryō hensanjo kenkyū kiyō 11 (2001): 75-98. Kuroda Toshio, James C. Dobbins, and Suzanne Gay, trans. "Shinto in the History of Japanese Religion." Journal of Japanese Studies 7, no. 1 (1981): 1-21. Kuroda Toshio, and Fabio Rambelli. "The Discourse on the "Land of Kami" (Shinkoku) in Medieval Japan: National Consciousness and International Awareness." Japanese Journal of Religious Studies 23, no. 3/4 (1996): 353-85. Kuroda Toshio 黒田俊雄. "Chūsei jisha seiryoku ron 中世寺社勢力論." In Iwanami kōza Nihon rekishi: Chūsei 2 岩波講座日本歴史：中世 2, edited by Asao Naohiro 朝尾直弘 and Naoki Kō jirō 直木孝次郎, Vol. 6, 245-95. Tokyo: Iwanami Shoten, 1975. ———. "Chūsei no shinkoku shisō: Kokka ishiki to kokusai kankaku 中世の神国思想：国家意識と国際感覚." In Kuroda Toshio chosakushū 黒田俊雄著作集, 139-74. Kyoto: Hō zō kan, 1994. "Kuwanomidera engi 桑実寺縁起." In Zoku gunsho ruijū. Shake bu 續群書類從. 釈家部, edited by Hanawa Hokinoichi 塙保己一 and Oota Toshiro 太田藤四郎, Vol. 28 (jo), 122-27. Tokyo: Zoku Gunsho Ruijū Kanseikai, 1957. "Kyūchūji engi emaki 橋柱寺縁起絵巻." http://mahoroba.lib.nara- wu.ac.jp/y21/kyouchuji_engi/index.html: Daichiji 大智寺. "Kyūchūji engi 橋柱寺縁起." In Kizu chōshi: Shiryō hen 木津町史：史料篇, 3 vols. Vol. 1. Kyōto-fu Sō raku-gun Kizu-chō : Kizu-chō , 1984. LaFleur, William R. The Karma of Words: Buddhism and the Literary Arts in Medieval Japan. Berkeley: University of California Press, 1983. Latour, Bruno. Reassembling the Social: An Introduction to Actor-Network-Theory. New York: Oxford University Press, 2005. ———. Science in Action: How to Follow Scientists and Engineers through Society. Cambridge, MA: Harvard University Press, 1987. ———. "“Thou Shall Not Freeze-Frame,” or, How Not to Misunderstand the Science and Religion Debate." In Science, Religion, and the Human Experience, edited by James D. Proctor, 27-48. Oxford: Oxford University Press, 2005. 320 ———. "Why Has Critique Run out of Steam? From Matters of Fact to Matters of Concern." Critical Inquiry 30, no. 2 (2004): 225-48. Linda, M. G. Zerilli. "This Universalism Which Is Not One." Review of Emancipation(s), Ernesto Laclau. Diacritics 28, no. 2 (1998): 3-20. The Lotus Sutra. Translated by Burton Watson. New York: Columbia University Press, 1993. Lovell, Nadia. "Introduction." In Locality and Belonging, edited by Nadia Lovell, 1-22. New York: Routledge, 1998. Ludvik, Catherine. Sarasvati: Riverine Goddess of Knowledge, From the Manuscript-Carrying Vina-Player to the Weapon-Wielding Defender of the Dharma. Leiden: Brill, 2007. Makino Kazuo 牧野和夫. Engyōbon "Heike monogatari" no setsuwa to gakumon 延慶本「平家物語」の說話と学問. Kyoto: Shibunkaku Shuppan, 2005. ———. "Nagatobon "Heike monogatari" kan go 'Itsukushima shidai no koto' wo meguru ikkō satsu: "Kamado san hōman dai bosatsu ki" wo kaishite 長門本『平家物語』巻五「厳嶋次第之事」をめぐる一考察 : 『竈門山寶満大菩薩記』を介して." Jissen koku bungaku 50 (1996): 76-98. ———. "Nagatobon "Heike monogatari" to "Itsukushima Daimyō jin nikki": Nagato-bon "Heike monogatari" seiritsu no ichi katei wo "Itsukushima Daimyō jin nikki" shishū ni saguru 長門本「平家物語』と『厳島大明神日記』：長門本『平家物語』生成の一過程を『厳島大明神日記』四周に探る." In Nagatobon Heike monogatari no sōgō kenkyū: Kōchū hen 長門本平家物語の総合研究：校注篇, edited by Asahara Yoshiko 麻原美子 and Nanami Hiroaki 名波弘彰, 227-53. Tokyo: Benseisha, 1998. Mass, Jeffrey P. "Patterns of Provincial Inheritance in Late Heian Japan." Journal of Japanese Studies 9, no. 1 (1983): 67-95. Massey, Doreen B. For Space. Thousand Oaks, CA: Sage, 2005. Matsubara Kazuyoshi 松原一義. "Jinkeishō" no kenkyū 「塵荊抄」の硏究. Tokyo: Ō fū, 2002. Matsui Teruaki 松井輝昭. "Itsukushima jinja ni okeru Taira no Kiyomori shinkō no genryū ni tsuite: 'Imayashiro' no seiritsu wo chūshin ni 厳島神社における平清盛信仰の源流について：｢今社｣の成立を中心に." In Gakusei sanka ni yoru sekai isan Miyajima no kasseika: Gakusei ga Miyajima no miryoku o saihakkenshi, sekai ni hasshinsuru, Saishū hōkokusho 学生参加による世界遺産宮島の活性化 : 学生が宮島の魅力を再発見し, 世界に発信する : 最終報告書, 184-98. Hiroshima-shi: Kenritsu Hiroshima Daigaku Ningen Bunka Gakubu, 2009. 321 ———. "Itsukushima jinja no Benzaiten shinkō no seiritsu to sono seikaku 厳島神社の弁財天信仰の成立とその性格." Kenritsu Hirshomia Daigaku ningen bunka gaku-bu kiyō, no. 8 (2013): 148-37. ———. Itsukushima monjo denrai no kenkyū: Chūsei monjo kanri shiron 厳島文書伝来の研究: 中世文書管理史論. Tokyo: Yoshikawa Kōbunkan, 2008. ———. "Sengoku daimyō Mōri shi to Itsukushima jinja: Tanamori Fus aaki no katsudo wo tegakari toshite 戦国大名毛利氏と厳島神社：棚守房顕の活動を手掛かりとして." In Mōri Motonari to chiiki shakai 毛利元就と地域社会, edited by Kishida Hiroshi 岸田裕之, 139-66. Hiroshima: Chūgoku Shinbunsha, 2007. Matsumoto Ikuyo 松本郁代. "Shinsen'en to 'ryūō': Muromachi jidai 'rei jō' no isō 神泉苑と「龍王」 : 室町時代「霊場」の位相." Art research 6 (2006): 43-58. Matsumoto Kō ichi 松本公—. "Shintō mandara no kosumoroji: Miya mandara no shōchō suru sekaikan ni tsuite 神道曇茶羅のコスモ口ジー：宮曼荼羅の象徴する世界観について." In Kami to Hotoke no kosumologie 神と仏のコスモロジー. Nihon no Bukkyō , 148-63. Kyoto: Hō zō kan, 1995. Matsumoto Ryūshin 松本隆信. Chūsei ni okeru honjimono no kenkyū 中世における本地物の研究. Tokyo: Kyūko Shoin, 1996. ———. "Zōtei Muromachi jidai monogatari genzonbon kanmei mokuroku 増訂室町時代物語類現存本簡明目錄." In Otogizōshi no sekai 御伽草子の世界, edited by Nara Ehon Kokusai Kenkyū Kaigi 奈良絵本国際研究会議, 51-136. Tokyo: Sanseidō , 1982. Matsumoto Shinsuke 松本真輔. Shōtoku Taishi den to kassentan 聖徳太子伝と合戦譚. Tokyo: Bensei Shuppan, 2007. Matsuo Ashie 松尾葦江. Gunki monogatari genron 軍記物語原論. Tokyo: Kasama Shoi, 2008. ———. Heike monogatari ronkyū 平家物語論究. Tokyo: Meiji Shoin, 1985. Matsuoka Hisato 松岡久人. Aki Itsukushimasha 安芸厳島社. Kyoto: Hō zō kan, 1986. ———. "Itsukushima monzenchō no keisei 厳島門前町の形成." In Seto Naikai chiiki no shakaishiteki kenkyū 瀨戶內海地域の社會史的研究, edited by Uozumi Sō gorō 魚澄惣五郎, 35-80. Kyoto: Yanagihara Shoten, 1952. McCarty, Michael. "Divided Loyalties and Shifting Perceptions: The Jōkyū Disturbance and Courtier Warrior Relations in Medieval Japan." Ph.D. diss., Columbia University, 2013. McCormick, Melissa. Tosa Mitsunobu and the Small Scroll in Medieval Japan. Seattle: University of Washington Press, 2009. 322 Meeks, Lori Rachelle. Hokkeji and the Reemergence of Female Monastic Orders in Premodern Japan. Honolulu: University of Hawaii Press, 2010. Michichika Koga. "Takakura-in Itsukushima gokō ki (Account of the Journey of the ex -Emperor Takakura to Itsukushima)." Translated by Herbert E. Plutschow. In Four Japanese Travel Diaries of the Middle Ages, 25-43. Ithaca, N.Y.: China-Japan Program, Cornell University, 1981. "Mihara shibashi 三原志橋." In Mihara shishi 三原市史, 7 vols. Vol. 4, 3-189. Hiroshima-ken Mihara-shi: Mihara Shiyakusho, 1970. Minamoto Akikane 源顕兼, Kawabata Yoshiaki 川端善明, and Araki Hiroshi 荒木浩, eds. Kojidan, Zoku Kojidan 古事談. 続古事談, Shin Nihon koten bungaku taikei, vol. 41. Tokyo: Iwanami Shoten, 2005. Minamoto Kenichiro 源健一郎. "Chiba Myōken no hontai, honji setsu: Genpei tōjōroku to Chiba Myoken sha kankei shiryō to no aida 千葉妙見の本体・本地説：源平闘諍録と千葉妙見社関係資料との間." Junreiki kenkyū 3 (2006): 67-85. "Mineaiki 峯相記." In Zoku gunsho ruijū 續群書類從.釈家部, edited by Hanawa Hokinoichi 塙保己一 and Oota Toshiro 太田藤四郎. Vol. 28 (jo), 217-53. Tokyo: Zoku Gunsho Ruijū Kanseikai, 1957. Minobe Shigekatsu. "The World View of "Genpei Jō suiki"." Japanese Journal of Religious Studies 9, no. 2/3 (1982): 213-33. Miura Kunio 三浦国雄. "Dōten f ukuchi shōron 洞天福地小論." Tōhō shūkyō 61 (1983): 1-23. "Miyadera enji shō 宮寺縁事抄." In Shintō taikei. Jinja hen, Iwashimizu 神道大系. 神社編：石清水, edited by Murata Masashi 村田正志, Vol. 7, 1-676. Tokyo: Shintō Taikei Hensankai, 1988. Moerman, D. Max. "The Buddha and the Bathwater: Defilement and Enlightenment in the Onsenji engi." Japanese Journal of Religious Studies 42, no. 1 (2015): 71-87. ———. Localizing Paradise: Kumano Pilgrimage and the Religious Landscape of Premodern Japan. Cambridge, MA: Harvard University Asia Center : Distributed by Harvard University Press, 2005. Monkan 文観 "Kinpusen himitsu den 金峰山秘密伝." In Nihon daizōkyō 日本大蔵経, edited by Suzuki Gakujutsu Zaidan, Vol. 93, 3-36. Tokyo: Suzuki gakujutsu zaidan, 1976. Mori Masato 森正人. Ba no monogatariron 場の物語論. Tokyo: Wakakusa Shobō , 2012. Morrell, Robert E. Sand and Pebbles (Shasekishū): The Tales of Mujū Ichien, A Voice for Pluralism in Kamakura Buddhism. Albany: State University of New York Press, 1985. 323 Mototsuna 元網. "Rokuon’in Saigoku gekō -ki 鹿苑院西国下向記." In Sankeiki 參詣記, edited by Shinjō Tsunezō 新城常三. Shintō taikei. Bungaku hen, 151-74. Tokyo: Shintō Taikei Hensankai, 1984. Mujū Ichien 無住一円. Shasekishū 沙石集. Shinpen Nihon koten bungaku zenshū. Edited by Kojima Takayuki 小島孝之. Tokyo: Shōgakkan, 2001 . Mukaishima chōshi. Tsūshi hen 向島町史. 通史編. Hiroshima-ken Mitsugi-gun Mukaishima- chō : Mukaishima-chō , 2000. Mukasa Akira 武笠朗. "Taira Kiyomori no shinkō to heishi no zōji, zōbutsu (jō) 平清盛の信仰と平氏の造寺・造仏( 上) ". Jissen joshi daigaku bijutsu-shi gaku 13 (1998): 21-46. Mukoda Hiroshi 向田裕始,. "Seto nai no Ichikishima hime densetsu to kangensai 瀬戸内の市杵島姫伝説と管弦祭." In Itsukushima shinkō jiten 厳島信仰事典, edited by Nozaka Motoyoshi 野坂元良. Shinbutsu Shinkō Jiten Shirī zu, 214-24. Tokyo: Ebisu Kōshō Shuppan, 2002. Mukozaka Takuya 向坂卓也. "Sōshū Enoshima ue no miya engi ni tsuite: Honkoku to shō kai 相州得瑞嶋上之宮縁起について：翻刻と紹介." Kanazawa bunko kenkyū 319 (2007): 34-43. Munakata shishi. tsūshi hen 宗像市史. 通史編. 2 vols. Vol. 2, Munakata: Munakata-shi, 1999. Murakami Manabu 村上学. "Shintō shū." In Hotoke to kami 仏と神 edited by Konno Tō ru 今野達, Satake Akihiro 佐竹昭広 and Ueda Shizuteru 上田閑照. Iwanami kō za Nihon bungaku to Bukkyō , 107-32. Tokyo: Iwanami shoten, 1994. Murakami Mitsunori 村上光徳. "Akama Jingū shozō gojūni gō monjo no imi: Nagatobon Heike monogatari kenkyū no isshu ken toshite 赤間神宮所蔵五十二号文書の意味：長間本平家物語研究の一手懸として." Komazawa tandai kokubun 6 (1975): 49-64. Murata Shin'ichi 村田真一. Usa Hachiman shinwa gensetsu no kenkyū: "Hachiman Usagū gotakusenshū" o yomu 宇佐八幡神話言說の研究：「八幡宇佐宮御託宣集」を読む. Kyoto: Hō zō kan, 2016. Myōkū 明空. "Enkyokushū 宴曲集." In Zoku gunsho ruijū. Yūgi bu 続群書類従.遊戯部, edited by Hanawa Hokinoichi 塙保己一 and Oota Toshiro 太田藤四郎, Vol. 19 (ge), 76-108. Tokyo: Zoku Gunsho Ruijū Kanseikai, 1957. Nakajima Ayaka 中島彩花. "Chūsei Benzaiten shinkō to zuzō ni kansuru kenkyū: Benzaiten mandara no tenkai ni miru shinbutsu kan 中世弁才天信仰と図像に関する研究 : 弁才天曼荼羅の展開にみる神仏観." Joshi bijutsu daigaku, 2013. 324 Nakamichi Gō ichi 中道豪一. "Kyū Hiroshima-shi ni okeru Itsukushima kangensai ni matsuwaru sairei gyōji ni tsuite: Kindai ni okeru Takachōchin, hifuri, otomon -bune no yōsō to kinsairei gyō ji no hassei 旧広島市域における嚴島管絃祭にまつわる祭礼行事について : 近代における高ちょうちん・火振り・御供船の様相と新祭礼行事の発生." Hiroshima shūdai ronshū 59, no. 2 (2019): 131-55. Nakano Hatayoshi 中野幡能. Usagū 宇佐宮. Tokyo: Yoshikawa Kō bunkan, 1985. Nakano Maori 中野真麻理. "Ichijō shūgyokushō shoin waka kō 『一乗拾玉抄』所引和歌攷." Kokubungaku kenkyū shiryōkan kiyō 23 (1997): 167-205. Nakara Naritomo 中原業倫. "Chōkan kanmon 長寛勘文." In Gunsho ruijū. Zatsu bu 群書類從. 第 26 輯 ( 雑部), edited by Hanawa Hokinoichi 塙保己一. Vol. 26, 241-62. Tokyo: Yagi Shoten kosho shuppanbu 2013. Nanami Hiroaki 名波弘彰. "Nanto bon "Heike monogatari" Tsunemasa Chikubushima mōde to Hiyoshi seijo no miya no biwa hōshi: Hiei san shinkō -ken ni okeru Uga Benzaiten shinkō wo megutte 南都本『平家物語』経正竹生島詣と日吉社聖女宮の琵琶法師 : 叡山信仰圏における宇賀弁財天信仰をめぐって." Bungei gengo kenkyū 11 (1987 1987): 164-204. Nichiren 日蓮. Nichiren: Shugo kokkaron, Hōon shō, hoka nijūichi-hen 日蓮 : 守護国家論, 報恩抄, 他二十一篇. Genten Nihon bukkyō no shisō 9. Edited by Tokoro Shigemoto 戶頃重基 and Takagi Yutaka 高木豊. Tokyo: Iwanami Shoten, 1991. Nihon kiryaku 日本紀略. Kokushi taikei 國史大系. Tokyo: Keizai Zasshisha, 1897. Nihon kokugo daijiten 日本国語大辞典. Shogakkan. Nihonmatsu Yasuhiro 二本松康宏. "Suwa engi no henyō : Suwa daiō kara Kō ga Saburō he 諏訪縁起の変容：陬波大王から甲賀三郎へ ". In Suwa shinkō no Chūsei: Shinwa, denshō, rekishi 諏訪信仰の中世 : 神話・伝承・歴史, edited by Fukuda Akira 福田晃, Tokuda Kazuo 徳田和夫 and Nihonmatsu Yasuhiro 二本松康宏, 111-35. Tokyo: Miyai Shoten, 2015. "Nikkōsan engi 日光山縁起." In Jisha engi 寺社緣起, edited by Sakurai Tokutarō 桜井徳太郎, Hagiwara Tatsuo 萩原龍夫 and Miyata Noboru 宮田登, 275-89. Tokyo: Iwanami Shoten, 1975. Nishiguchi Junko 西口順子. Heian jidai no jiin to minshū 平安時代の寺院と民衆. Kyoto: Hō zō kan, 2004. 325 Nōin 能因. "Nōin utamakura (kōhon) 能因歌枕（広本）." In Nihon kagaku taikei 日本歌学大系, edited by Sasaki Nobutsuna 佐佐木信綱, 10 vols. Vol. 1, 73-107. Tokyo: Kazama Shobō , 1973. Nora, Pierre. "Between Memory and History: Les lieux de mémoire." Representations 26 (1989): 7-24. "Nosaka monjo 野坂文書." In Hiroshima kenshi: Kodai chūsei shiryō hen 広島県史. 古代中世資料編, edited by Hiroshima-ken 広島県, 5 vols. Vol. 3, 425-1104. Hiroshima: Hiroshima, 1978. Nō tomi Jō ten 納富常天. Kanazawa Bunko shiryō no kenkyū 金沢文庫資料の研究. Kyoto: Hō zō kan, 1982. Novetzke, Christian Lee. "Bhakti and Its Public." International Journal of Hindu Studies 11, no. 3 (2007): 255-72. Ōchi Tokuko 大知徳子. "Mōri Terumoto shojō t o oshima meguri 毛利輝元書状と御嶋廻." Miyajimagaku senta nenpō 1 (2010): 43-49. Ogawa Eiichi 小川栄一. "Nihongo shiryō toshite no Nagatobon Heike mon ogatari 日本語史料としての長門本平家物語 ". Musashi daigaku jinbun gakkai zasshi 41, no. 3 (2010): 456-24. Ō hashi Naoyoshi 大橋直義. "Jisha no kūkan to gensetsu: "Jishaken" toshite no Nanto ni oyobu 寺社の空間と言説：「寺社圏」としての南都に及ぶ." In Chūsei jisha no kūkan, tekusuto, gigei: "Jishaken" no pā supekutivu 中世寺社の空間・テクスト・技芸 : 「寺社圈」のパースペクティヴ, edited by Ō hashi Naoyoshi, Fujimaki Kazuhiro, and Takahashi Yūsuke, 9-23. Tō kyō : Bensei Shuppan, 2014. Ono Takashi 小野尚志. Hachiman gudōkun shohon kenkyū: Ronkō to shiryō 八幡愚童訓諸本研究：論考と資料. Tokyo: Miyai Shoten, 2001. "The Origins of Hashidate." Translated by David Atherton. In Monsters, Animals, and Other Worlds: A Collection of Short Medieval Japanese Tales, edited by R. Keller Kimbrough and Haruo Shirane, 100-23. New York: Columbia University Press, 2018. Ō shima Yukio 大島由紀夫. Chūsei shūsho no bungei bunka: Engi, setsuwa, monogatari no enpen 中世衆庶の文芸文化：縁起・說話・物語の演変. Tokyo: Miyai Shoten, 2014. Oyler, Elizabeth. "Daimokutate: Ritual Placatory Performance of the Genpei War." Oral Tradition 21, no. 1 (2006): 90-118. "The Palace of the Tengu (Tengu no dairi)." Translated by R. Keller Kimbrough. In Monsters, Animals, and Other Worlds: A Collection of Short Medieval Japanese Tales, edited by R. 326 Keller Kimbrough and Haruo Shirane, 124-48. New York: Columbia University Press, 2018. Patton, Kimberley C. "Juggling Torches: Why We Still Need Comparative Religion." In A Magic Still Dwells: Comparative Religion in the Postmodern Age, edited by Kimberley C. Patton and Ray Benjamin C., 153-71. Berkeley, CA: University of California Press, 2000. Pechilis Prentiss, Karen. The Embodiment of Bhakti. New York: Oxford University Press, 1999. Piggott, Joan R., ed. Capital and Countryside in Japan, 300-1180: Japanese Historians in English. Ithaca, NY: East Asia Program, Cornell University, 2006. Plutschow, Herbert E. A Reader in Edo Period Travel. Folkestone, Kent, UK: Global Oriental, 2006. Primiano, Leonard Norman. "Manifestations of the Religious Vernacular: Ambiguity, Power, and Creativity." In Vernacular Religion in Everyday Life : Expressions of Belief, edited by Marion Bowman and Ulo Valk, 382-94. New York: Routledge, 2014. ———. "Vernacular Religion and the Search for Method in Religious Folklife." Western Folklore 54, no. 1 (1995): 37-56. Quijano, Aníbal. "Coloniality and Modernity/Rationality." Cultural Studies 21, no. 2-3 (2007): 168-78. Quinter, David. "Localizing Strategies: Eison and the Shōtoku Taishi Cult." Monumenta Nipponica 69, no. 2 (2014): 153-98. Rambelli, Fabio. "Before the First Buddha: Medieval Japanese Cosmogony and the Quest for the Primeval Kami." Monumenta Nipponica 64, no. 2 (2009): 235-71. ———. "The Idea of India (Tenjiku) in Pre-Modern Japan: Issues of Signification and Representation in the Buddhist Translation of Cultures." In Buddhism Across Asia: Networks of Material, Intellectual and Cultural Exchange, edited by Tansen Sen, 259-90. Singapore: ISEAS–Yusof Ishak Institute, 2014. ———. "Re-positioning the Gods: "Medieval Shintō" and the Origins of Non -Buddhist Discourses on the "Kami"." Cahiers d'Extrême-Asie 16 (2006): 305-25. ———. "The Story of Prince Rama in Japan: Sources and Transformations." In Ramayana in Focus: Visual and Performing Arts of Asia, edited by I. Made Bandem and Gauri Parimoo Krishnan, 28-37. Singapore: Asian Civilisations Museum, 2010. Rao, Ajay K. Re-figuring the Rā mā yaṇa as Theology: A History of Reception in Premodern India. New York: Routledge, 2015. Rappo, Gaétan. "Essentialism in Early Shinto Studies." In Exploring Shinto, edited by Michael Pye, 34-56. Bristol, CT: Equinox, 2020. 327 "Reikiki 麗気記." In Shintō taikei. ronsetsu hen, Shingon shintō 神道大系. 論說編, 真言神道, edited by Wada Shūjō 和多秀乗, 2 vols. Vol. 1, 1-118. Tokyo: Shintō Taikei Hensankai, 1993. Robson, James. "Buddhist Sacred Geography." In Early Chinese Religion: Part Two, Period of Division (220-589 AD), edited by John Lagerwey and Lü Pengzhi, 1353-98. Leiden: Brill, 2010. Rodriguez Medina, Leandro. Centers and Peripheries in Knowledge Production. New York: Routledge, 2014. "Rokudai shōjiki 六代勝事記." In Gunsho ruiju. Teiōbu 群書類從.帝王部, edited by Hanawa Hokinoichi 塙保己一, Vol. 3, 403-424. Tokyo: Yagi Shoten, 2013. "Ruiju kigenshō 類聚既験抄." In Chūsei shōdō shiryōshū 中世唱導資料集, edited by Okabe Kaien 岡部快圓, Matsuno Yō ichi 松野陽一, Abe Yasurō 阿部泰郎 and Yamasaki Makoto 山崎誠, 235-68. Kyoto: Rinsen Shoten, 2000. Ruppert, Brian O. "Buddhist Rainmaking in Early Japan: The Dragon King and the Ritual Careers of Esoteric Monks." History of Religions 42, no. 2 (2002): 143-74. Ryōhen 良遍. "Nihon shoki kan daiichi, daini kikigaki (Tawa bunko zō) 日本書紀巻第一・第二聞書( 多和文庫蔵)." In Sonarejō: Murasame hen 磯馴帖：村雨篇, edited by Itō Masayoshi 伊藤正義. Koten kenkyū shiryō shū, 2 vols. Vol. 2, 335-402. Osaka: Izumi Shoin, 2002. ———. "Jingi keizu 神祇系図." In Tendai Shintō 天台神道, edited by Tamura Yoshirō 田村芳朗, Sueki Fumihiko 末木文美士 and Satō Masato 佐藤真人. Shintō taikei. Ronsetsu hen, 2 vols. Vol. 1, 595-619. Tokyo: Shintō Taikei Hensankai, 1990. Ryūen 隆円, and Iwasa Miyoko 岩佐美代子, ed. Bunki dan zenchūshaku 文机談全注釈. Tokyo: Kasama Shoin, 2007. Saeki Shin'ichi 佐伯真一. "'Heike monogatari' to 'Hōryaku kanki': Shibubon, Seisuiki kyōtsu sohon no sōtei 『平家物語』と『保暦間記：四部本・盛衰記共通祖本の想定." Chūsei bungaku 40 (1995): 89-98. ———. "'Hōryaku kanki' no rekish i jojutsu 『保暦間記』の歴史叙述." Denshō bungaku kenkyū 46 (1997): 47-57. "Sakakiba no nikki さかき葉の日記." In Gunsho ruijū. Jingibu 群書類従.神祇部, edited by Hanawa Hokinoichi 塙保己一, Vol. 2. Tokyo: Yagi Shoten, 2013. 328 Sakurai Tokutarō 桜井徳太郎. "Engi no ruikei to tenkai 縁起の類型と展開." In Jisha engi 寺社緣起, edited by Sakurai Tokutarō 桜井徳太郎, Hagiwara Tatsuo 萩原龍夫 and Miyata Noboru 宮田登, 445-78. Tokyo: Iwanami Shoten, 1975. Sanjō nishi Sanetaka 三條西實隆. Sanetaka Kōki 實隆公記. Third ed. 13 vols. Tokyo: Zoku Gunsho Ruijū Kanseikai, 1979-1980. "Sannō hiki 山王秘記." In Zoku Tendaishū zensho, Shintō 續天台宗全書ー神道, edited by Tendai Shūten Hensanjo, Vol. 8, 223-34. Tokyo: Shunjūsha, 1987-. “Sannō koshiki 山王講式.” In Gyosan soshō 魚山叢書. Vol. 68-9, 89-99. Doc. 6186-26. Tokyo University Historiographical Insitute. Sano Midori. "The Narration of Tales, The Narration of Paintings." Translated by Sinead Kehoe. In Reading the Tale of Genji: Its Picture Scrolls, Texts and Romance, edited by Richard Stanley-Baker, Murakami Fuminobu and Jeremy Tambling, 37-60. Leiden: Global Oriental, 2009. Sasaki Hachirō 佐々木八郎. Heike monogatari hyōkō 平家物語評講. 2 vols. Tokyo: Meiji Shoin, 1963. Sasaki Yasuyuki 佐々木康之, ed. Mizu: Shinpi no katachi = Prayers to water 水 : 神秘のかたち Tokyo: Santorī Bijutsukan, 2015. Satō Hiroo 佐藤弘夫. "The Emergence of Shinkoku (Land of the Gods) Ideology in Japan." In Challenging Paradigms, Buddhism and Nativism: Framing Identity Discourse in Buddhist Environments, edited by Henk Blezer and Mark Teeuwen, 29-49. Leiden: Brill, 2013. ———. Kami, hotoke, ōken no chūsei 神・仏・王権の中世. Kyoto: Hōzōk an, 1998. ———. "Wrathful Deities and Saving Deities." In Buddhas and Kami in Japan: Honji Suijaku as a Combinatory Paradigm, edited by Mark Teeuwen and Fabio Rambelli, 95-114. New York: RoutledgeCurzon, 2003. Scheid, Bernhard. "Two Modes of Secrecy in the Nihon shoki Transmission." In The Culture of Secrecy in Japanese Religion, edited by Bernhard Scheid and Mark Teeuwen, 284-306. London: Routledge, 2006. Seki Hideo 関秀夫. Kyōzuka no shosō to sono tenkai 經塚の諸相とその展開. Tokyo: Yūzankaku, 1990. Senoo Yoshinobu 妹尾好信. "Ryaku engi ‘Itsukushima yurai ’ kō: Fuku, eiin, to honkoku 略縁起『いつくしま由来』考：附・影印と翻刻." Itsukushima kenkyū 3 (2007): 6-25. 329 Seo Shūsō 妹尾周三. "Aki, Itsukushima ni okeru shin hakken no sairei iseki: Misen no chūfuku de hakken sareta gankai gun no kentō 安芸、厳島における新発見の祭祀遺跡：弥山の中腹で発見された岩塊群の検討." Museum, no. 639 (2012): 5-21. Shigematsu Akihisa 重松明久, ed. Hachiman Usagū gotakusenshū 八幡宇佐宮御託宣集. Tokyo: Gendai Shichōsha, 1986. Shimasue Kiyoshi 島居清. "Shoshin kongenshō, Shoshinki, Shosha kongenki ni tsuite (2): Shoshinki to Shoshin kongenshō 諸神根源抄・諸神記・諸社根元記に就いて(2) ：諸神記と諸神根源抄." Biburia, Tenri toshokan hō 31 (1965): 52-61. ———. "Yoshida bunko no Kanemigi jihitsubon ni tsuite: Jinja jinmyō saishi bukki nado ni kansuru mono (ni) 吉田文庫の兼右自筆本に就いて：神社神名祭祀服忌等に関するもの（二）." Biburia, Tenri toshokan hō 27 (1964/04 1964): 46-58. Shimazu, Naoko. "Reading the Diaries of Japanese Conscripts: Forging National Consciousness during the Russo-Japanese War." In Nationalisms in Japan, edited by Naoko Shimazu, 41-66. New York: Routledge, 2006. Shimazu Tadao 島津忠夫. Heike monogatari shiron 平家物語試論. Tokyo: Kyūko Shoin, 1997. Shimizu Minoru 清水実. "Hie Sannō mandala kō: Sannō Shintō shiryō kara no yomitoki 日吉山王曼荼羅考：山王神道史料からの読み解き." Mitsui bijutsu bunkashi ronshū 6 (2013): 11-67. Shimomukai Tatsuhiko 下向井龍彦. "Taira no Kiyomori Ondo seto 'Hi maneki' kaisaku densetsu no keisei to shintō 平清盛音戸瀬戸「日招き」開削伝説の形成と浸透." Geibi chihōshi kenkyū 282-283 (2012): 46-76. Shinma Mio 新間水緖. Shinbutsu setsuwa to setsuwashū no kenkyū 神仏説話と説話集の研究. Osaka: Seibundō , 2008. Shinohara, Koichi. "The Story of the Buddha's Begging Bowl: Imagining a Biography and Sacred Places." In Pilgrims, Patrons, and Place: Localizing Sanctity in Asian Religions, edited by Phyllis Granoff and Shinohara Koichi, 68-106. Vancouver: UBC Press, 2003. "Shinshutsu Itsukushima monjo 新出厳島文書." In Hiroshima kenshi: Kodai chūsei shiryō hen 広島県史. 古代中世資料編, edited by Hiroshima-ken 広島県, 5 vols. Vol. 3. Hiroshima: Hiroshima-ken, 1978. Shintōshū 神道集. Shintō taikei. Bungaku hen 神道大系. 文學編. Edited by Masao 岡見正雄 Okami and Takahashi Kiichi 高橋喜一. Tokyo: Shintō Taikei Hensankai, 1988. "Shōbōrinzō 聖法輪蔵." In Shinshū shiryō shūsei 眞宗史料集成, edited by Ishida Mitsuyuki 石田充之 and Chiba Jō ryū 千葉乘隆, 13 vols. Vol. 4, 434-566. Kyoto: Dō hō sha, 1974. 330 Shō kai 聖戒, and En'i 円伊. Ippen Shōnin eden 一遍聖人絵伝. Nihon emaki taisei 日本絵卷大成 Edited by Komatsu Shigemi 小松茂美, Murashige Yasushi 村重寧 and Furuya Minoru 古谷稔. Tokyo: Chūō Kō ronsha, 1978. "Shosha kongen ki 諸社根元記." In Jingi zensho 神祇全書, edited by Saeki Ariyoshi 佐伯有義, Vol. 1, 1-69. Tokyo: Kō ten Kō kyūjo, Jingū Hō saikai, 1906. "Shoshin kongenshō 諸神根元抄." National Institute of Japanese Literature: Hizen Shimabara Matsudaira bunko 肥前島原松平文庫. Simpson, Emily Blythe. "Crafting a Goddess: Divinization, Womanhood and Genre in Narratives of Empress Jingū." Ph.D. diss, University of California, Santa Barbara, 2019. Smith, Jonathan Z. Drudgery Divine: On the Comparison of Early Christianities and the Religions of Late Antiquity. Chicago: University of Chicago Press, 1990. Smits, Gregory. "Conduits of Power: What the Origins of Japan's Earthquake Catfish Reveal about Religious Geography." Japan Review 24 (2012): 41-65. Spafford, David. A Sense of Place: The Political Landscape in Late Medieval Japan. Cambridge, MA: Harvard University Press, 2013. Steininger, Brian. "The Scribal Imaginary in Medieval Japanese Paratexts." The Journal of Japanese studies 45, no. 2 (2019): 241-67. Stone, Jacqueline Ilyse. Original Enlightenment and the Transformation of Medieval Japanese Buddhism. Honolulu, Hawaii: University of Hawai'i Press, 1999. Sueki Fumihiko 末木文美士 ed. Shintō taikei: Ronsetsu hen, Tendai Shintō (ge) 神道大系.：論說編 , 天台神道. 下. Tokyo: Shintō Taikei Hensankai, 1993. Sugano Fumi 菅野扶美. "'Gyōshō Shōnin' no katararekata to Amano sha yonjo myōjin 「行勝上人」の語られ方と天野社四所明神." Junreiki kenkyū 7 (2010): 19-35. Suishin 水心. "Jitokushū 寺徳集." In Zoku gunsho ruiju. Shakke bu (jō) 續群書類從. 釈家部（上） , edited by Hanawa Hokinoichi 塙保己一 and Oota Toshiro 太田藤四郎. Vol. 28, 1-25. Tokyo: Yagi Shoten, 2013. Suitō Makoto 水藤真. Munafuda no kenkyū 棟札の研究. Kyoto: Shibunkaku Shuppan, 2005. Sunagawa Hiroshi 砂川博. Heike monogatari no keisei to biwa hōshi 平家物語の形成と琵琶法師. Tokyo: Ō fū, 2001. "Susa jinja engi 須佐神社縁起." In Setouchi jisha engishū 瀬戶内寺社緣起集, edited by Wada Shigeki 和田茂樹, Tomohisa Takefumi 友久武文 and Takemoto Hiroo 竹本宏夫, 183- 96. Hiroshima-shi: Hiroshima Chūsei Bungei Kenkyūkai, 1967. 331 Suzuki Akira 鈴木彰. Heike monogatari no tenkai to chūsei shakai 平家物語の展開と中世社会. Tokyo: Kyūko Shoin, 2006. Suzuki Hideyuki 鈴木英之. Chūsei gakusō to Shintō: Ryōyo Shōgei no gakumon to shisō 中世学僧と神道 : 了誉聖冏の学問と思想. Tokyo: Benseishuppan, 2012. Suzuki Masataka 鈴木正崇. "Shugen reizan no engi ni kansuru kōsatsu: 'Hikosan ruki' wo yomu 修験霊山の縁起に関する考察 : 『彦山流記』を読む." Girei bunka gakkai kiyō 3 (2016): 51-70. Tada Keiko 多田圭子. "Chūsei ni okeru Jingū kōgō zō n o tenkai: Engi kara 'Taiheiki' he 中世における神功皇后像の展開：縁起から『太平記』へ." Kokubun mejiro 31 (1991): 191- 203. Taira Nobunori 平信範. Hyōhanki 兵範記. Zō ho shiryō taisei 増補史料大成. 5 vols. Kyoto: Rinsen Shoten, 1965. Takagishi Akira 髙岸輝, and Heather Blair, trans. "The Reproduction of Engi and Memorial Offerings: Multiple Generations of the Ashikaga Shoguns and the Yūzū nenbutsu engi emaki." Japanese Journal of Religious Studies 42, no. 1 (2015): 157-82. Takahashi Miyuki 高橋美由紀. Ise Shintō no seiritsu to tenkai 伊勢神道の成立と展開. Tokyo: Taimeidō , 1994. Takahashi Shūzō 高橋修三. "Itsukushima no rekishi wo megutte: Ichi, shibai, soshite tomikuji 厳島の歴史をめぐって：市・芝居、そして富籤." In Nihon sankei e no izanai: Matsushima, Amanohashidate, Itsukushima 日本三景への誘い : 松島・天橋立・厳島, edited by Shimao Arata 島尾新 and Hasegawa Seiichi 長谷川成一, 108-17. Osaka: Seibundō Shuppan, 2007. "Takakura-in Itsukushima gokō ki 高倉院厳島御幸記 ". In Gunsho ruijū. Kikō bu 群書類従.紀行部, edited by Hanawa Hokinoichi 塙保己一, Vol. 18. Tokyo: Gunsho Ruijū Kankō kai Kantō sha Nai, 1952. Takayama Toshihiro 高山利弘, ed. Kundoku Shibu kassenjō-bon Heike monogatari 訓読四部合戦状本平家物語. Tokyo: Yūseidō , 1995. Takenishi Hiroko 竹西寬子. Kangensai 管絃祭. Tokyo: Shinchō sha, 1978. ———. "Shinto and Shoguns: A Return to Itsukushima." The Unesco Courier 54, no. 2 (2001): 41- 43. Takeuchi Rizō 竹內理三, ed. Kamakura ibun 鎌倉遺文. 42 vols. Tokyo: Tokyōdō Shuppan, 1971. 332 Takeya Yukie 竹谷靱負. Fujisan no saijinron 富士山の祭神論. Tokyo: Iwata Shoin, 2006. Tamori Masakazu 田森雅一. "Nihon ni okeru Ramayana no juyō to hen'yō 日本におけるラーマーヤナの受容と変容." In Rā mā yana no uchū: Denshō to minzoku zōkei = Cosmos of Ramayana: Tradition and Ethno-forms ラーマーヤナの宇宙 : 伝承と民族造形, edited by Kaneko Kazushige 金子量重, Sakata Teiji 坂田貞二 and Suzuki Masataka 鈴木正崇, 198-220. Tokyo: Shunjūsha, 1998. Tamura Hiroshi 田村裕. "Itsukushima sharyō shōen no keisei to kurashiki ni tsuite 厳島社領荘園の形成と倉敷について." In Naikai chiiki shakai no shiteki kenkyū 內海地域社会の史的研究, edited by Matsuoka Hisato 松岡久人, 65-86. Yamaguchi-ken Tokushima-shi: Matsuno Shoten, 1978. Tanaka Hisao 田中久夫. Ujigami shinkō to sosen saishi 神信仰と祖先祭祀. Tokyo: Meicho Shuppan, 1991. Tanaka Takako 田中貴子. Gehō to aihō no chūsei 外法と愛法の中世. Tokyo: Sunagoya Shobō, 1993. ———. "Keiran shūyōshū" no sekai 「渓